의과대학 교육에서 보건AI윤리 교육 필요성(Adv Health Sci Educ Theory Pract. 2021)
The need for health AI ethics in medical school education
Gali Katznelson1 · Sara Gerke2

소개
Introduction

인공지능(AI)이 의료 서비스를 혁신하고 있습니다. 2018년 4월, 미국 식품의약국(FDA)은 안과 질환인 당뇨망막병증에 대한 진단 결정을 내리는 최초의 '자율' AI 소프트웨어 시스템인 IDx-DR의 시판을 허가했습니다(Digital Diagnostics, 2020; FDA, 2018a). 2019년 9월, 의료 신경 기술 회사인 BrainScope는 결과와 조치를 제공하여 뇌진탕 및 경미한 외상성 뇌손상 진단에 도움을 주는 AI 기반 의료 기기 BrainScope TBI(모델: Ahead 500)에 대한 FDA 허가를 받았습니다(FDA, 2019a). 이 외에도 심장 환자를 위한 최초의 AI 기반 의료 영상 플랫폼인 Arterys(FDA, 2017), 손목 골절을 감지하도록 설계된 AI 소프트웨어인 OsteoDetect(FDA, 2018b), 뇌졸중과 관련된 지표에 대한 컴퓨터 단층 촬영 결과를 분석하기 위해 AI를 사용하는 임상 의사 결정 지원 소프트웨어인 Viz.AI(FDA, 2018c) 등이 이미 FDA의 허가를 받거나 승인된 건강 AI입니다. 
Artificial Intelligence (AI) is transforming health care. In April 2018, the U.S. Food and Drug Administration (FDA) permitted marketing of the first “autonomous” AI software system, called IDx-DR, to provide a diagnostic decision for the eye disease diabetic retinopathy (Digital Diagnostics, 2020; FDA, 2018a). In September 2019, the medical neurotechnology company BrainScope received FDA clearance for its AI-based medical device BrainScope TBI (model: Ahead 500) that helps in the diagnosis of concussion and mild traumatic brain injury by providing results and measures (FDA, 2019a). Other health AIs that have already been cleared or approved by the FDA include Arterys, the first medical imaging platform powered by AI for cardiac patients (FDA, 2017), OsteoDetect, an AI software designed to detect wrist fractures (FDA, 2018b) and Viz.AI, a clinical decision support software that uses AI to analyze computed tomography results for indicators associated with a stroke (FDA, 2018c).

지금까지 160개 이상의 의료용 AI 기반 기기가 FDA의 허가를 받았거나 승인되었으며(Ross, 2021), 현재 개발 파이프라인에 더 많은 제품이 있습니다. 의료 AI의 미래를 예측하기는 어렵지만 가상 비서, 개인 맞춤형 의료, 자율 로봇 수술은 의료 시스템의 미래 모습에 대한 가능성의 영역에서 벗어나지 않습니다. 의료 AI는 의료를 더 나은 방향으로 변화시킬 수 있는 엄청난 잠재력을 가지고 있지만, 사전 동의, 편견, 안전, 투명성, 환자 개인정보 보호, 할당에 이르기까지 윤리적 문제도 제기합니다.  
Over 160 health AI-based devices have been cleared or approved by the FDA so far (Ross, 2021), and there are currently many more products in the development pipeline. It is difficult to predict what the future of health AI will bring, but virtual assistants, personalized medicine, and autonomous robotic surgeries are not outside the realm of possibilities of how the health care system could look like in the future. Health AI has tremendous potential to transform medicine for the better, but it also raises ethical issues, ranging from informed consent, bias, safety, transparency, patient privacy to allocation.

의료 AI는 이미 의료 서비스를 변화시키고 있고 앞으로도 계속 변화할 것이지만, 의료 AI 윤리는 아직 의과대학 커리큘럼에 표준 과목으로 포함되어 있지 않습니다. 이러한 격차는 현재의 문헌에도 반영되어 있습니다: AI와 의학교육에 초점을 맞춘 출판물은 거의 없으며, 윤리를 전혀 언급하지 않은 출판물은 더 적습니다(Kolachalama & Garg, 2018; Wartman & Combs, 2018; Wartman & Combs, 2019; Paranjape et al., 2019; AMA, 2019). 
Although health AI is already shaping health care and will continue to do so, health AI ethics is not yet a standard course in medical school curriculums. This gap may also be reflected in the current literature: Only few publications focus on AI and medical education, and even fewer mention ethics at all (Kolachalama & Garg, 2018; Wartman & Combs, 2018; Wartman & Combs, 2019; Paranjape et al., 2019; AMA, 2019).

먼저 의료 인공지능 윤리가 무엇인지 설명한 다음 의과대학에서 이러한 교육이 필요한 이유를 논증할 것입니다. 그런 다음 의대 교육이 필요한 의료 AI가 제기하는 6가지 주요 윤리적 문제인 사전 동의, 편향성, 안전, 투명성, 환자 개인정보 보호, 할당에 대해 집중적으로 논의할 것입니다. 이러한 윤리적 이슈는 임상 의료 윤리와 관련된 AI의 프레임워크를 형성합니다. 실제 사례를 사용하여 응용적인 맥락에서 의료 AI 윤리를 가르치는 것이 유용할 수 있습니다. 
We will first explain what health AI ethics is and will then argue why this education is needed in medical schools. Our focus will then be devoted to a discussion of six key ethical issues raised by health AI that require medical school education: informed consent, bias, safety, transparency, patient privacy, and allocation. These ethical issues form a framework for AI that is pertinent to clinical medical ethics. We suggest that it may be useful to teach Health AI Ethics in an applied context using real-life examples.

의료 AI 윤리란 무엇인가요?
What is health AI ethics?

우리는 건강 AI 윤리'AI가 관여하는 건강 분야의 맥락에 윤리를 적용하고 분석하는 것'으로 정의합니다. 건강이 여러 가지 방식으로 정의되어 온 것처럼(Huber 등, 2011), 1955년 McCarthy 등(2006)에 의해 처음 사용된 인공지능(AI)이라는 용어에 대한 단일 정의도 존재하지 않습니다. 의료 분야에서 의료 전문가가 AI와 그 응용 분야를 이해하는 것이 점점 더 중요해지고 있습니다.
We define Health AI Ethics as the ‘application and analysis of ethics to contexts in health in which AI is involved’. Just as health has been defined in several ways (Huber et al., 2011), there is also no single definition for the term artificial intelligence (AI), which was first coined in 1955 by McCarthy et al. (2006). In medicine, it is becoming increasingly important for health care professionals to understand AI and its applications.

예를 들어, 미국의학협회(AMA)는 AI를 다음과 같이 정의합니다:
For example, the American Medical Association (AMA) defines AI as:

일반적으로 이성적인 인간과 관련된 방식으로 작업을 완료하는 컴퓨터의 능력, 즉 개체가 환경에서 적절하게 기능하고, 선견지명을 가지고 기능할 수 있도록 하는 자질입니다. 진정한 AI는 인공지능이 인간과 구별할 수 없는 지능적인 행동을 보일 수 있어야 한다는 튜링 테스트(2018)를 이길 수 있는 프로그램 또는 알고리즘으로 널리 알려져 있습니다.
The ability of a computer to complete tasks in a manner typically associated with a rational human being—a quality that enables an entity to function appropriately and with foresight in its environment. True AI is widely regarded as a program or algorithm that can beat the Turing Test, which states that an artificial intelligence must be able to exhibit intelligent behavior that is indistinguishable from that of a human (2018).

AI를 구성하는 요소에 대한 이해는 수년에 걸쳐 발전해 왔습니다. 1970년대에 의료 분야의 AI는 심전도 해석과 같은 규칙 기반 알고리즘으로 구성되었으며, 이는 오늘날에도 여전히 일반적이지만 일반적으로 "진정한 AI"라고 불리는 알고리즘은 아닙니다(Kundu 외., 2000; Yu 외., 2018). 최근 AI의 발전은 알고리즘을 사용하여 데이터에서 패턴을 찾는 컴퓨터 과학의 한 분야인 머신 러닝(ML)으로 이루어졌습니다(Murphy, 2012; Yu et al., 2018). ML은 지도형 또는 비지도형일 수 있습니다. 지도 ML은 학습 중에 입출력 데이터의 상관관계를 기반으로 출력을 예측하는 데 사용되는 반면, 비지도 ML은 레이블이 지정되지 않은 데이터에서 패턴을 식별합니다(Yu et al., 2018). 예를 들어, 지도 머신러닝은 이미지에서 특정 유형의 암 유무를 감지하도록 학습되지만, 비지도 머신러닝은 많은 수의 이미지를 평가하여 패턴을 감지하고 그에 따라 이미지를 그룹화합니다. 특히 대규모 데이터 세트에 다층 인공 신경망을 사용하는 ML의 한 분야인 딥러닝의 발전은 AI 분야를 더욱 발전시켰습니다(Yu et al., 2018). 임상 치료에 사용되는 대부분의 AI는 아직 인간의 지능을 능가하지 못하며, 여전히 협소하고 특정 업무에 국한되어 있습니다(Hosny 외., 2018). AMA는 "증강 지능"이라는 용어를 사용하여 의사의 지능을 대체하는 것이 아니라 의사의 지능을 증강하는 현재 AI의 능력을 강조합니다(2018). 
The understanding of what constitutes AI has evolved over the years. In the 1970s, AI in health care consisted of rule-based algorithms, such as electrocardiogram interpretation, which remain commonplace today but are not typically those referred to as “true AI” (Kundu et al., 2000; Yu et al., 2018). Recent developments in AI have consisted of Machine Learning (ML), a branch of computer science that finds patterns in data using algorithms (Murphy, 2012; Yu et al., 2018). ML can be supervised or unsupervised. Supervised ML is used to predict outputs based on correlations of input–output data during training, while unsupervised ML identifies patterns in data that is not labelled (Yu et al., 2018). For example, while supervised ML would be trained to detect the presence or absence of a certain type of cancer on an image, unsupervised ML would evaluate a large number of images to detect patterns and group images accordingly. In particular, developments within Deep Learning, a field within ML that uses multi-layered artificial neural networks on large datasets, has propelled the field of AI forward (Yu et al., 2018). Most AI that is used in clinical care has yet to surpass human intelligence, and remains narrow and task-specific (Hosny et al., 2018). The AMA uses the term “Augmented Intelligence” to highlight current AI’s ability to augment the physician’s intelligence rather than to replace it (2018).

윤리는 간단히 말해서 옳고 그름의 문제를 고려하는 철학의 한 영역입니다. 윤리는 규범적 윤리, 메타윤리, 응용 윤리로 나눌 수 있습니다(Fieser, 1999). 

  • 규범 윤리는 올바른 행동 방식에 대한 질문에 답하려고 합니다. 규범적 윤리 이론에는 의무론, 결과론, 덕 윤리가 포함되지만 이에 국한되지는 않습니다.
  • 메타윤리학옳고 그름의 본질에 대한 질문을 다루는 철학의 한 분야입니다.
  • 응용 윤리는 윤리적 이론이나 원칙을 구체적인 실생활 문제에 적용하는 것을 다룹니다. 

많은 의대생들은 보챔프와 차일드레스(2012)가 제시한 자율성, 비악성, 유익성, 정의라는 네 가지 생의학 윤리의 원칙을 잘 알고 있을 것입니다. 원칙주의로 알려진 이 다원주의 이론은 의학의 윤리적 딜레마를 해결하기 위한 실용적인 접근 방식을 안내합니다. 의료 인공지능 윤리 교육은 사례 기반의 응용적 방법을 통해 윤리에 접근해야 하지만 규범 윤리와 메타 윤리의 요소도 포함해야 합니다. 정보에 입각한 동의, 편견, 안전, 투명성, 환자 프라이버시, 배분 등 6가지 이슈로 구성된 의료 AI 윤리를 학습하는 프레임워크가 의료 학습자에게 가장 적합하다고 제안합니다.
Ethics is the area of philosophy that, in simplest terms, considers questions of right and wrong. Ethics can be divided into normative ethics, metaethics, and applied ethics (Fieser, 1999).

  • Normative ethics tries to answer questions about the right way to act. Normative ethical theories include, but are certainly not limited to deontology, consequentialism, and virtue ethics.
  • Metaethics is the branch of philosophy that addresses questions about the nature of right and wrong.
  • Applied ethics deals with applying ethical theories or principles to specific, real-life issues.

Many medical students will be familiar with Beauchamp and Childress’ (2012) four principles of biomedical ethics: autonomy, non-maleficence, beneficence, and justice. This pluralistic theory, known as principlism, is meant to guide a pragmatic approach to resolving ethical dilemmas in medicine. A Health AI Ethics education should approach ethics through an applied, case-based method, but should also include elements from normative ethics and metaethics. We suggest that a framework to study Health AI Ethics consisting of the following six issues is most pertinent for medical learners: informed consent, bias, safety, transparency, patient privacy, and allocation.

의과대학 교육에 의료 AI 윤리가 필요한 이유는 무엇인가요?
Why is health AI ethics needed in medical school education?

의료 AI가 의료 분야에 미칠 영향력이 커질 것이라는 징후가 몇 가지 있습니다. 첫째, 전 세계 의료 시스템 전반에 걸쳐 엄청난 양의 데이터가 수집되고 있습니다. 한 추정에 따르면 '글로벌 데이터스피어'는 2025년까지 175제타바이트까지 증가할 것으로 예상됩니다(Reinsel et al., 2018). 이러한 전례 없는 양의 데이터는 의료 AI와 같이 데이터에 의존하는 분야에서 혁신적인 발전으로 이어질 수 있습니다. 
Several signs point toward the increasing impact that health AI will have in medicine. First, there are gigantic amounts of data being collected across the global health care system. According to one estimate, the “global datasphere” is expected to grow to 175 zettabytes by 2025 (Reinsel et al., 2018). These unprecedented amounts of data may lead to revolutionary developments in fields that rely on data, such as health AI.

둘째, 의료 AI 시장은 2020년 49억 달러에서 2026년 452억 달러로 증가할 것으로 예상됩니다(Markets & Markets, 2020). 특히 구글, 마이크로소프트, IBM과 같이 전통적으로 의료와 관련이 없었던 거대 기술 기업들도 빠르게 성장하는 이 시장의 경제적 인센티브에 자극을 받아 의료 AI 시장의 주요 플레이어로 부상하고 있습니다. 
Second, it is estimated that the market for health AI will increase from 4.9 billion USD in 2020 to 45.2 billion USD in 2026 (Markets & Markets, 2020). In particular, tech giants that are traditionally not health-related, such as Google, Microsoft, and IBM, are key players in the health AI market and are probably motivated by the economic incentives of this rapidly growing market.

셋째, 환자들은 AI를 사용하는 연결된 세상에 살고 있다는 사실을 알게 되었습니다. 환자들은 더 이상 의사 진료실에만 국한되지 않고, '닥터 구글'에게 조언을 구하고, 피트니스 트래커, 건강 AI 앱, 디지털 알약 등을 사용하여 건강을 모니터링하고 관리합니다(Gerke et al., 2019). 이러한 활동의 대부분은 환자 데이터를 수집하며, 데이터는 경제적 가치로 인해 오늘날 경제에서 "새로운 석유"라고 불리고 있습니다(The Economist, 2017). 한편, 병원은 데이터와 증거를 통해 개선된 환자 치료 개발을 추진하는 '학습형 의료 시스템'으로 거듭나도록 장려되고 있습니다(Olsen et al., 2007). 의료 AI는 새로운 방식으로 대량의 데이터를 분석할 수 있는 역량을 통해 병원이 이러한 목표를 달성하는 데 도움을 줄 수 있는 잠재력을 가지고 있습니다. 방대한 데이터 생성, 데이터 시장의 성장, 환자 치료 환경의 변화는 의료 AI의 의료 혁신 잠재력에 대한 환영할 만한 징후입니다. 
Third, patients have found themselves to be living in a connected world that uses AI. Health care is no longer confined to the doctor’s office—patients are targeted for health adds, are quick to consult ‘Dr. Google’ for advice, and use fitness trackers, health AI apps, and digital pills (Gerke et al., 2019) to help them monitor and manage their health. Many of these activities collect patient data, and data has been dubbed “the new oil” in today’s economy for its economic value (The Economist, 2017). Meanwhile, clinics have been encouraged to become “learning healthcare systems,” systems in which data and evidence are intended to drive the development of improved patient care (Olsen et al., 2007). Health AI has the potential to help hospitals achieve this goal through its capacity to analyze large quantities of data in novel ways. The massive generation of data, increasing market for data, and changing patient care landscape serve as welcoming indications for health AI’s potential to transform medicine.

그러나 이러한 변화에는 많은 윤리적 문제가 수반됩니다(아래 참조). 예를 들어, 아니타 호(2019)와 주나이드 나비(2018)는 헤이스팅스 센터 보고서에서 의료 분야에서 AI가 어떻게 윤리적으로 설계되고 구현되어야 하는지에 대한 글을 썼습니다. Ho는 의료 AI가 현재의 요구에 부응하고 그 과정에서 발생하는 문제에 대해 열린 자세를 유지하는 의료 서비스 품질 개선 문화 속에서 개발되어야 한다고 주장합니다. 나비는 의료 AI는 생명윤리 원칙을 지침으로 삼아 설계되어야 한다고 주장합니다. 미래의 중요한 의료 이해당사자인 의대생이 의료 AI 윤리에 대한 적절한 교육 없이는 의료 AI의 설계, 구현 및 사용에 참여하기 어렵다는 것이 우리의 주장입니다. 
Such changes, however, bring with them many ethical issues (see below). For example, both Anita Ho (2019) and Junaid Nabi (2018) have written in the Hastings Center Report about how AI should be designed and implemented ethically in medicine. Ho argues that health AI should be developed within a culture of quality improvement in health care that responds to current needs and remains open to issues that arise along the way. Nabi argues that health AI should be designed with bioethical principles as a guidance. We argue that, for medical students, as important future stakeholders in medicine, it will be difficult to participate in the design, implementation, and use of health AI without adequate training in health AI ethics.

일부 학자들은 이미 AI 시대에 의과대학 커리큘럼 개혁을 촉구한 바 있습니다(Kolachalama & Garg, 2018; Wartman & Combs, 2018; Wartman & Combs, 2019; Paranjape et al., 2019). 예를 들어, Kolachalama와 Garg(2018)는 의과대학이 AI의 '이점, 위험 및 윤리적 딜레마'에 중점을 둔 '머신러닝 리터러시'를 위해 노력할 것을 권장합니다. 2019년 6월, AMA는 증강 지능을 의학교육에 통합하는 것에 관한 정책을 발표했습니다. AMA는 무엇보다도

  • 의과대학 교수진에 데이터 과학자와 엔지니어를 포함시킬 것
  • 증강 지능 애플리케이션의 편견과 불균형을 해소하는 교육 자료를 제공할 것을 권장합니다.
  • 또한 협회는 인증 및 라이선스 기관에 증강 지능을 인증 및 라이선스 표준에 통합하는 방법을 고려할 것을 촉구합니다. 

Some scholars have already called for a medical school curriculum reform in the age of AI (Kolachalama & Garg, 2018; Wartman & Combs, 2018; Wartman & Combs, 2019; Paranjape et al., 2019). For example, Kolachalama and Garg (2018) encourage medical schools to strive for “machine learning literacy,” which includes an emphasis on “the benefits, risks, and the ethical dilemmas” of AI. In June 2019, the AMA released a policy regarding the integration of augmented intelligence into medical education. The AMA recommends, among other things,

  • the inclusion of data scientists and engineers on medical school faculties, as well as
  • educational materials that address bias and disparities of augmented intelligence applications.
  • The Association also calls on accreditation and licensing bodies to consider how augmented intelligence should be integrated into accreditation and licensing standards.

우리는 이러한 요구를 지지하며, 의료 인공지능의 윤리적 차원을 포괄적으로 다루는 의과대학 커리큘럼의 필요성을 강조합니다. 의료 전문가가 의료 AI에 적절히 대비할 수 있도록 의대생은 데이터 과학에 대한 지식뿐만 아니라 AI가 제기하는 윤리적 문제에 대한 미묘한 이해와 인식도 키워야 합니다. 오늘날의 의대생은 의료 서비스에 통합된 AI의 개발, 구현 및 평가에 참여하게 될 것입니다. 이러한 과정에 의미 있게 참여하려면 AI 윤리에 대한 탄탄한 기초가 필요합니다.
We support these calls and highlight the need for a medical school curriculum that comprehensively addresses the ethical dimensions of health AI. To ensure that health care professionals are adequately prepared for health AIs, medical school students do not only need to become knowledgeable in data science but also need to develop a nuanced understanding and awareness of the ethical issues raised by them. Today’s medical students will be involved in the development, implementation, and evaluation of AI that is integrated into health care. They require a solid foundation in AI ethics to engage meaningfully in these processes.

의대 교육에 필요한 윤리적 이슈는 무엇인가요?
What are the ethical issues requiring medical school education?

특히 의과대학 교육이 필요한 6가지 주요 윤리적 이슈는 의료 AI에 의해 제기됩니다:

  • 1. 사전 동의,
  • 2. 편견,
  • 3. 안전,
  • 4. 투명성,
  • 5. 환자 개인정보 보호,
  • 6. 할당.

우리는 모든 의료 전문가가 이러한 이슈가 AI와 어떻게 상호작용하는지에 대한 윤리적 이해가 있어야 한다고 생각합니다. 의대생은 이미 공부하면서 이러한 문제를 접했을 수도 있지만, AI는 이러한 문제를 복잡하게 만들고 특별한 주의가 필요한 고유한 우려를 제시합니다. 각 문제에 대해 사례 기반 방식으로 커리큘럼에 통합할 수 있는 AI의 예를 제공합니다. 
There are, in particular, six key ethical issues raised by health AI that require medical school education:

  • 1. informed consent,
  • 2. bias,
  • 3. safety,
  • 4. transparency,
  • 5. patient privacy, and
  • 6. allocation.

We believe that all health professionals should have an ethical understanding of how these issues interplay with AI. Although medical students may already come across these issues in their studies, AI complicates these issues and presents unique concerns that warrant special attention. For each issue, we provide examples with AI that can be incorporated into the curriculum in a case-based method.

정보에 입각한 동의
Informed consent

새로운 의료 AI를 임상 진료에 배치할 때 정보에 입각한 동의에 대한 긴급한 질문이 제기됩니다(Gerke et al., 2020c). 의사가 치료 계획을 결정하기 위해 의료 AI를 사용하는 경우 환자는 의사 결정 과정에 AI가 관여했다는 사실을 알 권리가 있나요? 만약, 그렇다면 환자에게 어떤 구체적인 정보를 알려야 할까요? (Cohen, 2020). 예를 들어, 임상의는 알고리즘을 훈련하는 데 사용되는 데이터가 전자 건강 기록 데이터인지 또는 인위적으로 생성된(합성) 데이터인지와 같은 데이터에 대해 환자에게 알려야 하나요?  
The deployment of new health AIs in clinical practice raises urgent questions about informed consent (Gerke et al., 2020c). If a doctor uses a health AI to determine a treatment plan, does the patient have a right to know that the AI was involved in the decision-making process, and if so, what specific information should the patient be told? (Cohen, 2020). For example, does the clinician need to inform the patient about the data that is being used to train the algorithm, such as whether it is electronic health record data or artificially created (synthetic) data?

환자의 자율성을 존중하기 위해 의료계는 일반적인 치료 동의서에 서명하는 것 외에 정보에 입각한 동의가 필요한 건강 AI 사용에 대해 결정할 필요가 있습니다. 이 문제에 대한 답은 특정 의료 AI와 그 적용 분야에 따라 달라질 수 있습니다. 가상의 미래 AI가 환자의 치료 과정을 완전히 자율적으로 결정하는 경우에는 구체적인 사전 동의가 필요할 수도 있습니다. 저희는 의사가 사전 동의를 둘러싼 윤리와 특정 의료 AI의 개발 방식, 작동 방식, 사용 의도를 이해해야만 이 문제에 대해 비판적인 의견을 제시할 수 있다고 믿습니다. 이해관계자들은 더 많은 의료 AI가 임상 진료에 도입되기 전에 지금부터 사전 동의의 윤리적 문제에 대한 논의를 시작해야 합니다. 이 논의를 거친 후 두 번째 단계로, 특정 의료 AI에 맞는 사전 동의에 관한 의사를 위한 가이드라인과 커뮤니케이션 계획이 개발되어 임상에서 유용한 도구로 사용될 수 있습니다. 의대생들은 이 분야를 주시할 필요가 있으며, 이러한 논의를 형성하는 데 중요한 역할을 할 수 있습니다. 
In an effort to respect patient autonomy, the medical community needs to decide if there is something essential about using health AIs that requires informed consent (in addition to signing the general treatment consent form). The answer to this issue will likely depend on the particular health AI and its application. Perhaps a hypothetical future AI that fully autonomously decides on the treatment course of a patient will warrant a specific informed consent. We believe that only by understanding the ethics surrounding informed consent, as well as how a specific health AI is developed, how it works, and how it is intended to be used, can physicians formulate a critical opinion on this matter. Stakeholders should start the discussion today on the ethical issue of informed consent before even more health AIs enter clinical practice. As a second step (after having this discussion), guidelines and communication plans for physicians on informed consent that are tailored to specific health AIs could then be developed and serve as a useful tool in clinical practice. Medical students need to watch this space and can also play a role in shaping this discussion.

편견
Bias

편견에는 다양한 유형이 있으며, 의료 전문가는 의료 AI에 존재하는 편견의 유형을 인식하고 환자를 대할 때와 의료 시스템 수준에서 이러한 편견을 완화하는 방법에 대해 생각하는 것이 중요합니다. 편향된 데이터로 의료 AI를 학습시키면 기존의 건강 격차가 줄어들기는커녕 오히려 확대될 수 있습니다. 예를 들어, 의사가 주로 백인 피부를 대상으로 학습된 의료 AI 암 검진 기기를 신뢰하여 흑인 환자의 피부가 어두운 사람에게 가장 흔한 흑색종(Villines, 2019)을 놓치는 시나리오가 발생할 수 있습니다(Adamson & Smith, 2018). 지도 머신러닝을 기반으로 하든 비지도 머신러닝을 기반으로 하든, 백인 피부만을 입력 데이터로 학습한 AI는 다른 피부 타입의 암을 진단하는 방법을 학습할 수 없습니다. 
There are different types of biases, and it is important for health care professionals to be aware of the types of biases that exist with health AI, and to think about how to mitigate such biases both within their patient encounters and on a health care system level. If health AIs are trained on biased data, existing disparities in health may be augmented rather than reduced. For example, the scenario could be that a Black patient’s acral lentiginous melanoma, the most common type of melanoma in individuals with darker skin (Villines, 2019), is missed because a physician trusts a health AI cancer screening device that was primarily trained on white skin (Adamson & Smith, 2018). Whether this AI is based on supervised or unsupervised ML, if it is only trained on white skin as input data, it cannot learn how to diagnose cancer in any other skin type.

다른 경우에는 알고리즘에 편견이 있을 수 있습니다. 예를 들어, 최근 대규모 의료 시스템과 보험사가 매년 미국 내 약 2억 명에게 적용되는 의료 결정을 안내하는 데 사용하는 알고리즘에서 흑인 환자에 대한 심각한 인종적 편견이 있음을 보여준 Obermeyer 등(2019)의 연구 결과가 있습니다. 이 알고리즘은 데이터 세트의 흑인 환자가 훨씬 더 아팠음에도 불구하고 백인 환자와 동일한 수준의 위험을 흑인 환자에게 잘못 할당했습니다(Obermeyer et al., 2019). 이러한 인종적 편향은 알고리즘이 건강 요구 수준을 대신하여 (질병 대신) 의료 비용을 사용했기 때문에 발생했습니다(Obermeyer 외., 2019). 흑인 환자의 건강에 더 적은 비용이 지출되었기 때문에 알고리즘은 흑인 환자가 더 건강하다고 잘못 결론을 내렸습니다(Obermeyer et al., 2019). 알고리즘 편향은 사용된 라벨에 대해 신중하게 생각함으로써 완화될 수 있지만(Obermeyer 등, 2019), 편향되지 않은 모델에 대해 학습할 수 있는 새로운 알고리즘도 필요합니다(Wiens 등, 2020). 
In other cases, the algorithm may exhibit bias. For example, Obermeyer et al., (2019) recently showed a major racial bias against Black patients in an algorithm used by large health systems and payers to guide health decisions applied to about 200 million people in the U.S. every year. The algorithm falsely assigned the same level of risk to Black as to White patients, even though Black patients in the dataset were much sicker (Obermeyer et al., 2019). Such a racial bias arose because the algorithm used health care costs (instead of illness) as a proxy for the level of health needs (Obermeyer et al., 2019). Since less money was spent on Black patients’ health, the algorithm incorrectly concluded that Black patients were healthier (Obermeyer et al., 2019). Algorithmic biases may be mitigated by carefully thinking about the labels used (Obermeyer et al., 2019), but new algorithms are also needed that can be trained on unbiased models (Wiens et al., 2020).

따라서 목표는 "제품이 설계되고 테스트된 후가 아니라 설계 단계에서부터 윤리를 고려하는 것"이어야 합니다(Gerke et al., 2019). 이해관계자, 특히 AI 제조업체는 의료 AI와 함께 존재할 수 있는 편견의 유형을 인식하고 제품 개발 과정에서 가능한 한 빨리 편견을 완화하기 위해 노력해야 합니다. 발생할 수 있는 편견의 유형에 대한 인식이 있다면 의대생들이 AI 개발 팀에 합류하여 미래의 의료 AI 설계에 기여할 수 있습니다. 또한, 미래의 직업에 대비하여 환자 치료를 위해 임상에서 사용하기 전에 의료 AI를 비판적으로 면밀히 검토할 준비가 될 것입니다. 
Consequently, the goal should be “ethics by design—rather than after a product has been designed and tested” (Gerke et al., 2019). Stakeholders, particularly AI makers, must be aware of the types of biases that can exist alongside health AI and try as early as possible in the development process of their products to mitigate biases. With an awareness of the types of biases that can arise, medical students may be encouraged to join AI development teams, and contribute to the design of future health AIs. Moreover, they will be prepared for their future profession and will be ready to critically scrutinize a health AI before using it in clinical practice for the treatment of their patients.

안전
Safety

속 쓰림을 호소하는 여성에게 위식도 역류 질환(흔히 위산 역류로 알려져 있음)이 있다고 가정하고 단순히 제산제를 추천하는 온라인 챗봇을 생각해 보세요. 숙련된 의사는 이 환자에게 심장마비를 즉시 배제하는 것이 중요하다는 것을 알고 있어야 합니다. 많은 AI 앱과 챗봇은 불필요한 병원 방문을 제한하도록 설계되었지만, 지속적인 업데이트, 점검 또는 규제가 이루어지지 않으면 소비자에게 심각한 피해를 입힐 수 있습니다. 
Consider an online chatbot that assumes that a woman with a sensation of heartburn has gastroesophageal reflux disease, commonly known as acid reflux, and simply recommends an antacid. A trained physician should know the importance of immediately ruling out a heart attack for this patient. Many AI apps and chatbots are designed to limit unnecessary doctor visits, but some of them can also cause serious harm to consumers if they are not continuously updated, checked, or regulated.

FDA와 같은 규제 당국은 최근 일부 의료용 AI(의료용 소프트웨어, 즉 SaMD, "하드웨어 의료 기기의 일부가 아니면서 이러한 목적을 수행하는 하나 이상의 의료 목적으로 사용되는 소프트웨어")를 규제하는 새로운 접근 방식을 도입하여 성능을 개선하고 안전성과 효과를 보장하기 위한 조치를 취하고 있습니다(IMDRF, 2013). 예를 들어, 소프트웨어 사전 인증 프로그램을 개발하려는 FDA의 노력(2019b), 소위 "업데이트 문제"와 "적응형" 알고리즘과 "잠금형" 알고리즘의 처리 문제를 해결하기 위한 최근 토론 보고서(2019c), 최근 실행 계획(2021)이 있습니다. 특히 FDA와 같은 규제 기관은 현재 시간이 지남에 따라 계속 학습하고 변화하는 알고리즘의 마케팅을 허용해야 하는지에 대한 문제에 직면해 있습니다(FDA, 2019c; Babic et al., 2019). 이러한 적응형 알고리즘의 마케팅을 허용하기로 결정할 경우, 이러한 AI/ML 기반 SaMD가 지속적으로 안전하고 효과적이라는 것을 어떻게 보장할 수 있는지에 대한 후속 문제에 직면하게 됩니다. 규제 당국은 이러한 새로운 혁신적 규제 모델의 세부 사항을 파악해야 하지만, 이러한 모델의 핵심 구성 요소는 기기 관점이 아닌 시스템 관점뿐만 아니라 AI/ML 시스템 고유의 기능으로 인한 위험에 초점을 맞춘 지속적인 위험 모니터링 접근 방식이 필수적입니다(Babic 외., 2019; Gerke 외., 2020a).  
Regulators like the FDA have recently initiated steps toward new approaches of regulating some health AIs—those that are classified as Software as a Medical Device (i.e., SaMD, “software intended to be used for one or more medical purposes that perform these purposes without being part of a hardware medical device”) (IMDRF, 2013)—to improve their performance and ensure their safety and effectiveness. Examples include the FDA’s effort to develop a Software Precertification Program (2019b), its recent discussion paper (2019c) to address the so-called “update problem” and the treatment of “adaptive” versus “locked” algorithms (Babic et al., 2019), and its recent Action Plan (2021). In particular, regulators like the FDA currently face the problem of whether they should permit the marketing of algorithms that continue to learn and change over time (FDA, 2019c; Babic et al., 2019). If they decide to permit the marketing of such adaptive algorithms, they face the follow-up issue of how they can safeguard that such AI/ML-based SaMD are continuously safe and effective. Regulators still need to figure out the details of such new innovative regulatory models, but it is essential that a key component of such models will be a continuous risk monitoring approach that focuses on risks due to features that are specific to AI/ML systems as well as a system view rather than a device one (Babic et al., 2019; Gerke et al., 2020a).

학생들이 이러한 노력을 이해하는 것이 중요하며, 특히 여전히 남아있는 의료 AI 안전에 대한 격차와 장벽에 주의를 기울여야 합니다. 지속적인 규제 이니셔티브에도 불구하고 미래의 의료 전문가는 특정 임상 의사 결정 지원 소프트웨어와 많은 AI 앱 및 챗봇과 같이 현재 많은 의료 AI가 FDA 검토 대상이 아니라는 사실을 알아야 합니다(미국 연방 식품, 의약품 및 화장품법, 520(o)(1)). FDA 시판 전 검토를 받아야 하는 의료용 AI의 경우, 안전성과 효과에 대한 합리적인 보증을 제공하기 위해 다양한 통제가 필요한 여러 경로가 적용될 수 있습니다(FDA, 2018d). 의대생은 복잡한 규제 환경을 인식하고 해당 분야의 새로운 발전과 프로세스에 계속 참여하는 것이 중요합니다. 
It is important for students to understand these efforts, with particular attention to the gaps and barriers to health AI safety that remain. Even with ongoing regulatory initiatives, future health professionals need to know that many health AIs are currently not subject to FDA review, such as certain clinical decision support software and many AI apps and chatbot (U.S. Federal Food, Drug, and Cosmetic Act, s. 520(o)(1)). Of those health AIs that do need to undergo FDA premarket review, they can be subject to different pathways requiring different controls to provide reasonable assurance of their safety and effectiveness (FDA, 2018d). It is important for medical students to be aware of the complicated regulatory landscape and to remain engaged with such processes and new developments in the field.

환자들이 진료실 밖에서 찾을 수 있는 의료 솔루션을 점점 더 많이 수용함에 따라 의료 전문가는 소비자 건강 AI의 잠재적 위험에 대해 환자에게 교육해야 합니다. 또한 안전을 염두에 두고 설계된 AI 앱과 챗봇을 추천할 수 있어야 합니다. 이미 시장에 압도적으로 많은 앱과 챗봇이 출시되어 있고 매일 더 많은 앱과 챗봇이 추가되고 있기 때문에 이는 의료 전문가에게 현실적인 도전이 될 것입니다. 또한 이러한 앱과 챗봇은 자주 업데이트되기 때문에, 여전히 신뢰할 수 있고 추천할 만한지 판단하기가 더욱 어렵습니다. 또한 AI가 환자 치료에 관여할 때 누가 책임을 져야 하는지에 대한 우려도 상당합니다(Price et al., 2019, 2021). 의대생들이 이러한 문제에 대한 인식을 갖게 되면 해당 분야의 전문가들과 협력하여 건강 앱과 챗봇의 안전성과 효과를 평가하는 강력한 윤리 및 법적 프레임워크를 개발하는 등 문제 해결에 참여하도록 장려할 수 있습니다. 
As patients embrace health care solutions that can be increasingly found outside of the doctor’s office, health professionals will need to educate patients about the potential risks of consumer health AIs. They will need to be able to recommend AI apps and chatbots that are designed with safety in mind. This will be a practical challenge for health professionals since there is already an overwhelming number of apps and chatbots available on the market, and more are added every day. In addition, these apps and chatbots are frequently updated, which makes it even harder to determine whether they are still trustworthy and recommendable. There are also significant concerns regarding who is liable when AI is involved in patient care (Price et al., 2019, 2021). By gaining an awareness of these problems, medical students might be encouraged to participate in solving them, such as by engaging with experts in the field to develop robust ethical and legal frameworks that evaluate the safety and effectiveness of health apps and chatbots.

투명성
Transparency

사전 동의와 안전 문제를 복잡하게 만들기 위해 '블랙박스'인 의료 AI를 사용하면 의사가 본질적으로 완전히 투명하지 않은 시스템으로 작업하면서 환자에게 어떻게 투명성을 유지할 수 있는지에 대한 의문이 제기됩니다. '블랙박스'는 일반적으로 의사의 환자 치료를 돕기 위해 설계되었지만 입력 데이터가 어떻게 분석되어 의사 결정에 도달하는지 설명하지 않는 소프트웨어로 설명할 수 있습니다(Daniel et al., 2019). 이러한 설명 불가능성은 의사가 쉽게 이해할 수 없는 복잡한 AI/ML 모델 또는 알고리즘이 독점적인 것으로 간주되기 때문에 발생할 수 있습니다(Daniel et al., 2019). 의사가 특정 환자에 대해 의료 AI/ML이 어떻게 또는 왜 결정을 내렸는지 이해할 수 없는 경우, 환자에게 알려야 할 내용은 물론 해당 소프트웨어에 의존해야 하는지 여부를 고려하는 것이 중요합니다. 데이터 과학자들이 AI/ML의 블랙박스를 열기 위해 노력하고 있지만(Lipton, 2016), 의사와 환자가 얼마나 많은 정보를 가져야 하는지에 대한 문제는 설명 가능한 AI와 함께 남아있을 것입니다. 
To complicate the matters of informed consent and safety, the use of health AIs that are ‘black boxes’ raises the question of how physicians can remain transparent with patients while working with systems that are, by nature, not fully transparent. ‘Black boxes’ can be described as software that is usually designed to help physicians with patient care, but that does not explain how the input data is analyzed to reach its decision (Daniel et al., 2019). This inexplicability may result from complicated AI/ML models that cannot be easily understood by physicians or due to the algorithm being considered proprietary (Daniel et al., 2019). If physicians cannot comprehend how or why a health AI/ML has arrived at a decision for a particular patient, it is important to consider whether they should be relying on the software, let alone what they should be informing their patients. Although data scientists are working on opening the black box of AI/ML (Lipton, 2016), the issue of how much information physicians and their patients should have will remain with explainable AI.

새로 배포된 의료 AI/ML이 환자의 사망 날짜를 예측했지만, 의사는 의료 AI/ML이 이 날짜를 어떻게 계산했는지 이해할 방법이 없는 상황을 생각해 보십시오. 사망률을 예측하는 ML 기법은 이미 여러 그룹에서 설명한 바 있습니다(Motwani 외., 2017; Shouval 외., 2017; Weng 외., 2019). 의료 AI/ML이 환자가 몇 시간 더 생존하지 못할 것이라고 계산했다고 상상해 보세요. 한편 의사는 제한적이기는 하지만 다른 환자에 대한 임상 경험을 바탕으로 환자의 예후를 훨씬 더 좋게 예측할 수 있습니다.

  • 사는 환자와 가족에게 건강 AI의 계산 결과를 알려야 할까요?
  • 의사가 정확히 뭐라고 말해야 할까요?

이제 대부분의 경우 건강 AI/ML이 올바른 것으로 판명되었다고 가정해 보겠습니다.

  • 이제 의사는 이 AI/ML에 전적으로 의존해야 할까요?
  • 의사가 임상에서 블랙박스형 의료 AI/ML 모델을 전혀 사용해야 할까요? 

Consider a situation in which a newly deployed health AI/ML predicts the date of a patient’s death, but the physician has no way of understanding how the health AI/ML has calculated this date. ML techniques to predict mortality have already been described by several groups (Motwani et al., 2017; Shouval et al., 2017; Weng et al., 2019). Imagine that the health AI/ML has calculated that the patient will not survive a few more hours. Meanwhile, based on clinical experience, albeit limited, with other patients, the physician would estimate a much better prognosis for the patient.

  • Should the physician inform the patient and their family about the health AI’s calculation?
  • What exactly would the physician say?

Now consider that the health AI/ML turns out to be correct in most such cases.

  • Should a physician now rely on this AI/ML fully?
  • Should physicians use black-box health AI/ML models in clinical practice at all?

이러한 질문에 대한 답은 의사의 법적 책임 위험과 같은 다양한 고려 사항에 따라 달라질 수 있습니다. 또한 AI/ML 제조업체가 무작위 임상시험 등을 통해 해당 기기가 안전하고 효과적이라는 충분한 증거를 제시했는지 여부에 따라 달라질 수 있습니다. 의대생이라면 알겠지만, 의학에서 사용되는 많은 약물은 초기에 완전히 이해되지 않았을 수 있습니다. 예를 들어, 임상의들은 약 70년 동안 아스피린이 해열, 진통, 항염증 효과가 있다는 것을 알고 있었지만 그 근본적인 메커니즘은 알지 못한 채 아스피린을 처방해 왔습니다(London, 2019). 아스피린이 사이클로옥시게나제와 결합하여 프로스타글란딘 생성을 억제하여 이러한 효과를 생성하는 경로는 나중에야 밝혀졌습니다(Vane & Botting, 2003). 따라서 신뢰할 수 있고 정확하며 장기와 같은 희소 자원을 할당하는 데 사용되지 않는다는 충분한 증거가 있는 한, 의사가 임상에서 일부 블랙박스 건강 AI/ML 모델을 사용할 수 있고 사용해야 한다고 주장할 수도 있습니다(Babic et al., 2020). 의대생들이 가까운 미래에 의사로서 블랙박스 의료 AI/ML 모델과 함께 일하게 된다면, 병원이 상당한 비용을 들여 구입하여 사용을 권장하고 있는 블랙박스 의료 AI/ML에 대해 병상에서 반대하는 자신을 발견하기 전에 윤리적으로 그렇게 하는 방법에 대한 논의가 시작되어야 합니다. 
The answers to such questions may depend on different considerations, such as the physician’s liability risk. They may also depend on whether the AI/ML maker has shown sufficient proof that the device is safe and effective such as through randomized clinical trials. As medical students might know, many of the drugs used in medicine may not have been fully understood initially. For example, clinicians prescribed Aspirin for about 70 years, knowing that the drug had antipyretic, analgesic, and anti-inflammatory effects, but without knowing its underlying mechanism (London, 2019). The pathway by which aspirin binds cyclooxygenase to inhibit prostaglandin production to produce these effects was only later understood (Vane & Botting, 2003). Thus, one might argue that physicians can (and perhaps should) use some black-box health AI/ML models in clinical practice, as long as there is sufficient proof that they are reliable and accurate and are not used for allocating scarce resources such as organs (Babic et al., 2020). If medical students are to be working alongside black-box health AI/ML models as physicians in the near future, the discussions about how to do so ethically should begin before they find themselves at the bedside disagreeing with a black-box health AI/ML that the hospital purchased for a considerable amount of money and thus is likely encouraging them to use.

환자 개인정보 보호
Patient privacy

임상 환경을 넘어 웨어러블과 건강 앱을 통해 일상 생활에서 환자로부터 수집되는 대량의 건강 데이터로 인해 환자 개인정보 보호가 중요한 고려 사항으로 떠올랐습니다. 전례 없는 양의 건강 데이터가 축적되면 환자가 어느 정도인지 인지하지 못한 채 환자의 개인정보가 침해될 수 있습니다. 
With the large amounts of health data collected from patients, beyond clinical settings and in daily life through wearables and health apps, patient privacy has emerged as an important consideration. The accumulation of unprecedented amounts of health data may compromise patient privacy, without patients even realizing to what extent.

예를 들어, 최근의 소송인 Dinerstein 대 Google 소송(2019)은 개인의 데이터 보호 및 개인정보 보호에 대한 새로운 관심을 반영하고 있습니다. 이 소송은 시카고대학교 의료센터의 환자였던 매트 디너스타인이 개인 자격으로, 그리고 비슷한 처지에 있는 다른 모든 환자를 대신하여 Google, 시카고대학교, 시카고대학교 의료센터를 상대로 제기한 소송입니다. 2017년 시카고대학교 의료센터와 Google은 병원 재입원과 같은 의료 이벤트를 예측하기 위해 새로운 머신러닝 기법을 사용하기 위한 파트너십을 선언했고(Wood, 2017), 1년 후 연구 결과가 발표되었습니다(Rajkomar 외., 2018). Dinerstein은 2009년부터 2016년 사이에 시카고대학교 의료센터가 환자의 명시적 동의 없이 자유 텍스트 메모와 데이터 스탬프가 포함된 수십만 건의 의료 기록을 Google에 전송했다고 주장했습니다(Dinerstein v. Google; Rajkomar 외., 2018). 그러나 이 소송은 2020년 9월 일리노이주 연방 판사가 Dinerstein이 손해를 입증하지 못했다는 이유로 기각했습니다. 이 사례는 구글과 같은 거대 기술 기업과 환자 데이터를 공유하는 병원을 상대로 소송을 제기하는 데 따르는 어려움을 강조하고, 의료 데이터 개인정보 보호가 불충분하다는 것을 보여줍니다(Becker, 2020). 
For example, the recent lawsuit, Dinerstein v. Google, has reflected the emerging concern for individuals’ data protection and privacy (2019). The lawsuit was by Matt Dinerstein, a patient of the University of Chicago Medical Center, individually and on behalf of all other patients similarly situated, against Google, the University of Chicago, and the University of Chicago Medical Center. In 2017, the University of Chicago Medical Center and Google proclaimed a partnership to use new ML techniques to predict medical events, such as hospital readmissions (Wood, 2017), and the study results were published a year later (Rajkomar et al., 2018). Dinerstein claimed that between 2009 and 2016, the University of Chicago Medical Center transferred hundreds of thousands of medical records to Google, which included free-text notes and datestamps without obtaining patients’ express consent (Dinerstein v. Google; Rajkomar et al., 2018). However, this lawsuit was dismissed in September 2020 by a federal judge in Illinois on the grounds that Dinerstein failed to demonstrate damages. This case highlights the challenges of pursuing claims against hospitals that share patient data with tech giants such as Google, and shows the insufficient protection of health data privacy (Becker, 2020).

이 사례를 통해 학생들은 데이터 처리 방법 및 공유 대상과 같은 데이터 공유 문제환자가 자신의 데이터에 대해 가져야 하는 권리의 유형에 대해 배울 수 있습니다. 이 사례는 다음과 같은 새로운 환자 개인정보 보호 문제에 대한 토론을 촉진하는 데 도움이 됩니다. 

  • 데이터 삼각 측량을 통한 재식별 문제(Cohen & Mello, 2019; Price & Cohen, 2019; Gerke 외., 2020b),
  • 의료 AI 혁신과 민감한 환자 데이터 간의 상호 작용,
  • 병원 시스템과 타사 의료 AI 개발자 간의 관계 

Through this example, students can learn about data sharing issues such as how data should be handled and with whom it should be shared, as well as the types of rights patients should have regarding their data. This example helps to facilitate a discussion on emerging patient privacy concerns, such as

  • the issue of reidentification through data triangulation (Cohen & Mello, 2019; Price & Cohen, 2019; Gerke et al., 2020b),
  • the interplay between health AI innovation and sensitive patient data, and
  • the relationship between hospital systems and third party health AI developers.

환자 개인정보 보호는 의학계에서 오랫동안 지켜온 원칙이며, 모든 의료 AI의 개발과 구현에 있어 핵심적인 고려사항으로 남아 있어야 합니다. 의대생은 의료 분야에서 데이터 프라이버시의 윤리적, 법적 차원에 대한 이해와 함께 새로운 의료 AI 개발을 접하여 환자 프라이버시를 위협하는 새로운 의료 기술의 구현을 비판적으로 평가할 수 있어야 합니다.
Patient privacy has been a longstanding principle in medicine, and should remain a key consideration with the development and implementation of all health AIs. Medical students should be confronted with new health AI developments alongside an understanding of the ethical and legal dimensions of data privacy in health care so they can critically appraise the implementation of new health technologies that threaten patient privacy.

할당
Allocation

마지막으로 할당 문제가 있습니다. 자원의 공정한 배분은 의료 분야에서 지속적으로 제기되는 문제이며, 의료 AI의 등장으로 더욱 복잡해졌습니다. 실제 사례 중 하나는 알고리즘을 기반으로 간병 자원을 할당하는 것입니다. 2016년에 주 프로그램의 일환으로 주당 56시간의 간병을 배정받았던 아칸소 주 출신의 뇌성마비 여성 태미 돕스(Tammy Dobbs)는 주정부가 알고리즘에 의존해 간병을 배정하기로 결정한 후 갑자기 32시간만 배정받았습니다(Lecher, 2018). 돕스 씨에 따르면 이 시간은 충분하지 않았고 알고리즘이 어떻게 결정에 도달했는지에 대한 정보도 제공되지 않았습니다(Lecher, 2018). AI는 이러한 결정을 보이지 않게 만들 수 있기 때문에 할당 결정을 복잡하게 만듭니다. 의대생들은 태미 돕스처럼 자동화된 시스템으로 인해 의료 자원에 대한 접근성이 저하될 수 있는 미래의 환자들을 옹호할 수 있도록 AI가 자원 할당과 어떻게 상호 작용하는지에 대한 인식을 가져야 합니다. 
Finally, there is the issue of allocation. The just allocation of resources is an ongoing issue in health care, made more complicated by the advent of health AI. One real-life example involves the allocation of caregiving resources based on an algorithm. In 2016, Tammy Dobbs, a woman from Arkansas with cerebral palsy who was initially allocated 56 h of care per week as part of a state program, was suddenly allotted just 32 h after the state decided to rely on an algorithm to allocate its caregiving (Lecher, 2018). According to Ms. Dobbs, these hours were insufficient, and she was not given any information about how the algorithm reached its decision (Lecher, 2018). AI complicates allocation decisions because it can make these decisions invisible. Medical students must gain an awareness of how AI interfaces with resource allocation so they can advocate for future patients, who, like Tammy Dobbs, may find their access to health care resources undermined by automated systems.

임상 환경에서 의료 AI를 사용하는 것은 이러한 도구의 급여화에 달려 있습니다. 미래의 많은 환자들, 아마도 가장 취약한 환자들은 보험이 적용되지 않는다면 의료 AI 진단 또는 의료 AI 치료 도구에 접근하지 못할 수도 있습니다. 학생들과 함께 논의해야 할 질문은 다음과 같습니다:

  • 의료 AI의 맥락에서 정의란 무엇인가?
  • 정의를 파괴하지 않고 증진하기 위해 의료 AI를 어떻게 설계하고 사용할 수 있을까요?
  • 의사, 의료 보험 회사, 공중 보건 당국은 어떻게 의료 AI의 공정한 배분을 보장할 수 있을까요? 

The use of health AIs in clinical settings also depends on the reimbursement of such tools. Many future patients—perhaps the most vulnerable ones—may not have access to health AI diagnostic or health AI treatment tools if their insurance does not cover them. Questions that should be discussed with students include:

  • What is justice in the context of health AI?
  • How can health AI be designed and used to promote justice rather than to subvert it?
  • How can physicians, health insurance companies, and public health authorities ensure the just allocation of health AI?

결론
Conclusion

의료 시스템의 중요한 미래 이해당사자인 오늘날의 의대생은 환자와의 일상적인 임상적 상호 작용에서, 그리고 의료에 AI를 통합하는 것에 대한 광범위한 정책 논의에서 의료 AI 사용과 관련된 중요한 결정을 내려야 할 것입니다. 최근의 사건들은 의대생들에게 의료 AI 윤리를 가르쳐야 할 필요성을 강조하고 있습니다. 의과대학에서 의료 AI에 대한 윤리적 논의를 시작하여 학생들이 의료 AI의 지속적인 발전에 주의를 기울일 뿐만 아니라, 윤리적 렌즈를 통해 이러한 발전을 이해하고 의료 AI 또는 기타 디지털 의료 기술로 인해 새롭게 제기되는 윤리적 문제를 처리하는 데 필요한 도구를 개발할 수 있도록 장려해야 합니다. 
As important future stakeholders in the health care system, today’s medical students will have to make important decisions related to the use of health AI, both in every day clinical interactions with patients, and in broader policy discussions about the emergent integration of AI into health care. Recent events are highlighting the need to teach health AI ethics to medical students. We must begin the ethical discussion about health AI in medical schools so that students will not only be encouraged to pay attention to the ongoing developments in health AI but also will develop the tools needed to understand these developments through an ethical lens and be able to deal with new emerging ethical issues raised by health AI or other digital health technologies.

이 커리큘럼을 구성하는 윤리적 이슈는 사전 동의, 편견, 안전, 투명성, 환자 개인정보 보호 및 할당입니다. 이러한 문제에 대해 도전적인 질문을 제기하고 사례 연구의 맥락에서 제시할 수 있는 최신 의료 AI 이벤트가 부족하지 않습니다. 의대생은 이러한 문제를 제기하는 실제 사례를 바탕으로 한 사례 연구를 접함으로써 이미 의료 현장에 영향을 미치고 있는 의료 AI의 윤리적 문제를 이해하고 해결할 수 있는 기술을 습득할 수 있습니다.

The ethical issues that should constitute this curriculum are informed consent, bias, safety, transparency, patient privacy, and allocation. There is no shortage of current health AI events that raise challenging questions about these issues, and that can be presented in the context of case studies. With exposure to case studies based on real-life examples that raise these issues, medical students can gain the skills to appreciate and solve the ethical challenges of health AI that are already affecting the practice of medicine.

 


Adv Health Sci Educ Theory Pract. 2021 Oct;26(4):1447-1458. doi: 10.1007/s10459-021-10040-3. Epub 2021 Mar 3.

The need for health AI ethics in medical school education

Affiliations

1Schulich School of Medicine and Dentistry, Western University, 1151 Richmond St, London, ON, N6A 5C1, Canada.

2The Petrie-Flom Center for Health Law Policy, Biotechnology, and Bioethics at Harvard Law School, Harvard University, 23 Everett Street, Cambridge, MA, 02138, USA. sgerke@law.harvard.edu.

PMID: 33655433

DOI: 10.1007/s10459-021-10040-3

Abstract

Health Artificial Intelligence (AI) has the potential to improve health care, but at the same time, raises many ethical challenges. Within the field of health AI ethics, the solutions to the questions posed by ethical issues such as informed consent, bias, safety, transparency, patient privacy, and allocation are complex and difficult to navigate. The increasing amount of data, market forces, and changing landscape of health care suggest that medical students may be faced with a workplace in which understanding how to safely and effectively interact with health AIs will be essential. Here we argue that there is a need to teach health AI ethics in medical schools. Real events in health AI already pose ethical challenges to the medical community. We discuss key ethical issues requiring medical school education and suggest that case studies based on recent real-life examples are useful tools to teach the ethical issues raised by health AIs.

Keywords: Artificial intelligence; Curriculum; Ethics; Health; Medical education.

프레임워크 방법을 사용하여 다분야 보건연구에서 질적자료 분석하기(BMC Med Res Methodol. 2013)
Using the framework method for the analysis of qualitative data in multi-disciplinary health research
Nicola K Gale1*, Gemma Heath2, Elaine Cameron3, Sabina Rashid4 and Sabi Redwood2

 

질적 데이터의 관리 및 분석을 위한 프레임워크 방법은 1980년대부터 사용되어 왔습니다[1]. 이 방법은 대규모 사회 정책 연구에서 시작되었지만 의료 및 보건 연구에서도 점점 더 대중적인 접근법이 되고 있지만, 그 잠재적 적용과 한계에 대해 약간의 혼란이 있습니다. 이 글에서는 프레임워크 방법을 사용하는 것이 적절한 시기와 다른 질적 분석 방법과 비교하는 방법에 대해 논의합니다. 특히 다학제 보건 연구팀에서 이 방법을 어떻게 사용할 수 있는지 살펴봅니다. 응용 보건 연구에서 다학제 및 혼합 방법 연구가 점점 더 보편화되고 있습니다. 간호학, 심리학, 사회학 등 질적 연구에 익숙한 분야뿐만 아니라 역학자, 보건 경제학자, 경영 과학자 등이 팀에 포함되는 경우가 많습니다. 또한 응용 보건 연구에는 종종 임상적 대표성이 있으며, 점점 더 많은 환자와 대중이 참여하고 있습니다[2]. 우리는 경험이 풍부한 질적 방법론자의 리더십이 의심할 여지없이 필요하지만, 더 넓은 팀의 비전문가도 분석 과정에 참여할 수 있고 참여해야 한다고 주장합니다. 그런 다음 프레임워크 방법의 적용에 대한 단계별 가이드를 제시하며, 프로세스의 주요 단계를 설명하기 위해 발표된 연구[3]의 작업 예시(추가 파일 1 참조)를 사용하여 설명합니다. 기술 용어는 용어집(아래)에 포함되어 있습니다. 마지막으로 이 접근 방식의 강점과 한계에 대해 논의합니다. 
The Framework Method for the management and analysis of qualitative data has been used since the 1980s [1]. The method originated in large-scale social policy research but is becoming an increasingly popular approach in medical and health research; however, there is some confusion about its potential application and limitations. In this article we discuss when it is appropriate to use the Framework Method and how it compares to other qualitative analysis methods. In particular, we explore how it can be used in multi-disciplinary health research teams. Multi-disciplinary and mixed methods studies are becoming increasingly commonplace in applied health research. As well as disciplines familiar with qualitative research, such as nursing, psychology and sociology, teams often include epidemiologists, health economists, management scientists and others. Furthermore, applied health research often has clinical representation and, increasingly, patient and public involvement [2]. We argue that while leadership is undoubtedly required from an experienced qualitative methodologist, non-specialists from the wider team can and should be involved in the analysis process. We then present a step-by-step guide to the application of the Framework Method, illustrated using a worked example (See Additional File 1) from a published study [3] to illustrate the main stages of the process. Technical terms are included in the glossary (below). Finally, we discuss the strengths and limitations of the approach.

프레임워크 방식에 사용되는 주요 용어 용어집
Glossary of key terms used in the Framework Method

  • 분석 프레임워크: 분석에 참여하는 연구자들이 공동으로 개발한 카테고리로 구성된 일련의 코드로, 데이터를 관리하고 구성하는 데 사용할 수 있습니다. 프레임워크는 연구 질문에 답하는 데 도움이 되는 방식으로 데이터를 요약/축소하는 데 도움이 되는 새로운 데이터 구조(참가자가 제공한 전체 원본 설명이 아닌)를 생성합니다.
    Analytical framework: A set of codes organised into categories that have been jointly developed by researchers involved in analysis that can be used to manage and organise the data. The framework creates a new structure for the data (rather than the full original accounts given by participants) that is helpful to summarize/reduce the data in a way that can support answering the research questions.
  • 분석 메모: 특정 개념, 주제 또는 문제에 대한 서면 조사로, 분석 프로세스를 포착하는 데이터의 새로운 이슈를 반영합니다(추가 파일 1, 섹션 7 참조).
    Analytic memo: A written investigation of a particular concept, theme or problem, reflecting on emerging issues in the data that captures the analytic process (see Additional file 1, Section 7).
  • 카테고리: 분석 과정에서 코드는 유사하고 상호 연관된 아이디어 또는 개념을 중심으로 클러스터로 그룹화됩니다. 카테고리와 코드는 일반적으로 분석 프레임워크에서 트리 다이어그램 구조로 배열됩니다. 카테고리는 원시 데이터와 밀접하고 명시적으로 연결되어 있지만, 카테고리를 개발하는 것은 데이터를 추상화하는 프로세스를 시작하는 방법입니다(즉, 특정 또는 일화적인 것이 아니라 일반적인 것을 향해).
    Categories: During the analysis process, codes are grouped into clusters around similar and interrelated ideas or concepts. Categories and codes are usually arranged in a tree diagram structure in the analytical framework. While categories are closely and explicitly linked to the raw data, developing categories is a way to start the process of abstraction of the data (i.e. towards the general rather than the specific or anecdotal).
  • 차트 작성: 요약된 데이터를 프레임워크 메서드 매트릭스에 입력합니다(추가 파일 1, 섹션 6 참조). 
    Charting: Entering summarized data into the Framework Method matrix (see Additional File 1, Section 6).
  • 코드: '코딩'이라는 프로세스에서 원시 데이터의 발췌에 할당되는 설명적 또는 개념적 레이블입니다(추가 파일 1, 섹션 3 참조). 
    Code: A descriptive or conceptual label that is assigned to excerpts of raw data in a process called ‘coding’ (see Additional File 1, Section 3).
  • 데이터: 정성적 데이터는 일반적으로 분석하기 전에 텍스트 형식이어야 합니다. 이러한 텍스트는 도출 텍스트(음식 일기 등 연구를 위해 특별히 작성된 텍스트) 또는 기존 텍스트(회의록, 정책 문서 또는 웹로그 등 기존 텍스트)일 수도 있고, 인터뷰 또는 포커스 그룹 데이터를 필사하거나 참여자 관찰을 수행하거나 대상 또는 사회적 상황을 관찰하면서 '현장' 노트를 작성하여 생성할 수도 있습니다.
    Data: Qualitative data usually needs to be in textual form before analysis. These texts can either be elicited texts (written specifically for the research, such as food diaries), or extant texts (pre-existing texts, such as meeting minutes, policy documents or weblogs), or can be produced by transcribing interview or focus group data, or creating ‘field’ notes while conducting participant-observation or observing objects or social situations.
  • 색인화: 합의된 분석 프레임워크의 코드를 전체 데이터 세트에 체계적으로 적용하는 것입니다(추가 파일 1, 섹션 5 참조).
    Indexing: The systematic application of codes from the agreed analytical framework to the whole dataset (see Additional File 1, Section 5).
  • 매트릭스: 스프레드시트에는 요약된 데이터를 코드(열)와 사례(행)별로 입력하는 수많은 셀이 포함되어 있습니다(추가 파일 1, 섹션 6 참조).
    Matrix: A spreadsheet contains numerous cells into which summarized data are entered by codes (columns) and cases (rows) (see Additional File 1, Section 6).
  • 주제: 전체 데이터 집합 분석의 최종 결과물로, 데이터의 측면을 묘사하거나 설명하는 해석적 개념 또는 명제입니다. 테마는 사례 간 및 사례 내 비교를 통해 데이터 범주를 조사하여 명확하게 표현하고 개발합니다. 일반적으로 여러 범주가 각 테마 또는 하위 테마에 속합니다[3].
    Themes: Interpretive concepts or propositions that describe or explain aspects of the data, which are the final output of the analysis of the whole dataset. Themes are articulated and developed by interrogating data categories through comparison between and within cases. Usually a number of categories would fall under each theme or sub-theme [3].
  • 트랜스크립트: 인터뷰나 대화와 같은 구두 상호 작용에 대한 축어적(단어 하나하나) 기록입니다.
    Transcript: A written verbatim (word-for-word) account of a verbal interaction, such as an interview or conversation.
12874_2012_991_MOESM1_ESM.docx
0.16MB

배경
Background

프레임워크 방법은 흔히 주제별 분석 또는 질적 콘텐츠 분석이라고 하는 광범위한 분석 방법군에 속합니다. 이러한 접근 방식은 정성적 데이터의 공통점과 차이점을 파악한, 다음 데이터의 여러 부분 간의 관계에 초점을 맞추어 주제를 중심으로 한 기술적 또는 설명적 결론을 도출하고자 합니다. 프레임워크 방법은 1980년대 후반 영국 국립사회연구센터의 질적 연구 부서의 연구자인 제인 리치와 리즈 스펜서에 의해 대규모 정책 연구에 사용하기 위해 개발되었습니다[1]. 현재는 건강 연구를 비롯한 다른 분야에서도 널리 사용되고 있습니다[3-12]. 요약된 데이터의 행(사례), 열(코드), '셀'로 구성된 매트릭스 아웃풋은 연구자가 데이터를 체계적으로 축소하여 사례별, 코드별로 분석할 수 있는 구조를 제공합니다[1]. 대부분의 경우 '사례'는 개별 인터뷰 대상자이지만, 미리 정의된 그룹이나 조직과 같은 다른 분석 단위에도 적용될 수 있습니다. 전체 데이터 세트에서 주요 주제에 대한 심층 분석이 이루어질 수 있지만, 각 연구 참여자의 견해는 매트릭스 내에서 해당 account의 다른 측면과 연결되어 있으므로 개인의 견해에 대한 맥락이 손실되지 않습니다. 데이터를 비교하고 대조하는 것은 질적 분석에 매우 중요하며, 프레임워크 방법의 구조와 프로세스에는 개별 사례 내뿐만 아니라 사례 간 데이터를 쉽게 비교할 수 있는 기능이 내장되어 있습니다. 
The Framework Method sits within a broad family of analysis methods often termed thematic analysis or qualitative content analysis. These approaches identify commonalities and differences in qualitative data, before focusing on relationships between different parts of the data, thereby seeking to draw descriptive and/or explanatory conclusions clustered around themes. The Framework Method was developed by researchers, Jane Ritchie and Liz Spencer, from the Qualitative Research Unit at the National Centre for Social Research in the United Kingdom in the late 1980s for use in large-scale policy research [1]. It is now used widely in other areas, including health research [312]. Its defining feature is the matrix output: rows (cases), columns (codes) and ‘cells’ of summarised data, providing a structure into which the researcher can systematically reduce the data, in order to analyse it by case and by code [1]. Most often a ‘case’ is an individual interviewee, but this can be adapted to other units of analysis, such as predefined groups or organisations. While in-depth analyses of key themes can take place across the whole data set, the views of each research participant remain connected to other aspects of their account within the matrix so that the context of the individual’s views is not lost. Comparing and contrasting data is vital to qualitative analysis and the ability to compare with ease data across cases as well as within individual cases is built into the structure and process of the Framework Method.

프레임워크 방법은 따라야 할 명확한 단계를 제공하며, 요약된 데이터의 고도로 구조화된 결과물을 생성합니다. 따라서 여러 연구자가 프로젝트를 진행하는 경우, 특히 모든 구성원이 질적 데이터 분석 경험이 없는 다학제 연구팀에서 전체 데이터 집합에 대한 총체적이고 설명적인 개요를 얻는 것이 바람직한 대규모 데이터 집합을 관리하는 데 유용합니다. 그러나 이 방법은 모든 유형의 정성적 데이터를 분석하거나 모든 정성적 연구 질문에 답하는 데 적합한 도구가 아니며, 정량적 연구자를 위한 '쉬운' 정성적 연구 버전도 아니므로 이 방법을 선택하기 전에 주의를 기울이는 것이 좋습니다. 중요한 점은 프레임워크 방법은 매우 이질적인 데이터를 수용할 수 없다는 것입니다. 즉, 데이터는 유사한 주제나 주요 이슈를 다루고 있어야 분류가 가능하다는 것입니다. 물론 개별 인터뷰 대상자는 각 주제와 관련하여 매우 다른 견해나 경험을 가지고 있을 수 있으며, 이를 비교하고 대조할 수 있습니다. 프레임워크 방법은 반구조화된 인터뷰 녹취록의 주제별 분석에 가장 일반적으로 사용되지만, 원칙적으로 회의록이나 일기[12], 관찰 현장 노트[10] 등 다른 유형의 텍스트 데이터[13]에도 적용될 수 있습니다.
The Framework Method provides clear steps to follow and produces highly structured outputs of summarised data. It is therefore useful where multiple researchers are working on a project, particularly in multi-disciplinary research teams were not all members have experience of qualitative data analysis, and for managing large data sets where obtaining a holistic, descriptive overview of the entire data set is desirable. However, caution is recommended before selecting the method as it is not a suitable tool for analysing all types of qualitative data or for answering all qualitative research questions, nor is it an ‘easy’ version of qualitative research for quantitative researchers. Importantly, the Framework Method cannot accommodate highly heterogeneous data, i.e. data must cover similar topics or key issues so that it is possible to categorize it. Individual interviewees may, of course, have very different views or experiences in relation to each topic, which can then be compared and contrasted. The Framework Method is most commonly used for the thematic analysis of semi-structured interview transcripts, which is what we focus on in this article, although it could, in principle, be adapted for other types of textual data [13], including documents, such as meeting minutes or diaries [12], or field notes from observations [10].

질적 연구자와 함께 일하거나 처음으로 질적 연구를 탐구하는 정량적 연구자에게는 프레임워크 방법의 체계적인 프로세스와 '스프레드시트' 접근 방식이 정량적 패러다임에 더 밀접하게 부합하는 것처럼 보이기 때문에 매력적으로 다가옵니다[14]. 프레임워크 방법은 다루기 어려워 보일 수 있는 정성적 데이터를 분류하고 정리하는 매우 체계적인 방법이지만, (분석 선택을 내리고 해석 전략을 가시화하고 감사할 수 있게 만드는 방법 등) 정성적 데이터 분석과 관련된 일반적인 문제에 대한 만병통치약은 아닙니다. 매트릭스를 적절하게 해석하고 설명, 범주, 설명 및 유형화를 용이하게 생성하기 위해서는 질적 연구 스킬이 필요합니다. 또한 다른 질적 방법에서와 마찬가지로 프레임워크 방법에서도 반성성, 엄격성 및 품질이 요구됩니다. 따라서 분석에 프레임워크 방법을 사용하는 연구는 숙련된 질적 연구자가 감독하는 것이 필수적이지만, 그렇다고 해서 질적 연구를 처음 접하는 연구자가 더 넓은 연구팀의 일원으로 분석에 기여하는 것을 배제하는 것은 아닙니다. 
For quantitative researchers working with qualitative colleagues or when exploring qualitative research for the first time, the nature of the Framework Method is seductive because its methodical processes and ‘spreadsheet’ approach seem more closely aligned to the quantitative paradigm [14]. Although the Framework Method is a highly systematic method of categorizing and organizing what may seem like unwieldy qualitative data, it is not a panacea for problematic issues commonly associated with qualitative data analysis such as how to make analytic choices and make interpretive strategies visible and auditable. Qualitative research skills are required to appropriately interpret the matrix, and facilitate the generation of descriptions, categories, explanations and typologies. Moreover, reflexivity, rigour and quality are issues that are requisite in the Framework Method just as they are in other qualitative methods. It is therefore essential that studies using the Framework Method for analysis are overseen by an experienced qualitative researcher, though this does not preclude those new to qualitative research from contributing to the analysis as part of a wider research team.

질적 데이터 분석에는 담화 분석[15] 및 민족지학[16]과 같이 언어와 사회적 상호작용에서 언어가 어떻게 사용되는지에 주목하는 접근법, 현상학[17, 18] 및 내러티브 방법[19]과 같이 경험, 의미 및 언어에 관심을 갖는 접근법, 근거 이론[20, 21]과 같이 일련의 절차와 상호 연결된 단계를 통해 데이터에서 도출된 이론을 개발하고자 하는 접근법 등 다양한 접근법이 존재합니다. 이러한 접근법 중 다수는 특정 분야와 연관되어 있으며 분석 과정을 형성하는 철학적 아이디어에 의해 뒷받침됩니다[22]. 그러나 프레임워크 방법은 특정 인식론적, 철학적 또는 이론적 접근 방식과 일치하지 않습니다. 오히려 테마를 생성하는 것을 목표로 하는 다양한 질적 접근 방식과 함께 사용할 수 있도록 조정할 수 있는 유연한 도구입니다. 
There are a number of approaches to qualitative data analysis, including those that pay close attention to language and how it is being used in social interaction such as discourse analysis [15] and ethnomethodology [16]; those that are concerned with experience, meaning and language such as phenomenology [17, 18] and narrative methods [19]; and those that seek to develop theory derived from data through a set of procedures and interconnected stages such as Grounded Theory [20, 21]. Many of these approaches are associated with specific disciplines and are underpinned by philosophical ideas which shape the process of analysis [22]. The Framework Method, however, is not aligned with a particular epistemological, philosophical, or theoretical approach. Rather it is a flexible tool that can be adapted for use with many qualitative approaches that aim to generate themes.

테마의 개발은 질적 데이터 분석의 공통적인 특징으로, 조사 대상 현상을 밝힐 수 있는 완전한 설명을 생성하기 위해 패턴을 체계적으로 검색하는 것을 포함합니다. 특히, 많은 질적 접근 방식은 근거 이론의 일부로 개발된 '지속적 비교 방법'을 사용하는데, 이는 각 주제를 구체화하기 위해 사례 간에 체계적으로 비교하는 것을 포함합니다[21, 23]. 근거 이론과 달리 프레임워크 방법은 사회 이론을 생성하는 데는 관심이 없지만, 매트릭스 전반에 걸친 데이터 검토를 통해 지속적인 비교 기법을 크게 촉진할 수 있습니다. 
The development of themes is a common feature of qualitative data analysis, involving the systematic search for patterns to generate full descriptions capable of shedding light on the phenomenon under investigation. In particular, many qualitative approaches use the ‘constant comparative method’ , developed as part of Grounded Theory, which involves making systematic comparisons across cases to refine each theme [21, 23]. Unlike Grounded Theory, the Framework Method is not necessarily concerned with generating social theory, but can greatly facilitate constant comparative techniques through the review of data across the matrix.

프레임워크 방법이 매우 체계적이기 때문에, 다른 논평가들이 지적했듯이, 종종 질적 분석에 대한 연역적 접근법과 혼동되기도 합니다[13, 14]. 그러나 이 도구 자체는 귀납적 또는 연역적 주제 분석 중 어느 쪽에도 충실하지 않으며, 연구가 귀납적-연역적 연속체에서 어디에 위치하는지는 연구 질문에 따라 달라집니다.

  • '환자가 심혈관 질환 발병에 대해 정확한 생의학적 설명을 할 수 있는가'와 같은 질문은 본질적으로 예/아니오 질문이므로(설명의 정도나 적절한 용어 사용에 따라 미묘한 차이가 있을 수 있지만) 데이터 수집과 분석(예: 구조화 또는 반구조화 인터뷰, 지시적 질적 내용 분석[24]) 모두에 연역적 접근 방식이 필요합니다.
  • 마찬가지로 행동 변화 이론과 같은 기존 이론에 근거하여 분석하는 경우, 예를 들어 '계획된 행동 이론이 GP 처방을 설명하는 데 어떻게 도움이 되는가'와 같은 연구 질문의 경우 연역적 접근 방식을 취할 수 있습니다[11].
  • 그러나 '사람들은 심혈관 질환 발병에 대한 설명을 어떻게 구성하는가'와 같은 연구 질문은 예상치 못한 것을 허용하고 연구자가 미리 예측할 수 없는 문화적 신념, 음식 준비 습관, '운명'의 개념 또는 슬픔과 같은 삶의 다른 중요한 사건과의 연관성을 포함할 수 있는 인터뷰 대상자의 사회적 위치 응답[25]을 허용하는 보다 귀납적 접근 방식이 필요합니다(예: 인터뷰 대상자 주도 개방형 인터뷰 및 근거 이론 [20]).

Perhaps because the Framework Method is so obviously systematic, it has often, as other commentators have noted, been conflated with a deductive approach to qualitative analysis [13, 14]. However, the tool itself has no allegiance to either inductive or deductive thematic analysis; where the research sits along this inductive-deductive continuum depends on the research question.

  • A question such as, ‘Can patients give an accurate biomedical account of the onset of their cardiovascular disease?’ is essentially a yes/no question (although it may be nuanced by the extent of their account or by appropriate use of terminology) and so requires a deductive approach to both data collection and analysis (e.g. structured or semi-structured interviews and directed qualitative content analysis [24]).
  • Similarly, a deductive approach may be taken if basing analysis on a pre-existing theory, such as behaviour change theories, for example in the case of a research question such as ‘How does the Theory of Planned Behaviour help explain GP prescribing?’ [11].
  • However, a research question such as, ‘How do people construct accounts of the onset of their cardiovascular disease?’ would require a more inductive approach that allows for the unexpected, and permits more socially-located responses [25] from interviewees that may include matters of cultural beliefs, habits of food preparation, concepts of ‘fate’, or links to other important events in their lives, such as grief, which cannot be predicted by the researcher in advance (e.g. an interviewee-led open ended interview and grounded theory [20]).

이러한 모든 경우에 프레임워크 방법을 사용하여 데이터를 관리하는 것이 적절할 수 있습니다.

  • 연역적 접근법에서는 이전 문헌, 이전 이론 또는 연구 질문의 특성에 따라 주제와 코드가 미리 선택되는 반면,
  • 귀납적 접근법에서는 개방형(제한 없는) 코딩을 통해 데이터에서 주제를 생성한 후 주제를 구체화하는 방식으로 차이가 분명해집니다.

많은 경우, 프로젝트에 탐구해야 할 특정 이슈가 있지만 참가자의 경험이나 현상에 의미를 부여하는 방식에서 예상치 못한 다른 측면을 발견할 수 있는 여지를 남겨두려는 목적이 있는 경우 두 가지 접근 방식을 결합하는 것이 적절합니다. 요컨대, 프레임워크 방법은 연역적, 귀납적 또는 복합적 유형의 질적 분석에 맞게 조정할 수 있습니다. 그러나 사례와 주제별로 데이터를 분석하는 것이 적절하지 않은 연구 질문이 있으므로 프레임워크 방법을 피해야 하는 경우도 있습니다. 예를 들어,

  • 연구 질문에 따라 생활사 데이터는 내러티브 분석[19]을,
  • 환자와 의료진 간의 상담 기록은 대화 분석[26]을,
  • 임산부를 위한 자료와 같은 다큐멘터리 데이터는 담화 분석[27]을 사용하여 분석하는 것이 더 적합할 수 있습니다. 

In all these cases, it may be appropriate to use the Framework Method to manage the data. The difference would become apparent in how themes are selected:

  • in the deductive approach, themes and codes are pre-selected based on previous literature, previous theories or the specifics of the research question; whereas
  • in the inductive approach, themes are generated from the data though open (unrestricted) coding, followed by refinement of themes.

In many cases, a combined approach is appropriate when the project has some specific issues to explore, but also aims to leave space to discover other unexpected aspects of the participants’ experience or the way they assign meaning to phenomena. In sum, the Framework Method can be adapted for use with deductive, inductive, or combined types of qualitative analysis. However, there are some research questions where analysing data by case and theme is not appropriate and so the Framework Method should be avoided. For instance,

  • depending on the research question, life history data might be better analysed using narrative analysis [19];
  • recorded consultations between patients and their healthcare practitioners using conversation analysis [26]; and
  • documentary data, such as resources for pregnant women, using discourse analysis [27].

연구 설계나 데이터 수집을 심도 있게 고려하는 것은 이 백서의 범위에 속하지 않지만, 프레임워크 방법 분석 프로세스를 설명하기 전에 한 걸음 물러나서 분석을 시작하기 전에 어떤 일이 일어나야 하는지 간략하게 살펴볼 필요가 있습니다. 분석 방법의 선택은 연구 제안 단계에서 고려되어야 하며, 연구 질문과 연구의 전반적인 목표에 부합해야 합니다. 많은 질적 연구, 특히 귀납적 분석을 사용하는 연구는 본질적으로 새로운 것이므로 연구자는 앞으로 일어날 일에 대한 "상상적 리허설"만 제공할 수 있습니다[28]. 혼합 방법 연구에서는 프로젝트의 더 넓은 목표 내에서 질적 구성 요소의 역할도 고려해야 합니다. 데이터 수집 단계에서는 질적 인터뷰가 고도로 숙련된 활동이기 때문에 적절히 훈련된 연구원이 질적 인터뷰를 수행할 수 있도록 리소스를 할당해야 합니다. 경우에 따라 연구팀은 일반인, 환자 또는 동료를 인터뷰에 참여시키기로 결정할 수 있으며[29-32], 이 경우 적절한 훈련과 멘토링을 받아야 하므로 시간과 자원이 필요합니다. 이 초기 단계에서는 데이터 관리 및 분석에 도움을 줄 수 있는 컴퓨터 지원 질적 데이터 분석 소프트웨어(CAQDAS)를 사용할지 여부를 고려하는 것도 유용합니다.

It is not within the scope of this paper to consider study design or data collection in any depth, but before moving on to describe the Framework Method analysis process, it is worth taking a step back to consider briefly what needs to happen before analysis begins. The selection of analysis method should have been considered at the proposal stage of the research and should fit with the research questions and overall aims of the study. Many qualitative studies, particularly ones using inductive analysis, are emergent in nature; this can be a challenge and the researchers can only provide an “imaginative rehearsal” of what is to come [28]. In mixed methods studies, the role of the qualitative component within the wider goals of the project must also be considered. In the data collection stage, resources must be allocated for properly trained researchers to conduct the qualitative interviewing because it is a highly skilled activity. In some cases, a research team may decide that they would like to use lay people, patients or peers to do the interviews [2932] and in this case they must be properly trained and mentored which requires time and resources. At this early stage it is also useful to consider whether the team will use Computer Assisted Qualitative Data Analysis Software (CAQDAS), which can assist with data management and analysis.

모든 형태의 정성적 또는 정량적 분석은 순전히 기술적인 과정이 아니라 연구자의 특성과 학문적 패러다임의 영향을 받기 때문에 연구의 설계, 데이터의 구성 또는 수집, 분석 등 연구 과정 전반에 걸친 비판적 성찰이 무엇보다 중요합니다. 팀의 모든 구성원은 연구 일기를 작성하여 반성적 메모, 데이터에 대한 인상, 분석에 대한 생각 등을 연구 과정 전반에 걸쳐 기록해야 합니다. 숙련된 질적 연구자는 엄격하고 반성적인 방식으로 데이터를 선별하고 분석하는 데 더욱 능숙해집니다. 실제 사회생활의 복잡성을 포용하고 설명하며 복잡한 사회 문제에 적용할 수 있는 풍부하고 미묘한 연구 결과를 도출하기 위해서는 확실성에 너무 집착하지 않고 연구 전반에 걸쳐 유연성과 적응력을 유지해야 합니다. 프레임워크 방법을 사용할 때는, 데이터 수집과 데이터 분석이 연구 과정의 엄격하게 순차적이고 상호 배타적인 단계인 양적 연구와 달리, 질적 분석에서는 프로젝트에 따라 어느 정도는 데이터 수집, 분석, 이론 개발 사이에 지속적인 상호작용이 있다는 점을 기억하는 것이 중요합니다. 예를 들어, 참가자의 새로운 아이디어나 인사이트가 잠재적으로 유익한 탐구 방향을 제시하거나, 면밀한 분석을 통해 추가 조사가 필요한 미묘한 불일치를 발견할 수 있습니다. 
As any form of qualitative or quantitative analysis is not a purely technical process, but influenced by the characteristics of the researchers and their disciplinary paradigms, critical reflection throughout the research process is paramount, including in the design of the study, the construction or collection of data, and the analysis. All members of the team should keep a research diary, where they record reflexive notes, impressions of the data and thoughts about analysis throughout the process. Experienced qualitative researchers become more skilled at sifting through data and analysing it in a rigorous and reflexive way. They cannot be too attached to certainty, but must remain flexible and adaptive throughout the research in order to generate rich and nuanced findings that embrace and explain the complexity of real social life and can be applied to complex social issues. It is important to remember when using the Framework Method that, unlike quantitative research where data collection and data analysis are strictly sequential and mutually exclusive stages of the research process, in qualitative analysis there is, to a greater or lesser extent depending on the project, ongoing interplay between data collection, analysis, and theory development. For example, new ideas or insights from participants may suggest potentially fruitful lines of enquiry, or close analysis might reveal subtle inconsistencies in an account which require further exploration.

분석 절차
Procedure for analysis

1단계: 전사
Stage 1: Transcription

좋은 품질의 오디오 녹음이 필요하며, 이상적으로는 인터뷰의 축어적(단어 대 단어) 필사본이 필요합니다. 프레임워크 방법 분석의 경우, 내용이 주요 관심사이기 때문에 읽기 어려울 수 있는 대화 필사본의 규칙(예: 일시 정지 또는 두 사람이 동시에 대화하는 경우)을 반드시 포함할 필요는 없습니다. 트랜스크립트에는 나중에 코딩하고 메모할 수 있도록 여백이 넓고 줄 간격이 적절해야 합니다. 필사 과정은 데이터에 몰입할 수 있는 좋은 기회이므로 신입 연구자에게 적극 권장할 만합니다. 그러나 일부 프로젝트에서는 이 작업을 전문 전사자에게 아웃소싱하는 것이 더 나은 리소스 활용이라고 판단할 수 있습니다. 
A good quality audio recording and, ideally, a verbatim (word for word) transcription of the interview is needed. For Framework Method analysis, it is not necessarily important to include the conventions of dialogue transcriptions which can be difficult to read (e.g. pauses or two people talking simultaneously), because the content is what is of primary interest. Transcripts should have large margins and adequate line spacing for later coding and making notes. The process of transcription is a good opportunity to become immersed in the data and is to be strongly encouraged for new researchers. However, in some projects, the decision may be made that it is a better use of resources to outsource this task to a professional transcriber.

2단계: 인터뷰에 익숙해지기
Stage 2: Familiarisation with the interview

오디오 녹음 및/또는 녹취록과 면접관이 녹음한 상황별 또는 반성적 메모를 사용하여 전체 인터뷰에 익숙해지는 것은 해석에 있어 매우 중요한 단계입니다. 오디오 녹음의 전체 또는 일부를 다시 듣는 것도 도움이 될 수 있습니다. 다학제 또는 대규모 연구 프로젝트의 경우, 데이터 분석에 참여하는 사람과 인터뷰를 진행하거나 필사한 사람이 다를 수 있으므로 이 단계가 특히 중요합니다. 한 칸의 여백을 사용하여 분석 메모, 생각 또는 인상을 기록할 수 있습니다. 

Becoming familiar with the whole interview using the audio recording and/or transcript and any contextual or reflective notes that were recorded by the interviewer is a vital stage in interpretation. It can also be helpful to re-listen to all or parts of the audio recording. In multi-disciplinary or large research projects, those involved in analysing the data may be different from those who conducted or transcribed the interviews, which makes this stage particularly important. One margin can be used to record any analytical notes, thoughts or impressions.

3단계: 코딩
Stage 3: Coding

익숙해지면 연구자는 대본을 한 줄 한 줄 주의 깊게 읽으면서 구절에서 중요하다고 해석한 내용을 설명하는 문구나 라벨('코드')을 붙입니다. 귀납적 연구에서는 이 단계에서 가능한 한 다양한 관점에서 관련성이 있을 수 있는 모든 것을 코딩하는 '오픈 코딩'이 이루어집니다. 코딩은 다음 등을 나타낼 수 있습니다.

  • 실체적인 것(예: 특정 행동, 사건 또는 구조),
  • 가치(예: 근거 기반 의학 또는 환자 선택에 대한 믿음과 같이 특정 진술을 알리거나 뒷받침하는 것),
  • 감정(예: 슬픔, 좌절, 사랑) 및
  • 인상적/방법론적인 요소(예: 인터뷰 대상자가 설명하기 어려운 것을 발견했다, 인터뷰 대상자가 감정적이 되었다, 인터뷰자가 불편함을 느꼈다) [33] 

After familiarization, the researcher carefully reads the transcript line by line, applying a paraphrase or label (a ‘code’) that describes what they have interpreted in the passage as important. In more inductive studies, at this stage ‘open coding’ takes place, i.e. coding anything that might be relevant from as many different perspectives as possible. Codes could refer to

  • substantive things (e.g. particular behaviours, incidents or structures),
  • values (e.g. those that inform or underpin certain statements, such as a belief in evidence-based medicine or in patient choice),
  • emotions (e.g. sorrow, frustration, love) and
  • more impressionistic/methodological elements (e.g. interviewee found something difficult to explain, interviewee became emotional, interviewer felt uncomfortable) [33].

순수 연역적 연구에서는 코드가 미리 정의되어 있을 수 있으므로(예: 기존 이론 또는 프로젝트의 특정 관심 분야) 이 단계가 꼭 필요하지 않을 수 있으며 바로 색인화로 넘어갈 수도 있지만, 일반적으로 데이터의 중요한 측면을 놓치지 않으려면, 적어도 몇 개의 트랜스크립트에서 오픈 코딩을 수행하는 것이 광범위한 연역적 접근 방식을 취하는 경우에도 도움이 될 수 있습니다. 코딩은 모든 데이터를 분류하여 데이터 세트의 다른 부분과 체계적으로 비교할 수 있도록 하는 것을 목표로 합니다. 가능하면 최소 두 명의 연구자(또는 다학제 연구팀의 경우 각 분야 또는 전문 분야에서 한 명 이상)가 처음 몇 개의 기록물을 독립적으로 코딩해야 합니다. 환자, 대중 참여 대표 또는 임상의도 이 단계에 생산적으로 참여할 수 있는데, 이는 대안적인 관점을 제시하여 특정 관점이 지배적이지 않도록 보장할 수 있기 때문입니다. 
In purely deductive studies, the codes may have been pre-defined (e.g. by an existing theory, or specific areas of interest to the project) so this stage may not be strictly necessary and you could just move straight onto indexing, although it is generally helpful even if you are taking a broadly deductive approach to do some open coding on at least a few of the transcripts to ensure important aspects of the data are not missed. Coding aims to classify all of the data so that it can be compared systematically with other parts of the data set. At least two researchers (or at least one from each discipline or speciality in a multi-disciplinary research team) should independently code the first few transcripts, if feasible. Patients, public involvement representatives or clinicians can also be productively involved at this stage, because they can offer alternative viewpoints thus ensuring that one particular perspective does not dominate.

귀납적 코딩에서는 문자 그대로의 설명적인 방식으로 코딩하는 것이 아니라, 예상치 못한 것을 찾아내는 것이 중요하므로, 다양한 관점을 가진 사람들의 참여가 큰 도움이 될 수 있습니다. 한 줄 한 줄 코딩을 통해 전체적인 내용을 파악할 수 있을 뿐만 아니라, 명확하게 표현되지 않았거나 나머지 계정과 '맞지 않아서' 보통은 보이지 않는 부분을 고려하도록 연구자에게 경각심을 줄 수 있습니다. 이런 식으로 데이터의 이상 징후를 조정하고 설명하면 분석을 더욱 강력하게 만들 수 있습니다. 코딩은 새로운 코드를 자동으로 추적할 수 있는 유용한 방법인 CAQDAS를 사용하여 디지털 방식으로도 수행할 수 있습니다. 그러나 일부 연구자들은 코딩의 초기 단계를 종이와 펜으로 하는 것을 선호하며, 5단계(아래 참조)에 도달한 후에야 CAQDAS를 사용하기 시작합니다.
It is vital in inductive coding to look out for the unexpected and not to just code in a literal, descriptive way so the involvement of people from different perspectives can aid greatly in this. As well as getting a holistic impression of what was said, coding line-by-line can often alert the researcher to consider that which may ordinarily remain invisible because it is not clearly expressed or does not ‘fit’ with the rest of the account. In this way the developing analysis is challenged; to reconcile and explain anomalies in the data can make the analysis stronger. Coding can also be done digitally using CAQDAS, which is a useful way to keep track automatically of new codes. However, some researchers prefer to do the early stages of coding with a paper and pen, and only start to use CAQDAS once they reach Stage 5 (see below).

4단계: 작업용 분석 프레임워크 개발하기
Stage 4: Developing a working analytical framework

처음 몇 개의 트랜스크립트를 코딩한 후에는 관련된 모든 연구자가 모여 각자가 적용한 레이블을 비교하고, 이후의 모든 트랜스크립트에 적용할 코드 세트에 동의해야 합니다. 코드를 카테고리로 그룹화한 다음(도움이 된다면 트리 다이어그램을 사용) 명확하게 정의할 수 있습니다. 이렇게 하면 작업용 분석 프레임워크가 형성됩니다. 추가 코드가 나오지 않을 때까지 분석 프레임워크를 여러 번 반복해야 할 수도 있습니다. 각 범주 아래에 '기타' 코드를 두어 맞지 않는 데이터를 무시하지 않도록 하고, 마지막 기록이 코딩될 때까지 분석 프레임워크는 결코 '최종'이 아닙니다. 
After coding the first few transcripts, all researchers involved should meet to compare the labels they have applied and agree on a set of codes to apply to all subsequent transcripts. Codes can be grouped together into categories (using a tree diagram if helpful), which are then clearly defined. This forms a working analytical framework. It is likely that several iterations of the analytical framework will be required before no additional codes emerge. It is always worth having an ‘other’ code under each category to avoid ignoring data that does not fit; the analytical framework is never ‘final’ until the last transcript has been coded.

5단계: 분석 프레임워크 적용
Stage 5: Applying the analytical framework

그런 다음 기존 카테고리와 코드를 사용하여 후속 트랜스크립트를 색인화하여 작업용 분석 프레임워크를 적용합니다. 각 코드에는 일반적으로 쉽게 식별할 수 있도록 번호나 약어가 할당되며(따라서 코드의 전체 이름을 매번 적을 필요가 없음), 트랜스크립트에 직접 기록됩니다. 컴퓨터 지원 정성적 데이터 분석 소프트웨어(CAQDAS)는 이 단계에서 특히 유용하며, 프로세스 속도를 높이고 이후 단계에서 데이터를 쉽게 검색할 수 있도록 해주기 때문입니다. 정확한 지침에 따라 실제로 계산을 수행하는 통계 분석용 소프트웨어와 달리, 정성적 분석 소프트웨어 패키지에 데이터를 넣는 것은 데이터를 분석하는 것이 아니라 분석 프로세스에 액세스할 수 있도록 데이터를 저장하고 구성하는 효과적인 방법일 뿐이라는 점에 유의할 필요가 있습니다.

The working analytical framework is then applied by indexing subsequent transcripts using the existing categories and codes. Each code is usually assigned a number or abbreviation for easy identification (and so the full names of the codes do not have to be written out each time) and written directly onto the transcripts. Computer Assisted Qualitative Data Analysis Software (CAQDAS) is particularly useful at this stage because it can speed up the process and ensures that, at later stages, data is easily retrievable. It is worth noting that unlike software for statistical analyses, which actually carries out the calculations with the correct instruction, putting the data into a qualitative analysis software package does not analyse the data; it is simply an effective way of storing and organising the data so that they are accessible for the analysis process.

6단계: 프레임워크 매트릭스에 데이터 차트화
Stage 6: Charting data into the framework matrix

정성적 데이터는 방대하며(한 시간의 인터뷰로 15~30페이지의 텍스트가 생성될 수 있음), 데이터를 관리하고 요약(축소)하는 것은 분석 프로세스에서 매우 중요한 측면입니다. 스프레드시트를 사용하여 매트릭스를 생성하고 데이터를 매트릭스에 '차트화'합니다. 차트 작성에는 각 기록에서 카테고리별로 데이터를 요약하는 작업이 포함됩니다. 좋은 차트를 작성하려면 한편으로는 데이터를 줄이고, 다른 한편으로는 인터뷰 대상자의 말의 원래 의미와 '느낌'을 유지하는 것 사이에서 균형을 맞출 수 있어야 합니다. 차트에는 흥미롭거나 예시적인 인용문에 대한 참조가 포함되어야 합니다. CAQDAS를 사용하여 데이터를 관리하는 경우 자동으로 태그를 지정할 수 있으며(N-Vivo 버전 9 이상에는 프레임워크 매트릭스를 생성하는 기능이 있음), 그렇지 않은 경우 대문자 'Q', (익명화된) 트랜스크립트 번호, 페이지 및 줄 참조로 충분합니다. 여러 분야가 참여하는 팀에서는 분석 프로세스의 초기 단계에서 요약 스타일을 비교하고 대조하여 팀 내 일관성을 유지하는 것이 도움이 됩니다. 약어를 사용할 때는 팀원들이 동의해야 합니다. 팀원들이 분석 프레임워크에 익숙해지고 코딩과 차트 작성에 능숙해지면 평균적으로 1시간 분량의 요약본을 작성하는 데 반나절 정도 걸립니다. 초기 단계에서는 훨씬 더 오래 걸립니다. 
Qualitative data are voluminous (an hour of interview can generate 15–30 pages of text) and being able to manage and summarize (reduce) data is a vital aspect of the analysis process. A spreadsheet is used to generate a matrix and the data are ‘charted’ into the matrix. Charting involves summarizing the data by category from each transcript. Good charting requires an ability to strike a balance between reducing the data on the one hand and retaining the original meanings and ‘feel’ of the interviewees’ words on the other. The chart should include references to interesting or illustrative quotations. These can be tagged automatically if you are using CAQDAS to manage your data (N-Vivo version 9 onwards has the capability to generate framework matrices), or otherwise a capital ‘Q’, an (anonymized) transcript number, page and line reference will suffice. It is helpful in multi-disciplinary teams to compare and contrast styles of summarizing in the early stages of the analysis process to ensure consistency within the team. Any abbreviations used should be agreed by the team. Once members of the team are familiar with the analytical framework and well practised at coding and charting, on average, it will take about half a day per hour-long transcript to reach this stage. In the early stages, it takes much longer.

7단계: 데이터 해석
Stage 7: Interpreting the data

데이터에 대한 인상, 아이디어, 초기 해석을 기록할 수 있는 별도의 노트나 컴퓨터 파일을 준비해 두는 것이 연구 전반에 걸쳐 유용합니다. 흥미로운 아이디어, 개념 또는 잠재적 주제를 탐색하기 위해 어느 단계에서든 분석 메모[20, 21]를 작성하여 일반인 및 임상 구성원을 포함한 다른 연구팀원들과 논의하는 것이 좋습니다. 점차적으로 데이터의 특징과 데이터 간의 차이점을 파악하여 다음을 할  수 있습니다. 

  • 유형학을 생성하고,
  • 이론적 개념(이전 개념 또는 데이터에서 나온 개념)을 질문하거나,
  • 범주 간의 연결을 매핑하여 관계 및 인과관계를 탐색

데이터가 충분히 풍부하다면 이 과정을 통해 도출된 결과는 특정 사례에 대한 기술을 넘어 설명으로 확장될 수 있습니다. 

  • 현상의 출현 이유,
  • 조직이나 기타 사회적 행위자가 상황을 어떻게 선동하거나 대응할지 예측,
  • 조직이나 시스템 내에서 제대로 작동하지 않는 영역을 식별

이 단계는 예상보다 오래 걸리는 경우가 많으므로 프로젝트 계획에 회의와 연구자 개개인의 연구 결과 해석 및 작성에 충분한 시간을 할당해야 합니다(추가 파일 1, 섹션 7 참조). 
It is useful throughout the research to have a separate note book or computer file to note down impressions, ideas and early interpretations of the data. It may be worth breaking off at any stage to explore an interesting idea, concept or potential theme by writing an analytic memo [20, 21] to then discuss with other members of the research team, including lay and clinical members. Gradually, characteristics of and differences between the data are identified, perhaps

  • generating typologies,
  • interrogating theoretical concepts (either prior concepts or ones emerging from the data) or
  • mapping connections between categories to explore relationships and/or causality.

If the data are rich enough, the findings generated through this process can go beyond description of particular cases to explanation of, for example,

  • reasons for the emergence of a phenomena,
  • predicting how an organisation or other social actor is likely to instigate or respond to a situation, or
  • identifying areas that are not functioning well within an organisation or system.

It is worth noting that this stage often takes longer than anticipated and that any project plan should ensure that sufficient time is allocated to meetings and individual researcher time to conduct interpretation and writing up of findings (see Additional file 1, Section 7).

토론
Discussion

프레임워크 방법은 25년 이상 개발되어 연구에 성공적으로 사용되어 왔으며, 최근 질적 보건 연구에서 널리 사용되는 분석 방법이 되었습니다. 질적 연구에서 품질을 평가하는 방법에 대한 문제는 많은 논쟁이 있어 왔지만[20, 34-40], 분석의 엄격성과 투명성을 보장하는 것은 필수적인 요소입니다. 물론 여러 가지 방법이 있지만 프레임워크 방법에서는 다음과 같은 방법이 도움이 됩니다: 
The Framework Method has been developed and used successfully in research for over 25 years, and has recently become a popular analysis method in qualitative health research. The issue of how to assess quality in qualitative research has been highly debated [20, 3440], but ensuring rigour and transparency in analysis is a vital component. There are, of course, many ways to do this but in the Framework Method the following are helpful:

  • 차트를 작성하는 동안 데이터를 요약하는 것은 데이터를 줄이는 실용적인 방법일 뿐만 아니라 일반인, 임상 및 (정량적) 학계 구성원을 포함한 다분야 팀의 모든 구성원이 (모든 기록을 읽거나 분석의 보다 기술적인 부분에 관여하지는 않더라도) 분석 과정에서 데이터에 참여하고 각자의 관점을 제공할 수 있음을 의미합니다.
    Summarizing the data during charting, as well as being a practical way to reduce the data, means that all members of a multi-disciplinary team, including lay, clinical and (quantitative) academic members can engage with the data and offer their perspectives during the analysis process without necessarily needing to read all the transcripts or be involved in the more technical parts of analysis.
  • 또한 차트를 작성하면 연구자가 해석으로 넘어가기 전각 참여자의 주관적인 프레임과 표현을 사용하여 데이터를 설명하는 데 세심한 주의를 기울일 수 있습니다. 
    Charting also ensures that researchers pay close attention to describing the data using each participant’s own subjective frames and expressions in the first instance, before moving onto interpretation.
  • 요약된 데이터는 각 사례의 더 넓은 맥락에서 유지되므로 복잡한 의미와 이해의 층위에 주의를 기울이는 두꺼운 설명을 장려합니다[38].
    The summarized data is kept within the wider context of each case, thereby encouraging thick description that pays attention to complex layers of meaning and understanding [38].
  • 매트릭스 구조는 시각적으로 간단하며 모순되는 데이터, 비정상적인 사례 또는 빈 셀에 주의를 기울이는 등 연구팀의 모든 구성원이 데이터의 패턴을 쉽게 인식할 수 있습니다.
    The matrix structure is visually straightforward and can facilitate recognition of patterns in the data by any member of the research team, including through drawing attention to contradictory data, deviant cases or empty cells.
  • 체계적인 절차(이 문서에 설명되어 있음)를 통해 여러 분야의 팀이나 대규모 데이터 세트가 있는 경우에도 쉽게 따라할 수 있습니다.
    The systematic procedure (described in this article) makes it easy to follow, even for multi-disciplinary teams and/or with large data sets.
  • 인터뷰가 아닌 데이터(예: 인터뷰 중에 작성한 현장 메모 또는 반성적 고려 사항)도 매트릭스에 포함할 수 있을 만큼 유연합니다.
    It is flexible enough that non-interview data (such as field notes taken during the interview or reflexive considerations) can be included in the matrix.
  • 특정 인식론적 관점이나 이론적 접근 방식과 일치하지 않으므로 귀납적 또는 연역적 분석 또는 이 둘을 조합하여 사용할 수 있습니다(예: 기존의 이론적 구성을 연역적으로 사용한 다음 귀납적 측면으로 이론을 수정하거나 귀납적 접근 방식을 사용하여 데이터에서 주제를 식별한 다음 문헌으로 돌아가 연역적으로 이론을 사용하여 특정 주제를 추가로 설명하는 데 도움이 됨).
    It is not aligned with a particular epistemological viewpoint or theoretical approach and therefore can be adapted for use in inductive or deductive analysis or a combination of the two (e.g. using pre-existing theoretical constructs deductively, then revising the theory with inductive aspects; or using an inductive approach to identify themes in the data, before returning to the literature and using theories deductively to help further explain certain themes).
  • 주제를 설명하기 위해 관련 데이터 추출을 식별하고 제안된 주제에 대한 충분한 증거가 있는지 확인하는 것이 쉽습니다.
    It is easy to identify relevant data extracts to illustrate themes and to check whether there is sufficient evidence for a proposed theme.
  • 마지막으로, 원본 원시 데이터에서 최종 테마까지, 예시 인용문을 포함한 명확한 감사 추적이 있습니다.
    Finally, there is a clear audit trail from original raw data to final themes, including the illustrative quotes.

이 접근 방식에는 여러 가지 잠재적인 함정이 있습니다:
There are also a number of potential pitfalls to this approach:

  • 배경에서 언급했듯이 체계적인 접근 방식과 매트릭스 형식은 정량적으로 훈련된 사람들에게는 직관적으로 매력적이지만, '스프레드시트' 형식은 정성적 연구에 대한 심층적인 이해가 없는 사람들이 정성적 데이터를 정량화하려는 유혹을 더욱 증가시킬 수 있습니다(예: "20명의 참가자 중 13명이 X라고 답함"). 정성적 연구의 표본 추출은 더 넓은 집단을 대표하도록 설계된 것이 아니라 현상에 대한 다양성을 포착하기 위한 것이기 때문에 이러한 종류의 진술은 분명히 무의미합니다 [41]. 
    The systematic approach and matrix format, as we noted in the background, is intuitively appealing to those trained quantitatively but the ‘spreadsheet’ look perhaps further increases the temptation for those without an in-depth understanding of qualitative research to attempt to quantify qualitative data (e.g. “13 out of 20 participants said X). This kind of statement is clearly meaningless because the sampling in qualitative research is not designed to be representative of a wider population, but purposive to capture diversity around a phenomenon [41].
  • 모든 질적 분석 방법과 마찬가지로 프레임워크 방법도 시간과 리소스 집약적입니다. 응용 보건 연구의 모범 사례처럼 데이터 분석 및 해석에 여러 이해관계자 및 분야를 참여시킬 경우 필요한 시간이 연장됩니다. 이 시간은 펀딩 전 단계에서 프로젝트 제안서에 반영되어야 합니다. 
    Like all qualitative analysis methods, the Framework Method is time consuming and resource-intensive. When involving multiple stakeholders and disciplines in the analysis and interpretation of the data, as is good practice in applied health research, the time needed is extended. This time needs to be factored into the project proposal at the pre-funding stage.
  • 새로운 다분야 팀에서 이 방법을 성공적으로 사용하려면 많은 교육이 필요합니다. 분석에서 각자의 역할에 따라 연구팀 구성원은 데이터를 코딩, 색인화, 차트화하는 방법을 배우고, 자신의 정체성과 경험이 분석 과정에 어떤 영향을 미치는지 반성적으로 생각해야 하며, 데이터의 의미와 중요성을 정당하게 해석하기 위해 일반화 방법(통계적 일반화가 아닌 분석적 일반화 및 전이 가능성[41])에 대해 배워야 할 수도 있습니다. 
    There is a high training component to successfully using the method in a new multi-disciplinary team. Depending on their role in the analysis, members of the research team may have to learn how to code, index, and chart data, to think reflexively about how their identities and experience affect the analysis process, and/or they may have to learn about the methods of generalisation (i.e. analytic generalisation and transferability, rather than statistical generalisation [41]) to help to interpret legitimately the meaning and significance of the data.

프레임워크 방법은 비전문가도 데이터 분석에 참여할 수 있지만, 대규모 혼합 방법 연구의 전체 책임자가 다른 사람이더라도 숙련된 질적 연구자가 프로젝트를 이끄는 것이 이 방법을 성공적으로 사용하는 데 매우 중요합니다. 질적 연구 책임자는 최소한 질적 분석에 대한 사전 교육이나 경험이 있는 다른 연구자들과 함께 하는 것이 이상적입니다. 질적 연구 책임자의 책임은 연구 설계, 프로젝트 일정 및 리소스 계획에 기여하고, 주니어 질적 연구자를 멘토링하고, 임상, 일반 및 기타 (비질적) 학자들이 분석 과정에 적절히 기여할 수 있도록 교육하고, 데이터 및 다른 팀원들과 비판적이고 반성적인 참여를 장려하는 방식으로 분석 회의를 촉진하고, 마지막으로 연구 보고서 작성을 주도하는 것입니다. 
While the Framework Method is amenable to the participation of non-experts in data analysis, it is critical to the successful use of the method that an experienced qualitative researcher leads the project (even if the overall lead for a large mixed methods study is a different person). The qualitative lead would ideally be joined by other researchers with at least some prior training in or experience of qualitative analysis. The responsibilities of the lead qualitative researcher are: to contribute to study design, project timelines and resource planning; to mentor junior qualitative researchers; to train clinical, lay and other (non-qualitative) academics to contribute as appropriate to the analysis process; to facilitate analysis meetings in a way that encourages critical and reflexive engagement with the data and other team members; and finally to lead the write-up of the study.

결론
Conclusion

우리는 프레임워크 방법 연구가 의료 전문가, 심리학자, 사회학자, 경제학자, 일반인/서비스 사용자 등을 포함하는 다학제 연구팀에 의해 수행될 수 있다고 주장해 왔습니다. 다양한 관점을 포함한다는 것은 분석 과정에서 의사결정을 내리는 데 많은 시간과 리소스가 소요될 수 있음을 의미합니다. 인터뷰 대상자가 표현하고 녹취록에서 확인된 아이디어가 각 분야의 기존 개념 및 이론과 프로젝트가 해결하고자 하는 보건 시스템의 실제 '문제'와 어떻게 관련되어 있는지에 대해 광범위하고 반성적이며 비판적인 대화가 필요할 수 있습니다. 그러나 이러한 종류의 팀 노력은 연구 결과의 신뢰성과 관련성을 높여 '전체가 부분의 합보다 크다'는 것을 보장하기 위해 임상 및 일반인의 연구 참여뿐만 아니라 학제 간 협업을 촉진하는 훌륭한 포럼입니다.
We have argued that Framework Method studies can be conducted by multi-disciplinary research teams that include, for example, healthcare professionals, psychologists, sociologists, economists, and lay people/service users. The inclusion of so many different perspectives means that decision-making in the analysis process can be very time consuming and resource-intensive. It may require extensive, reflexive and critical dialogue about how the ideas expressed by interviewees and identified in the transcript are related to pre-existing concepts and theories from each discipline, and to the real ‘problems’ in the health system that the project is addressing. This kind of team effort is, however, an excellent forum for driving forward interdisciplinary collaboration, as well as clinical and lay involvement in research, to ensure that ‘the whole is greater than the sum of the parts’, by enhancing the credibility and relevance of the findings.

프레임워크 방법은 텍스트 데이터, 특히 인터뷰 녹취록의 주제별 분석에 적합하며, 여러 사례의 데이터를 주제별로 비교하고 대조하는 동시에 각 관점을 각 개인 기록의 다른 측면과의 연관성을 유지하여 맥락에 맞게 배치하는 것이 중요합니다. 숙련된 질적 연구자가 분석의 모든 측면을 주도하고 진행해야 하지만, 프레임워크 방법의 체계적인 접근 방식은 다분야 팀의 모든 구성원을 참여시키는 데 적합합니다. 엄격한 질적 분석을 위해서는 모든 팀원의 개방적이고 비판적이며 반성적인 접근 방식이 필수적입니다.
The Framework Method is appropriate for thematic analysis of textual data, particularly interview transcripts, where it is important to be able to compare and contrast data by themes across many cases, while also situating each perspective in context by retaining the connection to other aspects of each individual’s account. Experienced qualitative researchers should lead and facilitate all aspects of the analysis, although the Framework Method’s systematic approach makes it suitable for involving all members of a multi-disciplinary team. An open, critical and reflexive approach from all team members is essential for rigorous qualitative analysis.

고품질의 질적 연구를 위해서는 실제 보건 시스템의 복잡성과 건강 문제에 대한 다양한 관점의 존재를 인정하는 것이 필요합니다. 질적 연구가 잘 수행되면 중요한 현상에 대한 설명과 예측을 제시하고, 대규모 연구의 정량적 부분과 건설적인 관계를 맺으며, 보건 서비스 개선과 보건 정책 개발에 기여할 수 있습니다. 프레임워크 방법을 적절히 선택하고 실행하면 신뢰할 수 있고 관련성 있는 결과를 도출하여 이러한 목표를 달성하는 데 적합한 도구가 될 수 있습니다. 
Acceptance of the complexity of real life health systems and the existence of multiple perspectives on health issues is necessary to produce high quality qualitative research. If done well, qualitative studies can shed explanatory and predictive light on important phenomena, relate constructively to quantitative parts of a larger study, and contribute to the improvement of health services and development of health policy. The Framework Method, when selected and implemented appropriately, can be a suitable tool for achieving these aims through producing credible and relevant findings.

요약
Summary

  • 프레임워크 방법은 데이터를 관리하고 매핑하기 위한 체계적인 모델을 제공하기 때문에 주제별(질적 내용) 분석을 지원하는 데 탁월한 도구입니다.
  • 프레임워크 방법은 사례 내 및 사례 간 비교를 통해 주제를 생성하는 것이 바람직한 인터뷰 데이터 분석에 가장 적합합니다.
  • 프레임워크 방법은 매트릭스 형태로 요약된 데이터를 직관적으로 구조화한 개요를 제공하기 때문에 대규모 데이터 세트의 관리가 용이합니다.
  • 프레임워크 방법의 명확한 단계별 프로세스는 학제 간 및 협업 프로젝트에 적합합니다.
  • 이 방법의 사용은 숙련된 질적 연구자가 주도하고 촉진해야 합니다.
  • The Framework Method is an excellent tool for supporting thematic (qualitative content) analysis because it provides a systematic model for managing and mapping the data.
  • The Framework Method is most suitable for analysis of interview data, where it is desirable to generate themes by making comparisons within and between cases.
  • The management of large data sets is facilitated by the Framework Method as its matrix form provides an intuitively structured overview of summarised data.
  • The clear, step-by-step process of the Framework Method makes it is suitable for interdisciplinary and collaborative projects.
  • The use of the method should be led and facilitated by an experienced qualitative researcher.

BMC Med Res Methodol. 2013 Sep 18;13:117. doi: 10.1186/1471-2288-13-117.

Using the framework method for the analysis of qualitative data in multi-disciplinary health research

Affiliations collapse

1Health Services Management Centre, University of Birmingham, Park House, 40 Edgbaston Park Road, Birmingham B15 2RT, UK. n.gale@bham.ac.uk.

PMID: 24047204

PMCID: PMC3848812

DOI: 10.1186/1471-2288-13-117

Free PMC article

Abstract

Background: The Framework Method is becoming an increasingly popular approach to the management and analysis of qualitative data in health research. However, there is confusion about its potential application and limitations.

Discussion: The article discusses when it is appropriate to adopt the Framework Method and explains the procedure for using it in multi-disciplinary health research teams, or those that involve clinicians, patients and lay people. The stages of the method are illustrated using examples from a published study.

Summary: Used effectively, with the leadership of an experienced qualitative researcher, the Framework Method is a systematic and flexible approach to analysing qualitative data and is appropriate for use in research teams even where not all members have previous experience of conducting qualitative research.

역량-기반 평가: 인공지능을 활용하여 하위역량 내용 예측하기 (Acad Med, 2023)
Competency-Based Assessments: Leveraging Artificial Intelligence to Predict Subcompetency Content
Gregory J. Booth, MD, Benjamin Ross, MD, William A. Cronin, MD, Angela McElrath, MD, Kyle L. Cyr, MD, John A. Hodgson, MD, Charles Sibley, MD, J. Martin Ismawan, MD, Alyssa Zuehl, MD, James G. Slotto, MD, Maureen Higgs, MD, Matthew Haldeman, MD, Phillip Geiger, MD, and Dink Jardine, MD, MS

 

 

역량 기반 의학교육(CBME)은 의학교육에서 졸업후교육의 기본이 되는 표준 패러다임이 되었습니다.1 미국의 경우, 미국의학전문대학원교육인증위원회(ACGME) 인증 프로그램 중 현재 평가 프레임워크는 6개의 핵심 역량 영역으로 구성되며, 각 영역은 수련의의 궤적을 추적하기 위해 5단계의 점진적 척도로 이루어져 있습니다. ACGME는 각 전문과목이 초심자에서 독립 전문의가 되기까지의 특정 경로("마일스톤"이라고 함)를 정의할 수 있도록 합니다.2 기술과 의료 행위가 발전함에 따라 이러한 경로는 주기적으로 재평가되어 지속적인 발전과 진화하는 수련 표준에 발맞출 수 있도록 보장합니다. 
Competency-based medical education (CBME) has become the standard paradigm underlying postgraduate training in medical education.1 In the United States, among Accreditation Council for Graduate Medical Education (ACGME)-accredited programs, the current evaluation framework consists of 6 core competency domains, each with a progressive 5-level scale to track the trajectory of trainees. ACGME allows each specialty to define the specific pathways of progression from novice to independent (termed “Milestones”).2 As technology and medical practice progress, these pathways are periodically reassessed to ensure that they keep pace with ongoing developments and the evolving standard of training.

담당 의사는 수련의를 위한 성과 피드백을 생성하며, 이 피드백은 개인의 진전도를 평가하는 데 중추적인 역할을 합니다.3 CBME 프레임워크에서 이러한 피드백은 정량적 평가에만 국한되어서는 안 됩니다.4 마일스톤을 만든 이유 중 하나는 수치화된 평가 척도에서 벗어나기 위해서였으며,5 의학교육의 일부 리더들은 서술적 피드백이 수치화된 평가를 완전히 대체해야 한다고 주장합니다.6 ACGME에 따르면, 각 프로그램은 "후원 기관의 전반적인 사명, 해당 기관이 봉사하고 졸업생이 봉사하게 될 지역사회의 요구사항, 졸업하고자 하는 의사의 고유한 역량"을 고려해야 합니다.7 내러티브 피드백에서 프로그램이 사용하는 용어와 문구는 피드백 문화, 기관 관행, 환자 집단 등의 차이로 인해 매우 다양할 수 있습니다. 따라서 내러티브 피드백을 분석하는 데 사용되는 기법을 포함한 평가 방법은 각 프로그램에 맞게 조정되어야 합니다.
Staff physicians generate performance feedback for trainees that is pivotal in assessing their individual progress.3 In the CBME framework, this feedback should not be limited to quantitative ratings.4 One of the stated reasons for creating Milestones was to move away from numerical rating scales,5 and some leaders in medical education argue that narrative feedback should replace numerical ratings entirely.6 According to the ACGME, each program must consider the “overall mission of its Sponsoring Institution, the needs of the community it serves and that its graduates will serve, and the distinctive capabilities of physicians it intends to graduate.”7 The terms and phrases that programs use in narrative feedback can vary widely due to differences in feedback culture, institution practices, patient populations, and more. Therefore, evaluation methods, including techniques used to analyze narrative feedback, must be tailored to each program.

자연어 처리(NLP) 또는 컴퓨터 언어학은 컴퓨터 처리 방법을 사용하여 구어 또는 문어를 분석하여 인간의 능력에 근사치를 구하는 작업을 수행합니다.8 CBME 영역 내에서 NLP를 평가한 연구는 거의 없으며, 이러한 연구의 대부분은 연수생의 임상 기록을 검토하거나 특정 임상 경험에 대한 연수생의 노출을 분석하는 데 중점을 두었습니다.9 그러나 일부 연구에서는 NLP가 ACGME 역량과 관련된 언어를 식별할 수 있는지 여부를 조사했습니다. 

  • Neves 등은 NLP를 사용하여 수련의 평가에 대한 서술적 코멘트에 높은 품질과 높은 유용성 등 특정 피드백 특성이 포함되어 있는지 예측하고 전문성 또는 의사소통에 대한 내용이 포함되어 있는지 조사했습니다.10
  • Zhang 등은 NLP를 사용하여 수련의 평가의 정서를 조사하고 코멘트에 반영된 ACGME 핵심 역량을 예측하는 모델을 개발했습니다.11 Zhang 등은 향후 연구에서는 더 큰 데이터 세트를 사용하여 하위 역량을 식별하는 이러한 모델의 타당성에 초점을 맞추는 것이 좋다고 제안했습니다.

Natural language processing (NLP), or computational linguistics, uses computer processing methods to analyze spoken or written language to perform tasks that seek to approximate human abilities.8 Few published studies have evaluated NLP within the realm of CBME, and most of these studies focused on reviewing trainees’ clinical notes or analyzing trainees’ exposure to certain clinical experiences.9 But a small subset explored whether NLP can identify language related to ACGME competencies.

  • Neves et al used NLP to predict whether narrative comments on trainee evaluations contained specific feedback traits, including high quality and high utility, and explored whether they included content about professionalism or communication.10 
  • Zhang et al used NLP to investigate sentiment of trainee evaluations and developed models to predict ACGME Core Competencies reflected by comments.11 Zhang et al suggested that future work should use larger data sets to focus on feasibility of such models to identify subcompetencies.

NLP를 사용하여 ACGME 전문 분야에 특화된 피드백 언어를 학습하는 모델을 만든 다음, 특정 프로그램에 대한 서술 코멘트의 ACGME 하위 역량 내용을 예측하기 위해 고유한 기관에 맞게 미세 조정할 수 있습니다. 이러한 모델은 각 프로그램이 졸업하고자 하는 고유한 역량에 대한 교육생의 진전 궤적을 추적하는 각 프로그램의 기능을 강화할 수 있습니다. NLP 기술은 대량의 텍스트를 매우 빠르게 처리할 수 있도록 확장할 수 있으며, 다양한 시스템에 통합하여 교육생 평가의 프로그래밍 관리와 평가 언어에 대한 교수진의 피드백을 향상시킬 수 있습니다. 또한 자연어 처리 기술은 수련의 성별과 관련된 암묵적 편견과 같은 평가의 편견을 밝혀내어 의학교육의 특정 격차를 완화할 수 있는 잠재력을 가지고 있습니다.12 
NLP could be used to create models that learn the feedback language specific to ACGME specialties, which then could be fine-tuned to unique institutions to predict ACGME subcompetency content in narrative comments for specific programs. Such models could augment each program’s ability to track the trajectory of trainees’ progress toward the unique capabilities that the program intends to graduate. NLP techniques can scale to process volumes of text very rapidly and could be integrated into a variety of systems to enhance programmatic management of trainee evaluations and feedback to faculty on their evaluation language. NLP also holds promise to reveal biases in evaluation, such as implicit bias related to trainee gender, which could mitigate certain disparities in medical education.12

이 연구의 목적은 마취과 수련의에 대한 평가의 서술적 피드백을 적절한 ACGME 마일스톤 2.0 하위 역량 범주로 자동 분류하는 NLP 모델을 개발하고 평가하는 것이었습니다.
The aim of this study was to develop and evaluate an NLP model to automatically categorize narrative feedback from evaluations of anesthesiology trainees into the appropriate ACGME Milestone 2.0 subcompetency categories.

방법
Method

투명성 및 개방성 증진
Transparency and openness promotion

이 조사는 개인 예후 또는 진단을 위한 다변량 예측 모델의 투명한 보고13 가이드라인과 생의학 연구에서의 기계 학습 예측 모델 개발 및 보고 가이드라인을 따릅니다.14 NLP 모델 구축을 위한 소스 코드는 공개적으로 사용 가능하며, 최종 모델은 공개적으로 사용할 수 있습니다. 
This investigation follows the Transparent Reporting of a Multivariable Prediction Model for Individual Prognosis or Diagnosis13 guidelines and the Guidelines for Developing and Reporting Machine Learning Predictive Models in Biomedical Research.14 Source code for building the NLP model is publicly available (https://github.com/gb563/Milestone_NLP), and the final model is available for public use (http://nb-group.org/milestone/).

연구 설계 및 데이터 출처
Study design and data source

이 프로토콜은 해군 의료센터 포츠머스 기관생명윤리심의위원회의 승인을 받았습니다. 이 연구는 2개의 완전한 학년도(2019년 7월 1일 ~ 2021년 6월 30일, 적격 기록 수 = 6,494건) 동안 4개의 군 의학전문대학원(GME) 프로그램에서 마취과 수련의의 임상 성과 평가를 포괄하는 후향적 코호트 연구였습니다. 모든 사이트는 동일한 핵심 프로그램 목표를 공유합니다. 모든 데이터는 전자 평가 시스템에서 추출했습니다. 성과 평가에는 일일 임상 과제에서 직접 관찰한 피드백이 포함되었습니다. 평가 구조와 질문은 사이트마다 다르지만, 모두 서술적 설명을 위한 필드가 포함되어 있습니다. 2곳의 사이트에는 전반적인 의견을 위한 일반 항목이, 1곳에는 긍정적인 관찰과 건설적인 관찰 항목이, 1곳에는 전반적인 의견, 긍정적인 관찰, 건설적인 관찰 항목이 포함되었습니다. 코멘트이 없는 평가는 모두 제외했으며, 나머지 평가의 각 분야별 코멘트을 합산했습니다. 총 5,935개의 평가가 연구에 포함되었습니다. 이러한 평가는 25,714개의 개별 문장으로 분리되었으며, 169명의 수련의에 대한 207명의 마취과 전문의의 피드백을 나타냅니다. 문장 분리 및 추가 텍스트 전처리는 자연어 툴킷 3.6.15를 사용하여 수행되었습니다. 
This protocol was approved by the Naval Medical Center Portsmouth Institutional Review Board. This was a retrospective cohort study encompassing clinical performance evaluations of anesthesiology trainees from 4 military graduate medical education (GME) programs during 2 complete academic years (July 1, 2019, to June 30, 2021, n = 6,494 eligible records). All sites share the same core set of program aims. All data were extracted from electronic evaluation systems. Performance evaluations included direct observation feedback from daily clinical assignments. Evaluation structure and questions are site-specific, but all include fields for narrative comments. Two sites included a generic field for overall comments, 1 site included a field for positive observations and another for constructive observations, and 1 site include fields for overall comments, positive observations, and constructive observations. All evaluations without comments were excluded, and comments from each field in the remaining evaluations were pooled. A total of 5,935 evaluations were included in the study. These evaluations were separated into 25,714 individual sentences and represented feedback from 207 staff anesthesiologists on 169 trainees. Sentence separation and further textual preprocessing were performed using Natural Language Toolkit 3.6.15

이 조사는 Zhang 등11의 연구를 기반으로 하며 몇 가지 주요 차이점이 있습니다. 이 조사는 단일 기관의 699개 코멘트을 사용하여 ACGME 핵심 역량을 예측하도록 설계된 모델에 대한 내부 검증 메트릭을 보고했으며, 유사한 모델이 하위 역량을 예측할 수 있는지 평가하기 위해 더 큰 데이터 세트를 탐색해야 한다고 가정했습니다. 이 조사에는 여러 기관의 수천 개의 코멘트이 포함되었으며, 사전에 훈련 코호트검증 코호트로 나뉘어 하위 역량 예측을 수행했습니다. 
This investigation builds on the work of Zhang et al11 and has several key differences. Their investigation reported internal validation metrics on models designed to predict ACGME Core Competencies using 699 comments from a single institution and posited that a larger data set should be explored to assess whether similar models can predict subcompetencies. This investigation included several thousand comments from multiple institutions, split a priori into training and validation cohorts, and we performed subcompetency prediction.

본 조사는 의료 분야에서 책임감 있는 인공 지능을 목표로 하는 몇 가지 최신 전략을 사용하여 설계 및 구현되었습니다.16 여기에는 다음이 포함됩니다.

  • 엄격한 설계 및 보고 기준13,14;
  • 모델 개발에 사용된 코드 공개;
  • 모델을 웹 기반 애플리케이션으로 구축하여 공개적으로 탐색, 테스트 및 검증할 수 있도록 하는 노력.
  • 모델이 예측하는 방식을 더 잘 이해할 수 있도록 해석 가능성 그래픽을 포함시켰습니다.

또한, 하위 단어 또는 단어의 작은 부분을 잡아내어 학습 중에 한 번도 접해보지 못한 단어에 대해서도 예측할 수 있습니다. 이 기능은 비표준 단어, 약어, 철자 오류가 자주 발생하는 의료 분야에서 특히 중요합니다.
Our investigation was designed and implemented using several contemporary strategies aimed at responsible artificial intelligence in health care,16 including

  • rigorous design and reporting criteria13,14;
  • publishing the code that was used to develop models;
  • an effort to build the model into a web-based application to allow public exploration, testing, and validation; and
  • the inclusion of interpretability graphics to better understand how the model makes predictions.

Additionally, our model captures subwords, or smaller sections of words, so that it can make predictions on words that it never encountered during training. This feature is particularly important in health care where nonstandard words, abbreviations, and spelling errors are frequently encountered.

결과
Outcomes

이 조사의 1차 평가지표는 마취과 수련의의 성과 평가에 대한 서술적 코멘트를 통해 마일스톤 2.0 하위 역량을 예측하는 알고리즘의 개발 및 평가였습니다. 2차 결과에는 각 평가 문장에 하위 역량 레이블을 할당하는 전문가들의 평가자 간 합의와 전문가들이 동의하지 않을 때 가장 빈번하게 발생하는 하위 역량 레이블에 대한 탐색적 분석이 포함되었습니다.
The primary endpoint of this investigation was the development and evaluation of an algorithm to predict Milestone 2.0 subcompetencies from narrative comments on anesthesiology trainee performance evaluations. Secondary outcomes included inter-rater agreement for experts assigning subcompetency labels to each evaluation sentence and exploratory analysis of the most frequent co-occurring subcompetency labels when experts disagreed.

마일스톤 2.0 하위 역량을 평가에 할당하기
Assigning Milestone 2.0 subcompetencies to evaluations

4개의 군사 대학원 교육 프로그램에서 10명의 마취과 전문의가 모집되었습니다. 이 그룹에는 프로그램 책임자 2명과 현재 또는 과거의 부프로그램 책임자 8명이 포함되었습니다. 편향성을 완화하기 위해 난수 생성기를 사용하여 각 평가에 2명의 검토자를 배정했습니다. 어떤 리뷰어에게도 동일한 평가 세트를 배정하지 않았으며, 각 평가는 정확히 두 번 배정되었습니다. 모든 평가는 비식별화되었습니다. 비식별화를 달성하기 위해 평가 모음에 나타나는 모든 단어의 사전을 수동으로 검토했습니다. 이 사전 내의 모든 교육생 및 교수진 이름을 컴파일한 다음 프로그래밍 방식으로 평가에서 제거했습니다. 각 검토자에게는 개별 문장으로 분할된 평가 목록이 제공되었습니다. 그런 다음 검토자는 할당된 문장에 내용을 가장 잘 파악한 23개의 마취과 마일스톤 2.0 하위 역량 중 하나를 선택하거나, 인구통계학적 특성(예: "이것은 당직 근무에 대한 평가입니다.") 또는 유용하지 않은 경우(예: "잘했음.")를 나타내는 D 또는 N으로 레이블을 지정했습니다. 각 문장에는 고유 식별 번호가 부여되어 검토자가 어떤 문장이 같은 평가에서 나온 것인지 알 수 있도록 했습니다. 여러 문장이 있는 평가의 경우, 식별 번호는 검토자가 코멘트의 맥락을 이해하는 데 도움이 되었습니다.
Ten academic anesthesiologists from 4 military graduate education programs were recruited. This group included 2 program directors and 8 current or past associate program directors. To mitigate bias, a random number generator was used to assign 2 reviewers to each evaluation. No reviewers were assigned the same set of evaluations, and each evaluation was assigned exactly twice. All evaluations were deidentified. To achieve deidentification, a dictionary of all words that appear in the collection of evaluations was manually reviewed. All trainee and faculty names within this dictionary were compiled and then programmatically removed from evaluations. Each reviewer was provided a list of evaluations that were split into individual sentences. The reviewers then labeled their assigned sentences with 1 of the 23 anesthesiology Milestone 2.0 subcompetencies that best captured content, or D for demographics (e.g., “This is an evaluation for a call shift.”), or N for not useful (e.g., “Good job.”). Each sentence was given a unique identifying number so that reviewers knew which sentences came from the same evaluation. For evaluations that had multiple sentences, the identifier helped reviewers understand the context of comments.

연구 책임자(G.J.B.)는 각 검토자를 만나 마일스톤 2.0 하위 역량에 대해 논의하고, 지침을 설명하고, 검토자가 과제를 수행할 능력이 있다고 느낄 때까지 몇 가지 예제를 검토했습니다. 각 검토자에게는 ACGME 마취학 마일스톤 2.0 가이드가 제공되었지만, 결과 평가의 편견을 최소화하기 위해 명시적인 채점 규칙은 제공되지 않았습니다. 
The principal investigator (G.J.B.) met with each reviewer to discuss the Milestone 2.0 subcompetencies, explain the instructions, and go through several examples until reviewers felt competent performing the task. Each reviewer was given the ACGME Anesthesiology Milestone 2.0 guide, but no explicit scoring rules were provided to minimize bias in outcome assessment.

훈련 및 검증 코호트
Training and validation cohorts

사전 훈련 코호트는 사이트 1, 2, 4의 풀링된 데이터로, 외부 검증 코호트는 사이트 3으로 결정했습니다. 훈련 코호트는 두 리뷰어가 라벨에 동의한 문장으로 제한되었습니다(n = 11,960). 이러한 문장은 각 평가 내에서 동일한 레이블이 할당된 문장을 결합하는 방식으로 추가 처리되었습니다. 이 단계를 거친 최종 훈련 코호트에는 10,218개의 코멘트이 포함되었습니다(그림 1). 따라서 각 코멘트은 단일 문장 또는 단일 평가에서 나온 동일한 레이블을 가진 문장들의 조합을 나타냅니다. 이 단계는 다양한 구두점 사용을 제어하기 위해 수행되었습니다.

  • 어떤 평가자는 연속된 문장을 작성했고,
  • 어떤 평가자는 세미콜론으로 아이디어를 구분했으며,
  • 어떤 평가자는 대시 또는 아무것도 사용하지 않았습니다.

A priori, it was decided that the training cohort was the pooled data from Sites 1, 2, and 4, and the external validation cohort was Site 3. The training cohort was limited to sentences for which both reviewers agreed on the label (n = 11,960). These sentences were further processed by combining sentences within each evaluation that were assigned the same label. After this step, the final training cohort contained 10,218 comments (Figure 1). Therefore, each comment represents either a single sentence or a combination of sentences that come from a single evaluation and have the same label. This step was performed to help control for variable use of punctuation.

  • Some evaluators wrote run-on sentences,
  • others separated ideas with semicolons, and
  • others with dashes or nothing at all.

알고리즘은 평가를 문장으로 분할한 다음 분류하도록 설계되었기 때문에 이 처리 단계는 학습 데이터가 단순히 [별개의 문장]이 아닌 평가에서 [별개의 주제]를 포착하도록 하는 데 도움이 되었습니다. 검증 코호트의 경우, 연구 책임자가 라벨 할당에 대한 의견 불일치 시 동점자를 결정하는 역할을 했습니다. 최종 검증 코호트에는 2,255개의 문장이 포함되었습니다. 이 접근 방식에는 두 가지 장점이 있었습니다.

  • 첫째, 평가자 간 합의가 완벽한 훈련 데이터를 선택함으로써 숙련된 분류자를 구축하는 데 도움이 되었습니다.
  • 둘째, 검증 사이트의 모든 평가를 포함함으로써 프로그램의 모든 피드백을 대표하는 보이지 않는 데이터에 대한 분류기의 일반화 가능성을 평가하여 편향을 줄이는 데 도움이 되었습니다. 

Since the algorithm was designed to split evaluations into sentences and then classify them, this processing step helped ensure that training data captured distinct topics in evaluations rather than simply distinct sentences. For the validation cohort, the principal investigator served as the tie-breaker for any disagreements in label assignments. The final validation cohort contained 2,255 sentences. This approach had 2 advantages.

  • First, selecting training data with perfect inter-rater agreement helped build a skilled classifier.
  • Second, including all evaluations from the validation site helped reduce bias by assessing the generalizability of the classifier to unseen data representative of all feedback from a program.

 

예측자
Predictors

모델 입력은 내러티브 성능 평가의 원시 텍스트만 사용했습니다. 모든 텍스트는 소문자로 변환되었고 알파벳이 아닌 문자는 제거되었습니다. 텍스트 분류기에 거의 가치를 부여하지 않는 일반적인 단어, 즉 중단어(예: of, to, in)는 제거되었습니다. 하위 역량 레이블은 상당한 클래스 불균형을 보여주었습니다(표 1). 환자 관리(PC)6, PC9, 전문성(P)3, 대인 커뮤니케이션 및 기술(ICS)3, 실습 기반 학습 및 개선(PBLI)1, 시스템 기반 실습(SBP)1은 각각 전체 레이블의 1% 미만으로 포함되었습니다. 따라서 NLP 모델에 각 범주에서 적절한 샘플을 제공하기 위해 특정 하위 역량을 결합했습니다. PC6(현장 초음파)는 PC2(수술 전후 치료 및 관리)로, PC9(중환자 치료)는 PC7(상황 인식 및 위기 관리)로 재할당되었습니다. P, ICS, PBLI 및 SBP 내의 모든 하위 역량은 통합되었습니다. 따라서 분류 알고리즘이 예측할 수 있는 최종 카테고리는 16개였습니다: PC1, PC2, PC3, PC4, PC5, PC7, PC8, PC10, MK1, MK2, P, ICS, PBLI, SBP, D, N(표 2). 이러한 처리 기법은 훈련 및 검증 코호트에도 동일한 방식으로 적용되었습니다.

The only model inputs were raw text from narrative performance evaluations. All text was converted to lowercase, and nonalphabetic characters were removed. Common words that add little value to text classifiers, termed stopwords (e.g., of, to, in), were removed. Subcompetency labels demonstrated substantial class imbalance (Table 1). Patient care (PC)6, PC9, professionalism (P)3, interpersonal communication and skills (ICS)3, practice-based learning and improvement (PBLI)1, and systems-based practice (SBP)1 each contained < 1% of total labels. Therefore, to provide the NLP model with adequate samples in each category, certain subcompetencies were combined. PC6 (point of care ultrasound) was reassigned to PC2 (perioperative care and management) and PC9 (critical care) was reassigned to PC7 (situational awareness and crisis management). All subcompetencies within P, ICS, PBLI, and SBP were combined. Therefore, there were 16 final categories for the classification algorithm to predict: PC1, PC2, PC3, PC4, PC5, PC7, PC8, PC10, MK1, MK2, P, ICS, PBLI, SBP, D, N (Table 2). These processing techniques were applied in the same manner for the training and validation cohorts.

 

통계 분석
Statistical analysis

모델 개발.
Model development.

NLP 모델에는 FastText가 사용되었습니다. FastText는 맞춤법 오류와 의학 전문 용어 등 데이터 세트에 내재된 몇 가지 한계를 극복하는 오픈 소스 머신 러닝 알고리즘17입니다. FastText는 하위 단어(예: "intubate"의 하위 단어에는 "int", "ntu", "tub", "uba", "bat", "ate"가 포함될 수 있음)에 집중함으로써 이러한 영역에서 탁월한 성능을 발휘하며, 표준 영어 사전에 나타나지 않을 수 있는 철자와 단어 또는 약어에 대한 내성을 갖추고 있습니다. 또한 하위 단어 분석을 통해 분류기가 학습 과정에서 한 번도 보지 못한 단어가 포함된 문장을 예측할 수 있습니다. 마지막으로, FastText는 불균형한 클래스가 포함된 대규모 데이터 세트를 쉽게 처리할 수 있습니다. 
FastText was used for the NLP model. FastText is an open-source machine learning algorithm17 that overcomes several limitations inherent to the dataset, including spelling errors and medical jargon. FastText excels in these areas by focusing on subwords (e.g., subwords for “intubate” could include “int,” “ntu,” “tub,” “uba,” “bat,” “ate”), making it resistant to misspellings and words or abbreviations that may not appear in a standard English dictionary. Subword analysis also allows predictions on sentences containing words that the classifier has never seen in the training process. Finally, FastText is facile dealing with large datasets that contain imbalanced classes.

머신러닝에서 하이퍼파라미터는 학습 과정을 안내하는 알고리즘 피쳐입니다. 조사된 구체적인 하이퍼파라미터는 다음이 있다. 

  • 학습 속도(각 훈련 주기 동안 모델이 스스로 업데이트하는 정도),
  • 에포크(훈련 주기 수),
  • 하위 단어의 길이(고려해야 할 연속 글자 수),
  • 단어 n-그램(고려해야 할 연속 단어 수)

매크로 F1 점수를 최적화하는 하이퍼파라미터 세트를 결정하기 위해 5배 교차 검증을 사용하여 그리드 검색을 수행했습니다. F1 점수는 민감도와 양의 예측값의 균형을 추구하며, 매크로 F1은 각 클래스의 관찰 수를 고려하지 않고 각 클래스에 대한 F1 점수의 평균을 산출합니다. 이 접근 방식은 클래스 불균형의 영향을 완화하고 16개 카테고리를 모두 동등하게 중요하게 취급합니다.
In machine learning, hyperparameters are algorithm features that guide the learning process. Specific hyperparameters investigated were

  • learning rate (the extent to which the model updates itself during each training cycle),
  • epochs (number of training cycles),
  • length of subwords (number of consecutive letters to consider), and
  • word n-grams (number of consecutive words to consider).

Grid search was performed using fivefold cross-validation to determine the set of hyperparameters that optimized the macro F1 score. F1 score seeks to balance sensitivity and positive predictive value, and macro F1 averages F1 scores for each class without considering the number of observations in each class. This approach mitigates the impact of class imbalance and treats all 16 categories as equally important.

교차 검증을 사용하면 전체 훈련 세트가 80%:20% 코호트 또는 폴드로 분할됩니다. 모델은 80% 폴드에서 일련의 하이퍼파라미터를 사용하여 훈련되고, 훈련된 모델의 성능은 20% 폴드에서 평가됩니다. 다른 폴드를 사용하여 동일한 하이퍼파라미터로 이 과정을 4회 반복하여 주어진 하이퍼파라미터 조합에 대한 전체 학습 데이터 세트의 평균 매크로 F1 점수를 계산합니다. 이 과정은 가능한 모든 하이퍼파라미터 조합에 대해 반복됩니다. 그런 다음 가장 우수한 성능을 보인 하이퍼파라미터를 전체 훈련 데이터 세트를 사용하여 최종 모델을 훈련하는 데 사용했습니다.
With cross-validation, the entire training set is split into 80%:20% cohorts, or folds. The model is trained using a set of hyperparameters on the 80% fold, and performance of that trained model is assessed on the 20% fold. This repeats 4 times with the same hyperparameters using the other folds to calculate average macro F1 score across the entire training dataset for a given combination of hyperparameters. This process is repeated for all possible combinations of hyperparameters. The hyperparameters that yielded the best performance were then used to train the final model using the entire training dataset.

성능 메트릭 및 모델 검증.
Performance metrics and model validation.

최종 모델 파라미터가 결정되면 각 클래스에 대한 민감도, 양의 예측값, F1 및 수신기 작동 특성 곡선 아래 면적(AUC)을 계산하여 내부 검증을 평가했습니다. AUC는 한 클래스와 다른 모든 클래스를 고려하여 계산했습니다. 예를 들어, PC1에 대한 AUC는 코멘트을 PC1에 속하는 것과 다른 카테고리에 속하는 것을 구별하는 알고리즘의 능력을 측정합니다. 외부 검증은 검증 코호트에 대한 AUC를 계산하여 평가했습니다. 성능 지표의 평균 및 95% 신뢰 구간은 500개의 샘플로 계층화된 부트스트래핑을 통해 결정되었습니다. 계층화된 부트스트래핑은 각 훈련 샘플에 동일한 비율의 하위 역량이 포함되도록 하여 클래스 불균형을 고려합니다. 훈련 및 검증 코호트에서 보정 곡선을 통해 예측 신뢰도를 평가했습니다. 각 범주에 대한 관찰 확률과 모델 예측 확률을 비교하여 각 범주에 대한 보정 곡선을 생성했습니다. 잘 보정된 모델은 관찰된 확률과 매우 유사한 예측 확률을 보여야 합니다. 예를 들어, 모델이 어떤 코멘트을 70%의 확률로 PC1이라고 예측했다면, 리뷰어에 따르면 유사한 코멘트 10개 중 7개가 실제로는 PC1이어야 합니다.
Once the final model parameters were determined, internal validation was assessed by calculating sensitivity, positive predictive value, F1, and area under the receiver operating characteristic curve (AUC) for each class. AUC was calculated by considering one class vs all others. For example, AUC for PC1 measures the algorithm’s ability to discriminate a comment as belonging to PC1 vs any other category. External validation was assessed by calculating AUC on the validation cohort. Mean and 95% confidence intervals for performance metrics were determined by stratified bootstrapping with 500 samples. Stratified bootstrapping ensures that each training sample contains the same relative proportions of subcompetencies to account for class imbalance. Prediction reliability was assessed in the training and validation cohorts through calibration curves. Calibration curves were generated for each category by plotting observed probabilities vs model prediction probabilities for each class. A well-calibrated model should demonstrate prediction probabilities that are very similar to observed probabilities. For example, if the model predicts that a comment is PC1 with 70% probability, then 7 out of 10 similar comments should in fact be PC1 according to reviewers.

2차 결과 및 민감도 분석.
Secondary outcomes and sensitivity analyses.

리뷰어의 레이블은 개수와 빈도로 설명했습니다. 레이블은 순위나 순서가 없는 명목 변수로 간주했습니다. 모든 하위 역량 라벨에 대한 전반적인 평가자 간 일치도는 Fleiss의 카파를 사용하여 평가했습니다. 각 하위 역량에 대해 범주별 일치율을 계산하여 특정 하위 역량에 따라 평가자 간 동의가 달라지는지 조사했습니다. 리뷰어들이 동의하지 않을 때 가장 빈번하게 함께 나타나는 하위 역량 레이블을 조사하여 하위 역량에 구분하기 어려운 주제가 있는지 조사했습니다. 두 가지 민감도 분석이 수행되었습니다.

  • 첫째, 6개의 핵심 역량으로 묶은 후 평가자가 지정한 라벨에 대한 평가자 간 합의도를 조사했습니다.
  • 둘째, 외부 검증 데이터에 대한 성능 지표를 평가자 간 완벽한 합의가 있는 문장 하위 집합에 대해 계산하여 평가자 간 완벽한 합의가 포함된 학습 데이터로 개발되었기 때문에 동점자 불일치 또는 모델 예측에 편향이 있을 수 있는지 평가했습니다.

분석은 Python 3.8(델라웨어주 윌밍턴) 및 R 4.1(매사추세츠주 보스턴)을 사용하여 수행되었습니다.
Reviewers’ labels were described with counts and frequencies. Labels were considered nominal variables with no ranking or ordering. Overall inter-rater agreement for all subcompetency labels was assessed with Fleiss’ Kappa. Category-specific percent agreement was calculated for each subcompetency to investigate whether specific subcompetencies varied in reviewer agreement. The most frequent co-occurring subcompetency labels when reviewers disagreed were explored to investigate whether there were themes in subcompetencies that were difficult to differentiate. Two sensitivity analyses were performed.

  • First, inter-rater agreement on reviewer-assigned labels was explored after pooling them into the 6 Core Competencies.
  • Second, performance metrics on external validation data were calculated for the subset of sentences for which there was perfect inter-rater agreement to assess whether there could be bias in the process of tie-breaking disagreements or bias in model predictions since it was developed with training data that contained perfect inter-rater agreement.

Analyses were performed with Python 3.8 (Wilmington, Delaware) and R 4.1 (Boston, Massachusetts).

모델 해석 가능성.
Model interpretability.

대부분의 머신 러닝 모델과 마찬가지로 NLP 모델은 블랙박스 동작으로 인해 어려움을 겪습니다. 로지스틱 회귀의 확률 비율과 같이 쉽게 해석할 수 있는 계수가 없습니다. 최종 모델이 어떻게 예측을 했는지 이해하기 위해 로컬 해석 가능한 모델-무관용 설명(LIME)18을 사용했습니다. LIME은 다양한 머신러닝 알고리즘에 사용되어 예측을 할 때 어떤 특징이 중요한지 설명하는 데 도움이 됩니다. 3개의 예시 문장에 대해 모델 예측과 LIME 시각화를 수행했으며, 이 기능은 온라인 애플리케이션에 내장되었습니다.
NLP models, like most machine learning models, suffer from black-box behavior. There are no coefficients, such as odds ratios in logistic regression, that are easily interpretable. Local Interpretable Model-Agnostic Explanations (LIME)18 was used to understand how the final model made predictions. LIME can be used for a wide range of machine learning algorithms to help demonstrate what features are important while making predictions. Model predictions and LIME visualizations were performed for 3 example sentences, and this feature was built into the online application.

모델 구현.
Model implementation.

이 모델은 교육생 자가 평가 연습의 일부로 작동되었습니다. 모델을 구현하는 간단한 컴퓨터 프로그램을 만든 다음 모델 출력이 포함된 스프레드시트를 생성했습니다. 18명의 교육생에 대한 모든 평가는 연구 기간 후 6개월 동안 수집되었습니다. 이 모델은 각 교육생의 내러티브 코멘트를 읽고 개별 문장으로 분리한 다음 각각에 해당 카테고리에 레이블을 지정했습니다. 이렇게 분류된 코멘트는 각 교육생이 멘토와 함께 검토할 수 있도록 별도의 스프레드시트에 자동으로 정리되어 스스로 결정한 하위 역량에 대한 성과를 개선하기 위한 실행 계획을 안내하는 데 도움이 되었습니다.
The model was operationalized as part of a trainee self-assessment exercise. A simple computer program was created which implements the model and then produces spreadsheets containing model outputs. All evaluations on 18 trainees were collected from a 6-month window after the study period. The model read each trainee’s narrative comments, separated them into individual sentences, and labeled each with the corresponding category. These labeled comments were then automatically organized into separate spreadsheets for each trainee for review with their mentors to help guide action plans to improve performance in self-determined subcompetencies.

결과
Results

평가 라벨
Evaluation labels

전체 데이터 세트에 대한 검토자 라벨(두 검토자가 라벨에 동의하지 않은 의견을 제외하기 전)에는 PC 35.2%, MK 8.8%, P 10.1%, ICS 5.9%, PBLI 5.1%, SBP 3.8%, D 10.1%, N 21.0%가 포함되었으며, 훈련 코호트최종 검증 코호트 간에 검토자가 지정한 라벨의 분포에 유의미한 차이가 있었습니다(P < .01). 가장 큰 차이는 P와 N에서 관찰되었습니다(표 1).
Reviewer labels on the complete dataset (before excluding comments where both reviewers did not agree on the labels) included 35.2% PC, 8.8% MK, 10.1% P, 5.9% ICS, 5.1% PBLI, 3.8% SBP, 10.1% D, and 21.0% N. There were significant differences in the distribution of reviewer-assigned labels between the training and final validation cohorts (P < .01). The biggest differences were observed for P and N (Table 1).

평가자 간 합의
Inter-rater agreement

검토자들은 25,714개 문장 중 13,068개 문장에 대해 25개 레이블(23개 하위 역량, D, N)에 동의했습니다(전체 동의율 50.8%). PC10(말초 및 신경계 마취, 72.0%)이 가장 높은 동의율을 보였고 SBP3(의료 시스템에서의 의사의 역할, 9.9%)가 가장 낮은 동의율을 보였습니다(보충 디지털 부록 1). Fleiss의 카파는 25개 범주에 대해 0.44였습니다. 하위 역량을 핵심 역량으로 합산했을 때 Fleiss의 카파는 0.50이었습니다. 전문가들이 하위 역량에 대해 의견이 일치하지 않을 때 가장 빈번하게 동반되는 하위 역량 레이블은 PC4(수술 중 관리)였습니다. ICS와 SBP 간, 그리고 P와 N 간에는 빈번하게 공존하는 경향이 있었습니다(보충 디지털 부록 1).
Reviewers agreed on the 25 labels (23 subcompetencies, D, and N) for 13,068 of 25,714 sentences (overall agreement 50.8%). Percent agreement was highest for PC10 (peripheral and neuraxial anesthesia, 72.0%) and lowest for SBP3 (physician role in health care systems, 9.9%) (Supplemental Digital Appendix 1 at https://links.lww.com/ACADMED/B361). Fleiss’ Kappa was 0.44 for the 25 categories. When subcompetencies were aggregated into the core competencies, Fleiss’ Kappa was 0.50. The most frequent co-occurring subcompetency label when experts disagreed on any subcompetency was PC4 (intraoperative care). There tended to be frequent co-occurrence between ICS and SBP and between P and N (Supplemental Digital Appendix 1 at https://links.lww.com/ACADMED/B361).

모델 개발
Model development

432개의 하이퍼파라미터 조합(4개의 개별 학습률과 에포크, 3개의 단어 길이, 9개의 개별 하위 단어 길이)에 대해 그리드 검색을 수행했습니다. 모델 구축에 대한 자세한 내용은 여기에서 확인할 수 있습니다. 최종 모델의 하이퍼파라미터에는 2개의 단어 n-그램과 1~4자 길이의 하위 단어가 포함되었습니다. 
Grid search was performed on 432 combinations of hyperparameters (4 separate learning rates and epochs, 3 lengths for word n-grams, and 9 separate subword lengths). Details for building the model are available at https://github.com/gb563/Milestone_NLP. Hyperparameters for the final model included 2 word n-grams and subwords that were 1–4 characters long.

모델 성능
Model performance

내부 검증 결과, PC5, PC10, MK1, D, N은 우수한 판별력(AUC ≥ 0.8)을 보였고, PC1, PC3, PC4, P, ICS, PBLI는 양호한 판별력(AUC ≥ 0.7)을 보였습니다. SBP는 변별력이 가장 낮았습니다(AUC 0.59)(표 2). 양성 예측값은 PC5와 PC10(0.88)이 가장 높았고, SBP(0.31)가 가장 낮았습니다. 민감도는 N(0.89)이 가장 높았고 SBP(0.19)가 가장 낮았습니다. 훈련 및 검증 코호트에 대한 양성 예측값, 민감도, F1에 대한 전체 보고는 보충 디지털 부록 2와 3에서 확인할 수 있습니다. 보정 곡선은 예측 확률 범위에서 신뢰할 수 있는 예측을 보여주었습니다(추가 디지털 부록 4).
On internal validation, excellent discrimination (AUC ≥ 0.8) was observed for PC5, PC10, MK1, D, and N. Good discrimination (AUC ≥ 0.7) was observed for PC1, PC3, PC4, P, ICS, and PBLI. SBP had the worst discrimination (AUC 0.59) (Table 2). Positive predictive value was highest for PC5 and PC10 (0.88), and lowest for SBP (0.31). Sensitivity was highest for N (0.89) and lowest for SBP (0.19). Full reporting of positive predictive value, sensitivity, and F1 for training and validation cohorts can be found in Supplemental Digital Appendices 2 and 3 at https://links.lww.com/ACADMED/B361. Calibration curves demonstrated reliable predictions across the range of predicted probabilities (Supplemental Digital Appendix 4 at https://links.lww.com/ACADMED/B361).

모델 평가
Model evaluation

모델 변별력은 검증 데이터에서 매우 일관적이었습니다. 내부 검증 데이터와 비교했을 때 Professionalism을 제외하고는 성능에 차이가 없었습니다(표 2). 검토자의 완전한 동의를 얻은 문장만 검증 데이터에 포함시킨 민감도 분석에서도 내부 검증과 비교했을 때 전문성에 대한 AUC는 더 이상 차이가 없었습니다(보충 디지털 부록 5). 검증 데이터에 대한 모델 보정 결과, 모든 범주에서 예측이 과신하는 경향이 나타났습니다(보충 디지털 부록 6).
Model discrimination was very consistent on the validation data. There were no differences in performance compared with the internal validation data except professionalism (Table 2). On the sensitivity analysis in which only sentences with complete reviewer agreement were included in the validation data, there was no longer a difference in AUC for professionalism compared with internal validation (Supplemental Digital Appendix 5 at https://links.lww.com/ACADMED/B361). Model calibration on validation data demonstrated a tendency toward overconfident predictions for all categories (Supplemental Digital Appendix 6 at https://links.lww.com/ACADMED/B361).

모델 사양
Model specification

이 알고리즘은 독립적인 테스트와 검증을 위해 웹 기반 애플리케이션에 구축되었습니다. 누구나 자유 응답 평가 내러티브를 입력할 수 있으며, 모델은 해당 내러티브에 포함된 (하위) 역량을 예측합니다. 상위 3개 예측을 요약하고 모델이 이러한 예측을 하게 된 단어를 강조하는 LIME 그래픽이 표시됩니다. 예를 들어 다음과 같은 코멘트가 모델에 전달되었습니다:

  • "이 평가는 흉부외과에서의 하루를 다룹니다. 동의하는 동안 환자와 가족을 안심시키는 데 큰 역할을 했습니다. 응급 섬망에 대한 근거 기반 진료 권고사항을 통합했습니다. 중심정맥관 배치는 숙련되었으나 심폐우회술에서 분리하는 동안 인트로프로프 관리를 제대로 하지 못함. 다음 사례에 대비하여 분리 과정을 시각화하여 보다 원활하고 안전하게 분리할 수 있도록 하는 것이 좋습니다."

The algorithm was built into a web-based application for independent testing and validation. Anyone can type a free-response evaluation narrative, and the model predicts which (sub)competencies it contains. LIME graphics are displayed that outline the top 3 predictions and highlight which words drove the model to make those predictions. As an example, the following comment was passed to the model:

  • “This evaluation covers a day in the cardiothoracic room. Did a great job putting patient and family member at ease during consent. Incorporated evidence-based practice recommendations on emergence delirium. Skillful central line placement, but poor management of inotropes while separating from cardiopulmonary bypass. Consider visualizing the separation process in preparation for your next case so you can do this more smoothly and safely.”

이 모델은 단락을 개별 문장으로 분할한 다음, 각 문장이 어떤 (하위) 역량을 다룰지 예측합니다. 이 예는 모델이 자유 응답 내러티브를 분석하여 ACGME (하위) 역량 프레임워크로 구성하는 방법을 강조합니다. 모델 예측은 표 3에 나와 있습니다. LIME 해석 가능성 도표로 개선된 세 가지 추가 예는 부록 디지털 부록 7-9에 포함되어 있습니다.
The model splits the paragraph into individual sentences, then predicts which (sub)competencies each sentence addresses. This example highlights how the model can dissect a free-response narrative to organize it into an ACGME (sub)competency framework. Model predictions are presented in Table 3. Three additional examples, enhanced by LIME interpretability plots, are included in Supplemental Digital Appendices 7–9 at https://links.lww.com/ACADMED/B361.

acadmed_2023_01_19_booth_acadmed-d-22-00772_sdc1.pdf
0.51MB

모델 구현
Model implementation

컴퓨터 프로그램은 약 1분 만에 18명의 교육생에 대한 수백 개의 서술적 의견을 읽고, 라벨을 붙이고, 정리하여 교육생과 멘토가 검토할 수 있도록 개별화된 스프레드시트를 생성했습니다. 조직화된 피드백은 교육생이 마일스톤 2.0을 통해 스스로 결정한 하위 역량에 대한 진도를 향상시키기 위한 실행 계획을 세우도록 안내했습니다. 이 구현에 대한 영향 지표는 수집되지 않았습니다. 그러나 내부 피드백은 이 연습에 대해 매우 긍정적이었으며, "레지던트가 자신의 강점을 활용하고 성장 영역을 식별하기 위해 개별화된 학습 계획을 개발하도록 지원"하는 ACGME 공통 프로그램 요건을 달성했습니다.7
In approximately 1 minute, the computer program read, labeled, and organized several hundred narrative comments on 18 trainees and produced individualized spreadsheets for review by trainees and their mentors. The organized feedback guided trainees to make action plans to enhance their progress through Milestones 2.0 on self-determined subcompetencies. Impact metrics were not collected on this implementation. However, internal feedback was very positive on the exercise, and it achieved the ACGME Common Program Requirement to “assist residents in developing individualized learning plans to capitalize on their strengths and identify areas for growth.”7

토론
Discussion

이 연구는 인공지능이 GME 연수생에 대한 내러티브 피드백을 마일스톤 2.0의 하위 역량과 핵심 역량으로 정확하게 분류할 수 있음을 보여줍니다. NLP 모델은 3개 레지던트 프로그램의 마취과 전문의로부터 10,000개 이상의 코멘트을 사용하여 피드백 언어를 학습했습니다. 이 모델은 외부 평가 사이트에서도 잘 일반화되었습니다. 개념 증명 시험에서 이 모델은 최소한의 노력으로 대량의 내러티브 피드백을 빠르게 정리하고 마일스톤 자가 평가 연습에서 수련의를 성공적으로 안내하는 것으로 입증되었습니다. 이 연구에서 개발된 NLP 모델은 기존 전자 평가 시스템에 구축하여 수천 개의 평가를 쉽게 관리할 수 있도록 확장할 수 있으며, 이를 통해 교육생의 진행 상황에 대한 구조화된 피드백과 적시 평가를 제공하는 프로그램의 기능을 강화할 수 있습니다.
This study shows that artificial intelligence can accurately categorize narrative feedback on GME trainees into corresponding Milestone 2.0 subcompetencies and core competencies. An NLP model learned the feedback language from anesthesiologists at 3 residency programs using more than 10,000 comments. The model generalized well to an external evaluation site. A proof-of-concept trial demonstrated that the model rapidly organizes large volumes of narrative feedback with minimal effort, and it successfully guided trainees in a Milestone self-assessment exercise. An NLP model, such as developed in this study, could be built into existing electronic evaluation systems and scaled to manage thousands of evaluations with ease, thereby augmenting programs’ ability to provide structured feedback and timely assessments of trainee progress.

NLP 모델은 대부분의 하위 역량과 핵심 역량에 걸쳐 우수한 성능을 보였으며, 내부 및 외부 검증 모두에서 우수한 변별력을 보여준 여러 범주에서 우수한 성능을 보였습니다. 

  • PC3(훈련 데이터의 1.9%, AUC 0.76) 및 PBLI(훈련 데이터의 3.8%, AUC 0.75)와 같이 거의 사용되지 않는 몇 가지 범주에서도 우수한 성능을 보였습니다.
  • 이 모델은 인구통계학적 정보와 관련된 코멘트(예: "이것은 전기생리학 실험실에서 하루 동안의 평가입니다.", AUC 0.87)과 유용하지 않은 코멘트(예: "잘했어요.", AUC 0.87)을 예측하는 데 특히 우수한 성능을 보였습니다.

The NLP model performed well across most subcompetencies and core competencies with several categories demonstrating excellent discrimination on both internal and external validation.

  • It performed well even for several of the rarely used categories, such as PC3 (1.9% of training data, AUC 0.76) and PBLI (3.8% of training data, AUC 0.75).
  • The model performed particularly well for predicting comments related to demographic information (e.g., “This is an evaluation for a day in the electrophysiology lab,” AUC 0.87) and comments that were not useful (e.g., “Good job,” AUC 0.87).

SBP(AUC 0.59)의 성능이 최악이었습니다. Zhang 등은 교육생 평가의 피드백 코멘트에 포함된 감정과 관련 핵심 역량을 식별하려는 NLP 기법을 살펴본 결과, 마찬가지로 SBP가 예측하기 가장 어려운 역량이라는 사실을 발견했습니다.11 SBP에 대한 코멘트 분석의 어려움은 평가자 간 일치도 결과에서 잘 드러나는데, SBP 하위 역량 중 하나에 대한 평가자 간 일치도가 9.9%로 낮게 나타났습니다. 이 결과는 SBP와 관련된 의견을 예측하는 데 문제가 있는 것은 GME 커뮤니티가 SBP를 특성화하는 데 어려움을 겪고 있거나, 서술적 의견 이외의 평가 기법으로 SBP를 더 잘 평가할 수 있다는 것을 시사합니다. 
Performance was worst for SBP (AUC 0.59). Zhang et al explored NLP techniques that attempt to discern the sentiment and pertinent core competencies contained in feedback comments from trainee evaluations, and they similarly found that SBP was the most difficult competency to predict.11 The challenge of analyzing comments about SBP is underscored by our results for inter-rater agreement; inter-rater agreement was as low as 9.9% for one of the SBP subcompetencies. This finding suggests that the problem predicting comments related to SBP may be that the GME community has a hard time characterizing SBP or that SBP may be assessed better by evaluation techniques other than narrative comments.

전문가 라벨에 대한 전반적인 동의도는 보통 수준이었으며(Fleiss의 카파 0.44), 하위 역량 간 동의도에는 상당한 편차가 있었습니다. 일반적으로 PC 하위 역량에 대한 동의도가 가장 높았고, P와 SBP에 대한 동의도가 가장 낮았습니다. 하위 역량을 핵심 역량으로 통합했을 때 평가자 간 일치도는 소폭 개선되었지만 여전히 보통 수준이었습니다(Fleiss의 카파 0.50). 이러한 결과는 지속적인 교수진 개발의 필요성을 뒷받침하며, 이 프로젝트의 다음 반복에서는 트레이너 교육 환경에서 이러한 기회를 모색할 것입니다.
Overall agreement on expert labels was moderate (Fleiss’ Kappa 0.44), and there was substantial variability in agreement between subcompetencies. In general, agreement was highest for PC subcompetencies and lowest for P and SBP. When subcompetencies were pooled into core competencies, inter-rater agreement improved marginally but was still moderate (Fleiss’ Kappa 0.50). These results support the need for ongoing faculty development, and the next iteration of this project will seek to explore this opportunity in a train-the-trainer setting.

이 연구는 레지던트 프로그램의 지속적인 수련 과정 평가를 지원하기 위한 도구로 NLP를 사용하려는 이전의 노력을 기반으로 합니다. 여러 선행 연구에서 평가의 품질 또는 유용성을 예측하기 위해 NLP를 다루고 있지만,10,19,20 본 연구 결과에 따르면 교수진이 평가서를 작성하는 방법뿐만 아니라 평가서에 작성하는 내용을 대상으로 하는 것이 더 신중할 수 있음을 시사합니다. 유용하지 않거나 인구통계학적 내용에 관한 코멘트이 데이터의 30% 이상을 차지했습니다. CBME 평가를 목표로 하는 교수진 개발 이니셔티브는 필수적입니다.21 ACGME는 교수진 개발에는 교육기관에 맞는 구조화된 프로그램이 포함된다고 설명하며7 피드백 제공자를 위한 트레이너 양성 프로그램을 권장합니다.22 프로그램은 웹 기반 애플리케이션에 내장된 NLP 모델을 사용하여 교수진이 교육기관에 맞는 피드백 언어에 대해 구조적이고 신중한 연습을 할 수 있도록 지원할 수 있습니다.
This study builds on prior endeavors to use NLP as a tool to assist residency programs in the ongoing evaluation of trainee progress. While several prior studies address NLP to predict the quality or utility of evaluations,10,19,20 our results suggest that it may be as prudent to target what faculty write about in their evaluations, rather than only how they write them. Comments that were not useful or about demographic content made up more than 30% of the data. Faculty development initiatives aimed at CBME assessments are imperative.21 ACGME outlines that faculty development includes structured programming specific to an institution7 and recommends train-the-trainers programs for feedback providers.22 Programs could use NLP models built into web-based applications to facilitate structured, deliberate practice for faculty on feedback language consistent with their institution.

이 연구에는 몇 가지 한계가 있습니다. NLP 모델은 훈련에 사용된 데이터에 의해 제한됩니다. SBP와 같은 일부 범주에서는 평가자 간 일치도가 낮게 나타났습니다. 잘 훈련된 인공 지능 모델은 모호한 결과를 예측하는 데 제한적인 유용성을 가집니다. 이 연구에서 개발된 모델과 같은 모델은 ACGME 마일스톤에 대한 프로그램의 해석에 대해 직원을 교육하는 것을 목표로 하는 전문성 개발 활동과 병행해야 할 수도 있습니다. 또 다른 한계는 이 모델이 품질이나 유용성보다는 하위 역량과 역량을 학습하도록 훈련되었다는 점입니다. 따라서 알고리즘을 통합한 트레이너 교육 프로그램은 교수진이 형성적인 피드백보다는 특정 유행어나 문구를 작성하도록 장려할 수 있습니다. 데이터 세트에서 유용하지 않은 코멘트의 비율이 높다는 점을 고려할 때, 현재의 알고리즘은 평가자가 역량에 대해 더 구체적으로 작성하는 방법을 배우는 데 도움을 줌으로써 우선적으로 큰 가치를 더할 수 있습니다. 향후 연구에서는 (코멘트가 설명하는 역량과 무관하게 )실행 가능한 코멘트와 그렇지 않은 코멘트을 인식하는 별도의 모델을 쉽게 훈련할 수 있으며, 이를 현재 알고리즘과 결합하여 콘텐츠와 품질에 대한 동시 예측을 생성할 수 있습니다.
This study has several limitations. NLP models are limited by the data used to train them. Some categories, such as SBP, demonstrated poor inter-rater agreement. A well-trained artificial intelligence model has limited utility in predicting ambiguous outcomes. Models such as the one developed in this study may need to be paired with a professional development activity aimed at educating staff on a program’s interpretation of ACGME Milestones. Another limitation was that the model was trained to learn subcompetencies and competencies rather than quality or utility. Therefore, a train-the-trainer program incorporating the algorithm may encourage faculty to write certain buzzwords or phrases rather than formative feedback. Given the high proportion of comments that were not useful in the dataset, the present algorithm can add a great deal of value first by helping evaluators learn to write more specifically to competencies. A future study could easily train a separate model to recognize actionable comments vs not (independent of the competencies they may or may not describe), which could be paired with the present algorithm to produce simultaneous predictions on content and quality.

향후 작업에서 가장 중요하게 고려해야 할 사항은 이러한 모델을 책임감 있고 윤리적으로 배포하는 것입니다. 개념 증명 시험은 교육생이 자가 평가 연습을 할 수 있도록 설계되었습니다. 만약 이 모델이 훈련생의 진도에 대한 결정을 내리기 위해 설계된 방식으로 배포되었다면 알고리즘 개발의 특정 편향23,24 또는 내러티브 피드백 자체에 내재된 편향23-25으로 인해 재앙적인 결과를 초래할 수 있습니다. 
A major consideration for future work is responsible and ethical deployment of these models. The proof-of-concept trial was designed to guide trainees in a self-assessment exercise. If the model were deployed instead in a manner designed to make decisions on trainee progress, certain biases in algorithm development,23,24 or biases inherent in the narrative feedback itself,23–25 could have disastrous consequences.

결과 평가에 편향이 있을 가능성이 있었습니다. 라벨링 프로세스는 고유한 지식, 교육 및 GME 경험을 가진 10명의 리뷰어에게 의존했으며, 모델은 완전한 동의가 있는 데이터만을 사용하여 학습되었습니다. 이로 인해 평가 의견의 크기가 제한되고 모호성이 덜한 언어로 모델이 편향되었습니다. 그러나 최종 모델에는 여전히 10,000개가 넘는 개별 코멘트의 언어가 포함되었으며, 외부 검증을 통해 성능이 잘 일반화되었음을 확인할 수 있었습니다. 데이터 분석 수명 주기에서 많은 의료 서비스 관련 프로젝트에 부족한 중요한 단계는 모델 모니터링 및 피드백 루프입니다. 이 연구에서 설명한 것과 같은 알고리즘에 대한 향후 작업에서는 오분류의 영향을 완화하는 데 도움이 되는 다양한 데이터 세트와 모델 업데이트를 통한 지속적인 검증을 모색해야 합니다. 또한 모델 모니터링을 통해 기관 및/또는 프로그램 간의 일반화 가능성을 평가할 수 있으며, 성능 임계값을 정의하여 모델을 미세 조정하기 위한 재교육을 트리거할 수 있습니다. 또 다른 한계는 문장을 하나의 하위 역량으로만 제한하는 것입니다. 여러 문장이 두 개 이상의 하위 역량에 대해 말하고 있으며, 평가자에게는 레이블을 할당하는 구체적인 규칙이 제공되지 않았습니다. 따라서 여러 하위 역량이 적용되는 경우, 전문가가 특정 인지적 편견으로 인해 한 범주를 다른 범주보다 더 많이 선택했을 수 있습니다.
There was potential for bias in outcome assessment. The labeling process relied on 10 reviewers with unique knowledge, training, and GME experiences, and the model was trained only using data for which there was complete agreement. This restricted the size of evaluation comments and biased the model toward language that is less ambiguous. However, the final model still incorporated language from more than 10,000 individual comments, and external validation supports that performance generalizes well. A critical phase of the data analytics lifecycle that many health care–related projects lack is model monitoring and feedback loops. Future work on algorithms, such as described in this work, should explore ongoing validation with various datasets and model updating to help mitigate the impacts of misclassification. Model monitoring would also allow assessment of generalizability between institutions and/or programs, and performance thresholds could be defined to trigger retraining to fine-tune models. Another limitation is restricting sentences to just a single subcompetency. Several sentences speak to more than one subcompetency, and raters were not provided with specific rules to assign labels. Therefore, when multiple subcompetencies applied, certain cognitive biases may have contributed to experts choosing one category over another.

마지막으로, 마일스톤 2.0과 그에 수반되는 하위 역량이 포함 기간 동안 공개되었습니다. 대부분의 의견은 첫 번째 마일스톤 세트를 대상으로 했을 가능성이 높지만, 전문가들은 이를 마일스톤 2.0에 매핑했습니다. 이러한 접근 방식은 특정 범주에 편향성을 불러일으킬 수 있지만, 이러한 접근 방식은 추가적인 미래 지향적 검증을 용이하게 합니다.
Finally, Milestones 2.0 with their accompanying subcompetencies were released during the inclusion period. Most comments likely targeted the first set of Milestones, yet experts mapped them to Milestones 2.0. This approach may have introduced bias to certain categories, but this approach facilitates further prospective validation.

결론
Conclusions

이 연구는 4개의 교육 프로그램에 걸쳐 207명의 교수진이 169명의 마취과 수련의에 대한 25,714개의 서술적 코멘트를 검토했습니다. NLP 알고리즘은 3개의 마취과 GME 프로그램의 언어를 학습하도록 훈련되었으며, 대부분의 ACGME 하위 역량에 걸쳐 매우 우수한 성능을 보였습니다. 이 모델은 6개월간의 교육생 피드백을 신속하게 정리하여 자가 평가 연습을 안내하는 컴퓨터 프로그램에 구축되었습니다. 효과적인 자동화 프로세스를 개발하는 데 성공하면 마일스톤 2.0을 통해 개별 수련의가 발전함에 따라 레지던트 프로그램의 추적 기능을 향상시킬 수 있지만, 이러한 시스템이 책임감 있고 윤리적으로 배포되도록 하기 위해서는 상당한 작업이 필요합니다. 자연어 처리 방법은 내러티브 데이터를 복잡하게 분류하고 처리할 수 있으며, 이러한 방법을 CBME에 적용할 수 있는지에 대한 추가 조사가 필요합니다.

This study reviewed 25,714 narrative comments on 169 anesthesiology trainees from 207 faculty across 4 training programs. An NLP algorithm was trained to learn the language of 3 anesthesiology GME programs, and it performed very well across most ACGME subcompetencies. The model was built into a computer program which rapidly organized 6 months of trainee feedback to guide a self-assessment exercise. Success in developing an efficacious automated process could enhance the tracking capabilities of residency programs as individual trainees progress through Milestones 2.0, although substantial work is needed to make sure such systems are deployed responsibly and ethically. NLP methods are capable of complex categorization and processing of narrative data, and further investigation into these applications to CBME is warranted.


Acad Med. 2023 Apr 1;98(4):497-504. doi: 10.1097/ACM.0000000000005115. Epub 2022 Dec 5.

Competency-Based Assessments: Leveraging Artificial Intelligence to Predict Subcompetency Content

Affiliations collapse

1G.J. Booth is assistant professor, Uniformed Services University of the Health Sciences, and residency program director, Department of Anesthesiology and Pain Medicine, Naval Medical Center Portsmouth, Portsmouth, Virginia.

PMID: 36477379

DOI: 10.1097/ACM.0000000000005115

Abstract

Purpose: Faculty feedback on trainees is critical to guiding trainee progress in a competency-based medical education framework. The authors aimed to develop and evaluate a Natural Language Processing (NLP) algorithm that automatically categorizes narrative feedback into corresponding Accreditation Council for Graduate Medical Education Milestone 2.0 subcompetencies.

Method: Ten academic anesthesiologists analyzed 5,935 narrative evaluations on anesthesiology trainees at 4 graduate medical education (GME) programs between July 1, 2019, and June 30, 2021. Each sentence (n = 25,714) was labeled with the Milestone 2.0 subcompetency that best captured its content or was labeled as demographic or not useful. Inter-rater agreement was assessed by Fleiss' Kappa. The authors trained an NLP model to predict feedback subcompetencies using data from 3 sites and evaluated its performance at a fourth site. Performance metrics included area under the receiver operating characteristic curve (AUC), positive predictive value, sensitivity, F1, and calibration curves. The model was implemented at 1 site in a self-assessment exercise.

Results: Fleiss' Kappa for subcompetency agreement was moderate (0.44). Model performance was good for professionalism, interpersonal and communication skills, and practice-based learning and improvement (AUC 0.79, 0.79, and 0.75, respectively). Subcompetencies within medical knowledge and patient care ranged from fair to excellent (AUC 0.66-0.84 and 0.63-0.88, respectively). Performance for systems-based practice was poor (AUC 0.59). Performances for demographic and not useful categories were excellent (AUC 0.87 for both). In approximately 1 minute, the model interpreted several hundred evaluations and produced individual trainee reports with organized feedback to guide a self-assessment exercise. The model was built into a web-based application.

Conclusions: The authors developed an NLP model that recognized the feedback language of anesthesiologists across multiple GME programs. The model was operationalized in a self-assessment exercise. It is a powerful tool which rapidly organizes large amounts of narrative feedback.

주관주의자 귀납적 패러다임을 향한 자연어처리의 진화(Med Educ, 2023)
Evolving natural language processing towards a subjectivist inductive paradigm
Brian C. Gin

 

 

인공지능(AI)이 데이터 중심 사회에서 점점 더 큰 입지를 차지하면서 언어에 초점을 맞춘 파생 기술인 자연어 처리(NLP)가 보건 전문직 교육 연구(HPER)에 진출했습니다.1 NLP는 텍스트와 문서의 질적 분석을 위한 획기적인 검색 및 자동화 도구를 약속합니다. 그러나 탄탄한 이론적, 철학적 토대가 없다면 NLP를 통한 질적 결과는 불투명하거나 최악의 경우 오해의 소지가 있을 수 있습니다. 이번 호에서 Cleland 등은 HPER에서 전통적인(즉, NLP가 아닌) 질적 문서 분석(DA)의 실행에서 유사한 이론적 결함을 지적하고 그러한 DA 연구의 신뢰성을 증진하기 위한 표준을 제안합니다.2 이들의 접근 방식은 실제로 NLP를 사용한 연구를 강화하기 위해 확장될 수 있습니다. 
As artificial intelligence (AI) takes an ever-increasing foothold in our data-driven society, its language-focused derivative, natural language processing (NLP), has made headway into health professions education research (HPER).1 NLP promises groundbreaking discovery and automation tools for the qualitative analysis of text and documents. However, without sound theoretical and philosophical foundations, qualitative results facilitated by NLP stand to be opaque, or at worst, misleading. In this issue, Cleland et al note a similar deficiency of theory in the practice of traditional (i.e. non-NLP) qualitative document analysis (DA) in HPER and propose standards to promote the trustworthiness of such DA research.2 Their approach could indeed be extended to strengthen studies using NLP.

탄탄한 이론적, 철학적 토대가 없다면 NLP를 통해 얻은 질적 결과는 불투명하거나 최악의 경우 오해의 소지가 있을 수 있습니다.
Without sound theoretical and philosophical foundations, qualitative results facilitated by NLP stand to be opaque, or at worst, misleading.

HPER에서 빠르게 증가하는 연구 모음은 본질적으로 인간 코더의 범위를 대규모 데이터세트로 확장하기 위한 '자동화된 코딩 도우미'로 NLP를 사용했습니다. 이러한 연구들은 인간 연구자의 '훈련'에 의존하는 NLP '분류기' 알고리즘을 활용했습니다. 예를 들어,

  • 부스(Booth) 등은 NLP를 활용하여 ACGME 마일스톤 하위 역량별로 내러티브 평가를 자동으로 분류하고, 인간 코더가 데이터의 하위 집합(즉, '훈련 데이터 세트')에 붙인 레이블을 모방하도록 NLP 알고리즘을 효과적으로 훈련시켰습니다. 3
  • 내러티브 평가에서 성별 편향 가능성을 찾기 위해 Andrews 등은 인간 연구자가 코드북을 개발하는 데 사용하는 작은 조각으로 내러티브를 표현하여 내러티브의 정서(즉, 긍정적 또는 부정적 정서적 가치)를 결정하는 NLP 알고리즘을 만들었고, 이 알고리즘은 전체 데이터 세트에 자동으로 적용되었습니다.4
  • 비슷한 질문을 탐구하면서 Sarraf 등은 상업적으로 훈련된 NLP 알고리즘을 사용하여 추천서의 정서를 특성화했습니다.5

이러한 사례와 그 밖의 많은 최근 HPER의 NLP 분류 사례는 철학적 입장과 이론적 가정에 대한 추가 검토를 유도하는 'NLP 지원 코딩' 방법론의 새로운 패턴을 가리킵니다. 
A rapidly growing collection of studies in HPER has used NLP, in essence, as an ‘automated coding assistant’ to extend the reach of human coders to large datasets. These studies utilised NLP ‘classifier’ algorithms that rely on some form of ‘training’ by human researchers. For example,

  • Booth et al utilised NLP to automatically classify narrative evaluations by ACGME milestone subcompetency, effectively training their NLP algorithm to mimic the labels placed by human coders on a subset of the data (i.e. a ‘training dataset’).3 
  • Searching for possible gender bias in narrative evaluations, Andrews et al created an NLP algorithm to determine the sentiment (i.e. positive or negative emotional valence) of narratives by representing them in bite-sized pieces that human researchers used to develop a codebook, which the algorithm then automatically applied to the entire dataset.4 
  • Exploring a similar question, Sarraf et al employed commercially trained NLP algorithms to characterise the sentiment of letters of recommendation.5 

These and many other recent examples of NLP classification in HPER point to an emerging pattern of ‘NLP-assisted coding’ methodologies that invite further examination of their philosophical stances and theoretical assumptions.

Cleland의 DA 연구 결과와 유사하게, 연구자들이 NLP 지원 코딩을 사용하는 기본 연구 패러다임은 종종 불분명합니다. 성찰적 주제 분석, 근거 이론, 질적 내용 분석과 같은 유비쿼터스 질적 방법론은 모두 데이터 탐색, 코드 개발, 주제 종합 사이에 어떤 형태의 반복에 의존합니다.6, 7 그러나 NLP 지원 코딩을 사용하는 경우, 반복적인 탐구 과정이 아닌 전체 데이터 세트에 고정된 코드북을 적용하는 방식으로 배포되는 경우가 많습니다. 초기 연구자가 수행한 훈련 데이터 세트의 코딩은 반복적인 귀납적 접근 방식을 통해 이루어질 수 있지만, 더 큰 전체 데이터 세트에 NLP를 일회성으로 적용하면 코드 및 테마 형성에 대한 해당 데이터의 추가 기여를 놓치게 됩니다. 따라서 현재 HPER에서 NLP를 주로 구현하는 방식은 귀납적 접근 방식이 아닌 연역적 접근 방식에 속하는 것으로 보이며, 데이터를 사용하여 가설이나 이론을 생성하기보다는 정적 학습 데이터 세트에 코드화된 가설을 데이터에 적용하는 방식입니다. 
Similar to Cleland's findings in DA, the underlying research paradigms within which researchers employ NLP-assisted coding are often unclear. Ubiquitous qualitative methodologies such as reflexive thematic analysis, grounded theory and qualitative content analysis all rely on some form of iteration between exploring data, developing codes and synthesising themes.6, 7 When NLP-assisted coding is used, however, it is often deployed by applying a fixed codebook to the entire dataset, rather than within an iterative process of inquiry. While the initial researcher-performed coding of the training dataset may occur via an iterative inductive approach, the use of NLP as a one-time application to the larger complete dataset will miss additional contributions of those data to the formation of codes and themes. Thus, the predominant implementation of NLP in HPER currently appears to fall under a deductive rather than inductive approach, applying a hypothesis—codified in a static training dataset—to data rather than using the data to generate a hypothesis or theory.

현재 HPER에서 주로 구현되는 NLP는 귀납적 접근 방식이 아닌 연역적 접근 방식에 속하는 것으로 보이며, 데이터를 사용하여 가설이나 이론을 생성하기보다는 정적 훈련 데이터 세트에 코드화된 가설을 데이터에 적용합니다.
The predominant implementation of NLP in HPER currently appears to fall under a deductive rather than inductive approach, applying a hypothesis—codified in a static training dataset—to data rather than using the data to generate a hypothesis or theory.

그렇다고 NLP가 연역적 탐구에만 사용되거나 (포스트)실증주의 패러다임 내에서만 사용 가능하다는 것은 아닙니다. 귀납적 정성적 접근 방식을 지원할 수 있는 NLP의 잠재력은 존재하지만, 현재 기술적으로 까다롭고 HPE 내에서 상대적으로 미개척 분야로 남아 있습니다. 한 가지 접근 방식은 '비지도' 주제 모델링 전략으로, NLP 알고리즘이 코드화되지 않은 데이터 세트 내에서 의미 패턴을 자동으로 검색하여 새로운 코드를 개발하는 것입니다.8 '연역적'인 것처럼 보이지만 완전히 자동화된 비지도 접근 방식은 알고리즘과 인간 연구자 간의 반복적인 협업이 수반되지 않는 한 이상적이라고 할 수 없습니다. 보다 유망한 방향은 능동형 머신 러닝과 연구자 지원 주제 모델링과 같이 연구자가 수행하는 코딩과 NLP 자동 코딩을 혼합하는 것으로 보입니다. 인간과 기계 간의 미래 협업은 알고리즘이 학습하고 의사 결정을 내리는 방식에 대한 이해, 즉 '설명 가능한 AI'의 진전에 달려 있습니다. 최신 자연어 처리 알고리즘이 텍스트 의미를 표현하는 AI의 능력을 확장함에 따라 복잡성이 증가하면서 의사 결정 과정이 모호해지고 있습니다. 이러한 불투명성은 소위 '블랙박스' 현상으로, AI 알고리즘이 복잡한 행동을 모방하도록 훈련할 수는 있지만 그렇게 학습하는 방법에 대한 인사이트는 거의 드러나지 않습니다. 인간의 마음도 마찬가지이지만, 효과적인 인간과 기계의 협업은 최종 분류에 대한 지식만 얻는 것이 아니라 NLP 알고리즘이 내부적으로 의미적 의미를 어떻게 표현하는지를 이해하는 데 달려 있습니다.9 
This is not to say that NLP can only be used for deductive inquiry or only within a (post-)positivist paradigm. The potential for NLP to assist an inductive qualitative approach exists, but it currently appears technically challenging and remains relatively untapped within HPE. One approach involves an ‘unsupervised’ topic modelling strategy whereby an NLP algorithm automatically searches for patterns of meaning within an uncoded dataset to develop de novo codes.8 Although seemingly ‘deductive’, a completely automated unsupervised approach is not ideal unless it also involves iterative collaboration between the algorithm and human researchers. More promising directions appear to involve hybrids between researcher-performed and NLP-automated coding, such as active machine learning and researcher-assisted topic modelling. Future collaboration between human and machine also depends on making progress in ‘explainable AI’—the understanding of how algorithms learn and make decisions. As newer NLP algorithms expand AI's ability to represent textual meaning, their increasing complexity obfuscates their decision-making process. This opacity is the so-called ‘black box’ phenomenon whereby an AI algorithm can be trained to mimic complex behaviour but reveals little insight into how it learned to do so. While the same can be said of the human mind, effective human–machine collaboration will depend on understanding how NLP algorithms represent semantic meaning internally, rather than gaining knowledge only of the final classifications they make.9

효과적인 인간과 기계의 협업은 NLP 알고리즘이 내부적으로 의미적 의미를 표현하는 방식을 이해하는 데 달려 있습니다.
Effective human-machine collaboration will depend on understanding how NLP algorithms represent semantic meaning internally.

마지막으로, NLP를 이용한 정성적 문서 분석의 성찰성reflexivity 문제가 있습니다. 반사적 정성적 접근 방식은 의미를 만드는 데 있어 연구자의 기여를 인정하고 존중하지만, 현재의 NLP 알고리즘은 이러한 자기 인식이 부족합니다. 학습된 NLP 알고리즘은 학습 데이터를 코딩한 연구자의 반사성을 이어받을 수 있지만, 의도하지 않은 잠재적으로 원치 않는 코딩 경향이나 편견을 분석에 도입할 수도 있습니다. 또한 NLP 알고리즘은 종종 '전이 지식'(예: 별도의 말뭉치에 대한 학습을 통해 얻은 언어에 대한 기계의 이해)으로 보완되는데, 이는 외부 학습 데이터의 편견을 항상 그대로 전달할 수 있습니다. Sarraf 등이 수행한 것처럼 상업적으로 사전 학습된 NLP 알고리즘을 사용하는 경우, 이러한 편향(및 관련 반사성)은 연구자 자신의 반사성과는 완전히 외부에 있습니다. 이러한 알고리즘 편향을 식별하고 완화하는 것은 NLP 내에서 활발히 연구되고 있는 분야입니다. 그러나 '편향되지 않은' NLP 알고리즘의 이상은 필연적으로 객관주의에 귀를 기울일 수밖에 없으며 주관주의적 관점에서는 달성할 수도 없고 원하지도 않을 수 있습니다. 따라서 NLP 알고리즘의 '반사성'은 잘 정의되어 있지 않으며, 알고리즘이 분석에서 중요한 역할을 하는 경우 연구에 비판적 노출을 남길 수 있습니다. 질적 연구의 반사성에 대한 NLP의 기여도를 정의, 평가 및 조정하는 방법을 이해하는 것은 시급히 해결해야 할 과제입니다.10
Finally, there is the issue of reflexivity in NLP-assisted qualitative document analysis. While reflexive qualitative approaches have both acknowledged and enshrined researchers' contributions to making meaning, current NLP algorithms lack such self-awareness. While a trained NLP algorithm may carry forward the reflexivity of the researchers who coded the training data, it may also introduce unintended and potentially unwelcome coding tendencies or biases into the analysis. Further, NLP algorithms are often supplemented by ‘transfer knowledge’—for example, machine understanding of language afforded by training on a separate corpus—that invariably carries forward biases from that external training data. In the case of using a commercially pre-trained NLP algorithm as was done by Sarraf et al, such bias (and any associated reflexivity) is completely external to the researchers' own reflexivity. Identifying and mitigating such algorithmic biases is an active area of research within NLP. However, the ideal of ‘unbiased’ NLP algorithms inevitably hearkens to objectivism and may neither be achievable nor desired from a subjectivist perspective. As such, the ‘reflexivity’ of an NLP algorithm is not well defined and may leave a study critically exposed if the algorithm plays a significant role in the analysis. Understanding how to define, assess and tune the contribution of NLP to a qualitative study's reflexivity needs to be urgently addressed.10

질적 연구의 반사성에 대한 NLP의 기여도를 정의, 평가 및 조정하는 방법을 이해하는 것은 시급히 해결해야 할 과제입니다.
Understanding how to define, assess and tune the contribution of NLP to a qualitative study's reflexivity needs to be urgently addressed.

특히 임상 학습 환경을 사회적으로 구성된 것으로 간주할 때, 연역적 (후기) 실증주의 접근법을 지향하는 NLP 지원 DA 연구의 암묵적 경향은 많은 HPER 연구자들이 DA를 통해 답을 구하고자 하는 주관주의적 연구 질문과 상반될 수 있습니다. NLP를 주관주의 귀납적 패러다임으로 전환하는 것은 연구자와 AI '어시스턴트' 모두를 협력적이고 투명한 탐구 프로세스로 초대하는 새로운 NLP 전략을 개발하는 데 달려 있습니다. 한편, Cleland 등이 개발한 CARDA 체크리스트를 사용하면 AI 기술의 클라우드가 빠르게 진화하는 동안에도 NLP 지원 DA 연구가 이론적, 철학적 토대 위에 확고하게 자리 잡을 수 있습니다.
The implicit tendency of NLP-assisted DA studies towards a deductive (post-)positivist approach may stand in contrast to subjectivist research questions that many HPER researchers seek to answer via DA, particularly when considering the clinical learning environment as socially constructed. Moving NLP towards a subjectivist inductive paradigm will depend upon developing new NLP strategies that invite both researchers and their AI ‘assistants’ into a collaborative and transparent process of inquiry. In the meantime, use of the CARDA checklist by Cleland et al can keep NLP-assisted DA studies firmly situated on their theoretical and philosophical underpinnings while the clouds of AI technology rapidly evolve.

NLP를 주관주의 귀납적 패러다임으로 전환하는 것은 연구자와 AI '어시스턴트' 모두를 협력적이고 투명한 탐구 과정으로 초대하는 새로운 NLP 전략을 개발하는 데 달려 있습니다.
Moving NLP towards a subjectivist inductive paradigm will depend upon developing new NLP strategies that invite both researchers and their AI ‘assistants’ into a collaborative and transparent process of inquiry.


Med Educ. 2023 May;57(5):384-387. doi: 10.1111/medu.15024. Epub 2023 Feb 15.

Evolving natural language processing towards a subjectivist inductive paradigm

Affiliations collapse

1Department of Pediatrics, University of California San Francisco, San Francisco, California, USA.

PMID: 36739578

DOI: 10.1111/medu.15024

사회적 책무성 프레임워크와 그것이 의학교육과 프로그램 평가에 갖는 함의: 내러티브 리뷰(Acad Med, 2020)
Social Accountability Frameworks and Their Implications for Medical Education and Program Evaluation: A Narrative Review 
Cassandra Barber, MA, Cees van der Vleuten, PhD, Jimmie Leppink, PhD, and Saad Chahine, PhD

 

의과대학이 봉사하고자 하는 인구에 대한 사회적 책무성을 다해야 한다는 국제적인 요구가 계속 제기되어 왔습니다. 사회적 책무성은 많은 교육기관이 추구하는 이상이지만, 이를 측정하는 것은 전 세계적인 과제로 남아 있습니다. 투명성과 책임성에 대한 사회적 요구가 증가함에 따라 의과대학은 사회적 책무성에 대한 더 강력한 증거를 제시해야 한다는 압박에 직면해 있습니다.1,2 
There have been repeated international calls for medical schools to be socially accountable to the populations they intend to serve. While social accountability is an ideal that many institutions strive toward, measuring it remains a global challenge. With increasing societal demands for greater transparency and accountability, medical schools face growing pressures to produce stronger evidence of their social accountability.1,2

1995년 세계보건기구(WHO)는 사회적 책무성을 다음과 같이 정의했습니다: 
In 1995, the World Health Organization (WHO) defined social accountability as:

[의과대학은 교육, 연구 및 봉사 활동을 그들이 봉사해야 하는 지역사회, 지역 및 국가의 우선적인 건강 문제를 해결하는 방향으로 이끌어야 할 의무가 있습니다. 우선적 건강 요구는 정부, 의료 기관, 의료 전문가 및 대중이 공동으로 파악해야 합니다.3 
[T]he obligation of medical schools to direct their education, research and service activities towards addressing the priority health concerns of the community, region, and/or the nation they have a mandate to serve. The priority health needs are to be identified jointly by governments, healthcare organizations, health professionals and the public.3

그 이후로 사회적 책무성에 관한 문헌이 확대되고 이니셔티브의 수가 증가했습니다.4,5 많은 의과대학이 사명 선언문, 프로그램 목표 및 전략 계획에 사회적 책무성 정책을 포함시켰으며 일부 조직은 공식 인증 절차에 포함시켰습니다.5 그러나 증가하는 관심에도 불구하고 사회적 책무성이 측정 가능한 속성으로 운영되는 방법은 여전히 불분명하여 사회적 책무성을 객관적으로 평가하기 어렵습니다.6 
Since then, the literature surrounding social accountability has expanded and the number of initiatives has multiplied.4,5 Many medical schools have embedded social accountability policies in their mission statements, program objectives, and strategic plans, and some organizations have included them in formal accreditation processes.5 Yet despite the growing interest, how social accountability is operationalized into measurable attributes remains elusive, making social accountability difficult to evaluate objectively.6

의과대학의 사회적 책무성 평가를 지원하기 위해 다양한 정책과 프레임워크가 마련되었지만, 사회적 책무성 원칙, 지표 및 매개변수에 대한 설명은 주로 개념적인 수준에 머물러 있습니다. 위의 WHO의 사회적 책무성 정의는 의학교육의 세 가지 영역(교육, 연구, 봉사 활동)을 포괄하며, 이 검토에서는 교육 영역에 대해 다룹니다. 이 검토의 목적은 프로그램 평가 모델을 조직적 프레임워크로 사용하여 대규모 사회적 책무성 프레임워크 전반에서 공통 주제와 지표를 식별하고 문서화하는 것입니다. 이는 의학교육의 사회적 책무성을 평가하는 데 필요한 초기 운영 구조의 개발을 촉진하기 위한 것입니다.  
Although various policies and frameworks have been established to assist medical schools in the evaluation of social accountability, their descriptions of socially accountable principles, indicators, and parameters remain predominately conceptual in nature. The WHO’s social accountability definition, above, encompasses the 3 domains of medical education (education, research, and service activities), and this review addresses the educational domain. The purpose of this review is to identify and document common themes and indicators across large-scale social accountability frameworks, using a program evaluation model as an organizational framework. It is intended to facilitate the development of initial operational constructs needed to evaluate social accountability in medical education.

배경
Background

설명하다(account)라는 동사에서 파생된 책무성은 가장 단순한 형태의 답변성, 즉 자신의 행동에 대해 설명하고 책임을 져야 하는 의무를 의미합니다.7,8 교육에서 책무성은 기관의 효과성(즉, 기관이 목표를 얼마나 잘 달성했는지)을 평가하는 시스템으로 기능하여 기관이 결과에 대한 책임을 지고 교육 개선을 촉진합니다.9-12 이 시스템은 교육 기관이 자신의 행동에 대해 사회에 답변해야 하는 책임감, 투명성 및 공공 신뢰를 의미합니다.13,14 다양한 형태의 책임성은 존재하지만 모두 다음의 근본적인 질문을 다루고 있습니다:

  • 누가, 무엇에 대해, 누구에게, 어떤 수단을 통해 책임을 져야 하는가?7,10,15

Derived from the verb account, accountability in its simplest form means answerability, the obligation to provide an account and be held responsible for one’s actions.7,8 In education, accountability functions as a system to evaluate institutional effectiveness (i.e., how well institutions meet their goals), holding institutions responsible for results and promoting educational improvement.9–12 This system implies a sense of responsibility, transparency, and public trust, whereby educational institutions are obligated to answer to society for their actions.13,14 While many forms of accountability exist, they all address the following fundamental questions:

  • Who is held to account, for what, to whom, and through what means?7,10,15

모든 의과대학은 이러한 의무를 인정하거나 해결하기로 선택했는지 여부에 관계없이 대중에 대해 책임을 집니다.3 보건의료 전문직 교육 프로그램과 미래의 보건의료 인력을 준비하는 모든 교육기관은 다음에 대해 책임을 집니다. 

  • 의료계,
  • 대중(환자, 가족, 지역사회, 사회),
  • 교육 결과물(졸업생, 봉사 활동, 연구 활동),
  • 미래의 보건의료 수요

책임의 한 형태인 사회적 책무성은 의과대학이 지역사회의 변화하는 공공 의료 수요에 대응할 준비가 된 유능한 졸업생을 배출해야 한다는 점에서 암묵적, 명시적, 예상되는 것입니다.16-20 
All medical schools are accountable to the public, regardless of whether they choose to acknowledge or address this obligation.3 Health professions education programs and any educational institutions responsible for preparing the future health care workforce are accountable to

  • the medical profession;
  • the public (patients, families, communities, and society);
  • their educational products (graduates, service activities, and research activities); and
  • future health care needs.

As a form of accountability, social accountability is implicit, explicit, and anticipated, in that medical schools must produce competent graduates prepared to respond to the changing public health care needs within their local communities.16–20

의료계는 사회로부터 일정한 책임과 특권을 부여받았습니다. 의과대학은 법률, 규제 및 인가를 통해 사회의 요구를 충족할 준비가 된 유능한 의사를 배출하도록 위임받았습니다.21,22 이러한 사회적 역할은 의사와 사회 간의 본질적인 사회적 계약을 의미하는 큰 책임을 수반합니다.23 이러한 사회적 계약은 특히 의학교육이 정부 자금으로 지원되는 국가에서 더욱 강화됩니다. 그 결과, 의과대학은 긍정적인 사회적 환원에 대한 증거를 제공해야 한다는 사회적 압력에 직면해 있습니다.3,16 사회적 책무성은 의학과 사회 사이에 존재하는 전방위적인 사회적 계약을 나타냅니다.24-31 
The medical profession has been granted certain responsibilities and privileges by society. Through legislation, regulation, and accreditation, medical schools are entrusted to produce competent physicians who are prepared to meet the needs of society.21,22 This social role carries great responsibilities, signifying the intrinsic social contract between physicians and society.23 This social contract is specifically amplified in countries where medical education is government funded. As a result, medical schools face increasing societal pressures to provide evidence of a positive social return.3,16 Social accountability represents an omnipresent social contract that exists between medicine and society.24–31

일반적으로 사회적 책무성이란 기업이 자신의 행동, 행위 및 성과에 대해 봉사하고자 하는 사회에 대한 헌신을 의미합니다.32 WHO의 사회적 책무성에 대한 정의는 국제적으로 가장 널리 받아들여지고 있습니다. 2010년에 의과대학의 사회적 책무성에 대한 글로벌 컨센서스는 이 정의를 재확인하면서 사회적 책무성이 측정 가능한 활동임을 강조했습니다: 
Broadly, social accountability implies an entity’s commitment to the society it is intended to serve for its actions, conduct, and performance.32 The WHO’s definition of social accountability remains the most widely accepted internationally. In 2010, the Global Consensus for Social Accountability of Medical Schools reaffirmed this definition, emphasizing that social accountability is a measurable activity:

주요 이해관계자, 정책 입안자, 의료기관, 의료보험 제공자, 의료 전문가 및 시민사회와 협력하면서 사회의 현재 및 미래 건강 요구와 도전에 대응하기 위한 행동입니다.19 
[A]n action to respond to current and future health needs and challenges in society while working collaboratively with key stakeholders; policy-makers; healthcare organizations; health-insurance providers, health professionals and civil society.19

광범위한 책임성 문헌에서 책임성이라는 용어는 종종 개념적 우산13,33,34 으로 불리며 신뢰, 신뢰성 또는 투명성의 이미지를 묘사하기 위해 책임성, 답변 가능성 또는 효과성과 상호 교환적으로 사용됩니다. 그러나 의학교육 문헌에서 책임성, 책임감, 응답성이라는 용어는 동일하지 않습니다. 이들 간의 차이점은 Boelen과 Woollard의 사회적 의무 척도에 명확하게 정의되어 있습니다.32 이 분류법은 사회적 책무성을 달성하기 위한 선형적 진행을 나타냅니다:

  • 책임성이란 "사회의 요구에 부응해야 할 의무를 인식하는 상태"를 의미합니다; 
  • 반응성은 "사회의 요구에 반응하는 행동 과정"을 의미합니다. 
  • 책무성은 프로그램이 공중 보건에 긍정적인 영향을 미치기 위해 주요 이해관계자들과 협력하면서 사회의 우선적인 보건의료 요구선제적으로 충족한다는 증거를 제공하는 "측정 가능한 활동"을 의미합니다.32  

Within the broader accountability literature, the term accountability is often referred to as a conceptual umbrella13,33,34 and used interchangeably with responsibility, answerability, or effectiveness to portray an image of trust, trustworthiness, or transparency. However, in the medical education literature, the terms accountable, responsible, and responsive are not equivalent. Differences between them are clearly defined within Boelen and Woollard’s social obligation scale.32 Their taxonomy represents a linear progression toward achieving social accountability: 

  • responsibility refers to a “state of awareness of duties to respond to society’s needs”; 
  • responsiveness refers to “a course of action addressing society’s needs”; and 
  • accountability represents a “measurable activity” to provide evidence that programs proactively meet the priority health care needs of society while working alongside key stakeholders to positively impact public health.32 

방법
Method

프로그램 평가 모델은 사회 정책, 프로그램 및 개입에 대한 포괄적인 평가를 제공하기 위해 여러 분야에서 널리 사용됩니다.35-39 우리는 프로그램 평가 모델을 조직적 프레임워크와 체계화된 프로세스로 사용하여 대규모 사회적 책무성 프레임워크와 저널 논문 및 의학교육 문헌의 기타 문서를 검토하는 내러티브 검토40를 수행했습니다. 그런 다음 질적 접근법을 사용하여 핵심 개념을 종합했습니다.  
Program evaluation models are widely used in multiple fields to provide comprehensive evaluations of social policies, programs, and interventions.35–39 We conducted a narrative review40 using a program evaluation model as an organizational framework and a systematized process to review large-scale social accountability frameworks as well as journal articles and other documents from the medical education literature. We then synthesized key concepts using a qualitative approach.

조직 프레임워크
Organizational framework

우리는 사회적 책무성의 복잡한 요구, 지표, 결과를 체계적으로 파악하기 위한 평가 도구로 스터플빔의 맥락-입력-과정-산출물(CIPP) 모델을 선택했습니다.35 교육의 책무성을 높이기 위해 1960년대에 처음 개념화된 이 프로그램 평가 모델은 국제적으로 사용되는 책무성 모델이며 의학 교육에서 널리 수용되고 있습니다.35-37 그림 1에 표시된 대로 CIPP 모델은 프로그램 개선 및 책무성을 위한 방법으로 평가를 사용합니다. 이 모델은 상호 관련된 4개의 구성요소로 구성되어 있으며 평가 모델 전체에서 사용되는 지속적인 질 개선 피드백 루프를 통합합니다.35,36,38 
We selected Stufflebeam’s context–input–process–product (CIPP) model as the assessment tool to systematically identify social accountability complex needs, indicators, and outcomes.35 First conceptualized in the 1960s to provide greater accountability in education, this program evaluation model is an internationally used accountability model and widely accepted in medical education.35–37 As depicted in Figure 1, the CIPP model uses evaluation as a method for program improvement and accountability. It consists of 4 interrelated components and incorporates continuous quality improvement feedback loops to be used throughout the evaluation model.35,36,38

CIPP 모델에서35-37 

  • 맥락배경을 의미하며, 교육 기관의 요구, 목표 및 기회를 파악하는 데 사용되는 요구 평가이다
  • 투입은 교육기관이 효과적으로 기능하는 데 필요한 물적 및 인적 자원을 의미합니다. 투입은 프로그램 목표와 목적을 달성하는 데 필요한 적절한 행동 방침을 결정하는 데 사용됩니다. 
  • 프로세스는 프로그램 실행을 가이드하는 데 사용됩니다. 
  • 산출은 학생 학습의 질과 개인 및 사회에 대한 유용성을 나타냅니다. 산출은 결과를 측정하는 데 사용됩니다. 이후 CIPP 모델에서는 프로그램의 영향력, 효과성, 지속가능성, 전달성을 평가하기 위해 산출 구성요소를 4개의 하위 구성요소로 나누었습니다.36,38

CIPP 모델은 역동적이며 교육을 생산 기능으로 간주하여 교육 투입물이 교육 산출물로 전환되는 방식으로 접근합니다. 각 구성 요소는 독립적으로 평가할 수 있지만, 어떤 지표도 프로그램 성과를 절대적으로 나타내는 것은 아닙니다.36,38
In the CIPP model,35–37

  • Context refers to background—a needs assessment used to help identify needs, objectives, and/or opportunities of an educational institution. 
  • Inputs refer to material and human resources needed for effective functioning of an educational institution. Inputs are used to determine the appropriate course of action(s) required to achieve program goals and objectives. 
  • Processes are used to guide the implementation of a program. 
  • Products refer to the quality of student learning and its usefulness for the individual and for society. Products are used to measure outcomes. In later iterations of the CIPP model, the product component was divided into 4 subcomponents to assess a program’s impact, effectiveness, sustainability, and transportability.36,38 

The CIPP model is dynamic and views education as a production function, whereby educational inputs are transformed to educational outputs. While each component can be evaluated independently, no indicator independently represents an absolute measure of program performance.36,38

선택 및 검색 기준
Selection and search criteria

반복적인 프로세스를 사용하여 5개의 전자 서지 데이터베이스 및 플랫폼(PubMed, Embase, ERIC, Web of Science, Google Scholar)과 광범위한 월드 와이드 웹(Google 사용)에서 의학교육에 적용 가능한 사회적 책무성 프레임워크와 동료 평가 저널 논문 및 문서를 검색했습니다. 이러한 검색은 영어 문서로 제한되었습니다. 검색은 2018년 10월에 처음 수행된 후 2019년 3월 31일에 최신 문서를 포함하기 위해 반복되었습니다. 검색 전략에 사용된 키워드에는 사회적 책무성 또는 책임, 사회적 책무성 또는 책임, 사회 정책 등이 포함되었습니다. 이러한 단어는 의학교육, 의과대학, 의료 수련 프로그램, 보건 전문직 교육 주제 제목 용어와 함께 검색되었습니다. 데이터베이스 검색 전략 샘플은 부록 디지털 부록 1에 제공됩니다.  
Using an iterative process, we searched 5 electronic bibliographic databases and platforms (PubMed, Embase, ERIC, Web of Science, and Google Scholar) as well as the broader World Wide Web (using Google) for social accountability frameworks and peer-reviewed journal articles and documents applicable to medical education. These searches were limited to English-language documents. The searches were first conducted in October 2018 and then repeated on March 31, 2019, to include any more recent documents. Keywords used in the search strategies included social accountability OR responsibility, socially accountable OR responsible, and social policies. These words were searched in combination with medical education, medical schools, medical training programs, and health professions education subject heading terms. A sample database search strategy is provided in Supplemental Digital Appendix 1 at https://links.lww.com/ACADMED/B24.

포함 및 제외 기준
Inclusion and exclusion criteria

저희는 의과대학의 사회적 책무성에 초점을 맞추었습니다. 1990년(사회적 책무성이라는 용어가 의학교육 문헌에 명시적으로 등장한 시기)부터 2019년 3월까지 출판된 주요 영어 정책 프레임워크와 동료 검토 문서가 포함 대상에 포함되었습니다. 사회적 책무성 프레임워크에 대해 논의하지 않은 문서는 제외되었습니다. 검색에서 확인된 모든 문서는 연구팀의 포함 여부 검토 과정을 거쳤습니다. 저자 중 두 명(C.B., S.C.)이 검색에서 확인된 모든 문서를 선별했습니다. 전체 연구팀은 수시로 만나 문서를 검토하고 자격 요건에 관한 합의를 도출했습니다. 의과대학이 사회적 책무를 다하기 위해 노력할 수 있는 기본 가치, 원칙 및/또는 매개변수를 나타내고 사회적 책무를 개념화하기 위해 후속 논문에서 사용된 주요 출처의 사회적 책무 프레임워크가 검토에 포함되었습니다. 하위 프레임워크 및/또는 프로그램 또는 기관별 문서는 이전에 확립된 프레임워크를 기반으로 구축되어 일반화 가능성이 부족할 수 있으므로 제외되었습니다. 
Our focus was social accountability in medical schools. Key English-language policy frameworks and peer-reviewed documents published from 1990 (when the term social accountability explicitly emerged within the medical education literature) through March 2019 were eligible for inclusion. Documents that did not discuss social accountability frameworks were excluded. All documents identified in the searches underwent an inclusion review process by the research team. Two of the authors (C.B. and S.C.) screened all documents identified in the searches. The full research team met frequently to review the documents and come to consensus regarding eligibility requirements. Primary source social accountability frameworks which represented the foundational values, principles, and/or parameters of the attributes medical schools can strive toward to fulfill their social mandate and which were used in subsequent papers to conceptualize social accountability were included in the review. Subframeworks and/or program- or institution-specific documents were excluded as these built upon previously established frameworks and could lack generalizability.

분석
Analysis

주제별 종합41-44은 포함된 사회적 책무성 프레임워크 전반에 걸쳐 공통 요소와 고유 요소를 설명하는 데 사용되었습니다. 주제 종합은 귀납적 접근법을 사용하여 텍스트를 체계적으로 코딩하여 주제를 생성하는 것입니다.43,44 3단계 분석 프로세스는

  • 텍스트의 줄별 코딩으로 시작하여,
  • CIPP 모델의 4가지 차원을 조직적 프레임워크로 사용하여 특성화한 설명적 주제를 개발한 다음
  • 분석적 주제를 생성하는 것으로 이어집니다.

저자 중 두 명(C.B. 및 S.C.)이 포함된 문서를 독립적으로 코딩했습니다. 결과물은 두 명의 코더가 검토하고 코딩의 정확성과 포괄성을 보장하기 위해 합의에 도달할 때까지 연구팀 내에서 논의했습니다. 
Thematic synthesis41–44 was used to describe common and unique elements across the included social accountability frameworks. Thematic synthesis involves the systematic coding of text using an inductive approach to generate themes.43,44 The 3-stage analytical process starts with

  • line-by-line coding of text;
  • followed by the development of descriptive themes, which we characterized using the 4 dimensions of the CIPP model as an organizational framework; and
  • then the generation of analytical themes.

Two of the authors (C.B. and S.C.) coded the included documents independently. Resulting themes were reviewed by the 2 coders and discussed within the research team until consensus was reached to ensure coding accuracy and inclusivity.

결과
Results

33개의 초기 샘플 문서3,16,18-20,23,44-70에서 4개의 주요 대규모 사회적 책무성 정책 프레임워크3,16,18,19를 선정하여 검토에 포함시켰습니다(선정된 프레임워크의 개요는 표 1 참조). 이 4개의 주요 출처 문서는 의학교육에서 사회적 책무성의 기본 가치, 원칙 또는 매개변수를 나타냅니다. 또한 이 문서들은 모두 사회적 책무성을 개념화하기 위한 후속 논문에서 많이 인용되고 사용되었습니다. 또한 건강 형평성을 위한 교육 네트워크 평가 프레임워크47 및 다양한 기관별 교육, 연구, 서비스 활동의 정보로도 사용되었습니다. 
From the initial sample of 33 documents,3,16,18–20,23,44–70 we selected 4 key large-scale social accountability policy frameworks3,16,18,19 for inclusion in the review (see Table 1 for an overview of the selected frameworks). These 4 primary source documents represent the foundational values, principles, and/or parameters of social accountability in medical education. Additionally, these documents have all been highly cited and used in subsequent papers to conceptualize social accountability. They were also used to inform the Training for Health Equity Network evaluation framework47 as well as various institution-specific education, research, and service activities.


이러한 프레임워크에는 지역, 국가 및 국제 수준에서 사회적 책무성에 대한 정책, 정의, 적용 및 평가가 포함됩니다.71 이러한 프레임워크는 조금씩 다르지만 모두 [사회적 책무성을 입증하는 데 사용할 수 있는 특성]을 설명합니다. 공통점으로는

  • 지역 공중보건 수요에 대응하고,
  • 주요 이해관계자와 협력하여 기존 및 향후 사회적 공중보건 수요를 파악하고,
  • 주변 지역사회에 봉사하고,
  • 의사 부족 문제를 해결하고,
  • 지역 인구통계와 지리를 반영하여 입학 과정 내 다양성을 높이고,
  • 유능한 의료 전문가를 배출하고,
  • 교육과정에 우선적 보건 수요를 반영하는 것 등이 있습니다.3,16,18,19 

These frameworks include policy, definition, application, and evaluation of social accountability at the local, national, and international levels.71 Although these frameworks differ slightly, they all describe characteristics that can be used toward demonstrating social accountability. Commonalities include

  • responding to local public health needs;
  • working alongside key stakeholders in identifying existing and forthcoming societal public health needs;
  • servicing surrounding communities;
  • addressing physician shortages;
  • increasing diversity within the admissions process to reflect local demographics and geography;
  • producing competent medical professionals; and
  • ensuring the curriculum reflects priority health needs.3,16,18,19

주제별 종합에서는 공유 가치를 포함한 6가지 주제와 CIPP 평가 모델과 관련된 5가지 지표를 확인했습니다.

  • 맥락(프로그램 목표),
  • 투입물(활동),
  • 과정(활동),
  • 결과물(기관의 산출물/결과),
  • 사회 보건에 미치는 영향

영향 평가는 CIPP 모델에서 제품 평가의 하위 구성 요소이지만, 의학교육의 사회적 책무성이 실무에 미치는 영향과 공중 보건 개선에 중점을 두는 점을 고려하여 분석에서 영향 평가를 별도의 주제로 다루었습니다. 또한 아래 설명과 그림 2에 표시된 것처럼 각 테마 내에서 하위 주제를 식별했습니다. 주제와 하위 주제를 설명하기 위한 인용문은 표 2에 나와 있습니다. 
Our thematic synthesis identified 6 themes, including shared values and 5 indicators as they relate to the CIPP evaluation model:

  • context (program objectives),
  • inputs (actions),
  • processes (activities),
  • products (institutional outputs/outcomes), and
  • impacts on societal health.

While impact evaluation is a subcomponent of product evaluation in the CIPP model, given the emphasis of social accountability in medical education on impact in practice and improvement in public health, we treated impacts as a separate theme in our analysis. Additionally, we identified subthemes within each theme, as described below and depicted in Figure 2. A selection of quotes to illustrate the themes and subthemes is provided in Table 2.

 

공유 가치
Shared values

4개의 프레임워크는 모두 4가지 핵심 사회적 가치(관련성, 품질, 효과성, 형평성)를 강조합니다.3,16,18,19 이러한 광범위한 가치는 CIPP 모델의 모든 구성 요소에 걸쳐 있습니다. 일반적으로 핵심 사회적 가치는 맥락(프로그램 목표), 투입물(행동), 프로세스(활동), 결과물(기관의 산출물/결과)에 정보를 제공하기 위한 사회적 책무성의 개념적 이상과 잘 의도된 속성을 나타냅니다. 이는 행동 지향적이며 사회적 요구의 파악에 기반을 두고 있습니다. 이는 교육, 연구 및 서비스 전반에 걸쳐 의학교육 프로그램 활동을 안내하기 위한 것입니다.18 
All 4 frameworks emphasized the 4 core social values (relevance, quality, effectiveness, and equity).3,16,18,19 These far-reaching values extend across all components of the CIPP model. Generally, the core social values refer to the conceptual ideals and well-intended attributes of social accountability intended to inform context (program objectives), inputs (actions), processes (activities), and products (institutional outputs/outcomes). They are action oriented and grounded in the identification of societal needs. They are intended to guide medical education program activities in education, research, and service across the training continuum.18

핵심 사회적 가치는 의과대학이 사회적 책무성에 대한 진전을 평가하는 데 도움을 주기 위해 1995년 WHO3 에 의해 처음 개념화되었으며, 이후 후속 프레임워크에서 적용되었습니다.16,18-20,23,44-70

  • 관련성은 의학교육 프로그램이 교육, 연구 및 봉사 활동에서 체계적인 접근법을 사용하여 인구, 지역사회 또는 국가의 우선적인 건강 요구 또는 우려를 해결한다는 것을 의미합니다.3,16,18,19
  • 근거에 기반하고 포괄적이며 문화적으로 민감최상의 진료를 개인에게 제공하는 것을 말합니다.3,16,18,19
  • 효과성은 보건의료 자원(비용)을 활용하고, 자원을 최대한 활용하면서 공중보건에 가장 큰 영향을 미치는 것을 말합니다.3,16,18,19
  • 형평성보편적 접근성을 의미하며 모든 개인이 양질의 보건의료에 접근할 수 있도록 노력하는 것을 말합니다.3,16,18,19 

The core social values were originally conceptualized in 1995 by the WHO3 as a means to help medical schools evaluate their progress in addressing social accountability and have since been adapted by subsequent frameworks.16,18–20,23,44–70

  • Relevance implies that a medical education program addresses priority health needs or concerns of the population, community, or nation using a systematic approach in education, research, and service activities.3,16,18,19
  • Quality refers to providing individuals with the best possible care that is evidence based, comprehensive, and culturally sensitive.3,16,18,19
  • Effectiveness refers to the utilization of health care resources (costs) and ensuring that the greatest impact on public health is achieved while making the best use of resources.3,16,18,19
  • Equity refers to universal access and striving to ensure that all individuals have access to quality health care.3,16,18,19

이러한 핵심 사회적 가치들 간의 상호관계는 "지역사회 또는 국가의 필요와 관련이 있고, 비용 효율적이며, 공평한 양질의 진료를 제공하는 의료 시스템을 구축"하겠다는 보편적인 사회적 약속을 나타냅니다.3 교육, 연구, 봉사 및 보건 정책에서 의과대학 활동은 이러한 필요를 반영해야 하며, 인구의 우선적인 건강 요구와 관련되고, 이에 대응하며, 예측해야 합니다.3,16,18,19 
The interrelationship between these core social values represents a universal social commitment to “building a health care system that is relevant to the needs of the community or nation and provides high-quality care that is cost-effective and equitable.”3 Medical school activities in education, research, and service as well as health policies must be reflective of these needs—they must relate to, respond to, and anticipate priority health needs of the population.3,16,18,19

핵심 사회적 가치에 더하여, 포함된 3개의 프레임워크16,18,19 는 전문직업성의 가치와 윤리, 팀워크, 문화적 역량, 리더십, 의사소통, 평생 학습, 근거 기반 실천 등의 역량을 강조했습니다. 캐나다의 맥락에서는 학문적 자유와 임상 자율성의 가치도 강조되었습니다.16 
In addition to the core social values, 3 of the included frameworks16,18,19 emphasized the value of professionalism as well as the following competencies: ethics, teamwork, cultural competence, leadership, communication, lifelong learning, and evidence-based practice. In the Canadian context, the values of academic freedom and clinical autonomy were also highlighted.16

CIPP 모델
CIPP model

맥락.
Context.

맥락은 CIPP 모델의 첫 번째 구성 요소입니다. 프레임워크 전반에 걸쳐 반복적으로 등장하는 하위 주제에는 사명 선언문, 지역사회 파트너십, 보건의료 정책에 대한 적극적인 기여가 포함됩니다. 
Context is the first component in the CIPP model. Recurring subthemes that emerged across the frameworks included mission statements, community partnerships, and active contributions to health care policy.

[기관 또는 프로그램 사명 선언문, 의무, 정책, 목적 및 목표]는 사회적 책무성의 핵심 사회적 가치와 사회적 건강 요구 충족에 대한 명시적 약속을 반영해야 합니다.3,16,18,19 이러한 선언문은 공개적으로 게시하고 일반 대중이 쉽게 접근할 수 있도록 해야 합니다.16 또한 의과대학의 사명 선언문과 교육, 연구, 봉사 활동의 내용과 맥락의 특수성은 해당 기관이 서비스를 제공하는 지역사회 및 국가의 현재 및 예상되는 우선적 건강 요구 또는 관심사에 영감을 받고 이에 부합해야 합니다.3,19 이러한 사명 선언문은 요구 평가의 역할을 하며 기관의 교육, 연구, 봉사 활동이 사회에 대한 사회적 의무와 헌신을 입증하도록 안내하기 위한 것입니다.3 
Institutional or program mission statements, mandates, policies, objectives, and/or goals must reflect the core social values of social accountability and the explicit commitment to meeting societal health needs.3,16,18,19 These statements should be posted publicly and made easily accessible to the general population.16 Additionally, the content and context specificity of a medical school’s mission statement and activities in education, research, and service should be inspired by and aligned with the current and anticipated priority health needs or concerns of the community and/or nation the institution serves.3,19 These mission statements serve as needs assessments and are intended to guide institutions’ education, research, and service activities to demonstrate their social obligation and commitment to society.3

지역 보건 시스템 및 기타 이해관계자들과 효과적인 지역사회 파트너십을 개발하는 것도 중요합니다.3,16,18,19 의과대학은 다른 이해관계자들과 협력하여 우선순위를 설정하고 현재 및 미래의 보건 요구를 파악하면 그 효과를 개선할 가능성이 더 높습니다.18 지역사회는 모든 의과대학의 주요 이해관계자 역할을 합니다.19 따라서 학교는 보건의료 정책, 계획, 재정을 담당하는 지역 이해관계자들과 협력하여 우선순위 건강 요구와 최적의 환자 치료에 필요한 서비스 및 자원을 파악하는 것이 필수적입니다.3,16,18,19 관련 보건의료 기관, 전문가 그룹, 정부, 소비자, 시민사회와의 파트너십은 보건 계획, 정책 개발, 보건의료 전달 및 평가에 대한 공동 작업을 촉진하고 장려할 수 있습니다.16 
Developing effective community partnerships with local health systems as well as other stakeholders is also important.3,16,18,19 Medical schools are more likely to improve their effectiveness if they work collaboratively with other stakeholders to establish priorities and identify current and future health needs.18 The local community serves as the primary stakeholder of all medical schools.19 Therefore, it is imperative that schools work in partnership with local stakeholders responsible for health care policy, planning, and finance to identify priority health needs as well as services and resources required for optimal patient care.3,16,18,19 Partnerships with affiliated health care organizations, professional groups, governments, consumers, and civil society could facilitate and encourage shared work on health planning, policy development, health care delivery, and evaluation.16

의료 교육 프로그램도 보건의료 시스템을 형성하는 데 중요한 역할을 합니다. 지역사회 파트너십은 의과대학이 보건의료 정책에 적극적으로 기여할 수 있는 수단이 될 것입니다.3,16,18,19 의과대학은 변화의 촉매제 역할을 하고 보건의료 계획 및 전달, 정책 개발의 지속 가능성과 평가에 적극적으로 기여해야 합니다.16 
Medical education programs also play an important role in shaping the health care system. Community partnerships would serve as a means for medical schools to actively contribute to health care policy.3,16,18,19 Medical schools should act as catalysts of change and actively contribute to the sustainability and evaluation of health care planning and delivery, and policy development.16

입력.
Inputs.

인풋은 목표한 목표를 달성하기 위해 프로그램에서 취하는 조치입니다. 이러한 행동은 기관/프로그램의 의무와 사명 선언문에 의해 동기가 부여되며, 사회적 책무성의 핵심 사회적 가치를 반영합니다. 프레임워크 전반의 하위 주제에는 다음이 포함됩니다. 

  • 모집 및 선발(학생, 교수진, 교직원)의 다양성 및 형평성,
  • 지역사회 인구 건강 프로필

Inputs are actions taken by programs to meet targeted goals. These actions are motivated by institution/program mandates and mission statements, and they reflect the core social values of social accountability. Subthemes across frameworks included

  • diversity and equity in recruitment and selection (students, faculty, and staff) and
  • community population health profiles.

두 가지 프레임워크는 학생 모집 및 선발에서 다양성과 형평성의 중요성을 강조했습니다.18,19 핵심 사회적 가치와 사명 선언문에 포함된 사회적 약속을 이행하기 위해 의과대학은 모집 및 선발 정책을 조정하여 지원자의 다양성을 높여 소외된 인구와 취약 계층을 포함해야 합니다.18 학생들은 인종 및 민족, 가시적 소수 또는 원주민 신분, 사회경제적 지위, 성별 및 성적 지향, 종교를 포함한 일반 인구의 인구통계를 반영하고 농촌 및 소외된 지역사회 등 기타 취약 계층의 특성을 반영해야 합니다.19 또한 학교는 사회적 약자 지원자에게 동등한 기회를 보장하기 위해 지원 메커니즘(예: 재정 지원, 상담 서비스)뿐만 아니라 소외 계층을 위한 전략적 파이프라인 또는 할당제를 시행해야 합니다.19 의과대학은 또한 의학, 의료 서비스 전달 및 사회과학 부서의 교수진이 커리큘럼과 프로그램 의사 결정에 대표성을 갖고 참여하도록 해야 합니다.18 마지막으로 의과대학은 WHO 보고서에서 권고한 대로 일반의로서 진료할 가능성이 더 높은 학생을 입학시켜야 합니다.3 
Two frameworks emphasized the importance of diversity and equity in the recruitment and selection of students.18,19 To meet the social commitments embedded within the core social values and mission statements, medical schools must adapt their recruitment and selection policies to increase the diversity of accepted applicants to include individuals from underrepresented populations and disadvantaged groups.18 Students should reflect the demographics of the general population—including race and ethnicity, visible minority or indigenous status, socioeconomic status, gender and sexual orientation, and religious affiliation—and reflect other disadvantaged groups, such as rural and underserved communities.19 Additionally, schools should implement strategic pipelines and/or quotas for underrepresented groups as well as support mechanisms (e.g., financial aid, counseling services) to ensure equal opportunities for socially disadvantaged applicants.19 Medical schools should also ensure that faculty from medicine, health service delivery, and social science divisions are represented and involved in the curriculum and in programmatic decision making.18 Lastly, medical schools should matriculate students who are more likely to practice as generalists, as recommended by the WHO report.3

세 가지 프레임워크의 또 다른 핵심 주제는 의과대학이 대상 지역사회 또는 국가의 서비스 격차뿐만 아니라 인구 요구를 파악해야 한다는 것입니다.3,18,19 학교는 잘 정의된 인구 건강 연구와 포괄적인 지역사회 인구 건강 프로필 개발을 통해 이러한 요구를 파악하기 시작할 수 있습니다.3,18 이러한 프로필에는 지역사회의 사회 인구학적 및 지정학적 구성과 인구 건강 위험, 건강의 사회적 결정 요인, 서비스 접근 장벽이 반영되어야 합니다. 
Another central theme in 3 frameworks was the need for medical schools to identify population needs as well as service gaps of a targeted community and/or nation.3,18,19 Schools can begin to identify these needs through well-defined population health research and the development of a comprehensive community population health profile.3,18 These profiles must reflect the community’s sociodemographic and geopolitical composition as well as population health risks, social determinants of health, and barriers to accessing services.

프로세스.
Processes.

과정에는 교육 활동의 전체 스펙트럼이 포함됩니다.

  • 커리큘럼 내용 및 구조,
  • 교수법,
  • 지역사회 기반 임상 교육 기회 및 지역 인구와 의료 서비스 소외 지역에 대한 학습 노출,
  • 학습 평가,
  • 지속적인 전문성 개발,
  • 평가 시스템 등 

Processes include the entire spectrum of educational activities:

  • curricular content and structure;
  • teaching methods;
  • community-based clinical training opportunities and learning exposures to local populations and underserviced areas;
  • learning assessments;
  • continuing professional development; and
  • evaluation systems.

프레임워크 전반에 걸쳐 반복적으로 등장하는 하위 주제에는 다음이 포함되었습니다. 

  • 커리큘럼 활동
  • 지역사회 기반 임상 교육 기회 및 학습 노출

Recurring subthemes that emerged across frameworks included

  • curricular activities as well as
  • community-based clinical training opportunities and learning exposures.

의과대학은 우선순위 공중보건 요구를 해결하는 방향으로 교과과정 활동을 진행해야 합니다.3,16,18,19 교과과정 내용과 구조는 학생 중심 패러다임으로 접근해야 하며, 건강의 사회적 결정요인, 공중보건 위험, 인구, 지역사회, 국가의 지정학적, 사회인구학적, 역학적 특수성을 포함해야 합니다.3,19 또한 학교의 교과과정 활동은 강력한 지속적인 전문성 개발 프로그램을 통해 결함 학생, 졸업생, 직원에게 평생 학습 기회를 지원해야 합니다.16,18 
Medical schools must direct their curricular activities toward addressing priority public health needs.3,16,18,19 Curricular content and structure should be approached using a student-centered paradigm and must include the social determinants of health; public health risks; and the geopolitical, sociodemographic, and epidemiological specificities of a population, community, and/or nation.3,19 Additionally, schools’ curricular activities should support lifelong learning opportunities for faulty, graduates, and staff through the availability of robust continuing professional development programs.16,18

지역사회 기반 임상 교육 기회와 학습 노출인구 접근법을 사용하여 설계되어야 합니다.3,19 의과대학은 일차 진료를 장려하고 일차 진료 실습에 대한 학습 기회와 노출을 제공해야 합니다.3,16,19 또한 학교는 지역사회의 건강 요구와 관련된 종단적 지역사회 기반 학습 경험을 제공해야 합니다.3,19 마지막으로 학교는 농촌 의료 환경에서 학습 기회를 제공하고 불우하고 소외된 그룹에 노출되도록 해야 합니다.3,19 
Community-based clinical training opportunities and learning exposures should be designed using a population approach.3,19 Medical schools should promote primary care and provide learning opportunities and exposure to primary care practices.3,16,19 Additionally, schools should provide longitudinal community-based learning experiences that are relevant to the community’s health needs.3,19 Lastly, schools should provide learning opportunities in rural health care settings as well as exposure to disadvantaged and underserved groups.3,19

제품.
Products.

제품 평가는 프로그램 졸업생들의 유용성을 의미합니다. 프레임워크 전반에 걸쳐 반복적으로 등장하는 하위 주제에는 다음이 포함되었습니다. 

  • 의사 자원 계획,
  • 품질 보증,
  • 프로그램 평가 및 인증

Product evaluation refers to the usability of a program’s graduates. Recurring subthemes that emerged across frameworks included

  • physician resource planning,
  • quality assurance, and
  • program evaluation and accreditation.

4개의 프레임워크 모두 의사 자원 계획의 중요성을 강조했습니다. 의과대학은 학생의 적절한 구성을 결정하고 교육하며, 사회적 요구를 충족하는 데 필요한 졸업생의 분포, 배치 및 유지를 결정하는 데 적극적으로 참여해야 합니다.3,16,18,19 또한 학교는 일차 진료 의사를 위한 지역 취업 기회를 보장해야 합니다.18,19 
All 4 frameworks emphasized the importance of physician resource planning. Medical schools should be actively involved in determining and educating the right composition of students and in determining the distribution, deployment, and retention of graduates necessary to meet social needs.3,16,18,19 Additionally, schools must ensure local employment opportunities for primary care physicians.18,19

또 다른 핵심 주제는 프로그램 평가 및 인증의 중요성이었습니다.3,16,18,19 인증 기준 및 프로세스는 사회적 책무성 원칙을 통합해야 합니다.3,16,18,19 평가 및 인증은 정기적으로 수행되어야 하며,3,19 결과는 공개적으로 이용 가능하고 제도 개선에 사용되어야 합니다.19 또한 평가 및 인증 팀은 정책 입안자, 보건 전문가, 지역사회 구성원을 포함한 이해관계자를 폭넓게 대표할 수 있어야 합니다.3,16,18,19 
Another central theme was the importance of program evaluation and accreditation.3,16,18,19 Accreditation standards and processes should incorporate social accountability principles.3,16,18,19 Evaluation and accreditation must be conducted at regular intervals,3,19 and the results should be made publicly available and used for institutional improvement.19 Additionally, evaluation and accreditation teams should be widely representative of stakeholders, including policymakers, health professionals, and community members.19

마지막으로, 교육, 연구 및 서비스 제공에서 지속적인 품질 보증 프로세스를 수용하는 것의 중요성은 모든 프레임워크에서 강조되었습니다.3,16,18,19 이 프로세스는 투명해야 하며, 교육 개선을 촉진하기 위해 잘 정의된 표준을 사용하여 안내해야 합니다.3,19 또한 졸업생 역량을 정기적으로 평가하고, 잘 정의된 교육 기준을 반영하여 치료의 질을 보장하고, 졸업생이 변화하는 공중 보건 요구를 충족하는 데 필요한 기술을 갖추고 실무에 투입되도록 해야 합니다.3,16,18,19 
Lastly, the importance of embracing a continuous quality assurance process in education, research, and service delivery was emphasized across all frameworks.3,16,18,19 This process should be transparent and guided using well-defined standards to promote educational improvements.3,19 Additionally, graduate competencies must be assessed regularly and reflect well-defined educational standards to ensure quality of care and that graduates enter practice equipped with the skills required to meet changing public health needs.3,16,18,19

영향.
Impacts.

사회적 책무성의 전제는 의학교육 프로그램의 목적과 실천이 사회적 필요를 파악하는 데서 시작하여 그러한 필요를 충족하는 것으로 마무리될 것을 요구합니다.18 영향 평가는 제품 평가의 일부입니다.

  • 프레임워크 전반에서 강조되는 공통 주제는 지역사회 건강 결과의 전반적인 개선입니다.3,16,18,19
  • 또 다른 공통 주제는 지역사회 건강 위험과 지역사회 질병의 이환율 및 사망률의 감소 및 예방입니다.16,19 

The premise of social accountability requires that the purpose and practices of medical education programs commence in the identification of societal needs and conclude in meeting those needs.18 Impact evaluation is part of product evaluation.

  • A common theme highlighted across frameworks was overall improvement in community health outcomes.3,16,18,19 
  • Another common theme was reduction and prevention of community health risks and morbidity and mortality of community diseases.16,19

사회적 영향을 효과적으로 평가하기 위해 의과대학은 교육 연속체를 포괄하고 졸업생이 실제로 미치는 영향에 초점을 맞춘 표준을 개발해야 합니다.19 졸업생이 질병 부담을 줄이고 그들이 봉사하는 지역 사회의 건강을 개선하는 정도를 평가하는 지표를 개발해야 합니다.16 의과대학은 활동 결과가 지역 사회 건강에 긍정적인 영향을 미친다는 것을 입증할 수 있어야 합니다.3,16,18,19 졸업생이 지역사회 건강 위험과 지역사회 질병의 이환율 및 사망률을 줄여 공중 보건에 긍정적인 사회적 투자 수익을 얻도록 해야 할 의무가 있습니다.16,19 
To evaluate societal impacts effectively, medical schools must develop standards that span the educational continuum and focus on impacts of graduates in practice.19 They must develop metrics to assess the extent to which their graduates reduce the burden of illness and improve the health of the communities they serve.16 Medical schools must be able to demonstrate that the outcomes of their activities have positive impacts on community health.3,16,18,19 They have an obligation to ensure their graduates have a positive social return on investment to public health by reducing community health risks and the morbidity and mortality of community diseases.16,19

토론
Discussion

이 검토에서는 CIPP 평가 모델을 사용하여 4개의 대규모 사회적 책무성 정책 프레임워크3,16,18,19 에서 주요 주제와 지표를 확인했습니다.35 의학교육 문헌에서 정책 문서 전반에 걸쳐 사회적 책무성 지표를 식별하는 데 CIPP 모델이 사용된 적은 없지만, 이 검토는 의학교육의 사회적 책무성 평가를 위한 초기 운영 구성 개발에서 그 유용성을 입증합니다. 포함된 프레임워크에서 탐구된 주제는 의학교육의 광범위한 사회적 책무성 문헌과 일치합니다. 그러나 CIPP 모델은 책무성 시스템을 강화하기 위한 의학교육 프로그램의 평가 프레임워크를 제공합니다.35-38 또한, 이 검토에서는 책임에 대한 근본적인 질문도 의도치 않게 다루고 있습니다: 누가, 무엇에 대해, 누구에게, 어떤 수단을 통해 책임을 져야 하는가.7,10,15 이러한 질문은 책무성을 이해하는 데 중요하며, 의과대학의 사회적 책무성을 어떻게, 어떤 방식으로 평가할 수 있도록 사회적 책무성 프레임워크를 운영하는 데 도움이 될 수 있습니다. 
This review identified major themes and indicators across 4 large-scale social accountability policy frameworks3,16,18,19 using the CIPP evaluation model.35 The CIPP model has not been used previously in the medical education literature to identify social accountability indicators across policy documents, but this review provides evidence of its utility in the development of initial operational constructs to evaluate social accountability in medical education. The themes explored in the included frameworks are consistent with the broader social accountability literature in medical education. However, the CIPP model provides an evaluation framework for medical education programs to strengthen their accountability systems.35–38 Additionally, this review also inadvertently addresses the fundamental questions of accountability: Who is held to account, for what, to whom, and through what means?7,10,15 These questions are critical to understanding accountability and can be used to help operationalize social accountability frameworks to better evaluate how and in which ways medical schools are socially accountable.

이 검토는 주로 의과대학의 사회적 책무성에 초점을 맞추었지만, 사회적 책무성은 역동적인 과정이라는 점을 인식하는 것이 중요합니다. 이는 시민, 정부, 수련기관, 의료 교육자/공급자 간의 협력 관계를 통해 사회적 건강 요구를 체계적으로 파악하고 우선순위를 정하여 해결하는 과정입니다.3,72 의과대학의 사회적 책무성을 측정하고 체계적으로 평가하려면 개념적, 운영적 복잡성을 포착할 수 있는 강력한 평가 모델을 사용해야 합니다. 인증은 이러한 많은 문제를 해결할 수 있지만, 의과대학이 직장에서 일할 수 있는 유능한 졸업생을 배출하도록 하는 다른 목적도 있습니다. 이 경우 학교는 인증기관에 책임을 집니다. 캐나다와 호주는 인구의 우선적인 건강 문제를 해결하기 위한 의과대학의 노력을 평가하기 위한 수단으로 공식적인 사회적 책무성 기준을 인증 절차에 통합했습니다.19,72,73 이는 긍정적인 발전이지만, 사회적 책무성 결과에 대해 더 폭넓게 생각하고 교육 투입물, 산출물, 영향 간의 의미 있는 관계를 확립해야 합니다.19,73 
While this review focused primarily on social accountability of medical schools, it is important to acknowledge that social accountability is a dynamic process. It represents a collaborative relationship between citizens, government, training institutions, and health care educators/providers to systematically identify, prioritize, and address societal health needs.3,72 The measurement and systematic evaluation of social accountability in medical schools requires the use of a robust evaluation model to capture its conceptual and operational complexities. While accreditation may address many of these issues, it often serves a different purpose—ensuring medical schools produce competent graduates for the workplace. In this instance, schools are accountable to the accreditors. Canada and Australia have incorporated formal social accountability standards into their accreditation processes as a means to evaluate a medical school’s commitment to addressing the priority health concerns of the population.19,72,73 While this is a positive advancement, we need to continue to think about social accountability outcomes more broadly and establish meaningful relationships between educational inputs, outputs, and impacts.19,73

그러나 의과대학이 사회적 필요를 충족시킨다는 가정에 대한 연구가 부족합니다. Boelen에 따르면,72 사회적 책무성을 다하는 의과대학은 1%에 불과한 반면, 사회적 대응을 하는 의과대학은 9%, 사회적 책무성을 다하는 의과대학은 90%에 달한다고 합니다. 투명성 및 책임성 이니셔티브가 공공 서비스 개선을 위한 핵심 전략으로 부상했지만, 이러한 이니셔티브와 공중 보건에 미치는 영향 사이의 관계는 거의 알려지지 않았습니다.74 이 문제는 의학교육에만 국한된 것이 아닙니다.75 졸업생들이 실제로 지역사회에 미치는 사회적 영향을 평가 및 입증하고, 이론과 실무 사이의 연결 고리를 구축해야 할 필요성이 있습니다. 이러한 입증은 사회적 책무성에 대한 선한 의도와 헌신을 공개적으로 보여주는 것이 아니라 개념 증명에 가까워지고 있습니다.75 개별 의과대학이 사회적 책무성을 다하기 위한 진전을 확인하고자 하는 문헌이 증가하고 있습니다(체계적인 검토는 Reeve 외76 참조). 의과대학의 노력의 예로는 입학 절차를 통한 접근성 확대,77-82 건강의 사회적 결정 요인을 반영한 커리큘럼 개혁,83-86 지역사회 기반 임상 교육 기회 및 학습 노출,87-90 학습자의 위치 등이 있습니다.91-97 
There is, however, an understudied assumption that medical schools meet societal needs. According to Boelen,72 only 1% of medical schools are socially accountable, whereas 9% of medical schools are socially responsive and 90% are socially responsible. While transparency and accountability initiatives have emerged as a key strategy for improving public services, the relationship between these initiatives and their impacts on public health remains largely unknown.74 This issue is not specific to medical education.75 There is a need to evaluate and demonstrate the social impacts graduates have in practice on communities and establish a link between theory and practice. This demonstration becomes less about providing public displays of good intentions and commitment to social accountability and more about proof of concept.75 A growing body of literature seeks to affirm the progress of individual medical schools toward becoming socially accountable (see Reeve et al76 for a systematic review). Some examples of medical schools’ efforts include widening access through admissions processes,77–82 curricular reforms reflecting social determinants of health,83–86 community-based clinical training opportunities and learning exposures,87–90 and location of learners.91–97

일부 의학교육 환경에서 사회적 책무성 평가의 진전이 계속 확대되고 있지만,98 사회적 책무성 이니셔티브가 사회 건강에 미치는 영향은 거의 알려지지 않았습니다.99,100 그러나 소수의 경험적 논문에서 환자 건강 결과를 의사 교육 및 성과와 연관시키고 있으며101,102 일부 논평103-105에서는 의학교육 프로그램이 공중 보건 요구에 미치는 영향을 더 잘 이해하기 위해 국가 임상 데이터 세트를 사용하여 졸업생 결과를 환자 영향과 연결해야 한다는 점을 강조하고 있습니다. 
While progress in evaluating social accountability continues to expand in select medical education settings,98 the extent to which social accountability initiatives impact societal health remains largely unknown.99,100 However, a small number of empirical papers associate patient health outcomes with physician training and performance101,102 and some commentaries103–105 emphasize the need to link graduate outcomes with patient impacts using national clinical datasets to better understand the effects medical education programs have on public health needs.

한계점
Limitations

이 검토는 이전 연구를 확장한 것입니다.3,16,18-20,23,44-70 가능한 모든 사회적 책무성 지표의 포괄적인 목록을 제공하지는 않습니다. 여기에 제시된 주제와 지표는 주요 출처의 사회적 책무성 정책 프레임워크에 국한되어 있으며, 품질 평가에 사용되는 지표를 반드시 포함하지는 않습니다. 이 검토에서는 글로벌 건강 격차에 대한 우려가 커지는 등 최근의 글로벌 건강 운동은 다루지 않았습니다. 또한 CIPP 모델은 교육에 대한 하향식 시스템 접근 방식을 가정하여 교육 투입물이 제품으로 전환된다고 가정합니다. 또한 이 검토는 주로 의학교육에 초점을 맞추고 있으며, 사회적 책무성의 다른 상호 연관되고 상호 의존적인 프로그램 활동(예: 연구 및 서비스)은 포함하지 않았습니다. 이러한 관계를 더 자세히 조사하고 의과대학이 지역 보건 요구를 해결하고 이에 대응하는지 여부를 결정하기 위해서는 추가 연구가 필요합니다.  
This review extends earlier work.3,16,18–20,23,44–70 It does not provide a comprehensive list of all possible social accountability indicators. The themes and indicators presented here are limited to primary source social accountability policy frameworks and are not necessarily inclusive of metrics used to assess quality. This review does not address more recent global health movements, for instance, the growing concerns regarding global health disparities. Additionally, the CIPP model assumes a top-down systems approach to education, whereby educational inputs are turned into products. This review is also primarily on medical education, not other interrelated and interdependent program activities of social accountability (i.e., research and service). Further research is needed to examine these relationships in more detail and determine whether medical schools address and respond to local health needs.

결론
Conclusion

이 검토는 확립된 프로그램 평가 모델과 4가지 대규모 사회적 책무성 정책 프레임워크의 증거를 연결하여 의학교육 연속체 전반에 걸쳐 지표를 생성할 수 있도록 합니다. 프로그램 평가 모델은 교육기관이 원하는 목표와 목적을 향해 나아가는 과정을 모니터링하기 위한 체계적이고 쉽게 이해할 수 있는 실용적인 가이드를 제공합니다. 그러나 의과대학이 사회적 의무를 다하려고 노력하더라도 이러한 행동이 공중 보건에 긍정적인 영향을 미칠 것이라는 보장은 없습니다.3 
This review links an established program evaluation model and evidence from 4 large-scale social accountability policy frameworks, which may lead to the creation of indicators across the medical education continuum. Program evaluation models provide a systematic and easily understood practical guide for monitoring the progress of an institution toward desired goals and objectives. However, even when medical schools attempt to fulfill their social obligations, there is no guarantee that these actions will positively impact public health.3

사회적 책무성을 평가하는 일은 복잡합니다.65 의학교육 프로그램의 질을 평가하는 대부분의 이전 문헌은 주로 투입물과 과정에 초점을 맞추었습니다.72 사회적 책무성이 더욱 강조됨에 따라, 우리 커뮤니티는 [교육 투입물과 과정]에서 [산출물과 영향]에 초점을 전환해야 합니다. 다음의 것들 사이에 의미 있는 관계를 설정할 필요가 있습니다.32,74

  • 투입물(누가, 어디서, 어떤 교육을 받았는지),
  • 산출물(졸업생이 실제 진료 현장에서, 어떤 의료 전문 분야에서, 어떤 일을 하는지),
  • 영향(졸업생의 활동이 인구 건강을 어떻게 개선하는지) 

이러한 관계를 설정하기 시작하는 한 가지 방법으로 CIPP 프로그램 평가 모델을 사용할 것을 제안합니다. 
The task of evaluating social accountability is complex.65 Most of the previous literature assessing the quality of medical education programs has focused predominantly on inputs and processes.72 As more emphasis is placed on social accountability, it is imperative that we as a community shift our focus from educational inputs and processes to products and impacts. There is a need to establish meaningful relationships between program

  • inputs (who is trained and from where),
  • products (what graduates do in practice, in what medical specialty, and where), and
  • impacts (how graduates’ activities improve population health).32,74 

We suggest a way to begin to establish these links is through the use of the CIPP program evaluation model.


Acad Med. 2020 Dec;95(12):1945-1954. doi: 10.1097/ACM.0000000000003731.

Social Accountability Frameworks and Their Implications for Medical Education and Program Evaluation: A Narrative Review

Affiliations collapse

1C. Barber is a PhD candidate, School of Health Professions Education (SHE), Faculty of Health, Medicine and Life Sciences, Maastricht University, Maastricht, the Netherlands; ORCID: https://orcid.org/0000-0003-3605-8485.

2C. van der Vleuten is professor, Department of Educational Development and Research, Faculty of Health, Medicine and Life Sciences, Maastricht University, Maastricht, the Netherlands.

3J. Leppink is senior lecturer in medical education, Hull York Medical School, University of York, York, United Kingdom.

4S. Chahine is associate professor, Faculty of Education, Queen's University, Kingston, Ontario, Canada; ORCID: https://orcid.org/0000-0003-0488-773X.

PMID: 32910000

DOI: 10.1097/ACM.0000000000003731

Free article

Abstract

Purpose: Medical schools face growing pressures to produce stronger evidence of their social accountability, but measuring social accountability remains a global challenge. This narrative review aimed to identify and document common themes and indicators across large-scale social accountability frameworks to facilitate development of initial operational constructs to evaluate social accountability in medical education.

Method: The authors searched 5 electronic databases and platforms and the World Wide Web to identify social accountability frameworks applicable to medical education, with a focus on medical schools. English-language, peer-reviewed documents published between 1990 and March 2019 were eligible for inclusion. Primary source social accountability frameworks that represented foundational values, principles, and parameters and were cited in subsequent papers to conceptualize social accountability were included in the analysis. Thematic synthesis was used to describe common elements across included frameworks. Descriptive themes were characterized using the context-input-process-product (CIPP) evaluation model as an organizational framework.

Results: From the initial sample of 33 documents, 4 key social accountability frameworks were selected and analyzed. Six themes (with subthemes) emerged across frameworks, including shared values (core social values of relevance, quality, effectiveness, and equity; professionalism; academic freedom and clinical autonomy) and 5 indicators related to the CIPP model: context (mission statements, community partnerships, active contributions to health care policy); inputs (diversity/equity in recruitment/selection, community population health profiles); processes (curricular activities, community-based clinical training opportunities/learning exposures); products (physician resource planning, quality assurance, program evaluation and accreditation); and impacts (overall improvement in community health outcomes, reduction/prevention of health risks, morbidity/mortality of community diseases).

Conclusions: As more emphasis is placed on social accountability of medical schools, it is imperative to shift focus from educational inputs and processes to educational products and impacts. A way to begin to establish links between inputs, products, and impacts is by using the CIPP evaluation model.

 

인지과학의 인사이트를 임상 술기 교육에 사용하기: AMEE Guide No. 155 (Med Teach, 2023)
Using insights from cognitive science for the teaching of clinical skills: AMEE Guide No. 155
Dario Cecilio-Fernandes, Rakesh Patel & John Sandars

 

 

소개
Introduction

학생들이 임상 술기를 효과적으로 습득할 수 있도록 지원하는 것은 보건 전문직 교육(HPE) 교육자들에게 점점 더 어려운 과제가 되고 있습니다. 졸업 시점에 습득해야 하는 임상 술기의 수가 증가하고 시뮬레이션 및 임상 업무 환경에서 임상 술기를 배울 수 있는 기회가 점점 더 제한되고 있기 때문에 이러한 도전은 점점 더 커지고 있습니다. 특정 학부 강의 계획서에 따라 학생들이 습득해야 하는 임상 술기는 다양하며, 이러한 술기는 캐뉼레이션과 같은 간단한 절차적 술기부터 고급 생명유지술의 경우 보다 복잡한 진단 및 관리 추론 술기까지 다양합니다(Faustinella and Jacobs 2018). 또한 교육자들은 시간이 지남에 따라 임상 술기 감퇴의 중요성에 대해 점점 더 인식하고 있습니다(세실리오-페르난데스 외. 2018; 살 외. 2021). 
Supporting students in the effective acquisition of their clinical skills is becoming more of a challenge for health professions education (HPE) educators. The challenge is a growing one because of the increasing number of clinical skills required to be learned at the point of graduation, and also because of the increasingly limited opportunities for learning clinical skills in simulation and clinical workplace settings. There are many clinical skills that students need to acquire depending on the specific undergraduate syllabus, and these skills range from simple procedural skills, such as cannulation, through to more complex diagnostic and management reasoning skills in the case of advanced life support (Faustinella and Jacobs 2018). In addition, educators are becoming increasingly aware of the importance of clinical skill decay over time (Cecilio-Fernandes et al. 2018; Sall et al. 2021).

이 가이드의 목적은 학부, 대학원 및 평생 의학 교육에서 임상 술기 교육에 대한 인지과학적 관점을 제공하는 것입니다. 이러한 관점은 임상 술기를 보다 효과적으로 습득하고 술기 쇠퇴를 최소화할 수 있는 교육 방법에 대한 유용한 증거 기반 통찰력을 제공할 수 있습니다. 이 가이드는 두 섹션으로 나뉩니다. 

  • 첫째, 이 가이드에서는 전문성 및 숙련도 개발의 본질과 임상 술기 개발의 인지과학적 근거를 포함한 이론적 토대를 논의하여 술기 감퇴를 줄이면서 임상 술기를 효과적으로 습득할 수 있도록 합니다.
  • 둘째, 실제적인 시사점을 논의하고 이론적 토대를 현재 실무에 어떻게 통합할 수 있는지에 대한 실제적인 예시를 제시합니다.

The aim of this Guide is to provide a cognitive science perspective on teaching clinical skills in undergraduate, postgraduate, and continuing medical education. This perspective can provide useful evidence-based insights into how teaching can ensure the more effective acquisition of clinical skills and also minimize skill decay. The guide is divided in two sections.

  • First, the guide discusses the theoretical foundations, including the nature of expertise and mastery development and the cognitive science basis of clinical skills development to ensure the effective acquisition of clinical skills, with reduced skill decay.
  • Second, we discuss the practical implications and present a practical illustration of how the theoretical foundations can be integrated into current practices.

이론적 토대
Theoretical foundations

임상 술기 습득의 인지 과학적 기반
The cognitive science basis of clinical skills acquisition

HPE 교육자가 직면한 첫 번째 과제 중 하나는 '임상 술기'의 정의에 포함되는 기술이 매우 다양하다는 것입니다.

  • 신체 검사 기술실습 기술부터
  • 의사소통 기술, 치료 기술, 임상적 추론 또는 진단적 의사 결정 기술에 이르기까지 그 범위가 다양합니다.

이러한 기술 중 일부는 기술적 또는 절차적 성격이 강한 반면 일부는 지적 또는 인지적 성격이 강하기 때문에 업무 유형에 따라 그 차이가 큽니다. 또한 임상 술기에 어떤 영역을 포함할지에 대한 합의나 합의가 거의 없기 때문에 임상 술기 개발을 위한 최적의 교수 학습 전략에 대한 다양한 관점이 존재합니다(Michels Evans 및 Blok 2012). 그러나 인지과학적 관점에서 볼 때 모든 임상 술기의 기본 과정은 유사하며 선언적 지식과 절차적 지식의 효과적인 통합이 필요합니다.
One of the first challenges facing HPE educators is the wide variety of skills than come under the definition of ‘clinical skill’. The range is

  • from physical examination skills and practical skills through
  • to communication skills, treatment skills, and clinical reasoning or diagnostic decision-making skills.

The difference between all these skills in the type of task is significant since some are more technical or procedural in nature whereas some are more intellectual or cognitive in nature. There is also little consensus or agreement around which domains to include within clinical skills, therefore there are a number of different perspectives around the optimal teaching and learning strategies for developing clinical skills (Michels Evans and Blok 2012). However, from a cognitive science perspective, the underlying process of all clinical skills is similar and requires the effective integration of declarative and procedural knowledge.

선언적 지식은 사실이나 사건('무엇을 아는 것')을 말합니다(Anderson 1982). 임상 술기의 경우 선언적 지식에는 생의학 분야의 관련 사실 또는 개념적 지식과 실제 술기 수행 방법에 대한 기술적 측면이 포함됩니다. 

  • 예를 들어, 혈압 측정에 대한 선언적 지식은 '수축기'와 '이완기'라는 용어의 사실적 차이점과 혈액이 몸속으로 펌프질되는 여러 단계에 대한 '수축기'와 '이완기'의 개념적 차이점을 알아야 합니다. 마찬가지로, 혈압을 측정할 때 환자와 기계의 위치를 가장 잘 잡는 방법과 관련된 개념적 지식과 함께 사용할 장비에 대한 사실적 지식도 필요합니다.

Declarative knowledge refers to facts or events (‘knowing what’) (Anderson 1982). In the case of clinical skills, declarative knowledge includes relevant factual or conceptual knowledge in the biomedical sciences and technical aspects about how to undertake the skill in practice. For example, declarative knowledge about measuring blood pressure requires knowing the factual differences between the terms ‘systolic’ and ‘diastolic’ and the conceptual difference between ‘systolic’ and ‘diastolic’ in terms of the different phases when blood is pumped around the body. Likewise, factual knowledge about the equipment to be used is also necessary, alongside conceptual knowledge related to how best to position the patient and the machine when taking blood pressure.

절차적 지식('방법 알기')은 선언적 지식으로 행동을 자동화하는 것을 의미하며, 반복적인 기술 연습을 통해 발전합니다(Anderson 1982). 

  • 예를 들어, 절차적 지식은 환자의 혈압을 측정하고 그 절차를 수행하면서 환자에게 설명할 수 있는 능력을 말합니다. 시간이 지남에 따라 지식과 기술 모두에 의해 술기가 자동화되면 실제로는 더 빠른 수행으로 이어집니다. 

이 개념은 시간이 지남에 따라 기술이 쇠퇴하는 것을 이해하는 데 필수적입니다. 절차적 지식은 일반적으로 시간이 지나도 유지되는 반면, 선언적 지식은 시간이 지나면 쇠퇴할 수 있으며 반복적으로 사용하지 않으면 잊어버릴 수 있습니다(Anderson 1982; Anderson 외. 2004). 기술을 숙달했더라도 일정 기간 사용하지 않으면 전문가가 최적의 성과를 발휘하는 데 여전히 어려움을 겪을 수 있으며, 이는 선언적 지식의 쇠퇴와 관련이 있을 가능성이 높습니다.
Procedural knowledge (‘knowing how’) refers to the automatizing of our actions with declarative knowledge and develops with repeated practice of the skill (Anderson 1982).

  • For example, procedural knowledge refers to taking a patient’s blood pressure and being able to explain the procedure to them whilst undertaking it. The automatization of the skill over time, due to both knowledge and skill, leads to faster performance in practice.

This concept is essential for understanding skill decay over time. Whereas procedural knowledge is usually maintained over time, declarative knowledge can decay and may be forgotten without repeated use of the skill (Anderson 1982; Anderson et al. 2004). Even though a skill has been mastered, after a period of disuse, professionals may also still face some difficulties to demonstrate optimal performance, and again is more likely related to the decay of declarative knowledge.

임상 술기를 포함한 효과적인 술기 습득은 교사가 주도하는 특정 교육을 통해 선언적 지식을 구성하고 유지하는 것으로 시작해야 하며, 여기에는 학습자가 처음에 술기를 시연하는 교사를 관찰하는 것이 포함될 수도 있고 포함되지 않을 수도 있습니다. 시간이 지남에 따라 연습하면 절차화라는 과정을 통해 선언적 지식을 절차적 지식으로 변환하여 기술 습득을 증가시킬 수 있습니다(Taatgen and Lee 2003). 절차화는 개인이 기술 수행을 자동화함에 따라 학습자의 정신적 노력이 시간이 지남에 따라 감소하고, 그 과정에서 경험하는 인지적 요구와 오류의 양이 감소하기 때문에 효과적입니다(Anderson 1982). 
Effective skill acquisition, including clinical skills, should begin with constructing and maintaining declarative knowledge through specific teacher-led instruction, which may or may not include observation by learners of the teacher demonstrating the skill in the beginning. Practice over time will increase skill acquisition by transforming declarative knowledge into procedural knowledge through a process known as proceduralization (Taatgen and Lee 2003). The proceduralization is effective because the mental effort for learners reduces over time as individuals automatize the performance of the skill, thereby also reducing the cognitive demand experienced and the amount of errors in the process (Anderson 1982).

인지 과학을 숙달 학습과 의도적 연습에 적용하기
Applying cognitive science to mastery learning and deliberate practice

숙달 학습의도적 연습은 현재 임상 술기 교육에서 절차화를 통한 자동화를 개발하는 데 널리 사용되고 있습니다. 그러나 이러한 접근법의 교육 활동 설계에 인지 과학 증거 기반 교육 전략을 통합하면 숙련도 저하를 방지하기 위해 숙련도 습득 및 숙련도 유지에 대한 효과를 크게 향상시킬 수 있습니다. 이러한 전략 중 다수는 전통적으로 지식의 개발 또는 구성과 관련이 있지만, 술기 개발에는 다양한 유형의 지식이 필요하기 때문에 임상 술기 교육에도 적용할 수 있다고 생각합니다.
Mastery learning and deliberate practice are currently widely used to develop automatization through proceduralization in clinical skills teaching. However, the effectiveness for skill acquisition and skill retention over time to avoid skill decay can be greatly enhanced by also integrating cognitive science evidence-based teaching strategies into the design of the educational activities of these approaches. Although many of these strategies have been traditionally associated with the development or construction of knowledge, we consider that these can also be applied to the teaching of clinical skills because different types of knowledge are required for skills development.

숙달 학습
Mastery learning

숙달 학습은 구조화되거나 비계화된 교육에 의존하는 기술을 습득하기 위한 개별화된 접근 방식입니다(McGaghie 2015). 보다 구체적으로, 숙달 학습은 개인이 다음 교육 목표를 진행하기 전에 정의된 수준의 숙련도를 달성해야 합니다.

  • 이 접근 방식은 일반적으로 개인에 대한 기준선 또는 진단 테스트와 명확하게 정의된 학습 목표 설정으로 시작됩니다.
  • 그 다음에는 학습 목표를 달성하기 위해 난이도가 점점 높아지는 일련의 활동 단위가 이어집니다. 학습 목표가 구성되는 방식이나 활동 단위를 설계하고 개발하는 방식에는 공식적인 과정이나 절차가 없습니다.
  • 활동 단위는 학습자의 참여가 필요하며 미리 지정된 학습 목표에 도달하는 데 중점을 둡니다.
  • 활동 단위에는 개인이 다음 단계로 넘어가기 전에 달성해야 하는 최소 통과 기준이 있습니다.
  • 마찬가지로, 각 단원에는 개인이 숙달을 위한 최소 합격 기준을 어느 정도 달성했는지 측정하기 위한 평가 요소가 있습니다. 
  • 개인이 해당 활동 단위에서 성공했음을 입증하면 다음 활동 단위로 넘어갈 수 있습니다.
  • 개인은 숙달 표준에 도달할 때까지 단원을 진행하며 이동합니다.
  • 따라서 숙달 학습에서 역량 또는 성취도는 전적으로 지정된 기준 테스트에서 개인이 달성한 성과에 의해 평가됩니다. 

Mastery learning is an individualized approach for acquiring a skill that is dependent on structured or scaffolded instruction (McGaghie 2015). More specifically, mastery learning requires individuals to achieve a defined level of proficiency before proceeding onto the next instructional objective.

  • This approach usually commences with baseline or diagnostic testing of the individual and the setting of clearly defined learning objectives.
  • This is followed by a sequence of units of activity, which have increasing difficulty, to meet the learning objectives. In terms of the way in which learning objectives are constructed or units of activity are designed and developed, there is no formal process or procedure to follow.
  • Each unit of activities requires engagement by the learner and a focus on reaching the pre-specified learning objective.
  • There is a minimum passing standard for each unit of activity which individuals need to achieve before moving on.
  • Likewise, there is an assessment component to each unit to gauge the extent to which the individual has achieved a minimum passing standard for mastery.
  • Once individuals demonstrate their success in the unit of activity, they can advance to the next unit of activity.
  • Individuals progress to move through units until the mastery standard is reached.
  • Therefore in mastery learning, competence or achievement is evaluated entirely by the individual’s attained performance on a specified criterion test.

교육 방법으로서의 효과성에 대한 증거 측면에서, 기술 강화 시뮬레이션을 사용하여 의료 전문가를 대상으로 숙달 학습을 조사한 체계적인 검토 및 메타 분석에 따르면 이 접근 방식이 비숙달 교육보다 우수하지만 시간이 더 걸리는 것으로 확인되었습니다(Cook et al. 2013).
In terms of evidence for effectiveness as a training method, a systematic review and meta-analysis investigating mastery learning for health professionals using technology-enhanced simulation confirmed the approach was superior to non-mastery instruction but takes more time (Cook et al. 2013).

그러나 학습자가 숙달할 때까지 연습해도 임상 술기가 쇠퇴한다는 연구 결과가 있습니다(Higgins 외. 2021a, 2021b). 예를 들어, 중심정맥카테터 술기를 수행한 학습자는 6개월 및 12개월 후에 술기가 쇠퇴하는 것으로 관찰되었으며(Barsuk 등. 2010), 6개월 후 기본 생명유지술기에서도 유사한 관찰 결과가 나타났습니다(Wik 등. 2002; Srivilaithon 등. 2020). 이러한 연구 결과는 [임상 추론과 같은 개념적 임상 술기]뿐만 아니라 [캐뉼레이션 및 혈압 측정과 같은 기술적 술기]에 대한 지식이 중요하다는 것을 강조합니다. 또한, 이 연구 결과는 강의실 가용성 및 스케줄링보다는 이론과 근거에 기반하여 교육 시간과 시간표를 결정하는 것이 중요하다는 것을 보여줍니다. 이론과 증거의 관점에서 볼 때, 관찰된 차이는 학습자가 과제 수행 능력을 입증했음에도 불구하고 절차화를 달성할 수 있도록 지원하는 시간과 교육이 부족했기 때문일 가능성이 높습니다. 
However, research has identified decay of clinical skills even when learners practice until mastery (Higgins et al. 2021a, 2021b). For example, learners that performed central venous catheter performance were observed to demonstrate skills decay after six and 12 months (Barsuk et al. 2010), and similar observations have been noted with basic life support skills after 6 months as well (Wik et al. 2002; Srivilaithon et al. 2020). These findings highlight the importance of knowledge not just for conceptual clinical skills such as clinical reasoning but also technical ones such as cannulation and blood pressure measurement. Furthermore, the findings also demonstrate the importance of ensuring decisions around teaching time and timetabling are based on theory and evidence, rather than room availability and scheduling. From a theory and evidence perspective, the differences observed are likely due to a lack of time and training for supporting learners achieve proceduralization even though they were observed to have demonstrated competence on performing the task.

인지 과학적 관점에서 숙달은 개인이 필요한 모든 지식을 선언적 형태가 아닌 절차적 형태로 보유하는 것으로 이해되며, 절차적 형태의 지식은 선언적 지식과 운동 행동 사이의 연결로 존재합니다. 이러한 관점에서 볼 때, 기술 쇠퇴는 기억에 저장된 선언적 지식이 잊혀지고 그 결과 절차적 지식도 상실되는 현상입니다. 따라서 숙달을 달성하려면 선언적 지식과 절차적 지식 사이의 연결을 강화하여 기술 붕괴를 방지하거나 기술 붕괴가 일어나기 어렵게 만들어야 합니다.
From a cognitive science perspective, mastery is understood as when individuals possess all the necessary knowledge in the procedural form (as opposed to the declarative form), and that knowledge in procedural form exists as a connection between declarative knowledge and motor actions. Viewed through this perspective, skills decay is a phenomenon whereby declarative knowledge stored in memory is forgotten, and as a consequence, there is a loss of procedural knowledge as well. Therefore achieving mastery also involves preventing skills decay or certainly making it more difficult for skills decay to occur, through strengthening the connections between declarative and procedural knowledge.

언뜻 보기에는 단순 술기(예: 혈압 측정)부터 복합적 술기(예: 중심정맥관 삽관)복잡한 술기(예: 고급 생명유지술)까지 모든 임상 술기가 동일하지 않기 때문에 이것이 복잡하거나 불필요한 것처럼 보일 수 있습니다. 따라서 선언적 지식의 개발 및 통합에 대한 관심은 술기의 복잡성에 따라 비례하여 달라져야 합니다.

  • 혈압 측정과 같은 간단한 술기는 완료해야 할 작업이 적고 각 작업은 일반적으로 임상 실습 중 개인이 처할 수 있는 다양한 상황과 환경에 걸쳐 상당히 일관성이 있기 때문에 고급 생명 유지와 같은 복잡한 술기에 비해 선언적 및 절차적 지식이 덜 필요합니다.
  • 고급 생명 유지술의 경우 여러 술기와 관련된 여러 작업이 있으며, 각 작업은 임상 실습의 여러 상황과 환경에 따라 크게 달라집니다. 임상 술기로서 고급 생명유지술은 혈압 측정에 비해 개인이 업무에서 보여줘야 하는 술기의 질과 양이 훨씬 더 많기 때문에 훨씬 더 많은 선언적 및 절차적 지식이 필요합니다.

선언적 지식과 절차적 지식의 양은 임상 술기마다, 그리고 전문성 수준에 따라 다를 수 있지만, 안전한 술기 개발은 선언적 지식과 절차적 지식을 기억 속에 효과적으로 통합하는 데 달려 있습니다.
At first glance, this may appear complicated or unnecessary since not all clinical skills are the same, with simple ones (e.g. blood pressure measurement) through to complicated (e.g. central line cannulation) and complex ones (e.g. advanced life support). It follows that the amount of attention given to developing and consolidating declarative knowledge should be proportionate and vary according to the complexity of the skill.

  • A simple skill, such as measuring blood pressure, requires less declarative and procedural knowledge compared to a more complex skill, such as advanced life support, since there are less tasks to complete and each task is generally fairly consistent across different situations and circumstances individuals may find themselves in during clinical practice.
  • In the case of advanced life support, there are multiple tasks involving multiple skills, and each task varies significantly across different situations and circumstances in clinical practice. A lot more declarative and procedural knowledge is necessary for advanced life support as a clinical skill compared to blood pressure measurement due to the quality and quantity of skill necessary for individuals to demonstrate on task.

Although the amount of declarative and procedural knowledge may differ across clinical skills, and across different levels of expertise, secure skills development is dependent on effective integration between declarative and procedural knowledge in memory.

의도적인 연습
Deliberate practice

에릭슨이 설명하는 또 다른 개별화된 교육 방법을 전문성 수행 접근법이라고 합니다(에릭슨과 차네스 1994). 에릭슨은 먼저 '좋은 캐뉼레이션이 어떻게 생겼는지' 또는 '좋은 리서스가 어떻게 생겼는지'와 같이 전문가 또는 기술을 숙달한 사람이 수행한 실제 세계에서 재현 가능한 우수한 수행을 식별하는 것이 필요하다고 제안했으며, 이는 종종 인지 과제 분석을 사용하여 수행됩니다(Clark et al. 2008). 그 후 교육자의 과제는 시뮬레이션 실험실과 같이 통제된 조건에서 검사할 수 있는 표준화된 과제를 통해 이러한 성과를 캡처하고 재현하는 것입니다. 이 프로세스의 목적은 작업에서 더 나은 결과를 이끌어내는 주요 행동과 그렇지 않은 행동을 식별하여 개인에게 다시 반영할 수 있도록 하는 것입니다. 이 정보를 통해 개인은 이러한 행동의 생성 또는 선택을 매개하는 인지 프로세스에 대한 작업를 시작할 수 있으며, 이를 통해 향후에는 더 우수한 행동을 선택하고 과제에 대한 전반적인 성과를 개선할 수 있습니다. 이 방법을 교사의 감독과 지도를 받는 훈련 접근법으로 적용하는 경우 이를 '의도적 연습'이라고 합니다.
An alternative individualized method of instruction described by Ericsson is called the expertise performance approach (Ericsson and Charness 1994). Ericsson proposed that it is first necessary to identify reproducibly superior performance in the real world, for example, ‘what good cannulation looks like’ or ‘what good resus looks like’ as undertaken by an expert or someone who has mastered the skill, which is often done using cognitive task analysis (Clark et al. 2008). Thereafter, the challenge for educators is to then capture and reproduce this performance, ideally with standardized tasks that can be examined in controlled conditions such as a simulation laboratory. The purpose of this process is to identify key actions that lead to better outcomes on tasks, as well as those which do not, allowing these to be reflected back to individuals. With this information, individuals can begin to work on cognitive processes that mediated the generation or selection of those actions, so that in the future, they can select superior actions and improve their overall performance on the task. When this method is applied as a training approach, supervised and guided by a teacher, it is called ‘deliberate practice’.

의도적 연습은 반복과 연속적인 개선을 통해 개인의 수행 능력의 특정 측면을 향상시키기 위해 코치나 교사가 특별히 고안한 개별화된 훈련 활동을 말합니다(Ericsson and Lehmann 1996). 

  • 그러나 교사의 지도와 피드백을 최대한 활용하려면 개인이 집중력을 발휘하여 훈련을 모니터링해야 합니다. 
  • 또한, 완전한 집중력을 가지고 연습하는 것은 힘든 일이기 때문에 개인별 훈련 시간은 주어진 세션 길이에 따라 신중하게 관리되어야 합니다. 
  • 또한 의도적인 연습 접근 방식은 개인이 즉각적인 피드백을 받거나 스스로 생성하는 것이 특징입니다. 
    • 외부 피드백은 오류나 실수를 식별하고 개선 방법에 대한 후속 조언을 제공하는 데 중요합니다. 
    • 마찬가지로, 자체적으로 생성된 피드백은 전문가로서 독립적으로 수행하는 데 중요한 내부 표현 개발의 품질을 평가하는 데 중요합니다. 
  • 신중한 연습의 마지막 구성 요소는 개인이 반복적으로 연습할 수 있는 기회를 제공하고 연습을 시도할 때마다 점진적으로 개선할 수 있도록 연습을 구조화하는 훈련 과제의 가용성입니다.

이러한 모든 측면은 접근 방식이 전문성 수행 원칙에 기반한 유형인 '의도적 연습'으로 간주되기 전에 반드시 필요합니다.
Deliberate practice refers to individualized training activities specially designed by a coach or teacher to improve specific aspects of an individual’s performance through repetition and successive refinement (Ericsson and Lehmann 1996).

  • However, to receive maximal benefit from teacher instruction and feedback, individuals have to monitor their training with full concentration.
  • Furthermore, practicing with full concentration is effortful so the duration of training for individuals needs to be carefully managed in any given session length.
  • In addition, deliberate practice approaches are characterized by individuals receiving or self-generating immediate feedback.
    • External feedback is important for identifying errors or mistakes, with follow-up advice about ways to make improvements.
    • Likewise, self-generated feedback is important for evaluating the quality of developing internal representations, which are critical for performing independently as experts.
  • The final component of deliberate practice is the availability of training tasks the offer individuals opportunities to engage in repetitions and structure practice in such a way that individuals make gradual improvements with each attempt.

All these aspects are necessary before the approach can be considered ‘deliberate practice’, certainly the type based on the principles of expertise performance.

지난 20년간의 연구에 따르면 최적의 기술 습득을 위해 교사 주도의 코칭 및 가이드 연습과 결합된 반복 연습의 중요성이 입증되었습니다(Ericsson 외. 1993; Ericsson 2004; McGaghie 외. 2011). 의도적인 연습이 기술 습득과 유지를 향상시킨다는 사실은 여러 연구에서 확인되었지만, 최근의 검토에서는 (1) 비교 그룹의 부재와 (2) 2주 후부터 90일 후 점점 더 감소한다는 두 가지 중요한 문제가 강조되었습니다(Higgins 외. 2021a, 2021b). 이러한 문제의 원인은 연구 설계가 개발되는 방식과 실제로 교육이 제공되는 방식을 포함하여 여러 가지 요인이 있습니다.
Research over the past two decades has demonstrated the importance of repeated practice that is combined with teacher-led coaching and guided practice for optimal skill acquisition (Ericsson et al. 1993; Ericsson 2004; McGaghie et al. 2011). Although studies have identified that deliberate practice improved skill acquisition and retention, recent reviews have highlighted two important issues: (1) the lack of a comparison group and (2) a decay after 2 weeks and more increasingly after 90 days (Higgins et al. 2021a, 2021b). The reason for these issues are multi-factorial, including the specific to way in which the study design was developed, and the way in which the training was delivered in practice.

숙달 학습과 의도적 연습과 같은 현재의 교육 기법은 인지 과학의 증거 기반 전략을 통해 기술 유지를 극대화하고 기술 감퇴를 줄일 수 있다고 생각합니다. 숙달 학습은 효과적인 것으로 나타났지만, 멘탈 모델 개발에 더 중점을 두는 것이 좋습니다. 마찬가지로, 의도적 연습은 교육과 교사-학습자 관계를 강조하기 때문에 교육자들 사이에서 특히 인기가 있지만, 지식과 기술의 붕괴를 방지하기 위해 효과적인 지식 구성에도 더 중점을 두면 더 효과적일 수 있다고 권장합니다.
We consider that current instructional techniques, such as mastery learning and deliberate practice, can be informed by cognitive science evidence-based strategies to maximize skill retention and reduce skill decay. Mastery learning has been shown to be effective, however, we recommend greater emphasis on the development of mental models. Likewise, deliberate practice is particular popular among educators due to the emphasis on instruction and teacher-learner relationships, but we recommend that it could be more effective if the greater emphasis was also given to effective knowledge construction, with the aim to prevent the decay of knowledge and skills.

인지 과학 증거 기반 전략
Cognitive science evidence-based strategies

간격 연습
Spaced practice

임상 술기를 가르치는 데 자주 사용되는 접근 방식은 교육자가 세션 시작 시 술기를 설명하고 시범을 보인 후 학습자에게 세션이 끝날 때쯤 연습하고 질문할 기회를 제공하는 것입니다(Bullock 외. 2015). 대량 실습이라고 하는 이 접근 방식은 한 번에 모든 교육을 제공합니다. 예를 들어, 고급 생명 유지 술기는 일반적으로 단일 교육 과정의 일부로 2일 또는 3일에 걸쳐 연속적인 대량 실습 교육 세션으로 진행됩니다. 고급 생명 유지 술기 교육에 대량 실습을 사용하는 이유는 교육 과정의 교육 설계를 위한 근거 기반 또는 이론 중심이라기보다는 한 번에 세션을 모두 진행할 수 있다는 편의성과 관련이 있는 것으로 보입니다. 마찬가지로, 전부는 아니더라도 대부분의 임상 술기 교육에서 직면하는 어려움은 제한된 교수진으로 학습자 집단에게 교육을 제공하는 물류 문제입니다.

A frequently used approach for teaching a clinical skill is for educators to describe and demonstrate the skill at the start of a session, before offering learners the opportunity to practice and ask questions towards the end of it (Bullock et al. 2015). This approach, called massed practice, provides teaching all at only one time. For example, advanced life support skills are generally taught as part of a single training course with consecutive teaching sessions of massed practice over a duration of 2 or 3 days. The reason for using a massed practice for advanced life support skills training appears to be as much to do with the convenience of timetabling sessions all in one go, rather than any evidence-based or theory-driven basis for the instructional design of courses. Likewise, the challenge with most, if not all, clinical skills training in HPE invariably includes the logistics of delivering teaching to cohorts of learners with a limited teaching faculty.

그러나 인지 과학의 증거에 따르면 교육 간격을 두거나 여러 세션에 걸쳐 교육 시간을 분산하는 것이 학습 유지에 더 효과적이라는 사실이 입증되었습니다(Bjork and Allen 1970; Dempster 1989; Cepeda 외. 2006). 예를 들어, Cepeda 등(2006)은 184개의 논문을 대상으로 메타분석을 실시하여 한 세션(대규모)의 실습과 간격을 두는 실습을 비교했습니다. 그 결과, 간격을 두고 학습한 그룹의 학습자가 한꺼번에 학습한 그룹보다 학습 내용을 더 잘 유지한다는 사실을 발견했습니다. 간격 효과에 대한 대부분의 연구는 선언적 지식에 크게 의존하는 과제와 관련된 연구에서 나온 것일 수 있지만(검토는 Cepeda 외. 2006, Carpenter 외. 2012 참조), 임상 술기에 대한 간격 연습의 이점을 입증하는 증거도 있습니다(검토는 Cecilio-Fernandes 외. 2018 참조). 교육 세션 사이의 최적 간격은 특정 임상 술기에 필요한 지식의 유형에 따라 다릅니다. 고급 생명유지술과 같이 선언적 지식을 기반으로 하는 술기의 경우, 최적의 세션 간 간격은 [유지 간격의 10%에서 15% ] 사이입니다(Carpenter 외. 2012). 유지 간격은 마지막 교육 세션과 최종 테스트 사이의 시간입니다. 그러나 학습자가 필요한 사고가 필요할 때 행동할 수 있어야 하는 실제 상황에서는 유지 간격이 불가능할 수 있습니다. 그러나 교육자는 개별 교육 세션이 끝난 후에도 학습자가 기술을 지속적으로 유지한다고 가정하기보다는 학습자가 기술을 유지하는 시간을 인식하고 감사하는 태도를 가져야 합니다. 
However, evidence from cognitive science demonstrates that spacing the teaching or spreading training time over multiple sessions is more effective for retention (Bjork and Allen 1970; Dempster 1989; Cepeda et al. 2006). For example, Cepeda et al. (2006) conducted a meta-analysis with 184 articles comparing the practice in one session (massed) to space. They found that learners in the spaced group retained better than those in the massed group. Whilst much of the research into the spacing effect may come from studies involving tasks that heavily relies on declarative knowledge (for a review see Cepeda et al. 2006; Carpenter et al. 2012), there is also evidence demonstrating the benefit of spaced practice for clinical skills (for a review see Cecilio-Fernandes et al. 2018). The optimal interval between training sessions varies depending on the type of knowledge required for the specific clinical skill. For skills that are heavily based on declarative knowledge such as advanced life support, the optimal interval between sessions range from 10% to 15% of the retention interval (Carpenter et al. 2012). The retention interval is the time between the last training session and the final test. However, in real-life situations in which learners need to be able to act when required thinking in a retention interval may not be feasible. However, educators need to have awareness and appreciation of the time length over which skills are retained by learners, rather than assume they are constantly maintained after individual training sessions.

 

검색 연습
Retrieval practice

임상 술기를 가르치는 또 다른 자주 사용되는 방법은 한 번 가르친 다음, 그 직후 또는 어느 시점에 요약 또는 고난도 평가의 일부로 학습자가 술기를 어느 정도 개발했는지 평가하는 것입니다. 그러나 인지 심리학의 증거에 따르면 임상 술기를 테스트하지 않거나 한 번만 테스트하는 것보다 반복적으로 테스트하는 것이 장기적인 술기 유지에 더 효과적이라고 합니다. 
Another frequently used way of teaching a clinical skill is to teach it once, and then assess the extent to which learners have developed the skill either immediately afterward or at some point as part of a summative or high-stakes assessment. However, evidence from cognitive psychology suggests repeated testing of clinical skills, rather than no or single event testing, is more effective for long-term skills retention.

테스트 효과(Roediger and Karpicke 2006a, 2006b; Karpicke and Roediger 2008)는 기본 생명유지술(Li 외 2011), 고급 생명유지술(Kromann 외 2009), 방사선 사진 판독(Boutis 외 2019), 임상 추론(Raupach 외 2016) 등 여러 임상 술기를 습득하는 데 효과적인 전략입니다. 또한 선언적 지식에 더 많이 의존하는 술기의 유지를 향상시킨다는 증거도 있습니다(Larsen 외. 2009; Larsen, Butler, Lawson 외. 2013; Larsen, Butler, Roediger 2013). 
The testing effect (Roediger and Karpicke 2006a, 2006b; Karpicke and Roediger 2008) is an effective strategy for acquiring several clinical skills, including basic life support (Li et al. 2011), advanced life support (Kromann et al. 2009), radiograph interpretation (Boutis et al. 2019), and clinical reasoning (Raupach et al. 2016). There is also evidence for improving the retention of skills that relies more on declarative knowledge (Larsen et al. 2009; Larsen, Butler, Lawson, et al. 2013; Larsen, Butler, Roediger 2013).

시험 효과가 중요한 이유는 '시험'이 전통적인 의미의 형식적인 수행 평가나 '형성 평가'를 의미하는 것이 아니라 기술 쇠퇴를 최소화하기 위해 고안된 실제 교수 전략을 의미하기 때문입니다. 또한 의도적인 교육 전략으로서의 테스트 효과는 학습자에게 주어진 임상 술기를 연습할 수 있는 기회를 더 많이 제공하는 것과는 구별됩니다. 특히 임상 술기 맥락에서 테스트 효과는 직접 관찰 중인 학습자가 과제를 수행하는 과정을 지켜보면서, 선언적 지식과 절차적 지식을 통합할 수 있는 기회를 제공하고, 이에 대한 수행 결과 또는 이후 개선 전략에 대한 피드백을 제공하는 것을 말합니다.
The testing effect is important because ‘testing’ does not refer to the formal assessment of performance or ‘formative assessment’ in a traditional sense, but an actual teaching strategy designed to minimize skills decay. Furthermore, the testing effect as a deliberate teaching strategy is distinct from just providing more opportunities for learners to practice a given clinical skill. Specifically, the testing effect in a clinical skills context refers to learners under direct observation being watched undertaking a task, with the opportunity to integrate their declarative and procedural knowledge, and given feedback in terms of performance outcome on it, or strategies for improvement afterward.

교육자는 다양한 접근 방식을 사용하여 학습자를 테스트할 수 있으며, 학습자가 기술 습득을 진행하는 즉시 테스트를 교수 전략으로 사용해야 합니다. 그러나 커리큘럼 전반에 걸쳐 구현하려면 테스트피드백 제공을 위한 시간과 공간, 리소스가 필요합니다. 또한 테스트는 교육자가 시간을 두고 관찰하고 피드백을 기반으로 구축할 수 있는 방식으로 구성되어야 합니다. 교수 전략에 단일 테스트가 포함되거나 너무 빨리 테스트를 실시하는 경우, 학습자는 당장의 성과가 시간이 지남에 따른 학습을 제대로 예측하지 못하기 때문에, 지식에 대한 환상을 갖게 될 수 있습니다. 이러한 상황에서 '시험'은 단지 단기 기억력의 유창함을 평가하는 것이지, 시간이 지나면서 지식과 기술을 기억하는 장기적인 능력, 맥락에 따라 이를 전이하는 능력, 잊어버리는 것을 방지하는 능력은 평가하지 않습니다.
Educators can test learners using a variety of approaches and testing should be used as a teaching strategy as soon as learners progress with skill acquisition. However, for implementation across a curriculum, time and space as well as resources are necessary for testing, as well as for giving feedback. Testing also needs to be organized in such a way that educators make observations over time and can build on feedback. Where teaching strategies involve single tests or testing too soon, learners may develop the illusion of knowledge since performance in the moment is a poor predictor of learning over time. In these situations, the ‘test’ is merely assessing the fluency of short-term retention, and not their long-term ability to recall knowledge and skills over time, with transfer these across contexts, as well as their ability to prevent forgetting.

테스트 효과의 또 다른 특징은 피드백의 역할과 피드백의 구성 방식과 관련이 있습니다. 시험 후 피드백은 종종 교사가 과거 수행과 함께 관찰한 내용을 삼각 측량하지 않고 제공하지만, 복습 연습에서 교육자는 이전에 시도한 과제에 적극적으로 주의를 기울이고 피드백에 과제에 대한 [연속적인 시도에 대한 보다 전체적인 평가]가 포함되도록 해야 합니다. 동료 간 테스트전략의 필수 구성 요소가 수행 자체(결과)가 아닌 연습 중 테스트 기회(과정)이므로 교육자 주도 활동의 대안이 될 수 있습니다. 시험은 종종 기술 훈련의 작은 부분인 단원의 마지막에 이루어지기 때문에 의도적인 연습 및 숙달 학습과 같은 교육 설계가 손상될 수 있습니다. 
Another feature of the testing effect involves the role of feedback and the way it is constructed. Feedback after a test is often given by teachers without triangulating observations with past performance, but in retrieval practice educators are required to proactively give attention to previous attempts on a task and ensure feedback includes a more holistic evaluation of progress made over successive attempts on the task. Peer-to-peer testing may be an alternative to educator–led activities since the essential component of the strategy is the opportunity for testing during practice (the process) and not the performance in and of itself (the outcome). This may compromise instructional designs such as deliberate practice and mastery learning because test often happens at the end of a unit, which is a smaller part of the skill training.

과잉 학습
Overlearning

과잉학습숙련도competence를 달성한 후에도 반복적이고 추가적인 연습에 참여하는 개인을 말하며, 절차적 지식의 통합을 통해 기술 쇠퇴를 줄이는 등의 이점이 있습니다(Shibata 외. 2017).

  • 과잉학습에 참여하는 개인은 잘못된 접근 방식을 계속해서 반복할 수 있으므로 의도적인 학습 전략으로서의 과잉 학습을 '연습, 연습, 연습'에 참여하는 개인과 혼동해서는 안 됩니다.
  • 마찬가지로, 과잉학습은 종종 훈련의 마지막 단계에서 절차화를 위해 적용되기 때문에 교육적 접근 방식으로서의 의도적 연습과도 구별됩니다

Overlearning refers to individuals engaging in repeated and further practice after competence is achieved, and the benefits include less skill decay through increased consolidation of procedural knowledge (Shibata et al. 2017).

  • Overlearning as a deliberate learning strategy should not be confused with individuals engaged in ‘practice, practice, practice’ since individuals engaged in repeated practice may be repeating the wrong approach over and over again.
  • Likewise, overlearning is also distinct from deliberate practice as an instructional approach since overlearning is often applied for proceduralization in the final part of the training.

전략으로서 과잉학습의 전반적인 효과는 기술의 유형과 학습자가 반복 연습의 질과 양 모두에 참여하는 정도에 따라 달라집니다(Driskell 외. 1992). 효과를 극대화하려면 유능한competent 수준의 성과를 달성하는 데 필요한 기간의 두 배에 해당하는 시간 동안 반복 학습을 지속해야 합니다(Driskell 외. 1992; Krueger 1929; Rohrer 외. 2005). 개인에 따라 유능한 수준의 성과에 도달하는 데 걸리는 시간이 임상 술기에 따라 다르다는 것이 HPE의 과제입니다. 마찬가지로, 역량에 대한 판단은 평가자마다 신뢰할 수 없으며, Ericsson(2004)에 따르면 다양한 임상 영역에서 전문가 성과를 측정하는 기준은 아직 개발되지 않았습니다. 따라서 학습자는 이 전략에 충분히 참여하기 위해 자기 평가 및 자기 조절 학습과 관련된 다른 기술도 개발해야 합니다. 일부 학습자는 어느 정도 자율적으로 과잉 학습에 참여할 가능성이 높지만, 임상 기술에 어려움을 겪는 학습자는 과잉 학습을 덜 할 가능성이 높으며 교육자가 과잉 학습의 목적을 설명해 주는 지원이 필요합니다. 
The overall effect of overlearning as a strategy depends on the type of skill and the extent to which learners engage in both the quality and quantity of repeated practice (Driskell et al. 1992). To maximize effect, overlearning should continue for twice the period of time necessary to achieve a competent level of performance (Driskell et al. 1992; Krueger 1929; Rohrer et al. 2005). The challenge in HPE is the time taken for some individuals to achieve a competent level of performance varies across and within different clinical skills. Likewise, judgments about competence are unreliable across raters and as Ericsson (2004) states, the yardstick to measure expert performance across the different clinical domains is yet to be invented. Therefore, learners need also have developed other skills around self-assessment and self-regulated learning, in order to engage fully with this strategy. Whilst some learners likely engage in overlearning independently to some degree, learners who struggle with clinical skills are likely to do so less and require support from educators to explain the purpose of it.

과잉 학습은 교육자가 실습을 관찰할 때 개인이 자신의 수행 수준에 대해 더 많은 피드백을 얻을 수 있는 기회를 제공할 뿐만 아니라 과제 수행에 방해가 되기 전에 수행 능력의 격차를 해결할 수 있는 기회를 제공합니다. 또한 과잉 학습은 자동화에 걸리는 시간을 단축하고, 개인이 수행의 기술적 측면에 투자해야 하는 집중적인 노력의 양을 줄임으로써 수행을 모니터링할 수 있는 전반적인 인지 능력을 확보할 수 있습니다(Driskell 외. 1992). 이는 특히 단순하고 복잡한 임상 술기 과제에서 초심자 간 수행 차이를 파악할 때 특히 유용합니다. 고성과자는 평가 시 다른 사람들 앞에서 임상 술기를 수행해야 하는 어려움을 관리할 수 있지만, 저성과자는 여러 가지 이유로 어려움을 겪는데, 그 중 하나는 자동성을 달성하지 못했고 추가적인 인지 과부하로 인해 수행이 저하된다는 사실입니다. 따라서 인지 과학적 관점에서 볼 때, 과잉 학습은 역량을 달성하기 위한 전략이 아니라 절차적 지식을 공고화consolidate하기 위한 전략으로 사용되어야 합니다. 또한, 역량을 갖추기 전에 사용하는 과잉 학습은 이해가 아닌 모방을 통해 숙달을 유도할 수 있으며, 이는 개인이 지식에 대한 환상을 갖는 것과 관련된 또 다른 징후입니다. 이러한 맥락에서, 특정 임상 술기에 대한 역량을 보이는 개인에게 기대되는 행동을 구체적으로 모방하고 접근 방식을 반복함으로써, 특히 저성과자는 임상 술기 자체에 대한 지식이나 이해 부족에 대한 우려 없이 평가에서 발전하기 위해 모방에 참여하게 됩니다. 
Overlearning generates opportunities for the individual to obtain more feedback about their level of performance when the practice is observed by an educator, as well as address any developing gaps in performance before they inhibit performance on a task. Overlearning also enhances the time to automaticity and reduces the amount of concentrated effort that individuals have to invest into the technical aspects of their performance, thereby freeing up overall cognitive capacity for monitoring their performance (Driskell et al. 1992). This is particularly relevant when making sense of performance differences between novices at summative assessments, especially on simple and complex clinical skills tasks. Whilst high-performing individuals are able to manage the challenge of performing clinical skills in front of others at assessment, low performers struggle for many reasons, once of which is the fact they have not achieved automaticity and their performance suffers as a consequence of the additional cognitive overload. Therefore, from a cognitive science perspective, overlearning should be used as a strategy to consolidate procedural knowledge, rather than a strategy to achieve competence. Furthermore, overlearning employed before competence is achieved, may lead mastery, not through comprehension, but through mimicry which is another manifestation related to individuals having an illusion of knowledge. In this context, by repeating the approach over and over again, and specifically mimicking the behaviors expected of individuals demonstrating competence on a given clinical skill, low performers in particular engage in mimicry in order to progress at assessment, with little concern about their lack of knowledge or understanding about the clinical skill itself.

인터리빙
Interleaving

HPE에서 지지하는 전문 지식의 대부분은 다양한 상황적 맥락에서 임상 술기를 시연하는 데서 비롯되지만, 기존 교육 프로그램에서 학습자에게 제공되는 실습 기회 유형에는 변동성이 거의 없을 수 있습니다. 예를 들어,

  • 캐뉼레이션을 학습할 때 일반적인 실습 기회는 플라스틱 마네킹과 같은 제한된 범위의 실습 재료를 사용하여 이 술기의 운동 측면을 개선할 수 있는 기회로만 구성되는 경우가 많습니다.
  • 봉합 및 정맥천자부터 기본 및 고급 생명 유지 술기에 이르기까지 다양한 임상 술기에 대해 일반적으로 유사한 접근 방식이 제공됩니다.

언뜻 보기에는 한 번에 하나의 술기에 집중하는 것이 합리적이고 논리적으로 보이지만, 인지 과학 문헌의 증거는 시간이 지남에 따라 지식과 술기 개발을 늘리기 위한 전략으로서 교차 학습의 이점을 일관되게 입증하고 있습니다. 특히, 전체 스킬의 다양한 구성 요소 또는 서로 '얼마나 관련성이 있든 없든' 서로 다른 스킬을 같은 세션에서 연습하는 스킬 연습의 가변성variability은 가변성이 제한적이거나 전혀 없는 연습과 비교할 때 스킬 감퇴를 최소화합니다. 또한, 전통적인 기술 연습에 참여하는 개인은 기술 개발 전략으로 인터리빙을 사용하는 개인에 비해 기술 감퇴율이 더 높습니다(Spruit 외. 2014).

Although much of the expertise espoused in HPE comes from demonstrating clinical skills across a range of different situational contexts, there can often be little variability in the types of practice opportunities afforded to learners in traditional training programs. For example,

  • when learning cannulation, the usual practice opportunities involve chances to refine the motor aspects of this skill alone, often with a limited range of practice materials such as a plastic manikin.
  • A similar approach is usually provided for a range of clinical skills, from suturing and venepuncture through to basic and advanced life support skills.

At first sight, this focus on a single skill at a time would seem rational and logical but evidence from the cognitive science literature consistently demonstrates the benefits of interleaving as a strategy for increasing knowledge and skills development over time. Specifically, variability of skills practice – either different components of the overall skill or different skills (no matter ‘how related or unrelated’ to each other) practiced in the same session minimizes skill decay when compared with a practice that offers limited or no variability. Furthermore, individuals who engage in traditional skills practice have greater rates of decay when compared with individuals who engage with interleaving as a skills development strategy (Spruit et al. 2014).

실습 기회에서 가변성을 설계하는 방식도 중요합니다. 최소 침습 수술 술기의 습득 및 유지에 관한 연구에 따르면 단일 또는 정적인 교육 이미지보다는 무작위로 번갈아 가며 이미지를 사용하여 연습하는 것이 학습 과정을 향상시키는 것으로 나타났습니다(Jordan 외. 2000). 훈련 세션에서 이미지의 가변성을 높여 술기 개발을 연습한 개인은 제한된 수의 이미지로 훈련한 사람보다 훨씬 빠른 속도로 기본적인 수술 및 정신 운동 술기를 습득했습니다. 이러한 정도의 가변성은 실제 환경에서도 발견되지만 시뮬레이션에서도 이러한 유형의 경험을 제공할 수 있습니다. 가변성의 중요성은 운동 기술 문헌(Kerr and Booth 1978, Savion-Lemieux와 Penhune 2005, Corrêa 외. 2014), 화가와 그림 스타일 간의 관계 학습(Kornell과 Bjork 2008), 문제 해결(Rohrer와 Taylor 2007)에서도 강조되고 있습니다. 
The way in which variability is also designed in the practice opportunities is important. Research into the acquisition and retention of minimally invasive surgery skills demonstrates practice with randomly alternating images rather than a single or static teaching image enhances the learning process (Jordan et al. 2000). Individuals who practiced skills development with greater variability of the image in training sessions acquired basic surgical and psychomotor skills at a much faster rate than those trained with a limited number of images used as part of their training. This degree of variability is found in real-life settings but simulation can also provide this type of experience. The importance of variability is also highlighted in the motor skills literature (Kerr and Booth 1978; Savion-Lemieux and Penhune 2005, Corrêa et al. 2014), learning the relation between painters and painting styles (Kornell and Bjork 2008) and problem-solving (Rohrer and Taylor 2007).

정교화 및 생성
Elaboration and generation

정교화란 새로운 지식을 이전에 기억에 저장된 지식과 연결하는 것을 말합니다(Bartsch 외. 2018; Holland 외. 2011). 예를 들어

  • 정맥 천자 또는 캐뉼레이션과 같은 간단한 임상 술기의 경우, 교육자는 혈액 샘플을 채취하거나 실제 플라스틱 기구를 환자의 손에 넣는 기술적 측면을 시연하기 전에 학습자가 팔 혈관의 해부학적 구조를 떠올리도록 지원할 수 있습니다.
  • 고급 생명 유지와 같은 복잡한 임상 기술의 경우, 심장 제세동과 관련된 리듬 인식에 대해 가르치기 전에 교사가 학습자가 심장 파형에 대한 지식을 기억하도록 지원할 수 있습니다.

Elaboration refers to the connection of new knowledge with previously stored knowledge in memory (Bartsch et al. 2018; Holland et al. 2011).

  • In the case of a simple clinical skill such as venepuncture or cannulation for example, this could involve educators supporting learners to recall the anatomy of the arm vessels prior to demonstrating the technical aspects of obtaining a blood sample or placing the actual plastic apparatus in the hand of a patient.
  • In the case of a complex clinical skill such as advanced life support, this could involve teachers supporting learners to recall knowledge about the cardiac waveform, prior to teaching them about rhythm recognition in the context of cardiac defibrillation.

정교화는 이전에 저장된 지식을 활성화하기 위한 의식적인 노력이며, 주로 검색 가능한 지식인 선언적 지식을 통해 이루어집니다. 임상 술기에는 선언적 지식과 절차적 지식이 모두 필요하므로, 정교화는 저장된 관련 선언적 지식을 활성화하고 새로운 지식을 활성화된 지식과 연결함으로써 임상 술기 습득에 도움이 될 수 있습니다. 피드백 및 성찰은 학습자가 알고 있는 것을 새로운 콘텐츠와 연결해야 하므로 정교화를 지원하는 방법으로도 사용할 수 있습니다. 학습자가 새로운 자료와 이전 지식을 연결할 수 있도록 지원하면 학습자의 기억력이 더욱 향상됩니다(Bartsch 외. 2018; Stark 외. 2002). Bartsch 등(2018)은 6개의 명사를 연속된 순서로 기억할 때 정교화와 비 정교화의 효과를 비교한 결과, 정교화 그룹의 학습자가 비 정교화 학습자보다 유지력 테스트에서 더 우수한 성적을 보였다고 합니다.
Elaboration is a conscious effort to activate previously stored knowledge and it occurs mainly through declarative knowledge, which is the knowledge that is possible to retrieve. Since clinical skills require both declarative and procedural knowledge, elaboration may also benefit the acquisition of clinical skills by activating related declarative knowledge that is stored and connecting the new knowledge with the activated knowledge. Feedback and reflection may also be used as a way to support elaboration since both require to connect what learners know with the new content. Supporting learners to make a connection between the new material with their previous knowledge will further improve their retention (Bartsch et al. 2018; Stark et al. 2002). Bartsch et al. (2018) compared the effect of elaboration with non-elaboration on remembering six nouns in serial order and the results indicated that learners from the elaboration group had a better performance in the retention test than the non-elaboration learners.

생성이란 교육자가 구체적인 지시나 정보를 제공하기 전에 학습자가 주어진 문제에 대한 해결책을 스스로 찾아내는 것을 말하며, 여기서는 임상 기술을 수행하는 것을 의미합니다(Jacoby 1978; Slamecka와 Graf 1978; Richland 외. 2009). 교육자는 이 전략을 단순히 학습자가 '먼저 해보도록' 촉진하는 것과 연관시킬 수 있지만, 생성에는 학습자에게 지시나 새로운 정보가 제공되기 전에 의도적으로 어떤 형태의 사전 지식(선언적 및 절차적 지식 모두)을 활성화하도록 하는 것도 포함됩니다. 이러한 활성화가 학습자에게 미치는 영향은 새로운 지식과 이전 지식을 더 쉽게 연결할 수 있다는 것입니다.
Generation refers to the specific enabling of learners to identify their own solutions to a given problem, in this context, undertaking a clinical skill, before specific instruction or information is given by the educator (Jacoby 1978; Slamecka and Graf 1978; Richland et al. 2009). Whilst educators may associate this strategy simply as facilitating learners to ‘have a go first’, generation also involves deliberately enabling learners to activate some form of prior knowledge – both declarative and procedural – before either instruction or new information is given to them. The impact of this activation for the learner is the easier association between new and previous knowledge.

임상 술기 교육의 맥락에서 교육적 접근 방식으로서의 생성은 광범위한 문헌에서 근거가 제한적이지만, HPE에서 사용되는 기존 교육학의 대부분은 이 개념에 뿌리를 두고 있습니다. 발견 학습 또는 발견 연습의 핵심 원칙은 학습자가 주어진 문제에 대한 답을 스스로 찾을 수 있도록 지원해야 한다는 개념입니다. 사실일 수도 있지만, 이 접근법에 대한 많은 리뷰에서는 초기 발견 단계가 끝난 후 학습자가 [직접 지도를 받는 역할]이 중요하다고 제안합니다. 초기 발견 단계는 이론적으로 관련 선언적 지식을 활성화하여 새로운 지식을 이전에 저장된 지식과 연결함으로써 해당 지식의 유지 및 검색에 도움이 되어야 합니다. 이 단계는 학습자가 먼저 임상 술기를 시도해 보도록 유도하여 모든 임상 교육을 시작할 때 사용할 수 있습니다. 예를 들어, 고급 생명유지술 교육에서는 학습자가 소생술이 필요한 환자를 돌봐야 하는 시뮬레이션 시나리오를 교육 첫 부분에 포함시켜야 합니다. 이렇게 하면 학습자가 보유하고 있는 모든 관련 지식을 활성화할 수 있습니다.
Although generation as an instructional approach in the context of clinical skills training has a limited evidence base in the wider literature, much of the existing pedagogy used in HPE draws roots from this concept. The central tenet of discovery learning or discovery practice is the notion that learners should be supported to find the answer to a given problem by themselves. Whilst this may be the case, many of the reviews in this approach suggest there is a strong role for learners to receive direct instruction after the initial discovery phase has expired. The initial discovery phase should in theory activate relevant declarative knowledge, benefiting the retention and retrieval of that knowledge by connecting the new knowledge with the previous stored knowledge. This could be used at the beginning of every clinical training by encouraging learners to first try the clinical skill. For example, in advanced life support training, the first part of the training should be a simulated scenario in which learners need to attend to a patient requiring resus. In that way, learners would activate all the related knowledge they possessed.

바람직한 어려움
Desirable difficulties

학습자가 학습 프로그램을 마칠 때까지 습득해야 하는 임상 술기에 대한 지식의 양은 상당하며, 학습자가 평가를 '통과하기 위해' 대량 실습에 참여할 위험은 상당합니다. 일부 상황에서는 과잉 학습이 유용한 전략이 될 수 있지만, 임상 술기를 습득하지 않고 평가에 합격하기 위한 목적으로 과잉 학습을 하면 학습자가 시간이 지남에 따라 지식과 술기를 잊어버릴 위험이 있습니다. 검색과 간격을 두고 연습하면 망각의 영향을 일부 완화할 수 있지만(Bjork and Bjork 2019; Soderstrom and Bjork 2015), 바람직한 난이도를 전략으로 사용하는 것은 학습자에게 기억과 앎 또는 이해를 혼동하는 문제를 상기시키는 데도 효과적일 수 있습니다. 지식의 착각학습자가 다른 사람이 눈앞에서 기술을 시연하는 것을 보는 것만으로, 또는 학습자가 시간이 지나거나 다른 상황에서 다시 시연할 필요 없이 면밀한 관찰 하에 기술을 연습하는 것만으로도, 자신이 기술을 알고 있거나 가지고 있다고 믿는 현상입니다.
The amount of knowledge for clinical skills required for learners to acquire by the end of a program of study is significant, and the risk of learners engaging in massed practice ‘just to get through’ assessments is significant. Whilst overlearning is a helpful strategy in some circumstances, overlearning for the purpose of passing the assessment and not mastering the clinical skill risks learners forgetting knowledge and skills decaying over time. Although, interleaving retrieval and spaced practice mitigate some of the effects of forgetting (Bjork and Bjork 2019; Soderstrom and Bjork 2015), the use of desirable difficulties as a strategy can be effective also for reminding learners about the problem of confusing remembering, with knowing or understanding. The illusion of knowledge is the phenomenon whereby learners believe they know or have skills, by simply watching others demonstrate skills in front of them, or learners practicing skills under close observation without necessarily demonstrating them again over time, or in other contexts.

바람직한 어려움은 학습자의 주의를 끌기 위해 교사가 사용할 수 있는 또 다른 전략이며, 간단한 질문을 사용하여 기억하기, 알기, 이해하기 간의 차이를 강화할 수 있습니다. '자전거를 그릴 수 있느냐'와 같이 일반인에게 간단한 지시를 내리는 것은 지식의 착각 현상을 설명하는 것으로 나타났습니다. 그 후 교사는 학습자에게 구체적인 과제를 제공함으로써 학습자의 주의를 끌 수 있으며, 이를 통해 학습자는 어려움을 겪는 경험을 할 수 있을 뿐만 아니라 나중에 성공하는 경험도 할 수 있습니다. 따라서 '어려움'은 교사의 지원, 발판 또는 성공 없이 '고군분투'라는 결과만 유발하는 것이 아니라 "고군분투와 성공"이라는 두 가지 결과를 모두 불러일으키는 경우에만 바람직합니다(Bjork and Bjork 2011).
Desirable difficulties is another strategy that teachers can employ for both getting the attention of learners but also use simple questions to reinforce the difference between remember, knowing, and understanding. Simple instructions to a general population such as ‘can you draw a bicycle’ have been shown to illustrate the illusion of knowledge phenomenon. Thereafter, teachers can build on the gaining of learner attention by providing individuals with specific tasks that induce the experience of struggling, but also enables them to also experience success afterward. Therefore difficulty is only desirable if it evokes both outcomes – struggling and success (Bjork and Bjork 2011) – rather than just the former without any form of support, scaffold, or success provided by the teacher.

또한 현재의 증거에 따르면 과제를 수행할 때 실수를 하거나 선언적 및 절차적 지식을 검색하는 데 어려움을 겪는 것은 시간이 지나도 지식과 기술을 장기적으로 잘 유지하도록 하는 데 필요합니다(Bjork and Bjork 2011; Steenhof Woods와 Mylopoulos 2020). 바람직한 어려움 사용에 대한 증거는, '교사의 존재로 인해 오류가 근절될 때 연습이 가장 좋다'는 다른 증거를 배경으로 존재합니다. 이러한 문헌은 더 나아가 실수를 통한 학습은 비생산적이며, 실수가 발생하더라도 이는 단순히 부적절한 교육 때문이라는 것을 암시합니다(Skinner 1958). 또한, 실수 없이 학습하는 것(오류 없는 학습errorless learnging이라고도 함)은 학습자가 실수 없이 발전하는 자신을 보기 때문에 '기분이 좋다'고 느낄 수 있으며, 이는 단순히 더 많은 것을 알고 있다는 환상을 불러일으킬 수 있습니다(Glenberg 외. 1982). 오류 없는 학습을 지지하는 접근 방식의 의도하지 않은 결과는 학습자(및 교수자)가 개인이 감독 없이도 적절하게 수행할 수 있다고 편안하게 여기는 것인데, 이는 단기적으로는 증거가 될 수 있지만 시간이 지나면 실제로 그렇지 않을 수 있습니다(검토는 Higgins 외. 2021a, 2021b 참조).
Current evidence also suggests that both making errors and difficulty retrieving declarative and procedural knowledge when undertaking a task are necessary for facilitating good long-term retention of knowledge and skills over time (Bjork and Bjork 2011; Steenhof Woods and Mylopoulos 2020). The evidence for using desirable difficulties exists against a backdrop of other evidence proposing that practice is best when errors are eradicated by the presence of a teacher. That body of literature goes further and implies that learning through making errors is counterproductive and in the event, mistakes happen, they are simply due to inadequate instruction (Skinner 1958). Furthermore, learning without making errors – also known as errorless learning – can ‘feel good’ for learners since they see themselves progressing without making mistakes, simply fueling the illusion of knowing even more (Glenberg et al. 1982). The unintended consequences of approaches that espouse errorless learning are learners (and instructors) comfortable that individuals can adequately perform without supervision, which may be evidence in the short-term but may not actually be the case over time (for a review see, Higgins et al. 2021a, 2021b).

피드백
Feedback

피드백은 가장 많이 연구된 교육적 개입 중 하나이며, 심리학 및 교육 문헌에서 학습을 향상시키는 데 있어 피드백의 역할에 대한 많은 증거가 있습니다(Wisniewski Zierer 및 Hattie 2019). 피드백 제공에 대한 많은 리뷰와 가이드가 있지만(예: Veloski 외. 2006, Wulf Shea와 Lewthwaite 2010, Tavakol 외. 2022 참조), 인지 과학적 관점에서 임상 기술 개발에 대한 피드백을 제공할 때 특별히 고려해야 할 사항이 있습니다. 
Feedback is one of the most studied educational interventions, and there is much evidence for the role of feedback in enhance learning within the psychology and education literature (Wisniewski Zierer and Hattie 2019). A number of reviews and guides on giving feedback are available (for example, see Veloski et al. 2006; Wulf Shea and Lewthwaite 2010; Tavakol et al. 2022), however, there are specific considerations when giving feedback about the development of clinical skills from a cognitive science perspective.

학습자에게 피드백을 제공하는 것은 술기 습득 기간 동안 필수적이지만, 시간이 지남에 따라 실습 중 피드백의 양을 줄여야 한다는 데 인지과학계에서 합의가 이루어지고 있습니다. 시간이 지남에 따라 피드백의 양이 감소함에도 불구하고 학습 성과는 실제로 증가한다는 증거가 있습니다(Kovacs and Shea 2011; Sulzenbruck and Heuer 2011). 시간이 지남에 따라 피드백의 양을 조절하는 이유는, 학습자의 학습 단계에 관계없이 무제한으로 피드백을 제공할 경우, 학습자가 상당한 양의 피드백을 받는 데 익숙해질 위험을 최소화하기 위해서입니다. 학습자는 발달 단계를 거치면서 더 큰 독립성과 자기 인식을 달성하기보다는(Schmidt 1991; Salmoni 외. 1984), 스스로를 확장하거나 독립적으로 발달을 진행하기 전에 승인이나 허가를 구하는 형태로 피드백을 받는 것에 지나치게 의존하게 됩니다. 
Although giving feedback to learners is essential during the period of skill acquisition, there is a consensus within cognitive science that the amount of feedback during practise should reduce over time. Evidence suggests that learning outcomes actually increase despite the amount of feedback decreasing over time (Kovacs and Shea 2011; Sulzenbruck and Heuer 2011). The rationale for modifying the volume of the feedback given over time is to minimize the risk that learners become conditioned to receiving a significant amount of feedback if unrestrained feedback irrespective of their stage of learning. Rather than moving through development stages and achieving both greater independence and self-awareness, (Schmidt 1991; Salmoni et al. 1984), learners become over-reliant on getting feedback as a form of seeking approval or permission before extending themselves or moving on with their development independently.

스캐폴딩의 한 형태인 피드백이 점진적으로 감소하면 학습자가 경험과 전문성을 모두 얻으면서 실수를 경험할 수 있는 기회가 생겨 학습에 있어 오류의 중요성을 다시 한 번 상기시키는 동시에 실제 임상 실습에서 오류가 발생할 수 있음을 염두에 두게 됩니다. 

  • 예를 들어, 고급 생명유지술 교육에서는 학습자가 새로운 지식과 다양한 기술을 동시에 배우게 되므로 교육 초반에는 피드백이 풍부하고 지속적으로 제공되어야 하며, 새로운 학습 및 실무 상황에서도 피드백을 받을 수 있습니다.
  • 그러나 교육 과정이 끝날 무렵에는 실제 임상 실습과 유사하게 면밀한 감독 없이도 지식과 기술을 시연할 수 있어야 하므로 피드백을 거의 받지 않아야 합니다.

A progressive decrease in feedback as a form of scaffolding also creates the opportunity for learners to experience making mistakes, as they gain both experience and expertise, again reminding individuals about the importance of errors for learning, but also being mindful for the occurrence of errors in actual clinical practice.

  • For example, during advanced life support training, feedback should be plentiful and constant at the beginning of the course, since learners are introduced to new knowledge and multiple skills at the same time; arguably both also received in a new learning and working context as well.
  • However, at the end of the training course, students should receive very little feedback, since they should be able to demonstrate knowledge and skills without the need for close supervision; similar to actual clinical practice.

임상 술기를 자주 연습하는 경우 피드백은 술기 습득을 지원해야 하며, 드물게 연습하는 경우 피드백은 술기 유지를 지원해야 합니다. 예를 들어, 세실리오-페르난데스 등(2020)은 경흉부 심초음파 술기의 습득과 유지에 대해 전문가 피드백, 증강 피드백, 이 두 가지를 결합한 피드백의 세 가지 유형을 비교했습니다. 획득하는 동안 모든 그룹의 학습자는 두 번 연속 실수하지 않고 이미지를 획득할 수 있을 때까지 연습했습니다. 전문가 그룹의 학습자가 다른 그룹에 비해 기술 습득 속도가 더 빨랐습니다. 그러나 유지력 테스트에서는 겨합 교육을 받은 학습자가 다른 두 그룹보다 더 좋은 품질의 이미지를 획득했습니다. 이러한 결과는 습득에는 한 가지 유형의 피드백이 최적일 수 있지만, 유지에는 다른 유형의 피드백이 최적일 수 있음을 나타냅니다.
When a clinical skill is practiced frequently, the feedback should support its acquisition, whereas when practiced infrequently, the feedback should support retention. For example, Cecilio-Fernandes et al. (2020) compared three different types of feedback, expert, augmented, and a combination of both, on the acquisition and retention of transthoracic echocardiogram skill. During the acquisition, learners from all groups practiced until they were able to acquire images without making any mistakes twice in a row. Learners in the expert group were faster in acquiring the skill than the other groups. However, at the retention test, learners from the combined instruction obtained better-quality of images than the other two groups. These findings indicate that for acquisition one type feedback may be optimal, while for retention another type of feedback is optimal.

실무적 시사점
Practical implication

교육자에게 현지 자원의 가용성과 기존 기술 교육 시스템을 고려하는 등 주요 인지 과학 전략을 현지 상황에 맞게 조정할 수 있는 기회를 제공할 것을 권장합니다(Cecilio-Fernandes와 Sandars 2021). 임상 술기 교육에 인지 과학의 인사이트를 적용하여 얻을 수 있는 이점을 완전히 실현하려면 교수진 개발이 필요합니다. 
We recommend that educators are provided with opportunities to adapt the main cognitive science strategies to their own local context, including consideration of the availability of local resources and existing systems of skills training (Cecilio-Fernandes and Sandars 2021). This requires faculty development if the benefit of applying insights from cognitive science for clinical skills training is to be fully realized.

이 가이드는 임상 술기 습득과 술기 쇠퇴 감소를 위한 선언적 및 절차적 지식의 중요성을 강조했습니다. 임상 술기 교육 시 선언적 지식과 절차적 지식을 가장 적절하게 조합하여 제공하기 위해서는 임상 술기 교육을 시작할 때 학습자의 술기 수준을 파악하는 것이 필수적이라고 권장합니다. 우리의 경험에 따르면 이 필수 단계는 임상 술기 교육에 포함되지 않는 경우가 많습니다. 학습자가 술기를 수행하는 동안 습득해야 하는 지식의 유형을 파악하면 교육자는 개별 학습자에게 교육을 제공할 수 있습니다.

  • 예를 들어, 선언적 지식이 충분하지 않은 학습자는 이러한 지식 격차를 미리 해결하지 않으면 가장 기본적인 술기조차도 개발하는 데 어려움을 겪을 수 있습니다. 직접 또는 사전 교육용 객관식 퀴즈를 통해 구체적인 질문을 하면 선언적 지식의 결핍을 파악할 수 있습니다.
  • 절차적 지식의 결핍은 개인이 기술을 수행하는 것을 관찰함으로써 파악할 수 있습니다.
  • 반대로 선언적 지식이 있는 학습자는 훈련을 계속 진행하면서 더 정교한 절차적 지식과 더 큰 도전 과제를 개발하도록 지원할 수 있습니다.
  • 일부 학습자는 선언적 지식과 절차적 지식이 모두 적절한 수준이지만 다양한 상황의 다양한 과제에서 두 가지 요소를 통합하는 데 어려움을 겪을 수 있습니다. 예를 들어, 혈압 측정을 시도하는 일부 학습자는 사지 기형이나 동정맥 누공이 있는 환자 등 신체적 특성이 다른 환자에 대해 이 기술을 올바르게 수행하지 못할 수 있습니다. 

The Guide has highlighted the importance of declarative and procedural knowledge for clinical skill acquisition and reduction in skill decay. In order to provide the most appropriate mix of declarative and procedural knowledge during clinical skills teaching, we recommend that identifying the skill level of the learner is essential at the start of any clinical skills teaching. Our experience is that this essential step is often not included in clinical skills teaching. Identifying what type of knowledge the learner needs to acquire during the performance of the skill allows the educator to provide training to the individual learner.

  • For example, learners without sufficient declarative knowledge may struggle with developing even the most basic skills without addressing those knowledge gaps beforehand. Asking specific questions, either direct or by a pre-training multiple-choice quiz, can identify deficits in declarative knowledge.
  • Deficits in procedural knowledge can be identified by watching individuals perform the skill.
  • In contrast, learners with declarative knowledge could be supported to move on with their training and develop more sophisticated procedural knowledge and greater challenge on tasks.
  • Some learners may have appropriate levels of both declarative and procedural knowledge but struggle to integrate both components across a range of tasks in different situations. For example, some learners attempting to measure blood pressure may be unable to correctly undertake the skill across patients with different physical attributes such as individuals with limb deformities or arteriovenous fistulae in situ.

많은 의료 교육자들이 흔히 하는 질문은 어떻게 시작해야 하는가입니다. 각 상황에 따라 다르므로 간단한 답은 없지만, 의학교육자는 가장 많은 근거가 있는 전략부터 시작하는 것이 좋습니다.
A common question of many medical educators is how to start. Although there is no simple answer as it will depend on each context, we recommend that medical educators start with the strategy with the most evidence.

피드백, 간격 및 검색 연습은 의학교육을 포함하여 가장 높은 수준의 근거를 가진 전략입니다. 더 중요한 것은 선언적 및 절차적 지식을 포함한 임상 술기 교육의 효과를 입증하는 연구가 있다는 것입니다. 선언적 지식에 크게 의존하는 술기에서는 반복 학습과 인터리빙이 더 많은 증거를 가지고 있으며, 대부분 인지 심리학에서 나온 것입니다. 정교화와 생성은 임상 술기 교육에 대한 근거가 가장 적은 두 가지 개념입니다. 또한 이 두 개념은 검색 연습 및 피드백과 밀접한 관련이 있으므로 두 개념 모두 독립적인 개념이 아니라 검색 연습 및 피드백의 일부로 사용하는 것이 가장 적합할 수 있습니다.
Feedback, spaced and retrieval practice are strategies with the highest level of evidence, including in medical education. More importantly, there are studies demonstrating the effect of clinical skills training, including declarative and procedural knowledge. Overlearning and interleaving have more evidence in skills that are heavily dependent on declarative knowledge and most of them are from cognitive psychology. Elaboration and generation are two concepts with the least amount of evidence for clinical skills training. Those concepts are also closely related to retrieval practice and feedback, thus both concepts may be optimal as part of retrieval practice and feedback, instead of standalone concepts.

이 글에서는 임상 술기 교육에서 근거 기반 인지 과학 전략을 어떻게 구현할 수 있는지에 대한 예시를 제공합니다:
We provide an illustrative example of how evidence-based cognitive science strategies can be implemented in clinical skills training:

임상 술기 교육의 예시입니다: 고급 생명 유지 교육
Illustrative example of clinical skills training: Advanced life support training

고급 생명 유지(ALS)에는 청진, 심전도 해석, 다양한 약물 처방 등 여러 임상 기술이 필요합니다. 학습자가 직면한 과제는 방대한 양의 선언적 지식과 절차적 지식을 통합하는 것입니다. 표 1에서는 루게릭병 교육에 증거 기반 인지 과학 전략을 실제로 구현할 수 있는 방법에 대한 개요를 제공합니다(표 1). 표 1은 가장 많은 근거를 가진 전략과 그 전략이 중요한 이유, 수행 방법에 대한 설명이 순서대로 나열되어 있습니다.
Advanced Life Support (ALS) requires multiple clinical skills, including auscultation, interpretation of electrocardiograms, and prescribing a variety of medications. The challenge for learners is the integration of a large volume of declarative and procedural knowledge. In Table 1, we provide an overview of how evidence-based cognitive science strategies can be practically implemented into ALS training (Table 1). Table 1 is ordered by the strategy with the most evidence, and a description of what to do, why it is important, and how to do.

 

그림 1은 왼쪽에 근거 기반 인지 과학 전략을 사용한 루게릭병 입문 과정의 교육 설계를 보여줍니다. 오른쪽은 동일한 과정의 전통적인 설계입니다. 이 과정은 4개의 스킬 로테이션과 워크샵으로 나누어 진행되었습니다:

  1. 관련 리저스 스킬
  2. 평가에 대한 ABCDE 접근 방식 IIO
  3. 리듬 인식
  4. 고품질 CPR 및 제세동

Figure 1 shows on the left side a training design of an introduction to ALS course using evidence-based cognitive science strategies. On the right side is the traditional design of the same course. This course was divided into four skill rotations and workshops:

  1. Associated Resus Skills
  2. The ABCDE Approach to assessment inc IIO
  3. Rhythm Recognition
  4. High-quality CPR and Defibrillation

또한 이 코스에는 6개의 심장마비 시나리오(CAS)가 있어 학생들은 시뮬레이션된 임상 시나리오를 연습할 수 있습니다.
Also, in the course there are 6 Cardiac Arrest Scenarios (CAS) in which students have the opportunity to practice simulated clinical scenarios.

 

추가 개발 및 평가
Further development and evaluation

인지 과학은 인지 모델링의 복잡성을 발전시키고 신경 과학과 명시적으로 연결되는 흥미진진하고 진화하는 분야입니다. 인지 모델은 이제 활성화될 뇌 영역을 예측하여 선언적 지식과 절차적 지식이 상호 작용하는 방식에 대한 정보를 제공하고 복잡한 기술을 습득하는 방식에 대한 시사점과 통찰력을 제공할 수 있습니다(Borst and Anderson 2013, Taatgen 2013). 수학 교육에 사용되는 지능형 튜터의 성공적인 개발(Anderson 외. 1995)에 이어, 이제 인지 과학 원리를 임상 술기 교육에 어떻게 구현할 수 있는지 고려할 수 있는 기회가 생겼습니다. 인지 과학은 임상 추론과 같은 과제를 위한 이론 중심의 지능형 튜터를 설계하는 데 도움이 될 수 있으며, 전문가가 항상 참석하여 성과를 관찰하고 피드백을 제공할 필요를 없애줍니다. 요즘에는 다양한 피드백 시스템이 포함된 시뮬레이터를 사용하여 임상 술기에도 이러한 작업이 가능할 수 있습니다. 예를 들어, 복강경 시뮬레이터에는 인지 과학의 증거를 사용하여 개선할 수 있는 햅틱 피드백 시스템이 포함되어 있습니다. 또한 시뮬레이터는 수행 수준을 측정하고 다른 훈련 세션이 필요한 시기를 알려줄 수도 있습니다. 인지 과학은 또한 학습자가 과제를 수행하는 데 필요한 인지 처리량이 개인적으로 투자해야 하는 양을 초과할 때 더 높은 수준의 어려움을 경험할 때 주어진 과제에 대한 이해를 높이는 데 도움이 될 수 있습니다(Korbach 외. 2017). 예를 들어, 학습자가 초음파의 모든 구성 요소를 숙지하지 않은 상태에서 비만 환자에 대한 초음파 검사를 수행하도록 요구하면 초음파 검사를 성공적으로 수행하지 못하고 환자 치료가 손상될 수 있습니다. 
Cognitive science is an exciting and evolving field that has advanced in the complexity of cognitive modeling as well as making an explicit connection with neuroscience. Cognitive models can also now predict the areas of the brain that will be activated, providing information on how declarative and procedural knowledge interact, with implications and insights for how complex skills are acquired (Borst and Anderson 2013; Taatgen 2013). Following the successful development of intelligent tutors used for teaching mathematics (Anderson et al. 1995), there is now an opportunity to consider how cognitive science principles can be implement to clinical skills training in HPE. Cognitive science may help in the design of theory-driven intelligent tutors for tasks such as clinical reasoning, removing the need for an expert to always be present for observing and giving feedback on performance. This may also be feasible for clinical skills with the use of simulators, which nowadays include many feedback systems. For example, laparoscopic simulators have included a haptic feedback system, which can be improved by using evidence from cognitive science. The simulators could also measure the level of performance and indicate when another training session is necessary. Cognitive science may also help increase understanding around given learners experience greater levels of difficulty when the amount of cognitive processing necessary to execute a task exceeds what they have to personally invest into it (Korbach et al. 2017). For example, requiring learners to perform an ultrasound exam on an obese patient without mastering all the components of the ultrasound will lead to an unsuccessful performance of the ultrasound while compromising patient care.

결론
Conclusion

임상 술기를 가르치고 배우는 것은 HPE의 필수적인 측면입니다. 그러나 현재의 접근 방식으로는 여전히 숙련도 저하를 방지할 수 없습니다. HPE 교육자들은 임상 술기 습득을 최적화하고 술기 감퇴를 방지하기 위해 현재 접근 방식에 다양한 근거 기반 인지 과학 전략을 구현할 것을 권장합니다.
Teaching and learning clinical skills is an essential aspect of HPE. However, current approaches still do not prevent skill decay. We encourage HPE educators to implement a variety of evidence-based cognitive science strategies with their current approaches for the optimization of clinical skill acquisition and to prevent skill decay.

 


 

Med Teach. 2023 Jan 23;1-10. doi: 10.1080/0142159X.2023.2168528. Online ahead of print.

Using insights from cognitive science for the teaching of clinical skills: AMEE Guide No. 155

Affiliations collapse

1Department of Medical Psychology and Psychiatry, School of Medical Sciences, University of Campinas, Campinas, Brazil.

2Nottingham Medical School, University of Nottingham, Nottingham, UK.

3Health Research Institute, Edge Hill University, Ormskirk, UK.

PMID: 36688914

DOI: 10.1080/0142159X.2023.2168528

Abstract

Students have to develop a wide variety of clinical skills, from cannulation to advanced life support, prior to entering clinical practice. An important challenge for health professions' educators is the implementation of strategies for effectively supporting students in their acquisition of different types of clinical skills and also to minimize skill decay over time. Cognitive science provides a unified approach that can inform how to maximize clinical skill acquisition and also minimize skill decay. The Guide discusses the nature of expertise and mastery development, the key insights from cognitive science for clinical skill development and skill retention, how these insights can be practically applied and integrated with current approaches used in clinical skills teaching.

Keywords: Clinical skill; cognitive science; skill acquisition; skill retention.

 

보건의료전문직 교육에서 학생참여: AMEE Guide No. 152 (Med Teach, 2022)
Student engagement in health professions education: AMEE Guide No. 152
Salah Eldin Kassaba,b , David Taylorc and Hossam Hamdyc

 

소개
Introduction

학생 참여는 보건 전문직 교육(HPE)의 질에 직접적인 영향을 미칩니다.

  • 이러한 이유로 학교 운영 및 교육 활동에 대한 학생의 참여는 세계의학교육연맹의 기본 인증 기준 중 하나입니다(WFME 2020 개정판).
  • 또한, 유럽의학교육협회(Association for Medical Education in Europe)의 ASPIRE 이니셔티브에서는 학생 참여를 보건 전문직 교육 학교의 기관 우수성 지표 중 하나로 선정했습니다(Harden and Roberts 2015).
  • 또한, 파트너십을 통한 학생 참여는 영국의 QAA(Quality Assurance Authority) 및 고등 교육의 우수성 증진을 위한 전문 기관인 Advance HE와 같은 여러 조직에서 지원하고 있습니다.

학생 참여의 중요성에 대한 보편적인 합의에도 불구하고, 이 새로운 구성 요소를 개념화하고 운영하는 데 있어 통일성이 부족합니다. 
Student engagement has a direct repercussion on the quality of health professions education (HPE).

  • For that reason, engagement of students in school management and educational activities is one of the basic accreditation standards of World Federation for Medical Education (WFME 2020 revision).
  • In addition, the Association for Medical Education in Europe’s ASPIRE initiative identified student engagement as one of the indicators of institutional excellence in health professions education schools (Harden and Roberts 2015).
  • Furthermore, student engagement through partnerships is supported by several organizations such as the Quality Assurance Authority (QAA) in the UK and Advance HE which is a professional organization for promoting excellence in higher education (https://www.advance-he.ac.uk/).

Despite the universal agreement about the significance of student engagement, there is lack of uniformity in conceptualizing and operationalizing this emerging construct.

HPE에서 학생 참여에 대한 관심이 높아지면서 다양한 각도에서 이 주제에 대한 개요를 제공하는 여러 연구 간행물이 발간되었습니다.

  • ASPIRE 수상 기관의 저자들은 모범 사례를 바탕으로 기관이 파트너십을 통해 학생을 참여시킬 수 있는 방법에 대한 조언을 제공했습니다(Peters 외. 2019).
  • 또 다른 호주 저자 그룹은 웨스턴 시드니 대학교 의과 대학에서 경험한 사례를 바탕으로 학생들을 지역사회 파트너십에 참여시키기 위한 12가지 팁을 제공했습니다(Marjadi 외. 2021).
  • 최근 AMEE 가이드에서는 교육 및 학습의 공동 창작에 대한 학생 참여에 중점을 둔 학생 참여를 소개했습니다(Konings 외. 2021).
  • 또한, 학부 의학교육에서의 학생 참여는 최근 범위 검토에서 다루어졌습니다(Kassab 외. 2022).

The evolving interest in student engagement in HPE culminated in several research publications providing an overview of the topic from different angles.

  • Authors from the ASPIRE award-winning institutions built on their best practices and provided pieces of advice on how institutions can engage their students through partnerships (Peters et al. 2019).
  • Another group of Australian authors used examples from their experience at Western Sydney University School of Medicine and provided twelve tips for engaging students in community partnerships (Marjadi et al. 2021).
  • Recently, an AMEE guide introduced student engagement with a focus on student involvement in co-creation of teaching and learning (Konings et al. 2021).
  • Furthermore, student engagement in undergraduate medical education has been recently covered in a scoping review (Kassab et al. 2022).

그러나 HPE 문헌에는 파트너로서의 학생 참여를 통합하는 학생 참여에 대한 포괄적인 이론적 프레임워크가 부족합니다. 따라서 고등 교육 및 HPE 관련 문헌을 바탕으로 학생 참여의 정의와 기본 이론적 관점을 설명하겠습니다. 이에 따라 참여의 [선행 요소, 매개자, 차원, 영역, 결과]를 포함한 다양한 구성 요소와 함께 통합된 중요한 프레임워크를 소개합니다. 그런 다음 이 가이드는 보건 전문직 교육에서 학생 참여를 개선하기 위한 실질적인 전략을 제공하고 향후 연구 방향을 제시하며 마무리합니다. 
However, HPE literature is lacking a comprehensive theoretical framework of student engagement that incorporates engagement of students as partners. Therefore, drawing from literature in higher education and HPE, we will describe the definitions and underlying theoretical perspectives of student engagement. Accordingly, we will introduce an integrated overarching framework with its different components including the antecedents, mediators, dimensions, spheres, and outcomes of engagement. The guide will then provide practical strategies for improving student engagement in health professions education and end by future research directions.

학생 참여란 무엇인가요?
What is student engagement?

교육 문헌에서 학생 참여의 의미에 대한 공통된 합의에 도달하는 것은 난제였습니다. 언어학적으로 참여는 '어떤 일에 관여하는 사실'(캠브리지 사전) 또는 '누군가 또는 무언가를 이해하기 위해 그/그것에 관여하는 것'(옥스퍼드 영어 사전)으로 정의됩니다. 학생 참여의 정의에 대한 합의가 부족하지만 다음은 몇 가지 선구적인 시도입니다:
Reaching a common agreement about the meaning of student engagement in the education literature has been a conundrum. Linguistically, engagement is defined as ‘the fact of being involved with something’ (Cambridge Dictionary, nd), or ‘being involved with someone or something in an attempt to understand them/it’ (Oxford English Dictionary, nd). Although there is lack of consensus about defining student engagement, the following are some pioneering attempts:

  • 학생이 학업 경험에 쏟는 신체적, 심리적 에너지의 양(Astin 1999).
  • 학생이 교실 안팎에서 교육적으로 건전한 행동과 행동에 바치는 시간과 활력, 그리고 교육기관이 학생이 이러한 활동에 참여하도록 설득하기 위해 적용하는 지침 및 관행. (Kuh 2009).
  • 학습 활동 중에 활성화되고, 노력하고, 몰입 할 수있는 학생의 심리적 활동 상태와 학교 공동체와의 학생의 연결 상태 (Wong and Liem 2021).
  • 동료, 교수진 및 대학 커뮤니티와의 상호작용을 통해 인지적, 행동적, 정서적 수준에서 학습, 교수 및 연구 분야에서 학생들의 학업적 경험을 의미합니다(Groccia 2018).
  • amount of physical and psychological energy that the student devotes to the academic experience (Astin 1999).
  • time and vigor students dedicate to educationally sound actions and deeds internally and beyond the classroom and the guidelines and practices that institutions apply to persuade students to participate in these activities. (Kuh 2009).
  • students’ psychological state of activity that affords them to feel activated, exert effort, and be absorbed during learning activities and students’ state of connection with the school community (Wong and Liem 2021).
  • academic experiences of students in learning, teaching, and research, at the cognitive, behavioral, and emotional levels through interactions with peers, faculty, and college community (Groccia 2018).

이 가이드의 개념적 프레임워크에 따르면 학생 참여는 학습, 교수, 연구, 거버넌스 및 커뮤니티 활동을 포함하는 학업 및 비학업적 경험에 대한 학생의 시간과 에너지 투자로 정의됩니다. 학생은 인지적, 정서적, 행동적, 주체적, 사회문화적 차원에서 이러한 측면에 참여합니다.
According to our conceptual framework in this guide, we define student engagement as the student investment of time and energy in academic and non-academic experiences that include learning, teaching, research, governance, and community activities. Students are involved in these aspects at the cognitive, affective, behavioral, agentic, and socio-cultural dimensions.

학생 참여에 대한 이론적 관점
Theoretical perspectives of student engagement

학생 참여의 개념화에 대한 문헌의 불일치에도 불구하고 학생 참여가 복잡한 메타 구조라는 데는 대체로 동의하고 있습니다. 학생 참여를 설명하는 세 가지 주요 관점은

  • (1) 참여도를 시간에 따라 변화하는 학생의 내적 심리 상태로 설명하는 심리적 관점(프레드릭스 외. 2004),
  • (2) 교육기관의 효과적인 교육 활동과 학생 행동을 연결하는 행동적 관점(쿠 2009),
  • (3) 보다 광범위한 사회적, 문화적, 정치적 맥락에서 학생 경험을 검토하는 사회문화적 관점입니다.

Despite the inconsistencies in the literature about the conceptualization of student engagement, there is a general agreement that student engagement is a complex meta-construct. There are three main perspectives that explain student engagement:

  • (1) psychological perspective, which explains engagement as an internal psychological state of the students that undergo changes over time (Fredricks et al. 2004),
  • (2) behavioral perspective, which connects the effective instructional activities of the institution with the student behavior (Kuh 2009),
  • (3) sociocultural perspective, which examines the student experience from a broader social, cultural, and political context.

학생 참여의 심리적 관점
Psychological perspective of student engagement

심리적 관점은 대학 전 교육 상황에서 가장 널리 사용되는 관점이며(프레드릭스 외. 2004), HPE를 비롯한 대학 수준에서 무비판적으로 대규모로 채택되고 있습니다.

  • 이 관점의 주요 장점은 참여의 지표, 동인, 결과를 구분할 수 있다는 점입니다. 또한 참여의 정서적, 행동적, 인지적 차원을 통합하여 학생들의 경험에 대한 풍부한 이미지를 묘사할 수 있습니다.
  • 그러나 이 관점의 단점은 학생 참여도를 형성하는 데 있어 맥락의 역할을 사소화한다는 점(Furlong 외. 2003)과 차원 수 및 각 차원 아래의 지표를 포함한 명확한 개념화가 부족하다는 점입니다.
  • 심리적 관점에서 학생 참여도를 설명할 수 있는 다섯 가지 이론적 모델은 다음과 같습니다. a) 참여-식별 모델, b) 자기 결정 이론, c) 흐름 이론, d) 학업 참여도 모델, e) 메타 구성으로서의 참여도 모델입니다.

The psychological perspective is the most prevailing in precollege education contexts (Fredricks et al. 2004), and has been uncritically adopted on a large scale at university levels including HPE.

  • The main advantage of this perspective is making the distinction between indicators, drivers, and outcomes of engagement. In addition, it portrays a rich image of students experience by incorporating the affective, behavioral, and cognitive dimensions of engagement.
  • However, disadvantages of this perspective are trivializing the role of context in shaping up student engagement (Furlong et al. 2003) and lack of clear conceptualization including the number of dimensions and the indicators under each dimension.
  • The following five theoretical models can explain student engagement from the psychological perspective: a) participation-identification models, b) self-determination theory, c) flow theory, d) schoolwork engagement model, and e) model of engagement as a meta-construct.

참여 - 식별 모델
Participation – identification models

학생 참여를 개념화하기 위해 등장한 첫 번째 모델은 참여 모델입니다(Natriello 1984). 이 모델에 따르면 참여는 단순히 학생이 학교에서 제공하는 활동에 참여하는 것으로 정의됩니다(Natriello 1984). 5년 후 Finn(1989)은 이 모델을 기반으로 학생의 학교 참여가 행동적 차원정서적 차원으로 구성된다는 가설을 세웠습니다(Finn 1989).

  • 행동적 차원은 학교 활동에 대한 학생의 참여를 나타내고,
  • 정서적 차원은 학교에 대한 소속감과 가치관을 나타냅니다(Finn 1989).

이 모델은 학생들의 참여가 학업 성취도 향상으로 이어지고, 이는 학생들이 학교 공동체의 중요한 구성원이라는 느낌(소속감)과 학교를 자기 계발의 장으로 인식(소중함)함으로써 학교와 동일시하게 되는 연속적인 순환을 나타냅니다. 대체로 HPE 학생들의 행동 참여를 측정한 연구들은 참여 관점에 기반하여 이를 운영했습니다(O'Malley 외. 2003; Giddens 외. 2012; Yu 외. 2018; Riddle and Gier 2019; Nguyen 외. 2020; Smeby 외. 2020; Siqueira 외. 2021).
The first model that emerged to conceptualize student engagement was the participation model (Natriello 1984). According to this model, engagement is simply defined as the student participation in activities offered by the school (Natriello 1984). Five years later, Finn (1989) built on that model and posited that student engagement in schools consists of behavioral and emotional dimensions (Finn 1989).

  • The behavioral dimension represents participation of students in school activities, while
  • the emotional dimension represents the sense of belonging and valuing the school (Finn 1989).

The model represented a continuous cycle where participation of students leads to enhanced academic performance, which drives the students to identify with the school by feeling of being a significant member of the school community (belongingness) and recognition of the school as a place for personal development (valuing). By and large, studies which measured behavioural engagement of HPE students operationalized it based on the participation perspective (O'Malley et al. 2003; Giddens et al. 2012; Yu et al. 2018; Riddle and Gier 2019; Nguyen et al. 2020; Smeby et al. 2020; Siqueira et al. 2021).

자기 결정 이론(SDT)
Self-determination theory (SDT)

SDT는 학생의 내재적 동기를 위해서는 자율성, 역량, 타인과의 관계라는 기본적인 심리적 욕구가 충족되어야 한다고 가정합니다(Ryan and Deci 2000).

  • 자율성의 욕구는 학생의 특정 행동 선택이 자기 자신으로부터 비롯되며, 학생은 자신에게 적합한 행동을 자유롭게 결정할 수 있다는 것을 의미합니다.
  • 유능감의 욕구는 어떤 행동을 하든 자신감과 효과에 대한 인식을 의미합니다.
  • 관계성의 욕구는 다른 개인 및 주변 커뮤니티와의 연결 및 소속감을 나타냅니다.

The SDT posits that intrinsic motivation of students requires the satisfaction of basic psychological needs of autonomy, competence, and relatedness to others (Ryan and Deci 2000).

  • The need for autonomy indicates that students’ choices of certain behaviors are emanating from self, and they are free to decide the appropriate action for them.
  • The need for competence refers to the perception of confidence and effectiveness in whatever actions they do.
  • The need for relatedness indicates the feeling of connection and belongingness with other individuals and the surrounding community.

SDT를 기반으로 한 동기 개발의 자기 체계 모델(SSMMD)은 사회적 맥락 요인이 자기 체계 과정과 그에 따른 학생 참여에 어떻게 영향을 미치는지에 대한 이해를 제공했습니다(Connell and Wellborn 1991). SSMMD에 따르면 학생이 기본적인 심리적 욕구(자율성, 유능성, 관계성)를 충족하면 참여가 일어나고, 이러한 욕구가 좌절되면 불만족으로 이어집니다(Connell and Wellborn 1991; Babenko 외. 2018; Yu 외. 2018).
Building on SDT, the self-system model of motivational development (SSMMD) provided an understanding on how social contextual factors influence the self-system process and subsequently student engagement (Connell and Wellborn 1991). According to SSMMD, when students fulfill the basic psychological needs (autonomy, competence, and relatedness), engagement will occur, while thwarting these needs leads to disaffection (Connell and Wellborn 1991; Babenko et al. 2018; Yu et al. 2018).

흐름 이론
Flow theory

플로우는 '인지적으로 효율적이고 동기 부여가 되며 행복감을 동시에 느끼는 심리적 상태'로 정의되었습니다. 플로우 상태는 다음으로 나타납니다(Csikszentmihalyi 1990; Nakamura와 Csikszentmihalyi 2014). 

  • 활동에 대한 강렬한 집중과 몰입,
  • 통제하고 있다는 인식,
  • 자의식의 상실,
  • 시간 인식의 왜곡(즉 시간이 날아가는 것 같은 느낌),
  • 강렬한 즐거움

학생들이 플로우에 몰입(최적의 경험이라고도 함)하려면 달성 가능한 목표가 명확하고, 활동의 과제와 학생의 기술 간에 균형이 잡혀 있어야 하며, 진행 상황에 대한 건설적인 피드백이 제공되어야 합니다(Nakamura와 Csikszentmihalyi 2014). 플로우에서는 학생의 참여도가 높고 역동적이며 짧은 시간 동안 존재하지만 목표가 달성되면 본질적으로 보람 있는 경험으로 이어집니다. 플로우 이론은 가상 환자 시뮬레이션 중 의대생의 참여도 측정(McCoy 외. 2016)과 360° 비디오를 사용한 임상 해부학 교육(Chan 외. 2021)에 제한적으로 적용되고 있습니다. 
The flow has been defined as ‘psychological state in which the person feels simultaneously cognitively efficient, motivated, and happy’. The state of flow is manifested by

  • intense concentration and absorption in the activity,
  • perception of being in control,
  • loss of self-consciousness,
  • distortion of time awareness i.e. time appears to fly, and
  • intense enjoyment (Csikszentmihalyi 1990; Nakamura and Csikszentmihalyi 2014).

To achieve students’ engagement in flow (also called optimal experience), there should be clear attainable goals, balance between the challenges of the activity and the skills of the student, and constructive feedback on their progress (Nakamura and Csikszentmihalyi 2014). During flow, students’ engagement is high, dynamic, exists for a short period of time, but leads to an intrinsically rewarding experience if the goals are achieved. There is a limited application of flow theory in measuring engagement of medical students during virtual patient simulations (McCoy et al. 2016) and teaching clinical Anatomy using 360° videos (Chan et al. 2021).

학업 참여 모델
Schoolwork engagement model

업무 몰입은 직업 심리학에서 번아웃과 개념적으로 연결된 긍정적인 구성 요소로 등장했습니다(Schaufeli 외. 2002). 이 주제는 학계 및 보건 전문가의 업무 참여와 관련하여 이전 AMEE 가이드에서 잘 다루고 있습니다(van den Berg 외. 2017). 학생의 관점에서 학업 몰입 헌신, 활기, 몰입을 특징으로 하는 학업 관련 성취감의 긍정적인 상태로 정의되며, 이는 오랫동안 유지됩니다(Salmela-Aro와 Upadaya 2012).

  • 헌신은 학생이 학교 관련 활동을 하는 동안 느끼는 자부심, 영감, 열정을 말합니다.
  • 활력은 학교 관련 활동 중 높은 수준의 에너지와, 어려움에 직면했을 때 회복력과 끈기를 보여주려는 의지를 의미합니다.
  • 몰입은 학생이 학교 활동에서 분리되기 어려울 정도로 완전히 집중하고 행복감에 깊이 몰입하는 것을 말합니다(Schaufeli 외. 2006; Schaufeli 2017).

Work engagement emerged in occupational psychology as a positive construct that is conceptually linked to burnout (Schaufeli et al. 2002). The topic is well-covered in a previous AMEE Guide in relation to work engagement in academics and health professionals (van den Berg et al. 2017). From the student perspective, schoolwork engagement is defined as a positive state of study-related fulfillment characterized by dedication, vigor, and absorption which is maintained for a long time (Salmela-Aro and Upadaya 2012). 

  • Dedication refers to the student feeling of pride, inspiration, and enthusiasm while doing school-related activities. 
  • Vigor refers to the high level of energy during school-related activities and willingness to show resilience and persistence when confronted with difficulties. 
  • Absorption refers to the full concentration and deep engrossment with happiness so that it is difficult to get the student detached from his school activities (Schaufeli et al. 2006; Schaufeli 2017).

이 모델은 학부 의대생(Wouters 외. 2017; Liu 외. 2018; Agarwal 외. 2020; Kakoschke 외. 2021; Puranitee 외. 2022) 및 보건과학 학생(Casuso-Holgado 외. 2013; Skodova 외. 2017)의 장기 몰입을 측정하는 데 적용되었습니다.
This model has been applied for measuring long-term engagement in undergraduate medical students (Wouters et al. 2017; Liu et al. 2018; Agarwal et al. 2020; Kakoschke et al. 2021; Puranitee et al. 2022) and health sciences students (Casuso-Holgado et al. 2013; Skodova et al. 2017).

메타 구조로서의 학생 참여 모델
Model of student engagement as a meta construct

이 모델은 학생 참여를 인지적, 행동적, 정서적이라는 서로 다른 세 가지 차원으로 구성된 메타 구조로 설명합니다(프레드릭스 외. 2004). 학생 참여 구성에 네 번째 차원이 추가되었으며, 각 차원 아래의 지표에는 약간의 차이가 있습니다.

  • 사회적 참여 차원에는 다른 사람의 아이디어를 구축 및 공유하고 학습에 어려움을 겪는 다른 사람을 돕는 것과 같은 지표가 포함되었습니다(Fredricks 외. 2016).
  • 다른 연구자들은 학습 구성에 있어 학생의 능동적인 역할을 입증하기 위해 네 번째 차원으로 주체적 참여를 포함시켰습니다(Reeve and Tseng 2011; Bhat and Gupta 2019).
  • 또한 Appleton 등(2006)은 학업 참여도를 네 번째 차원으로 추가하여 학생의 과제 수행 시간 및 졸업까지 이수한 학점과 같은 지표를 포함했습니다(Appleton 등, 2006). 

This model describes student engagement as meta construct which consists of three distinct but interrelated dimensions: cognitive, behavioral, and emotional (Fredricks et al. 2004). A fourth dimension was added to the student engagement construct, with some differences in the indicators under each dimension.

  • The social engagement dimension included indicators such as building and sharing others’ ideas and helping others who are struggling in learning (Fredricks et al. 2016).
  • Other researchers included agentic engagement as a fourth dimension to demonstrate the active role of students in construction of their learning (Reeve and Tseng 2011; Bhat and Gupta 2019).
  • In addition, Appleton et al. (2006) added academic engagement as a fourth dimension with indicators such as students’ time on task and credit hours towards graduation (Appleton et al. 2006).

HPE 연구에 포함된 참여도 차원은 다양하여 구성의 다차원성을 반영하지 못했습니다. 세 가지 차원을 측정한 연구에서도 개념화와 구성의 운영 사이에 일관성이 없었습니다(Kassab 외. 2022).
The number of engagement dimensions included in HPE studies has been diverse and did not reflect the multidimensionality of the construct. Even with studies which measured the three dimensions, there was no consistency between conceptualization and operationalization of the construct (Kassab et al.
 2022).

학생 참여의 행동적 관점
Behavioral perspective of student engagement

이 관점은 학생 만족도 및 성취도와 관련된 학생 행동 및 교육기관 관행의 지표로서 참여를 강조합니다(Kahu 2013). 따라서 이 관점에서의 학생 참여는 학생이 교육적으로 목적이 있는 활동에 투자하는 시간과 노력으로 정의됩니다(Kuh G 2009). 그러나 이 관점은 개인의 심리적 상태로서의 참여보다는 제도적 관행을 더 강조합니다. 따라서 이 접근법의 단점은 참여 구조와 이를 주도하는 제도적 요인을 구분하지 못한다는 점입니다(Kahu 2013). 또한 참여의 행동적 측면을 강조하기 때문에 학생 참여의 다른 차원을 설명하는 데 있어 이 관점의 유용성은 제한적입니다. 

  • 행동적 관점에서 학생 참여도를 측정하는 데 사용되는 주요 방법으로는 전국 학생 참여도 설문조사(NSSE)호주 학생 참여도 설문조사(AUSSE)가 있습니다. NSSE와 AUSSE는 모두 고등 교육 기관의 질과 벤치마킹의 지표로 대규모로 사용되어 왔습니다.
  • HPE 연구 맥락에서는 NSSE의 일부(Hopper 2016, Hopper and Brake 2018, Hopper and Kaiser 2018) 또는 전체 설문지를 사용한 연구도 있습니다(Popkess and McDaniel 2011, Shoepe 외. 2020, Clynes 외. 2020a, 2020b).

This perspective emphasizes engagement as an indicator of student behaviors and institutional practices that related to student satisfaction and achievement (Kahu 2013). Therefore, student engagement in this perspective is defined as the time and effort students devote to educationally purposeful activities (Kuh G 2009). However, this perspective emphasizes on institutional practices more than engagement as an individual psychological state. Accordingly, the disadvantage of this approach is the lack of distinction between the engagement construct and institutional factors driving it (Kahu 2013). Furthermore, because of the emphasis on the behavioral aspect of engagement, the utility of this perspective in explaining the other dimensions of student engagement is limited.

  • The main methods used for measuring student engagement from the behavioral perspective are the National Survey for Student Engagement (NSSE) and the Australian Survey of Student Engagement (AUSSE). Both NSSE and AUSSE have been used on a large scale as a proxy of institutional quality and benchmarking in higher education.
  • In the HPE research context, studies have used portions of NSSE (Hopper 2016; Hopper and Brake 2018; Hopper and Kaiser 2018) or the full-fledged questionnaire (Popkess and McDaniel 2011; Shoepe et al. 2020; Clynes et al. 2020a2020b).

학생 참여에 대한 사회 문화적 관점
Socio-cultural perspective of student engagement

사회 문화적 관점은 사회, 문화 및 정치적 측면이 학생 참여에 미치는 영향을 다룹니다. 학생이 학교 커뮤니티에 참여하려면 자신의 정체성을 개발할 수 있는 소셜 네트워크 내에서 자신을 확립해야 합니다. 학생들은 자신의 정체성을 개발하면서 학교 커뮤니티에 대한 소속감을 형성합니다(Trowler 2019). 이러한 소속감은 [학생의 기존 규범 및 가치]와 [학교 공동체의 가치] 사이의 일치 정도와 관련이 있습니다(Kahu와 Nelson 2018). 사회경제적 계층이 낮거나 통학생, 소수 민족 또는 장애를 가진 학생과 같은 '비전통적' 학생의 경우 이러한 정체성 간의 일치성을 개발하는 것이 더 어렵습니다(Kahu 2013).
The socio-cultural perspective addresses the impact of social, cultural, and political aspects on student engagement. For students to be engaged in the school community, they need to establish themselves within a social network where they can develop their identity. As the students develop their identity, they construct a sense of belonging to the school community (Trowler 2019). This belonging relates to the degree of congruence between the existing norms and values of the student with that of the school community (Kahu and Nelson 2018). Developing this congruence between identities is more challenging for ‘non-traditional’ students such as those with low socio-economic class, commuters, ethnic minorities, or with disabilities (Kahu 2013).

[사회 문화적 관점]은 파트너십을 통한 학생 참여를 이해하기 위한 포괄적인 관점을 제공합니다. 결과적으로 다음 세 가지 이론이 파트너로서 학생의 참여를 설명할 수 있습니다: 실천 공동체, 자기 결정 이론, 포지셔닝 이론

  • 실천 공동체동일한 관심 영역을 공유하고, 정기적인 상호작용을 통해 공동 활동이나 토론에 참여하며, 반복되는 문제를 해결하기 위해 공유된 실천 모음을 개발하는 개인들의 그룹입니다.
  • '실천공동체(CoP)' 이론에 따르면, 참여 파트너는 다음을 공유하게 됩니다(Wenger 1998, 2010).
    • 공동체감 (소속감으로서의 학습),
    • 실천 (행함으로써 학습),
    • 의미 (경험함으로써 학습),
    • 정체성 (공동체의 일원이 됨으로써 학습)
  • SDT 이론은 참여에 대한 내재적 동기는 파트너가 학생에게 제공하는 자율성, 역량 및 관련성의 충족 정도에 따라 결정된다고 가정합니다(Konings 외. 2021).
  • 마지막으로, 포지셔닝 이론교사와 학생이 자신의 세계 내에서 자신과 상대방을 위치시키기 위해 자신의 위치를 구성하는 관행과 상호 작용을 말합니다(Andreouli 2010). 학생과 교사 간의 관계의 권력 역학은 각 정체성 위치와 관련된 권리와 의무와 관련이 있습니다(Andreouli 2010). 학생과 교사의 정체성을 파트너로 받아들이려면 각각의 '새로운' 위치를 공동으로 재생산하기 위한 협력공유된 이해가 필요합니다(Konings 외. 2021).

The socio-cultural perspective offers an umbrella for understanding student engagement through partnerships. Consequently, the following three theories could explain engagement of students as partners: communities of practice, self-determination theory, and positioning theory.

  • A community of practice is a group of individuals who share the same domain of interest, engage in joint activities or discussions through regular interactions, and develop a shared collection of practice to address recurring problems. According to ‘Communities of Practice (CoP)’ theory, engaged partners share the feeling of
    • community (learning as belonging),
    • practice (learning by doing),
    • meaning (learning by experiencing), and
    • identity (learning by becoming a member of a community) (Wenger 1998, 2010).
  • The SDT theory posits that intrinsic motivation for engagement in determined by the degree of fulfillment of autonomy, competence, and relatedness provided to students by their partners (Konings et al. 2021).
  • Finally, positioning theory refers to the practices and interactions by which teachers and students construct their positions to situate themselves and the other within their world (Andreouli 2010). The power dynamics of the relationships between students and teachers is associated with rights and duties related to each identity position (Andreouli 2010). Accepting the identities of the students and teachers as partners requires collaboration and shared understanding to jointly reproduce the ‘new’ positions for each (Konings et al. 2021).

학생 참여의 통합 프레임워크
Integrated frameworks of student engagement

현재 통합 프레임워크
Current integrated frameworks

학생 참여에 대한 세 가지 이론적 관점을 비판적으로 분석한 후 Kahu(2013)는 통합 프레임워크를 제안했습니다(Kahu 2013). 이 프레임워크는 학생 참여를 제도적 요인학생 요인의 영향을 받고 더 넓은 사회문화적 맥락에 내재된 내적 심리 상태로 인정합니다.

  • 제도적 요인에는 구조적(문화, 정책, 커리큘럼, 평가) 및 심리사회적(교수, 교직원, 지원, 업무량) 요인이 포함됩니다.
  • 학생 요인에는 구조적(배경, 성별, 업무량) 및 심리사회적(성격, 기술, 정체성) 요인이 포함됩니다.
  • 또한 이 프레임워크는 단기적(학업 성취도, 웰빙, 만족도, 대인관계 및 팀워크 기술) 및 장기적(유지, 업무 성공, 개인 개발, 평생 학습) 결과를 포함한 참여의 결과를 식별합니다.

Following critical analysis of the three theoretical perspectives of student engagement, Kahu (2013) proposed her integrated framework (Kahu 2013). The framework acknowledges student engagement as an internal psychological state which is affected by institutional and student factors and embedded within a wider sociocultural context.

  • The institutional factors include structural (culture, policies, curriculum, assessment) and psychosocial (teaching, staff, support, workload) factors.
  • Student factors include structural (background, gender, workload) and psychosocial (personality, skills, identity) factors.
  • The framework also identifies the outcomes of engagement including
    • short-term (academic achievement, well-being, satisfaction, interpersonal and teamwork skills) and
    • long-term (retention, work success, personal development, lifelong learning) outcomes.

그런 다음 Kahu와 Nelson은 개선된 프레임워크를 개발하여 학생의 참여와 성공을 중재하기 위해 기관과 학생 요인 간의 상호작용을 위한 심리사회적 공간으로서 '교육 인터페이스'라는 개념을 포함시켰습니다(Kahu와 Nelson 2018).

  • 이 모델에서 그들은 이러한 상호작용을 위한 네 가지 매개 구성 요소, 즉 자기 효능감, 감정, 소속감, 웰빙을 제안했습니다(Kahu and Nelson 2018).
  • 4년 후, Trowler 등은 참여의 세 가지 차원(정치적, 사회문화적, 비판적)을 추가하고
  • 참여의 매개 요인으로 동기부여, 회복탄력성, 성찰성의 세 가지 심리사회적 구성을 추가하여 Kahu와 Nelson 모델을 더욱 정교화할 것을 제안했습니다(Trowler 외. 2022).

Then, Kahu and Nelson developed a refined framework and included the concept of educational interface as a psychosocial space for interactions between institutional and student factors to mediate student engagement and success (Kahu and Nelson 2018).

  • In their model, they proposed four mediating constructs for these interactions: self-efficacy, emotions, belonging, and well-being (Kahu and Nelson 2018).
  • Four years later, Trowler et al suggested more refinement of Kahu and Nelson model by including
    • three additional dimensions of engagement: political, sociocultural, and critical, and
    • added three psychosocial constructs as mediators of engagement: motivation, resilience, and reflectivity (Trowler et al. 2022).

기존 통합 프레임워크의 문제점
Problems with the existing integrated frameworks

기존 문헌의 통합 프레임워크는 학생 참여에 대한 이해에 크게 기여했습니다(Kahu 2013; Kahu와 Nelson 2018; Trowler 외. 2022).

  • 그러나 이러한 프레임워크는 대학교육의 시장화와 맞물려 있으며 '고객' 또는 외부 이해관계자로서의 학생의 역할에 초점을 맞추고 있습니다(Kahu 2013; Milburn-Shaw와 Walker 2017; Kahu와 Nelson 2018; Trowler 외. 2022).
  • 따라서 교육의 파트너로서 학생의 역할이 명시적으로 누락되어 있고 학생의 참여가 학교의 역할로 간주됩니다.
  • 또한, 우리는 감정과 회복탄력성과 같은 이러한 모델에서 참여의 매개 구성 요소 중 일부는 각각 정서적 및 인지적 참여의 지표로 간주된다고 주장합니다 (구성 내에서). 또한, 학생의 웰빙은 참여의 중재자이자 결과로서 제안됩니다(Kahu와 Nelson 2018). 이처럼 참여의 [지표, 매개변수, 결과]를 구분하지 않는 것은 이미 존재하는 참여 구성의 모호함을 야기할 수 있습니다.

The integrated frameworks in the extant literature contributed significantly to understanding of student engagement (Kahu 2013; Kahu and Nelson 2018; Trowler et al. 2022).

  • However, these frameworks go in line with the marketization of university education and focus on the role of the student as a ‘customer’ or an external stakeholder (Kahu 2013; Milburn-Shaw and Walker 2017; Kahu and Nelson 2018; Trowler et al. 2022).
  • Therefore, the role of students as partners in education is explicitly missing and engagement of students is considered as the role of schools.
  • In addition, we argue that some of the proposed mediating constructs of engagement in these models such as emotions and resilience are considered indicators (within the construct) of emotional and cognitive engagement, respectively.
  • Furthermore, student wellbeing is proposed both as a mediator and an outcome of engagement (Kahu and Nelson 2018). This lack of distinction between indicators, mediators, and outcomes of engagement could contribute to the already existing fuzziness of the engagement construct.

HPE에서 제안하는 학생 참여의 통합 프레임워크
A proposed integrated framework of student engagement in HPE

이 가이드에서는 이전에 발표된 프레임워크에 추가 구성 요소로 학생 참여의 영역을 포함하는 포괄적인 프레임워크를 제안합니다(그림 1). 이전 모델과 마찬가지로, 우리는 학생 참여학생 내부에 위치하며, 교육기관과 학생 간의 상호작용에 의해 촉진되고, 심리사회적 메커니즘에 의해 매개되며, 더 넓은 사회문화적 맥락에 내재된, 학생 내부의 심리사회적 상태로 간주합니다. 또한 참여 구조는 행동, 정서, 인지, 대리인, 사회문화의 다섯 가지 차원으로 구성될 것을 제안합니다. 
We propose in this guide a comprehensive framework that includes the spheres of student engagement as an additional component to previously published frameworks (Figure 1). In agreement with previous models, we consider student engagement as an internal psychosocial state which is situated within the student, facilitated by the interactions between the institution and the student, mediated by psychosocial mechanisms, and embedded within a broader sociocultural context. We propose also that the engagement construct consists of five dimensions: behavioral, emotional, cognitive, agentic, and sociocultural.

그러나 제안된 프레임워크의 독창성은 학생의, 학생과 함께, 학생에 의한 참여를 강조하기 위해 학생의 '파트너십을 통한 참여'라는 개념을 도입한 것입니다. 이에 따라 참여의 영역을 크게 다음으로 분류합니다.

  • (1) 교실, 실험실, 임상 현장, 숙제 또는 자율 학습 등 개별 학습 활동에 대한 참여를 포함하는 자기 학습 참여
  • (2) 교육 제공, 학술 연구, 학교 거버넌스 및 품질 보증, 지역사회 활동에서의 파트너십을 포함하는 파트너십을 통한 참여

However, the uniqueness of the proposed framework is introducing the concept of student ‘engagement through partnerships’ to emphasize engagement ofwith, and by, students. Accordingly, we classify the spheres of engagement into two main categories:

  • (1) engagement in own learning, which includes engagement in individual learning activities in the classroom, laboratory, clinical sites, homework, or self-study, and
  • (2) engagement through partnerships, which include partnerships in provision of education, scholarly research, school governance & quality assurance, and community activities.

파트너십을 통한 학생 참여의 분류는 이 분야에 대한 최신 문헌(Healey와 해링턴 2014)과 학생 참여에 대한 기관의 우수성 기준(ASPIRE 2020)에 따라 개념적으로 안내됩니다. 이 프레임워크가 학생 참여에 대한 향후 발전과 이해를 촉진할 수 있는 토론을 촉발할 수 있기를 바랍니다.
The classification of student engagement through partnerships is conceptually guided by the current literature on this field (Healey and Harrington 2014), and by the AMEE-ASPIRE criteria for institutional excellence in student engagement (ASPIRE 2020). We hope that this framework triggers a debate which can stimulate future developments and understanding of student engagement in HPE.




박스 1 통합 이론 모델을 HPE 실무에 적용한 예시.
Box 1 Example of applying the integrative theoretical model in HPE practice.

한 교수진이 한 학기 동안 의대생들을 위한 PBL 튜토리얼에서 퍼실리테이터 역할을 맡게 되었습니다. 그러나 학생들의 참여도가 낮다는 것을 알게 된 그녀는 소그룹 PBL 튜토리얼에서 학생들의 참여도를 높이기 위한 계획을 세웠습니다. 학생 참여의 통합 프레임워크에 대한 이해를 바탕으로 그녀는 다음과 같은 전략을 적용할 계획이었습니다:
A faculty member is assigned the role of a facilitator in PBL tutorials for medical students for one semester. However, she noticed that the level of engagement of students is low and planned to improve the engagement of students in small group PBL tutorials. From her understanding of the integrated framework of student engagement, she planned to apply the following strategies:

  1. 참여 선행 조건 수정: PBL 수업 내에서 
    1. 서로를 지지하고 위협하지 않는 문화를 조성하고,
    2. 학생들과의 협업과 소통을 촉진하기 위해 그룹 규칙을 정하고,
    3. 사전 지식을 자극하는 질문을 하고,
    4. 커리큘럼 내용과 평가 방법에 대해 학생들과 토론하고,
    5. 건설적인 피드백을 제공했습니다.
    6. 또한 학생들은 각 세션이 끝날 때 그룹 기능에 대해 반성할 수 있습니다.
    이러한 전략은 동기 부여, 소속감, 자기 효능감 및 성찰력을 향상시킴으로써 학생의 참여를 촉진할 것으로 기대됩니다.

    Modify engagement antecedents
    : by
    1. creating a supportive, non-threatening culture inside the PBL class,
    2. establishing group rules to facilitate collaboration and communication with the students,
    3. asking questions that stimulate prior knowledge,
    4. discussing with students the curriculum content and assessment methods, and
    5. providing constructive feedback.
    6. Students can also reflect on the group function at the end of each session.
      These strategies are expected to mediate engagement of students through enhancing their motivation, sense of belonging, self-efficacy, and reflectivity.
  2. 다양한 참여 차원을 목표로 함:
    1. 학생 상호 작용을 자극하고 (행동적),
    2. 사례 이해와 관련된 탐색적 질문을 하고 (인지적),
    3. 학생에게 환자 사례에 대한 자신의 감정을 성찰하도록 요청하고 (정서적),
    4. 학생에게 그룹 역학 및 학습의 질과 개선 방법에 대해 성찰하도록 요청하고 (주체적),
    5. 튜토리얼 시간 이후 학생들이 서로 다시 연결될 수 있도록 계획하는 등 (사회문화적).

    Target the multiple engagement dimensions: by
    1. stimulating student interactions (behavioral),
    2. asking probing questions related to understanding of the case (cognitive),
    3. asking the students to reflect on their feelings about the patient case (emotional),
    4. asking students to reflect on the quality of group dynamics and learning and how to improve it (agentic), and
    5. plan for students to reconnect with each other beyond the time of the tutorial (socio-cultural).
  3. 파트너로서 학생의 참여 장려: 학생에게 PBL 튜토리얼의 진행자로서 동료 튜터링을 번갈아 가며 하도록 요청합니다. 학생은 자가 평가 및 동료 평가를 수행할 수도 있습니다.
    Encourage students’ engagement as partners: by asking students to take turns in peer tutoring as facilitators of PBL tutorials. Students can also undertake self-assessment and peer assessment.

다음 섹션에서는 학생 참여를 위해 제안된 통합 프레임워크의 다섯 가지 구성 요소(선행 요소, 매개자, 차원, 영역 및 결과)에 대한 개요를 제공합니다.
In the following section, we provide an overview of the five components of the proposed integrated framework for student engagement (antecedents, mediators, dimensions, spheres, and outcomes).

학생 참여의 선행 조건
Antecedents of student engagement

조직 문화와 학생 참여
Organizational culture and student engagement

이전 연구에서는 학교 문화와 고등 교육에서의 학생 참여 사이의 관계를 입증했습니다(van der Velden 2012; Borhan 2020). 학생 참여에 도움이 되는 학교 문화는 다음과 같은 특징이 있습니다:

  • (a) 팀 빌더로 인식되는 친근한 리더,
  • (b) 학교 구성원이 신뢰하는 대인 관계로 가족 같은 느낌을 공유(Cameron and Quinn 2011),
  • (c) 학생이 커뮤니티의 파트너로 간주(Borhan 2020),
  • (d) 유연성과 적응력을 갖춘 창의성, 혁신 및 성장에 집중(Borhan 2020) 등이 있습니다. 그

러나 안정성과 규정 준수에 중점을 두는 위계질서나 관료주의 문화에서는 학생 참여가 감소합니다(van der Velden 2012; Borhan 2020).
Previous studies demonstrated the relationship between school culture and student engagement in higher education (van der Velden 2012; Borhan 2020). School cultures that are conducive for student engagement are characterized by the following:

  • (a) friendly leaders who are perceived as teambuilders,
  • (b) members of the school share the feeling of a family with trusting interpersonal relationships (Cameron and Quinn 2011),
  • (c) students are considered as partners in the community (Borhan 2020), and
  • (d) focus on creativity, innovation, and growth with flexibility and adaptability (Borhan 2020).

However, student engagement is decreased in hierarchy or bureaucratic cultures, which focus on stability and enforcing compliance (van der Velden 2012; Borhan 2020).

학생 요인
Student factors

학생 참여에서 성별의 역할을 조사한 연구에 따르면 여학생은 남학생보다 더 높은 수준의 학습 참여를 보이는 반면(Presoto 외. 2021), 남학생은 연구 및 과외 활동에 참여할 가능성이 더 높은 것으로 나타났습니다(Salgueira 외. 2012). 또한 대학 입학 전 학업 수준은 학생의 대학 참여에 대한 중요한 예측 변수입니다(Salgueira 외. 2012). 그러나 사회경제적 지위, 부모의 특성 또는 소수 집단 소속을 포함한 학생의 배경은 학생들의 대학 참여도에 영향을 미치지 않는 것으로 보입니다(Pike and Kuh 2005; Trowler and Trowler 2010; Karabchuk and Roshchina 2022). 반면에 시간제 취업 학생의 참여도는 비취업 학생과 비교했을 때 비슷하거나(Clynes 외. 2020a), 낮거나(Walpole 2003), 높거나(Kuh 2009) 일관되지 않은 결과를 나타냈습니다.
Studies examining the role of gender in student engagement demonstrated that female students exhibit higher level of learning engagement more than male students (Presoto et al. 2021), while male students are more likely to engage in research and extracurricular activities (Salgueira et al. 2012). In addition, pre-college entry academic level is a significant predictor for student engagement in university (Salgueira et al. 2012). However, the student background including socioeconomic status, parental characteristics, or belonging to minority groups, does not appear to play a role in the level of engagement of students in colleges (Pike and Kuh 2005; Trowler and Trowler 2010; Karabchuk and Roshchina 2022). On the other hand, engagement of part-time employed students has yielded inconsistent results with similar (Clynes et al. 2020a), lower (Walpole 2003), or higher (Kuh 2009) engagement compared with non-employed students.

관계
Relationships

긍정적인 학생-학생 및 학생-교수 관계는 학생 참여도를 향상시키는 것으로 나타났습니다(Leiken 2017; Banos 외. 2019; Geraghty 외. 2020; Shoepe 외. 2020). 이는 교사가 친절하고 접근하기 쉬우며 도전적인 학습 활동을 제공할 때 분명해집니다(Bryson and Hand 2007). 학생과 교직원 또는 또래 간에 신뢰 관계를 발전시키면 학생들은 타인의 판단으로부터 자유로워지고 교육 활동에 온전히 집중할 수 있어 심리적 안전이 증진됩니다(Tsuei et al. 2019).

  • 학생에게 심리적으로 안전한 관계의 결과에는 위험 감수, 권위 있는 사람에게 말할 수 있는 자유, 창의성 등이 있습니다(Edmondson and Lei 2014). 또한 심리적 안전은 팀 성과를 향상시키고 구성원의 지속 가능한 다양성을 보장합니다.

이러한 관계의 근간이 되는 메커니즘은 업무량에 대한 인식을 낮추고(Kember 2004; Xerri 외. 2018), 학생들의 학교 공동체에 대한 소속감을 높이는 것입니다(Milburn-Shaw and Walker 2017).
Positive student-student and student-faculty relationships appear to enhance student engagement (Leiken 2017; Banos et al. 2019; Geraghty et al. 2020; Shoepe et al. 2020). This becomes evident when teachers are friendly, approachable, and provide challenging learning activities (Bryson and Hand 2007). Developing trusting relationships between students and staff or with their peers promote psychological safety of students by feeling free from judgment by others while fully concentrating on engagement with educational activities (Tsuei et al. 2019).

  • The outcomes of psychologically safe relationships for students include risk-taking, freedom to speak-up with higher authority persons, and creativity (Edmondson and Lei 2014). In addition, psychological safety enhances team performance and ensures sustainable diversity of the members.

The mechanisms underlying these relationships could be lowering the perception of workload (Kember 2004; Xerri et al. 2018) and enhancing students’ sense of belonging to the school community (Milburn-Shaw and Walker 2017).

학생 참여의 중재자
Mediators of student engagement

동기 부여
Motivation

동기는 참여의 원동력으로 간주되며 활동 전에 발생하는 반면, 참여는 활동 자체에서 학생의 적극적인 참여를 나타냅니다(그림 2). 내재적 동기는 의대생의 학습 참여를 향상시키고(Berman and Artino 2018; Wu 외. 2020), 간호대생의 학습 참여와 긍정적인 상관관계가 있습니다(Bayoumy and Alsayed 2021).

  • 특히, 학생 동기인지적 참여를 유의미하게 예측하며(Walker 외. 2006), 행동적 참여의 매개 메커니즘으로 제안되고 있습니다(Trowler 외. 2022).

동기는 과제 가치를 높이고 숙달 목표를 촉진함으로써 학생의 참여도를 높일 수 있습니다(Trowler 외. 2022). 예를 들어, 새로운 수술 기술을 배우면서 [숙달 목표]를 설정한 의대생은 [성과 목표]를 설정한 학생에 비해 과제에 더 잘 몰입하는 것으로 나타났습니다(Gardner 외. 2016).
Motivation is considered as the fuel for engagement and occurs before an activity while engagement represents students’ active involvement during the activity itself (Figure 2). Intrinsic motivation enhances learning engagement of medical students (Berman and Artino 2018; Wu et al. 2020) and positively correlates with learning engagement of nursing students (Bayoumy and Alsayed 2021).

  • Specifically, student motivation significantly predicts cognitive engagement (Walker et al. 2006), and is proposed as a mediating mechanism for behavioral engagement (Trowler et al. 2022).

Motivation could increase student engagement through enhancing task value and promoting mastery goals (Trowler et al. 2022). For example, medical students prepared with mastery goals while learning new surgical skills exhibited better engagement in the tasks compared with those provided with performance goals (Gardner et al. 2016).

자기 효능감
Self-efficacy

자기효능감은 제도적 요인과 학생 참여 간의 상호작용을 매개하는 메커니즘 중 하나로 제안되었습니다(Kahu and Nelson 2018; Trowler 외. 2022). 이전 연구에서는 자기효능감과 학생 참여도 간에 긍정적인 유의미한 관계가 있음이 입증되었습니다(Walker 외. 2006; Koob 외. 2021). 또한 학생의 동기 부여와 자기 효능감은 모두 학습 참여도를 높여 의대생의 학업 성취도를 높입니다(Wu 외. 2020). 자기효능감의 증가는 과제 자원에 대한 인식을 높이고 학생의 주체성을 장려함으로써 학생의 참여를 향상시킬 수 있습니다(Trowler 외. 2022). 
Self-efficacy has been proposed as one of the mediating mechanisms of interactions between institutional factors and student engagement (Kahu and Nelson 2018; Trowler et al. 2022). Previous studies demonstrated a positive significant relationship between self-efficacy and student engagement (Walker et al. 2006; Koob et al. 2021). In addition, both student motivation and self-efficacy increase academic performance of medical students through enhancing learning engagement (Wu et al. 2020). The increase in self-efficacy could enhance student engagement by increasing awareness of task resources and encouraging student agency (Trowler et al. 2022).

소속감
Sense of belonging

소속감교직원과 동료들로부터 가치 있고, 받아들여지고, 포용되고, 격려받는다는 느낌입니다. 이 느낌은 학생이 학교에 대한 심리적 소속감 또는 학교 및 학교 내 개인과의 관계를 나타냅니다(Burkhardt 외. 2015). 소속감은 학생 참여, 특히 사회-문화적 참여(Trowler 외. 2022)를 매개하는 메커니즘으로 제안되었습니다(Kahu and Nelson 2018). 또한, 또래에 비해 높은 소속감을 경험하는 학생은 정서적 및 행동적 참여가 증가하는 경향이 있습니다(Gillen-O'Neel 2021). 소속감은 더 높은 수준의 학생 동기 부여를 유지함으로써 참여도 증가에 기여할 수 있으며(Ryan and Deci 2000), 학생 참여는 소속감도 증가시킬 수 있습니다(Finn 1989).
Sense of belonging is the feeling of being valued, accepted, included, and encouraged by staff and peers. This feeling represents a psychological membership of the students with their school or connection with the school and individuals within the school (Burkhardt et al. 2015). Belonging has been proposed as a mediating mechanism of student engagement (Kahu and Nelson 2018), especially the socio-cultural engagement (Trowler et al. 2022). In addition, students who experience higher sense of belonging compared with their peers tend to have increased emotional and behavioral engagement (Gillen-O’Neel 2021). The sense of belonging can contribute to increased engagement by maintaining a higher level of student motivation (Ryan and Deci 2000), while student engagement can also increase the sense of belonging (Finn 1989).

반성
Reflectivity

성찰은 학생 참여도를 향상시킬 수 있기 때문에 중요한 교육 전략으로 간주됩니다(Sandars 2009). 임상 문제를 해결하는 동안 의도적인 성찰은 의대생의 학습 활동 참여를 촉진하고 학습 성과를 증가시켰습니다(Ribeiro et al. 2019). 이러한 의도적인 성찰은 학생들이 문제에 대해 더 많이 알 수 있도록 상황적 흥미를 유발하여 참여도를 높일 수 있습니다(Linnenbrink-Garcia 외. 2013; Ribeiro 외. 2018).
Reflection is considered an important educational strategy because of its potential enhancement of student engagement (Sandars 2009). Deliberate reflection while solving clinical problems promoted engagement of medical students in a learning activity and increased learning outcomes (Ribeiro et al. 2019). This deliberate reflection may increase engagement by triggering situational interest for students to know more about the problem (Linnenbrink-Garcia et al. 2013; Ribeiro et al. 2018).

학생 참여의 차원
Dimensions of student engagement

  • 인지적 참여는 학생의 학습에 대한 심리적 투자를 의미합니다(Fredricks 외. 2004). 
    • 요구 사항을 넘어서고, 도전을 선호하며, 콘텐츠를 이해하고 숙달하는 데 노력을 기울이고, 학습에 메타인지 및 심층 전략을 사용하는 등 
  • 행동적 참여긍정적인 행동, 끈기, 과제 완수를 위한 노력, 참여, 질문, 집중, 주의 집중, 학교 기반 활동 참여를 말합니다.
  • 정서적 참여는 (즐거움, 흥미, 지루함, 불안, 행복, 슬픔과 같은) 교실, 학교 또는 교사에 대한 정서적 반응을 말합니다(Fredricks 외. 2004). 
  • 사회문화적 참여는 관점을 확장하고 다양한 배경과 관점에 대한 인식과 인정을 의미합니다(프레임워크 2022). 이 차원은 지역사회 서비스, 파트너십, 동료 교육, 기관 거버넌스와 같은 활동에서 분명하게 드러납니다. 사회문화적 참여의 지표에는 다음이 포함됩니다: 
    • (a) 사회적 상호작용의 질과 타인과의 상호작용을 바탕으로 정체성에 대한 새로운 관점을 얻는 것, 
    • (b) 다양한 관점을 가진 타인으로부터 배우는 것을 받아들이는 것
    • (c) 다양한 사회 및 문화 집단에서 개인의 행동과 그 함의를 예측하는 것(프레임워크 2022). 
  • 주체적 참여는 학생의 교육 경로, 미래의 삶, 기존 및 더 넓은 사회적 환경에 대한 학생의 영향력을 의미합니다(Klemenčič 2017). 
    • 교실 내 주체적 참여의 지표는 학습에 대한 학생의 기여도와 학생이 받는 교육을 통해 확인할 수 있습니다(Reeve and Tseng 2011). 
    • 교실 밖에서의 주체적 참여에는 커뮤니티 활동, 동료 교수/멘토링, 교육기관 거버넌스 및 품질 보증에 대한 학생의 적극적인 참여가 포함될 수 있습니다.
  • Cognitive engagement refers to the psychological investment in learning, where the student
    • goes beyond requirements, prefers challenges, directs the effort towards understanding and mastering content, and uses metacognitive and deep strategies in learning (Fredricks et al. 2004). 
  • Behavioral engagement refers to positive conduct, persistence, directing the effort towards completing learning tasks, participating, asking questions, concentrating, paying attention, and participation in school-based activities. 
  • Emotional engagement refers to emotional reactions to classroom, school, or teachers such as enjoyment, interest, boredom, anxiety, happiness, and sadness (Fredricks et al. 2004). 
  • Sociocultural engagement refers to expanding viewpoints and providing awareness of, and appreciation for, diverse backgrounds and perspectives (Framework 2022). This dimension is evident in activities such as community services, partnerships, peer teaching, and institutional governance. Indicators of sociocultural engagement include the following:
    • (a) quality of social interactions and gaining new perspectives on identity based on interactions with others,
    • (b) accepting to learn from others with diverse perspectives, and
    • (c) predicting individual actions and their implications in different social and cultural groups (Framework 2022). 
  • Agentic engagement refers to the students’ influence on their educational pathways, their future lives, and their existing and broader social surroundings (Klemenčič 2017).
    • Indicators of agentic engagement inside the classroom could be through students’ contributions to their learning and the instruction they receive (Reeve and Tseng 2011).
    • Agentic engagement outside the classroom could include students’ active involvement in community activities, peer teaching/mentoring, and membership in institutional governance and quality assurance.

위의 참여도 차원을 측정하는 방법은 참여도를 개념화하는 이론적 관점, 참여도 차원의 수, 각 차원 내의 지표, 측정의 입자 크기(활동 대 학교 수준, 단기 대 장기)에 따라 다릅니다. 학생 참여도를 측정하는 방법은 이 가이드의 범위에 포함되지 않지만, 이 주제에 대한 좋은 리뷰가 HE 문헌에 나와 있습니다(Henrie 외. 2015; Fredricks 외. 2019; Hofkens and Ruzek 2019). 학생의 HPE 참여도를 측정하는 가장 일반적인 방법은 심리적 관점에서 개념화하며 행동적 관점을 사용한 연구는 거의 없습니다. 반면에 에이전트 및 사회 문화적 관점에서 학생 참여를 측정하는 연구는 드뭅니다. 

The methods of measuring the above dimensions of engagement differ according to the theoretical perspective of conceptualizing engagement, number of engagement dimensions, indicators within each dimension, and grain size of measurement (activity vs school level, short-term vs long-term). The methods of measuring student engagement are not part of the scope for this guide, but there are good reviews on the topic in HE literature (Henrie et al. 2015; Fredricks et al. 2019; Hofkens and Ruzek 2019). The most prevailing methods for measuring student engagement in HPE conceptualize it from the psychological perspective with few studies using the behavioral perspective. On the other hand, studies which measure student engagement from the agentic and socio-cultural perspectives are sparse.

HPE에서 학생 참여도를 측정하는 데 일반적으로 사용되는 방법으로는 자가 보고 설문지, 직접 관찰, 실시간 측정, 인터뷰/포커스 그룹 등이 있습니다(표 1). 

  • 의학교육에서 학생 참여도를 측정하는 데 가장 많이 사용되는 자가 보고 설문지는 여러 가지가 출판되어 있습니다(Kassab 외. 2022). 자기보고의 특별한 짧은 변형은 시간 경과에 따른 학생의 '흐름' 상태를 측정하는 경험 샘플링 방법입니다(Chan 외. 2021). 
  • 직접 관찰은 개인 수준에서의 행동 참여 또는 학생 그룹의 참여를 측정할 수 있습니다. 
    • HPE에서 참여도를 측정하는 데 사용되는 주요 관찰 도구 중 하나는 STROBE (O'Malley 외. 2003; Kelly 외. 2005) 또는 수업 내 참여도 측정(IEM)이라고 하는 수정된 버전의 STROBE입니다(Alimoglu 외. 2014; 2017). STROBE는 말하기, 듣기, 읽기, 쓰기와 같은 학생 행동에 중점을 두고 학습자 간 참여도, 학습자 대 교수자 참여도 및 자기 참여도를 측정합니다(O'Malley 외. 2003; Kelly 외. 2005). 실시간 측정은 특정 학습 활동에서 진화하는 학생 참여의 역동성을 감지하는 데 사용됩니다. 
  • 학생 참여도를 실시간으로 측정하는 주요 방법으로는 로그 파일, 얼굴 표정 분석, 시선 추적 및 생리적 측정이 있습니다. 
    • 로그 파일에서 수집한 참여도 지표의 예로는 학습 과제에 소요된 시간(Ribeiro 외. 2019), 학생이 임상 사례에 대해 작성한 게시물의 수 또는 특정 게시물의 방문 빈도(Kay and Pasarica 2019), 관련 소셜 미디어 페이지의 클릭 수(Quesnelle and Montemayor 2020), 게시된 문서의 질문 또는 수정 횟수(Saperstein 외. 2015; Caton 외. 2021) 등을 들 수 있습니다. 
    • 시선 추적 기술과 자동화된 얼굴 표정 인식은 실시간 시스템에서 학생의 참여를 감지하는 데 사용됩니다(Miller 2015). 
    • 심박수의 변화도 인지적 참여의 지표로 사용되었습니다(Darnell and Krieg 2019). 
    • 또한, 갈바닉 피부 반응은 정서적 참여의 지표로서 뇌 각성을 측정하는 데 사용될 수 있습니다(Poh et al. 2010). 
  • 마지막으로, 인터뷰/포커스 그룹은 일반적으로 구조적 또는 반구조적 방식으로 설계된 개방형 질문에 응답하도록 학생들에게 요청하는 방식으로 진행됩니다. 
    • 인터뷰에서 나오는 주제에는 참여의 다른 차원이 포함될 수 있습니다(Janssen 외. 2015; Wang and Ji 2021). 학생들은 맥락적 요인의 역할을 포함하여 참여/참여하지 않는 이유를 설명할 수도 있습니다.

Commonly used methods for measuring student engagement in HPE are self-reported questionnaires, direct observation, real-time measures, and interview/focus groups (Table 1).

  • There are several published self-reported questionnaires which are the most used measures of student engagement in medical education (Kassab et al. 2022). A special short variant of the self-reports is the experience sampling method to measure students’ state of ‘flow’ over time (Chan et al. 2021). 
  • Direct observations can measure behavioral engagement at the individual level or engagement of a group of students.
    • One of the main observational instruments used for measuring engagement in HPE is the STROBE (O'Malley et al. 2003; Kelly et al. 2005) or a modified version of STROBE called in-class engagement measure (IEM) (Alimoglu et al. 2014; 2017). The STROBE measures learner to learner engagement, learner to instructor engagement, and self-engagement with a focus on student behaviours such as speaking, listening, reading, and writing (O'Malley et al. 2003; Kelly et al. 2005). 
  • Real-time measures are used for detecting the dynamicity of student engagement as it evolves in specific learning activities. The main real-time measures of student engagement are log files, analysis of facial expressions, eye-tracking, and physiological measures.
    • Examples of engagement indicators gathered from log files are
      • time spent on a learning task (Ribeiro et al. 2019),
      • number of posts written by the students on a clinical case or frequency of visits to a specific post (Kay and Pasarica 2019),
      • number of clicks on relevant social media pages (Quesnelle and Montemayor 2020), and
      • number of questions asked, or edits on a posted document (Saperstein et al. 2015; Caton et al. 2021).
      • Eye-tracking techniques and automated recognition of facial expressions are used by real-time systems for detecting student engagement (Miller 2015).
    • Changes in heart rate have been also used as an indicator of cognitive engagement (Darnell and Krieg 2019).
    • In addition, galvanic skin response can be used to measure brain arousal as an indicator of emotional engagement (Poh et al. 2010).
  • Finally, interviews/focus groups are usually conducted by asking students to respond to open-ended questions which are designed in a structured or semi-structured way.
    • Themes emerging from interviews may include other dimensions of engagement (Janssen et al. 2015; Wang and Ji 2021). Students can also describe the reasons for their engagement/disengagement including the role of contextual factors.

 

학생 참여의 영역
Spheres of student engagement

학생의 자기 주도적 학습 참여
Student engagement in own learning

학생이 스스로 학습에 참여하는 것은 학생 행동의 중요한 측면이자 학업 성공의 매개체입니다. 학습 상황은 대규모 수업, 소그룹 학습, 과제 또는 자가 학습 등 다양할 수 있습니다. 각 맥락에서 학생 참여의 차원은 다를 수 있습니다. 예를 들어, 소그룹 PBL 튜토리얼에 참여하는 학생은 참여의 인지적 및 행동적 차원 외에 사회적-정서적 차원에 더 중점을 둘 수 있습니다. 
Student engagement in own learning represents an important aspect of student behavior and a mediator for academic success. The learning context could vary from large classroom, small group learning, assignments, or self-learning. The dimensions of student engagement in each context could be different. For example, student engagement in small group PBL tutorials may focus more on the social-emotional dimensions in addition to the cognitive and behavioral dimensions of engagement.

학생 참여의 개념적 명확성을 확보하려면 인지적 참여자기조절 학습 간의 관계를 설명하는 것이 중요합니다. 일부 연구자들은 자기조절 학습 전략이 학생의 인지적 참여도를 나타내는 지표라고 제안했습니다(Horner and Shwery 2002; Furlong and Christenson 2008). 그러나 최근의 한 모델에서는 이러한 구성 요소가 상호 연관되어 있지만 별개의 실체라는 것을 보여주었습니다(Li and Lajoie 2022). 이 모델에서 SRL은 사전 사고, 수행 및 자기 성찰의 세 가지 연속적인 단계를 기반으로 개념화됩니다(Zimmerman 2000).

  • 사전 생각 단계에서 학생은 학습 과정과 학습에 필요한 정신적 노력의 양을 계획합니다.
  • 수행 단계에서 학생은 학습 활동 중에 사용할 인지 전략과 정신적 노력의 수준을 스스로 통제하고 모니터링합니다.
  • 자기 성찰 단계에서 학생은 계획된 학습 목표를 달성하는 데 있어 인지적 참여 수준의 효과를 평가하고 그에 따라 해당 수준을 조정합니다(그림 2).

이 모델은 활동의 유형과 계획된 학습 목표 달성에 대한 이러한 요소의 효과에 따라 정신적 노력의 양과 사용되는 학습 전략의 유형을 조정할 수 있는 인지적 참여의 역동적인 특성을 보여줍니다(Li and Lajoie 2022). 
To ensure the conceptual clarity of student engagement, it is important to delineate the relationship between cognitive engagement and self-regulated learning. Some researchers proposed that self-regulated learning strategies are indicators of student cognitive engagement (Horner and Shwery 2002; Furlong and Christenson 2008). However, a recent model demonstrated that these constructs are interrelated, but distinct, entities (Li and Lajoie 2022). In that model, SRL is conceptualized based on three consecutive phases: forethought, performance, and self-reflection (Zimmerman 2000).

  • In the forethought phase, students plan their learning process and the amount of mental effort in learning.
  • In the performance phase, students perform self-control and monitoring of the cognitive strategies and the level of mental effort to be used during the learning activity.
  • During the self-reflection phase, students evaluate the effectiveness of the level of cognitive engagement in achieving the planned learning goals and adjust that level accordingly (Figure 2).

This model demonstrates the dynamic nature of cognitive engagement where the amount of mental effort and the types of learning strategies used can be adjusted according to the type of activity and the effectiveness of these factors on achieving the planned learning goals (Li and Lajoie 2022).

파트너십을 통한 학생 참여
Student engagement through partnerships

"파트너로서의 학생"(SaP)의 참여는 고등 교육을 변화시킬 수 있는 잠재적인 방법으로 제안되었습니다(Healey 외. 2016). 파트너십에 대한 가장 널리 인용되는 정의는 다음과 같습니다: '모든 참여자가 커리큘럼 또는 교육학적 개념화, 의사 결정, 실행, 조사 또는 분석에 반드시 동일한 방식은 아니지만 동등하게 기여할 수 있는 기회를 갖는 협력적이고 호혜적인 과정'(Cook-Sather 외. 2014)입니다. 이러한 파트너십은 주로 학생과 학자 간에 이루어지며, 학생-학생 파트너십은 파트너십 구조의 두 번째로 일반적인 범주에 속합니다(Mercer-Mapstone 외. 2017). 파트너십은 기본적으로 결과보다는 참여의 과정이며, 다른 형태의 학생 참여를 보완합니다(Healey 및 Harrington 2014). 파트너십의 주요 기본 원칙은 상호 존중, 정직, 투명성, 파트너 간의 정기적인 연결, 공유된 가치와 목표입니다(Bretts 외. 2018). 파트너십을 뒷받침하는 공유 가치는 다음과 같습니다:

  • (a) 모든 파트너가 자신의 경험, 관점 또는 재능을 공유할 권리가 있는 포용성,
  • (b) 각 파트너가 프로세스를 통해 이익을 얻거나 배울 목표가 있는 상호성,
  • (c) 모든 파트너가 파트너십을 통해 일할 때 동등한 권한을 갖는 권한 부여,
  • (d) 파트너가 상호 존중을 기대하며 정직하고 개방적으로 소통하는 신뢰,
  • (e) 소속감과 자신이 기여한 것에 대한 감사함,
  • (f) 그룹으로서 하는 일과 개별적으로 기여한 것에 대한 책임감 (Healey and Harrington 2014). 

Engagement of “students as partners” (SaP) has been proposed as a potential method for transforming higher education (Healey et al. 2016). The most widely quoted definition of partnership is: ‘a collaborative, reciprocal process through which all participants have the opportunity to contribute equally, although not necessarily in the same ways, to curricular or pedagogical conceptualization, decision making, implementation, investigation, or analysis’ (Cook-Sather et al. 2014). This partnership is mainly between students and academics, with the student-student partnership coming as the second most common category of partnership structure (Mercer-Mapstone et al. 2017). Partnership is basically a process of engagement rather than an outcome and complements other forms of student engagement (Healey and Harrington 2014). The main underlying principles of partnerships are mutual respect, honesty, transparency, regular connections between partners, and shared values and goals (Bretts et al. 2018). The shared values underpinning the partnerships include the following:

  • (a) inclusivity, where every partner has the right to share his or her experience, perspectives, or talents; (b) reciprocity, where each partner has a goal to benefit or learn for the process;
  • (c) empowerment, where every partner has an equal power in working through partnerships;
  • (d) trust, where partners communicate with honesty and openness with an expected mutual respect;
  • (e) sense of belonging and feeling of appreciation of what they contribute; and
  • (f) responsibility for what they do as a group and for what they contribute individually (Healey and Harrington 2014).

파트너십을 통해 학생이 얻을 수 있는 혜택으로는 지속적인 참여에 대한 동기 부여 증가(Border 외. 2021), 자기 효능감 증가(Henderson 외. 2018), 자신감 및 의사소통 기술 향상(Qazi 외. 2019; Reeves 외. 2019), 학업 경험에 대한 이해도 향상 등이 있습니다. 2019), 학업 경험에 대한 이해도 향상, 학생-교직원 및 학생-학생 관계 개선, 메타인지 능력 향상(Geraghty 외. 2020; Gheihman 외. 2021), 소속감 증가(Bovill 외. 2021), 규율 학습 개선, 학생 복지 증진(Cook-Sather 외. 2014; Mercer-Mapstone 외. 2017) 등의 효과가 있습니다. 

  • 그러나 파트너십의 내용이나 과정에 대한 전문성 부족, 학생과 교수진 간의 권력 관계, 음성 피로로 인한 파트너십 지속 꺼려하기 등 학생에게는 어려움이 있습니다(Konings 외. 2021). 
  • 교사의 어려움에는 학생에게 권한을 양보하는 데 따른 보안 부족, 변화에 대한 저항, 파트너십 참여 방법 이해, 시간 제약 등이 있습니다(Bovill 외. 2015; Gheihman 외. 2021; Konings 외. 2021).

The student benefits from partnerships include increased motivation for continuous engagement (Border et al. 2021), increased self-efficacy (Henderson et al. 2018), improved self-confidence and communication skills (Qazi et al. 2019; Reeves et al. 2019), improved understanding of academic experiences, enhanced student-staff and student-student relationships, enhanced metacognitive skills (Geraghty et al. 2020; Gheihman et al. 2021), increased sense of belonging (Bovill et al. 2021), improved discipline learning, and increased student well-being (Cook-Sather et al. 2014; Mercer-Mapstone et al. 2017).

  • However, there are challenges for students that include lack of expertise in the content or process of partnerships, power relations between students and faculty, and reluctance to continue in partnerships because of voice fatigue (Konings et al. 2021).
  • Challenges for teachers include lack of security because of conceding authority to students, resistance to change, understanding how to engage in partnerships, and time constraints (Bovill et al. 2015; Gheihman et al. 2021; Konings et al. 2021).

파트너십에 대한 학생의 참여는 저강도, 저호혜성, 단기 참여에서 고강도, 고호혜성, 장기 참여에 이르기까지 다양합니다(Gheihman 외. 2021). 다섯 가지 영역 각각에서 학생 참여의 단계는 연속체를 따라 질적으로 다른 네 가지 형태의 참여를 나타낼 수 있습니다(Dunne and Zandstra 2011). 

  • 1단계(협의)는 학생이 자신의 의견, 견해 및 관점을 표현할 수 있는 기회를 제공하는 단계,
  • 2단계(참여)는 학생에게 보다 적극적인 역할을 제공하는 단계,
  • 3단계(파트너)는 학생과 교수진/직원/다른 학생 간의 공동 협력 과정,
  • 4단계(변화 주도자)는 학생이 전적으로 통제권을 갖는 단계를 의미합니다.

Participation of students in partnerships can range from low-intensity, low-reciprocity, short-term involvement to high-intensity, high reciprocity, long-term involvement (Gheihman et al. 2021). The stages of student engagement in each of the five spheres can represent four qualitatively different forms of participation along a continuum (Dunne and Zandstra 2011).

  • Stage 1 (consult) means provision of opportunities for students to express their opinions, views, and perspectives;
  • stage 2 (participate) is providing more active role for students;
  • stage 3 (partner) is a shared collaborative process between students and faculty/staff/other students; and
  • stage 4 (change agent) where students are in full control.

다음 섹션에서는 파트너십을 통한 학생 참여의 네 가지 영역인 (1) 교육 프로그램 제공, (2) 학술 연구, (3) 기관 거버넌스 및 질 향상, (4) 커뮤니티 활동에 대해 설명합니다.
In the following section, we describe the four areas of student engagement through partnerships: (1) provision of the education programme, (2) scholarly research, (3) institutional governance and quality enhancement, and (4) community activities.

교육 프로그램 제공에 대한 학생 참여
Student engagement in provision of the education programme

강력한 파트너십 유형은 학생들을 교육, 학습 및 평가에 참여시키는 것입니다. 교육 프로그램 제공의 파트너로서 학생의 참여에는 프로그램의 공동 제작자, 교사 및 평가자로서의 역할이 포함됩니다.
A powerful type of partnerships is engaging students in teaching, learning, and assessment. Involvement of students as partners in provision of the education programme include their role as co-creators, teachers, and assessors in the programme.

공동 제작자로서의 학생
Students as co-creators

교육과정 중심의 교육적 파트너십 활동에는

  • (1) 교육 과정 또는 프로그램을 가르치기 전에 공동 설계하는 것을 의미하는 교육과정의 공동 창작,
  • (2) 교육 과정 또는 프로그램이 전개되는 동안 공동 설계하는 것을 의미하는 교육과정의 공동 창작,
  • (3) 교육 과정 제공 후 재설계에 참여하는 것,
  • (4) 숨겨진 교육과정 발견의 네 가지가 있습니다(Bovill 외. 2015; Cook-Sather 외. 2014).

코스의 공동 생성은 이전에 해당 코스를 학습한 학생 또는 현재 해당 코스에 등록한 학생이 수행할 수 있습니다. 공동 창작 과정에서는 교수진과 학생 모두 학교에서의 기존 입장을 생각하고 수정해야 합니다(Cook-Sather 외. 2014). 공동 창작 과정에 다양한 이해관계자를 참여시키면 설계된 커리큘럼 또는 코스의 품질이 향상됩니다(Konings 외. 2021). 또한 공동 창작의 파트너십은 교수진의 교육 품질과 전문성 개발에 긍정적인 영향을 미칩니다(Cook-Sather 외. 2014).
There are four curriculum-focused pedagogical partnership activities:

  • (1) co-creation of the curriculum, which refers to co-design of a course or a program before it is taught,
  • (2) co-creation in the curriculum, which refers to co-design of a course or program as it unfolds,
  • (3) engagement in re-designing a course after it is delivered, and
  • (4) uncovering the hidden curriculum (Bovill et al. 2015; Cook-Sather et al. 2014).

The co-creation of a course can be conducted by students who previously studied the course or the current enrolled students in the course. The process of co-creation necessitates both faculty and students to think and revise their conventional positions in the school (Cook-Sather et al. 2014). Involving different stakeholders in the process of co-creation improves the quality of the designed curriculum or course (Konings et al. 2021). In addition, partnerships in co-creation positively affects the teaching quality and professional development of faculty (Cook-Sather et al. 2014).

또래 멘토/교사로서의 학생
Students as peer-mentors/tutors

교사로서의 학생의 역할에는 동료 멘토링과 동료 지원 학습(PAL)이 포함됩니다. 또래 멘토링을 실행하려면 멘토가 자신만의 멘토링 전략을 자유롭게 사용할 수 있도록 지원하고 멘토와 멘티 간의 커뮤니케이션 방식에 유연성을 허용해야 합니다(Reeves 외. 2019). 멘티와의 관계를 촉진하기 위해 또래 멘토는 자신의 역할을 '전문가'가 아닌 '지원'으로 인식해야 하며, 두 번째 역할은 일부 멘티가 또래 멘토를 강사의 연장선으로 느낄 때 불편함을 유발할 수 있기 때문입니다(Reeves 외. 2019). PAL 튜터에게는 더 나은 교수 기술 개발과 선행 학습의 통합이라는 이점이 있습니다(이와타 외. 2014; GMC 2018). 또한 PAL 튜터링은 튜터와 튜티 모두의 학업 성취도를 향상시킵니다(Iwata 외. 2014; GMC 2018). 교수진에게는 학생의 경험을 더 잘 이해하고, 학생의 관심사, 두려움, 문화, 포부를 더 잘 이해하며, 학생의 역량을 더 존중할 수 있다는 이점이 있습니다(Reeves 등. 2019). 
The role of students as teachers includes peer mentoring and peer-assisted learning (PAL). Implementing peer mentoring requires supporting mentors in the freedom to use their own strategies of mentoring and allowing flexibility in the mode of communication between mentors and mentees (Reeves et al. 2019). For facilitating the rapport with their mentees, peer mentors need to identify their role as ‘supporting’ students and not as ‘expert’ students as the second role can create discomfort for some mentees when they feel the peer mentors as an extension of their lecturer (Reeves et al. 2019). The benefits for PAL tutors include development of better teaching skills and consolidation of prior learning (Iwata et al. 2014; GMC 2018). PAL tutoring also improves academic performance of both tutors (Iwata et al. 2014; GMC 2018) and tutees (Brierley et al. 2022). The advantages for faculty are more understanding about students’ experience, more appreciation of the students concerns, fears, culture, aspirations, and more respect for their capabilities (Reeves et al. 2019).

평가자로서의 학생
Students as assessors

자가 및 동료 평가, 평가 방법 및 평가 기준의 공동 설계, 공동 채점, 피드백의 유형 및 형식 결정 등 평가의 파트너로서 학생을 위해 여러 가지 방법이 사용됩니다(Ní Bheoláin 및 O'Riordan 2020). 학생이 자신의 평가에 참여하면 참여도(Juan 2021), 학업 성취도, 평가에 대한 관심 및 동기가 향상되고(Cook-Sather 외. 2014), 과목 학습 결과에 대한 더 깊은 이해를 촉진합니다(Bergmark and Westman 2016). 권력과 위계의 역학 관계는 평가에서 학생 파트너십에서 다시 표면화되며, 이러한 문제를 관리하는 것이 이러한 파트너십의 성공을 위한 핵심 요소입니다. 
Several methods are used for students as partners in assessment such as self- and peer assessment, co-design of assessment methods and assessment criteria, collaborative grading and determining the type and format of feedback (Ní Bheoláin and O’Riordan 2020). Student participation in their own assessment enhances their engagement (Juan 2021), academic performance, their interest and motivation for this assessment (Cook-Sather et al. 2014), and promotes deeper understanding of the subject learning outcomes (Bergmark and Westman 2016). The dynamics of power and hierarchy come back on the surface in student partnership in assessment, and managing these challenges is a key factor for the success of these partnerships.

학술 연구에 대한 학생 참여
Student engagement in scholarly research

학부 연구에 참여하는 학생은 졸업 후 더 많은 연구 논문을 발표하고(Reinders 외. 2005), 연구 수행에 대한 자신감을 높이며(Pearson 외. 2017), 학계에 대한 관심을 높이고, 교수직을 얻을 확률이 높아지는 등 많은 이점을 누릴 수 있습니다(Solomon 외. 2003). 또한 학생의 연구 참여는 특히 의학 지식과 전문성(Zhang 외. 2022), 의학 문헌에 대한 비판적 사고와 비판적 평가(Laskowitz 외. 2010) 영역에서 학습 성과를 향상시킵니다. 학생 참여의 주요 원동력은 연구와 경력 발전에 대한 본질적인 관심입니다(Remes 외. 2000; Alamri 2019). 그러나 학생들의 연구 참여에 장애가 되는 요인으로는

  • 지도교수 찾기 어려움(Remes 외. 2000; Alexander 외. 2019),
  • 시간 제약(Remes 외. 2000; Pearson 외. 2017; Trethewey 외. 2018; Alamri 2019; Alexander 외. 2019),
  • 자금 부족(Pearson 외. 2017),
  • 연구 교육 부족(Hamdan and Kakarla 2013; Alexander 외. 2019),
  • 잠재적 연구 프로젝트에 대한 인식 부족(Hamdan and Kakarla 2013; Trethewey 외. 2018) 등이 있습니다.

Student engagement in undergraduate research has many advantages including publishing more research articles after graduation (Reinders et al. 2005), increased confidence in conducting research (Pearson et al. 2017), increased interest in academic career, and higher probability of getting an academic position (Solomon et al. 2003). In addition, engaging students in research improves their learning outcomes especially in the domains of medical knowledge and professionalism (Zhang et al. 2022), as well as in critical thinking and critical appraisal of medical literature (Laskowitz et al. 2010). The main driving factors for student engagement are intrinsic interest in research and career progression (Remes et al. 2000; Alamri 2019). However, the barriers to engagement of students in research include

  • difficulty in finding supervisors (Remes et al. 2000; Alexander et al. 2019),
  • time constraints (Remes et al. 2000; Pearson et al. 2017; Trethewey et al. 2018; Alamri 2019; Alexander et al. 2019),
  • lack of funding (Pearson et al. 2017),
  • lack of training in research (Hamdan and Kakarla 2013; Alexander et al. 2019), and
  • lack of awareness about potential research projects (Hamdan and Kakarla 2013; Trethewey et al. 2018).

거버넌스 및 품질 보증에 대한 학생 참여
Student engagement in governance and quality assurance

기관 거버넌스 및 품질 보증에 대한 학생 참여는 성숙한 형태의 참여를 나타냅니다(Naylor 외. 2021). 그러나 경제적, 문화적 요인으로 인해 교육기관 거버넌스에서 학생의 참여는 교육기관마다 큰 편차를 보이며 아직 미발달 상태입니다(Green 2019).

  • 기관 거버넌스에 학생을 참여시키는 근거는 민주적 가치에 대한 헌신, 학생의 목소리를 개선의 촉진 요인으로 인정하는 것, 또는 인증 시스템의 요구 사항을 충족하는 것 등 다양할 수 있습니다(Naylor 외. 2021).
  • 거버넌스 및 품질 보증에 대한 학생 참여의 이점은 직원 및 동료와의 커뮤니케이션 개선, 개인 및 전문성 개발, 리더십 기술 및 전문성 향상입니다(Meeuwissen 외. 2019). 또한 학생들은 메타인지적 사고, 주장하기, 토론, 로비 및 네트워킹 기술을 습득합니다(Meeuwissen 외. 2019).
  • 학교 거버넌스 및 품질 보증의 파트너로서 HPE 학생의 역할에 대한 연구 문헌은 아직 초기 단계에 있습니다.
    • 의과대학에서 보고된 학생-교수 파트너십에 대한 성공 사례의 예로는 실시간 피드백을 통해 학생의 커리큘럼 평가를 촉진하기 위한 Ed Reps 프로그램(Scott 외 2019), 모듈 공동 감독으로서의 학생(Milles 외 2019), 커리큘럼에 대한 직접적인 피드백을 제공하기 위한 학생 커리큘럼 위원회(SCB) 프로그램(Geraghty 외 2020) 등이 있습니다. 이러한 프로그램은 학생과 교수진 간의 협업을 강화하고(Geraghty 외. 2020), 커리큘럼과 학교 문화에 긍정적인 영향을 미치며(Scott 외. 2019), 학생의 역량을 강화하는 것으로 잘 알려져 있습니다.

Student participation in institutional governance and quality assurance represents a mature form of engagement (Naylor et al. 2021). However, due to economic and cultural factors, student agency in institutional governance is still underdeveloped with large variations between institutions (Green 2019).

  • The rationale for involving students in institutional governance could vary from commitment to democratic values, to acknowledging the student voice as a promoting factor for improvement, or just meeting the requirements of accreditation systems (Naylor et al. 2021).
  • The benefits of student engagement in governance and quality assurance are improved communication with staff and peers, personal and professional development, leadership skills, and professionalism (Meeuwissen et al. 2019). Students also acquire skills of metacognitive thinking, making an argument, debating, lobbying, and networking (Meeuwissen et al. 2019).
  • The research literature on the role of HPE students as partners in school governance and quality assurance is still in infancy.
    • Examples of the reported success stories from medical schools about student-faculty partnerships include the Ed Reps program that aimed to promote curriculum evaluation by students through real-time feedback (Scott et al. 2019), students as module co-directors (Milles et al. 2019), and Student Curricular Board (SCB) program for providing direct feedback about curriculum (Geraghty et al. 2020). These programs were well-perceived as enhancing collaborations between students and faculty (Geraghty et al. 2020), having positive impact on the curriculum and school culture (Scott et al. 2019), and empowering students.

커뮤니티 활동에 대한 학생 참여
Student engagement in community activities

학생이 지역사회에 참여하는 것은 학생들이 지역사회로부터 배우고 지역사회에 봉사하는 시너지 효과를 낼 수 있는 경험입니다(Talib 외. 2017). 지역사회 활동에 대한 학생 참여의 예로는

  • a) 지역사회 봉사 선택 과목 수강,
  • b) 학생이 운영하는 보건소에서의 자원봉사,
  • c) 기본 생활 지원 및 지역사회 기반 재난 지원 교육 제공 등 지역사회를 위한 교육 활동 조직(Shannon 2019),
  • d) 워크숍 및 공개 프레젠테이션과 같은 아웃리치 활동 등을 들 수 있습니다.

이러한 파트너십은 학교 기반 프로젝트를 통해 시작하거나 학생들이 독립적인 파트너십을 통해 시작할 수 있습니다. 학생 주도의 자원봉사 이니셔티브의 대상 지역사회는 일반적으로 노숙자(Batra 외. 2009), 이주 노동자(Sin 외. 2019)와 같이 의료적으로 소외되거나 불우한 지역사회입니다. 지역사회 활동에 참여하는 학생들은 의사소통 기술, 팀워크, 전문성(Nauhria 외. 2021), 대인관계 기술(Haidar 외. 2020), 공감, 의사결정, 리더십 기술(Loh 외. 2016)과 같은 역량을 습득합니다. 또한 봉사활동은 학생들의 이타심을 증진하고(Batra 외. 2009), 학업 성취도를 향상시키며(Blue 외. 2006), 사회적 결정요인과 건강 간의 관계에 대한 학생들의 이해를 높이고(Talib 외. 2017), 졸업 후 지역사회 봉사에 대한 헌신을 높입니다(Jones 외. 2014). 또한 종단적 봉사 학습은 취약하고 소외된 불우 환자에 대한 의대생들의 태도를 개선합니다(Arebalos 외. 2021). 

Student engagement with the community is a synergistic experience where students learn from, and provide service to, the community (Talib et al. 2017). Examples of student engagement in community activities include the following:

  • a) taking a community service elective,
  • b) volunteering at student-run health clinics,
  • c) organizing educational activities for the community such as providing training in basic life support and community-based disaster support (Shannon 2019), and
  • d) outreach activities such as workshops and public presentations.

These partnerships can be initiated through school-based projects or through independent partnerships by the students. Target communities for student-led volunteering initiatives are usually medically underserved or disadvantaged communities such as homeless population (Batra et al. 2009), and migrant workers (Sin et al. 2019). Students who engage in community activities acquire competencies such as communication skills, teamwork, professionalism (Nauhria et al. 2021), interpersonal skills, (Haidar et al. 2020), empathy, decision-making, and leadership skills (Loh et al. 2016). In addition, volunteering services promote altruism of students (Batra et al. 2009), improve academic performance (Blue et al. 2006), improve student understanding of the relationship between social determinants and health (Talib et al. 2017), and increase the commitment to community volunteering after graduation (Jones et al. 2014). Furthermore, longitudinal service learning improves the attitudes of medical students towards the vulnerable and marginalized underserved patients (Arebalos et al. 2021).

학생 참여의 결과
Outcomes of student engagement

단기적 성과
Short-term outcomes

HPE 학생의 참여도와 학업 성취도 간의 관계를 조사한 연구에서는 일관되지 않은 결과가 나왔습니다. 예를 들어, 지식 기반 시험에서의 성과와 인지적 참여(Wong 외. 2015, Rotgans 외. 2018, Bhat and Gupta 2019, Reinke 2019, Hadie 외. 2021)와 행동적 참여(수업 활동 참여로 측정)간에 긍정적인 관계가 있다는 연구 결과도 있습니다(Dewar 외. 2021, Grant 외. 2021). 그러나 여러 차원을 사용한 학생 참여도와 학업 성취도 간의 상관관계는 긍정적이거나(Bayoumy and Alsayed 2021) 유의미한 관계가 없는 것으로 나타났습니다(Hopper and Brake 2018, Hopper and Kaiser 2018, Pickering and Swinnerton 2019). 학생 참여는 정신 건강 및 심리적 웰빙을 개선하고(Steele and Fullagar 2009) 전반적인 삶의 만족도를 향상시킬 수 있습니다(Lewis 외. 2011). 또한 학생 참여는 교사의 동기 부여를 향상시켜 교육 과정에 긍정적인 영향을 미칩니다(Frenzel 외. 2009).
Studies examining the relationships between engagement of HPE students and academic performance have yielded inconsistent findings. For example, studies demonstrated a positive relationship between performance in knowledge-based examinations and cognitive engagement (Wong et al. 2015; Rotgans et al. 2018; Bhat and Gupta 2019; Reinke 2019; Hadie et al. 2021) as well as behavioral engagement (measured by participation in class activities (Dewar et al. 2021; Grant et al. 2021). However, the correlations between student engagement using multiple dimensions and academic performance yielded positive (Bayoumy and Alsayed 2021) or no significant relationships (Hopper and Brake 2018; Hopper and Kaiser 2018; Pickering and Swinnerton 2019). Student engagement can improve mental health and psychological well-being (Steele and Fullagar 2009) and improves overall life satisfaction (Lewis et al. 2011). In addition, student engagement enhances teacher motivation which positively affects the teaching process (Frenzel et al. 2009).

장기적인 결과
Long-term outcomes

학생 참여는 학생 유지율을 높이고, 소진이 학생 유지율에 미치는 영향을 줄이며(Abreu Alves 외. 2022), 개인 개발과 평생 학습을 향상시킵니다(Fredricks 외. 2004; Trowler 2010). 또한 교실 참여와 자기 주도적 학습 사이에는 긍정적인 관계가 있습니다(Rossi 외. 2021).
Student engagement increases student retention, reduces impact of burnout on student retention (Abreu Alves et al. 2022), and enhances personal development and lifelong learning (Fredricks et al. 2004; Trowler 2010). There is also a positive relationship between classroom engagement and self-directed learning (Rossi et al. 2021).

HPE에서 학생 참여도를 높이기 위한 실용적인 전략
Practical strategies for enhancing student engagement in HPE

학교 커뮤니티 문화 조성
Promote a culture of a school community

교육기관은 학생들이 커뮤니티의 일원이라는 느낌을 받을 수 있도록 환영하는 문화를 제공해야 합니다. 학생들은 학교에 입학한 첫날부터 이러한 느낌을 발전시켜야 합니다. 공동체 문화를 조성하기 위한 전략에는 다음이 포함될 수 있습니다:

  • a) 학생마다 '학생이 되는 방식'이 다르다는 개념을 받아들이고 포용적인 방식으로 수용(Bryson 2014; Ashwin and McVitty 2015),
  • b) 학생 그룹 및 기관 차원의 그룹에 가입하여 신입생을 학교 커뮤니티에 통합하여 역량과 소속감을 개발하도록 지원(Milburn-Shaw and Walker 2017),
  • c) 소속감을 높이기 위해 오리엔테이션 및 동료 지원 프로그램을 처음에 실시,
  • d) 저소득층 및 소수 민족 학생 등 취약 계층 학생에게 재정 지원 제공(Hu 2010) 등이 있습니다.

Institutions need to provide a welcoming culture to students to make them feel part of a community. Students should develop this feeling starting from the first day of admission to the school. Strategies for promoting a culture of community could include the following:

  • a) adopt the concept that students are different in their ‘ways of being a student’ and accommodate them in an inclusive way (Bryson 2014; Ashwin and McVitty 2015),
  • b) support integrating newcomers into the school community by joining student groups and institutional level groups (Milburn-Shaw and Walker 2017) to develop the sense of competence and belongingness,
  • c) implement orientation and peer support programs at the beginning to enhance the sense of belonging, and
  • d) provide financial assistance to students from disadvantaged populations such as low-income and minority students (Hu 2010).

심리적 안전 문화 조성
Foster a culture of psychological safety

학생들이 심리적 안정감을 느끼면 타인에게 유능한 이미지를 투영하기보다 학습 과제에 집중할 수 있습니다(Tsuei 외. 2019). 심리적 안전 문화를 조성하기 위해 학교는 다음과 같은 전략을 적용할 수 있습니다:

  • a) 의견을 말하는 학생에게 보상하고 학생과 교직원 간, 학생과 또래 간 쉬운 소통 채널을 구축(Tsuei et al. 2019),
  • b) 교사는 다가가고 자신의 실수를 인정하며 학생의 말을 주의 깊게 듣고 존중과 신뢰로 학생을 대해야 합니다(Konings et al. 2021),
  • c) 학생의 자율성과 창의성을 증진하고 안전한 환경에서 학생의 발달과 성장의 기회를 제공해야 하며,
  • d) 학생 평가는 채점보다는 건설적인 피드백을 통해 학생의 개인 개발에 초점을 맞춰야 합니다(Tsuei et al. 2019).

Students’ feeling of psychological safety makes them focus on engagement on the learning tasks rather than projecting an image of competence for others (Tsuei et al. 2019). To create a culture of psychological safety, schools could apply the following strategies:

  • a) reward students who speak-up and establish easy channels of communications between students and staff and between students and their peers (Tsuei et al. 2019),
  • b) teachers should be approachable, admit their own mistakes, listen attentively to their students, and treat them with respect and trust (Konings et al. 2021),
  • c) promote student autonomy and creativity and provide opportunities for their development and growth in a safe environment, and
  • d) student assessment should focus on personal development of students through constructive feedback rather than on grading (Tsuei et al. 2019).

파트너로서의 학생 참여 문화 확립
Establish a culture of student engagement as partners

파트너로서 학생의 참여를 지원하는 문화를 구축하는 것은 학교 리더, 교수진, 학생의 헌신과 공동의 노력이 필요한 복잡한 과정입니다.
Establishing a culture that supports engagement of students as partners is a complex process that requires commitment and collective effort from the school leaders, faculty, and students.

텍스트 상자 2는 HPE에서 파트너십을 통해 학생 참여를 개발하고 확립하기 위한 실용적인 팁을 제공합니다.Text Box 2 provides practical tips for developing and establishing student engagement through partnerships in HPE.

상자 2 학생 참여를 위한 파트너십을 개발하고 구축하기 위한 팁
Box 2 Tips for developing and establishing partnerships for student engagement in HPE.


1. 모든 관련 파트너가 공동으로 참여할 수 있는 프레임워크를 개발합니다.
2. 파트너십 프레임워크를 학교의 전략 계획과 연계합니다.
3. 파트너십 프로세스에 대비할 수 있도록 학생과 교직원 모두에게 역량 강화를 제공합니다.
4. 가능한 한 빨리 학생들을 파트너십 프로세스에 적극적으로 참여시킵니다(Peters 외. 2019).
5. 파트너십의 목적, 기대 결과, 범위, 일정, 기본 가치, 관계의 경계, 잠재적 과제에 대해 학생과 교직원 간에 상호 공개적으로 논의하여 파트너십을 시작하세요.
6. 교사가 접근하기 쉽고, 학생의 말을 주의 깊게 경청하고, 학생에 대한 기대치를 설정하고, 학생의 발언을 장려하고, 존중과 신7. 뢰로 학생을 대하는 파트너십에서 심리적 안전의 분위기를 보장합니다(Konings 외. 2021).
8. 파트너십의 지속 가능성을 위한 제도적 계획을 수립합니다.
9. 관리 리더의 참여를 촉진하고 관련 파트너의 역할을 명확히 하는 정책을 개발합니다.
파트너로서 교직원과 학생 모두 가치 있고 학습 과정을 보완하는 다양한 유형의 전문성을 보유하고 있음을 인정합니다(Englander 외., 2020).
10. 학생이 환자, 동료, 교사와 함께 관련 종단적 경험을 개발할 수 있는 기회를 최적화합니다(예: 종단적 통합 사무직)(Englander et al., 2020).
11. 학생의 기여를 기반으로 프로세스의 변화를 승인하여 학생 기관을 장려합니다.
12. 파트너십에 대한 학생의 참여를 인정하고 소중히 여깁니다.
13. 파트너십 실행 후 모든 관련 이해관계자로부터 피드백을 수집하고 필요한 조치를 취하여 루프를 닫습니다.
14. 학생과 교직원이 파트너십 경험을 되돌아볼 수 있는 시간을 갖도록 합니다.
  1. Develop a framework for engagement collaboratively by all concerned partners.
  2. Align the partnership framework with the strategic plan of the school.
  3. Provide capacity building for both students and staff to prepare them for the partnership process.
  4. Involve students proactively in the partnership process as early as possible (Peters et al. 2019).
  5. Start the partnerships by mutual and open discussion between students and staff about the purpose, expected outcomes, scope, timeline, underlying values, boundaries for the relationship, and potential challenges.
  6. Ensure a climate of psychological safety in partnerships where teachers are approachable, listen attentively to their students, set expectations for students, encourage them to speak-up, and treat them with respect and trust (Konings et al. 2021).
  7. Establish an institutional plan for sustainability of the partnerships.
  8. Develop policies to facilitate engagement of management leaders and have clear roles of involved partners.
  9. Acknowledge that both staff and students as partners possess a different type of expertise that is valuable and complement the learning process (Englander et al., 2020).
  10. Optimize opportunities for students to develop relevant longitudinal experiences with patients, peers, and teachers e.g. longitudinal integrated clerkships (Englander et al., 2020).
  11. Promote the student agency by endorsing the change in the process based on contributions from students.
  12. Acknowledge and value student participation in partnerships.
  13. Collect feedback from all relevant stakeholders following implementation of the partnership and take necessary actions to close the loop.
  14. Allow a dedicated time for students and staff to reflect on the partnership experience.

교육 프로그램 제공에 학생 참여 유도
Engagement in provision of the education programme

커리큘럼 공동 제작에 학생의 참여를 촉진하기 위해 다음과 같은 전략을 권장합니다(Konings et al. 2021) 

  • 가) 학생의 관심사에 맞는 영역에 참여하도록 권유하여 공동창작 과정에 참여하도록 동기를 부여하고 점진적으로 유도,
  • 나) 교사와 학생 간의 권력 차이를 줄이고 참여 중 학생의 심리적 안전 보장,
  • 다) 공동창작 역량을 강화하는 영역에서 학생을 훈련,
  • 라) 학생에게 권한을 부여하고 공동창작 과정을 가치 있는 경험으로 인식하도록 자극하여 공동창작 과정에 계속 참여하도록 동기를 보장한다.

To promote engagement of students in the co-creation of the curriculum, the following strategies are recommended (Konings et al. 2021):

  • a) motivate students to get involved in co-creation process by inviting them to participate in areas that fit with their interest and gradually induct them into the process,
  • b) reduce the power difference between teachers and students and ensure psychological safety of students during participation,
  • c) train students in areas that enhance their skills in co-creation, and
  • d) ensure motivation of students to continue in the co-creation process by empowering them and stimulating them to see the co-creation process as a valuable experience.

학생들의 또래 튜터링 참여를 촉진하기 위해 권장되는 전략은 다음과 같습니다. 

  • a) 학업적 전제 조건이 없는 학생의 자발적인 경험으로 또래 튜터링을 고려하고,
  • b) 또래 튜터링이 의무적으로 요구되는 경우 참여 동기를 높이기 위해 학생들에게 주제 목록에서 선택할 수 있는 선택권을 부여하고(Ross and Cameron 2007; Blanco 외. 2014), 또래 튜터링에 참여하는 학생의 동기를 강화하는 방법을 고려합니다. 2014),
  • c) 성인 학습의 기본 개념, 인지 과학 원리, 소그룹 교수법, 교실 상호작용 촉진과 같은 주제에 중점을 두고 동료 튜터에게 교수 기술을 향상시키기 위한 교육을 제공하고,
  • d) 교수진이 동료 튜터에게 교수 기술에 대한 피드백을 제공하여 교수를 향상시킵니다(Freret et al. 2017).

The following strategies are recommended to stimulate student engagement in peer tutoring:

  • a) consider having the peer tutoring as a voluntary experience for students with no academic pre-requisites,
  • b) in case of mandatory requirement for peer tutoring, consider giving students the choice to select from a list of topics to enhance their motivation for participation (Ross and Cameron 2007; Blanco et al. 2014),
  • c) train peer tutors to enhance their teaching skills with a focus on topics such as fundamental concepts of adult learning, cognitive science principles, small group teaching, and promoting classroom interactions (Freret et al. 2017) and,
  • d) provide feedback by faculty to peer tutors about their teaching skills to enhance their teaching (Freret et al. 2017).

평가에서 학생 파트너십을 촉진하기 위해 학교는 다음과 같은 전략을 고려해야 합니다.

  • (a) 평가 및 피드백에서 파트너십 원칙이 포함된 새로운 코스를 도입하고,
  • (b) 평가 정책 문서에 학생 파트너십을 포함하며(Bovill 외. 2021),
  • (c) 평가에서 학생 파트너십의 실행 및 주기적 검토를 지지하는 실천 커뮤니티를 구축합니다.

For promoting student partnership in assessment, schools should consider the following strategies:

  • (a) introduce new courses with the principles of partnership in assessment and feedback,
  • (b) embed student partnership in the assessment policy documents (Bovill et al. 2021), and
  • (c) establish a community of practice which will champion the implementation and periodic review of the student partnership in assessment.

학술 연구 파트너십 참여
Engagement in scholarly research partnerships

학생들은 연구 문화가 잘 확립된 학교에서 학술 연구 경험에 참여하도록 동기를 부여받고, 연구 참여의 관련성과 가치를 인식하도록 동기를 부여받습니다(Cornett 외. 2021). 학교는 다음과 같은 전략을 적용하여 학생의 학술 연구 참여를 촉진할 수 있습니다:

  • 첫째, 연구 기회에 대한 인식을 높입니다. 연구 계획의 가용성에 대한 광고는 연구 참여에 대한 학생의 동기를 높일 수 있습니다(Trethewey 2019). 관심 있는 학생은 교수진 또는 동료와의 연구 협력을 모색하거나 STARSurge 이니셔티브와 같은 국제 협력, 학생 주도 감사 및 연구 그룹에 참여할 수도 있습니다. 이러한 연구 그룹을 통해 학생들은 종합적인 연구 프로젝트에서 공동 작업을 수행함으로써 연구 기술을 습득할 수 있습니다(Trethewey 2019).
  • 둘째, 학생과 감독자를 위한 전용 시간을 할당합니다. 이렇게 하면 지도 교수의 질이 향상되고 연구 수행에 대한 학생의 자신감(자기 효능감)이 높아질 것입니다(Cornett 외. 2021).
  • 셋째, 학생의 연구 자율성을 장려합니다. 이는 연구 기회를 선택 또는 연구 기반 모듈로 제공하고, 학생들이 관심 있는 연구 프로젝트를 선택할 수 있는 기회를 제공함으로써 학생들의 연구 참여 동기를 높일 수 있습니다(Trethewey 외. 2018; Cornett 외. 2021).

Students are motivated for engagement in scholarly research experiences in schools with well-established research culture, and motivates the students to see the relevance and value of engagement in research (Cornett et al. 2021). Schools can promote student engagement in scholarly research by applying the following strategies: 

  • First, increase awareness for research opportunities. Advertisement for the availability of research plans could increase the student motivation to join the research (Trethewey 2019). Interested students can also explore research collaborations with faculty or colleagues or join international collaborative, student-led audit, and research groups such as STARSurge initiative (www.starsurg.org). Through these research groups, students can acquire research skills by collaborating in comprehensive research projects (Trethewey 2019). 
  • Second, allocate dedicated times for students and supervisors. This will improve the quality of supervision and increase students’ confidence (self-efficacy) in conducting research (Cornett et al. 2021). 
  • Third, promote student autonomy in research. This can be fulfilled by providing research opportunities as optional or elective research-based modules and giving students the opportunity to choose the research projects of their interest which can increase the student motivation for engagement in research (Trethewey et al. 2018; Cornett et al. 2021).

거버넌스 및 품질 보증에 대한 참여
Engagement in governance and quality assurance

교육기관은 다음과 같은 전략을 사용하여 거버넌스 및 품질 보증에 대한 학생의 참여를 향상시킬 수 있습니다:

  • a) 학생의 거버넌스 참여를 환영하는 학교 문화를 장려하고,
  • b) 모든 학생의 목소리를 담을 수 있는 대표성의 다양성과 충분한 수의 참여 학생을 확보하고(Naylor et al. 2021),
  • c) 학생들이 거버넌스에서 후임자를 선택할 수 있는 자율성을 제공하고,
  • d) 학생 대표의 참여, 인수인계, 코칭 및 교육을 위한 잘 설계된 계획을 수립하고,
  • e) 학생 대표 간의 정보를 조정하고 공유하며,
  • f) 모든 학생에게 조치와 결정을 전달하여 학생의 목소리가 잘 반영되도록 합니다(Meeuwissen 외. 2019).

Institutions can enhance the engagement of students in governance and quality assurance by using the following strategies:

  • a) promote a school culture that welcomes students’ participation in governance,
  • b) ensure diversity in representation and enough numbers of engaged students that capture the voices of all students (Naylor et al. 2021),
  • c) provide autonomy for students to select their successors in governance,
  • d) establish a well-designed plan for participation, handover, coaching, and training of student representatives,
  • e) coordinate and share information between student representatives, and f) communicate actions and decisions to all the students to ensure that student voice is well-represented (Meeuwissen et al. 2019).

지역사회 파트너십 참여
Engagement in community partnerships

지역사회 파트너십에 대한 학생 참여는 지역사회의 요구에 부응할 때 효과적입니다(Talib 외. 2017). 학생과 지역사회 구성원 간의 비공식적인 의사소통은 학생들이 봉사하는 지역사회를 이해하고 지역사회의 필요를 파악하는 데 도움이 됩니다(Talib 외. 2017). 지역사회의 요구사항 해결은 지역사회, 학생, 교수진 대표로 구성된 자문위원회를 통해 촉진될 수 있습니다(Marjadi 외. 2021). 지역 보건 시스템의 참여와 지원은 학생들의 지역사회 참여를 촉진할 것입니다. 또한, 지역사회 파트너십 경험이 HPE 학생들의 미래 진로와 관련되면 참여 동기가 강화될 것입니다(Mann 2011). 또한 커뮤니티 파트너십 경험은 전문가 간 학습을 제공할 수 있는 좋은 기회이기도 합니다(Fatima 외. 2018).
Student engagement with community partnerships is effective when they respond to the needs of the community (Talib et al. 2017). Informal communications between students and community members help them to understand the communities they serve and identify the community needs (Talib et al. 2017). Addressing the needs of the community can be promoted through an advisory committee with representatives from the community, students, and faculty (Marjadi et al. 2021). Involvement and support of the local health system will facilitate the engagement of students in the community. Furthermore, relevance of the community partnership experience to the future career of HPE students will enhance their motivation for engagement (Mann 2011). The community partnership experience is also a great opportunity for providing interprofessional learning (Fatima et al. 2018).

상자 3에는 HPE에 대한 학생 참여를 향상시키는 교육 방법의 목록이 나와 있습니다.
Box 3 provides a list of the instructional methods that enhance student engagement in HPE.


상자 3 HPE에서 학생 참여를 높이는 교육 방법.
Box 3 
Instructional methods that enhance student engagement in HPE.


- 능동적이고 협력적인 학습:

 - 문제 기반 학습
 - 팀 기반 학습
 - 사례 기반 교육
 - 플립형 강의실
 - 전문가 간 학습
 - 반성적 연습
• Active and collaborative learning:

 – Problem based learning
 – Team-based learning
 – Case based teaching
 – Flipped classroom
 – Interprofessional learning
 – Reflective practice
- 기술 강화 학습
 - 가상 환자 시뮬레이션
 - 마이크로 블로그
 - 소셜 미디어
 - 적응형 튜토리얼
 - 화상 회의 세션
 - 디지털 게임
 - 3차원 인쇄 모델
 - 360도 비디오를 사용한 가상 현실
 - 학습 대시보드
 - 청중 응답 시스템

• Technology-enhanced learning:
 – Virtual patient simulations
 – Microblogs
 – Social media
 – Adaptive tutorials
 – Videoconference sessions
 – Digital games
 – Three-dimensional printed models
 – Virtual reality using 360o videos
 – Learning dashboards
 – Audience response systems

능동적이고 협력적인 학습 사용 촉진
Promote the use of active and collaborative learning

관련 임상 문제를 기반으로 한 능동적 학습 방법과 탐구 자극을 통해 학생들의 동기를 강화합니다. 또한 협업 학습은 학생들 간의 사회적 연결과 소속감을 구축합니다. 학생들의 참여도를 높이기 위해 HPE 학교는 문제 기반 학습(O'Malley 외. 2003, Kelly 외. 2005, Alimoglu 외. 2014), 팀 기반 학습(Kelly 외. 2005, Cheng 외. 2014, Alimoglu 외. 2017, Hopper and Brake 2018, Smeby 외. 2020) 등의 교육적 방법을 적용할 수 있습니다. 2020; Ulfa 외. 2021), 사례 기반 학습(Alimoglu 외. 2014), 거꾸로 교실(Riddle and Gier 2019; Burkhart 외. 2020), 전문가 간 교육(IPE)(Lee 외. 2018; Wang and Ji 2021), 성찰적 실습(Lee 외. 2018; Wang and Ji 2021) 등이 있습니다. 
The use of active learning methods which are based on relevant clinical problems and stimulation of inquiry enhance the motivation of students. In addition, collaborative learning builds social connections between students and the feeling of belongingness. To enhance students engagement, HPE schools can apply pedagogical methods such as problem-based learning (O'Malley et al. 2003; Kelly et al. 2005; Alimoglu et al. 2014), team-based learning (Kelly et al. 2005; Cheng et al. 2014; Alimoglu et al. 2017; Hopper and Brake 2018; Smeby et al. 2020; Ulfa et al. 2021), case-based learning (Alimoglu et al. 2014), flipped classroom (Riddle and Gier 2019; Burkhart et al. 2020), interprofessional education (IPE) (Lee et al. 2018; Wang and Ji 2021), and reflective practice (Lee et al. 2018; Wang and Ji 2021).

기술 강화 학습 사용 촉진
Promote the use technology-enhanced learning

학교는 기술 강화 학습 방법을 최대한 활용하여 학생 참여를 촉진할 수 있습니다. 방법에는 가상 환자 시뮬레이션(McCoy 외. 2016, Berman and Artino 2018, Stokes-Parish 외. 2020, Wang and Ji 2021), 마이크로 블로그(Saperstein 외. 2015), 소셜 미디어(Hartnup 외. 2018, Bhat and Gupta 2019, Quesnelle and Montemayor 2020, Siqueira 외. 2021), 적응형 자습서(Wong et al. 2015), 화상 회의 세션(Kay and Pasarica 2019; Caton 외. 2021), 디지털 게임(Janssen 외. 2015), 3차원 인쇄 모델(Tripodi 외. 2020), 360° 비디오를 사용한 가상 현실(Harrington 외. 2018; Torda 2020; Chan 외. 2021), 학습 대시보드(de Leng and Pawelka 2020), 청중 반응 시스템(Funnell 2017) 등 다양한 분야에서 활용되고 있습니다.
Schools can make best use of technology-enhanced learning methods to promote student engagement. Methods can include virtual patient simulations (McCoy et al. 2016; Berman and Artino 2018; Stokes-Parish et al. 2020; Wang and Ji 2021), microblogs (Saperstein et al. 2015), social media (Hartnup et al. 2018; Bhat and Gupta 2019; Quesnelle and Montemayor 2020; Siqueira et al. 2021), adaptive tutorials (Wong et al. 2015), videoconference sessions (Kay and Pasarica 2019; Caton et al. 2021), digital games (Janssen et al. 2015), three-dimensional printed models (Tripodi et al. 2020), virtual reality using 360° videos (Harrington et al. 2018; Torda 2020; Chan et al. 2021), learning dashboards (de Leng and Pawelka 2020), and audience response systems (Funnell 2017).

향후 연구 방향
Future research directions

학생 참여와 관련된 다양한 영역을 다루는 연구 출판물이 빠른 속도로 증가하고 있음에도 불구하고, 문헌에서 다루어야 할 몇 가지 공백이 있습니다. 학생 참여 구조의 포괄적인 특성으로 인해 다차원성을 측정할 수 있는 심리측정 특성을 갖춘 도구를 개발하는 데 여전히 어려움이 있습니다. 파트너십에 대한 학생 참여는 많은 출판물에서 인정되었지만, 파트너십에서 학생 참여의 차원을 측정하는 것은 현재 불분명합니다. 최근 중국에서 수행된 대규모 연구에서는 연구 프로젝트에 참여하는 빈도로 학생의 연구 참여를 측정했습니다(Zhang 외. 2022). 특히 파트너십에서 참여 수준과 참여 차원을 측정하는 도구를 개발하는 것이 연구 의제에 포함되어야 합니다. 또한, 참여 이탈을 개념화하고 이를 측정하기 위한 적절한 도구를 개발하여 참여 이탈의 동인과 결과를 파악할 필요가 있습니다. 현재 학생 참여에 관한 문헌의 연구 설계는 대부분 횡단적 연구이며, 상황별 개입 후 참여도 또는 참여 이탈을 측정하기 위한 종단적 연구는 부족합니다. 마지막으로, 학생 참여 개입의 결과를 측정하는 것은 현재 인지적 차원에 초점을 맞추고 있으며 정신 운동 기술, 정서적 영역 및 일반 역량과 같은 역량의 다른 측면을 측정하는 연구는 부족합니다.

Despite the fast pace in research publications addressing different areas related to student engagement, there are several gaps in literature that need to be covered. With the comprehensive nature of the student engagement construct, the challenge is still existing about developing instruments with good psychometric properties to measure its multidimensionality. Although student engagement in partnerships have been recognized in many publications, measuring the dimensions of student engagement in these partnerships is currently unclear. A recent large-scale study conducted in China measured student engagement in research by the frequency of their involvement in research projects (Zhang et al. 2022). Specifically, developing tools for measuring the level of involvement and engagement dimensions in partnerships should be on the research agenda. Furthermore, there is a need for conceptualizing disengagement and therefore developing proper instruments for its measurement and identifying drivers and outcomes of disengagement . Most of the study designs in the current student engagement literature in HPE are cross-sectional and there is lack of longitudinal studies for measuring engagement or disengagement after contextual interventions. Finally, measuring the outcomes of student engagement interventions is currently focusing on the cognitive dimension and there is lack of studies measuring the other aspects of competence such as psychomotor skills, affective domain, and generic competencies.


 

Med Teach. 2022 Oct 28;1-17. doi: 10.1080/0142159X.2022.2137018. Online ahead of print.

 

Student engagement in health professions education: AMEE Guide No. 152

Affiliations collapse

1Department of Physiology, Faculty of Medicine, Suez Canal University, Ismailia, Egypt.

2Department of Basic Medical Sciences, College of Medicine, Gulf Medical University, Ajman, United Arab Emirates.

3Gulf Medical University, Ajman, United Arab Emirates.

PMID: 36306374

DOI: 10.1080/0142159X.2022.2137018

Abstract

This guide aims to support our colleagues to have comprehensive understanding of student engagement in health professions education. Despite the universal agreement about the significance of student engagement, there is lack of uniformity in conceptualizing and operationalizing this emerging construct. We review the theoretical basis explaining student engagement from three main perspectives: behavioral, psychological, and socio-cultural. In addition, we propose a contemporary and comprehensive framework for the student engagement in higher education, which is applicable to health professions education contexts. Drawing from this framework, we explain the conceptualization of the construct and its preceding factors, mediators, dimensions, spheres, and outcomes of student engagement. The proposed framework introduces student 'engagement through partnerships' as a novel component compared with the existing models of student engagement in higher education. This way, we are proposing a mixed model that not only considers the student as a 'customer' but also as a 'partner' in education. Engagement of students through partnerships include four areas: (1) provision of the education program, (2) scholarly research, (3) governance and quality assurance, and (4) community activities. This guide will provide practical applications on how to improve student engagement in health professions education. Finally, we highlight the current gaps in areas of research in the student engagement literature and suggested plans for future directions.[Box: see text].

Keywords: Collaborative/peer-to-peer; student engagement; students as partners.

 

보건의료전문직 교육에 투자에 대해 기관 리더가 가지고 있는 가치요인 (JAMA Netw Open. 2023)
Identifying Value Factors in Institutional Leaders’ Perspectives on Investing in Health Professions Educators
Ann Poncelet, MD; Sally Collins, MA, MSc; Darren Fiore, MD; Glenn Rosenbluth, MD; Helen Loeser,MD, MSc; George F. Sawaya, MD; Arianne Teherani, PhD; Anna Chang, MD

소개
Introduction

교육자, 교육 혁신 및 학술활동에 대한 투자는 모든 의과대학과 의료계의 핵심 사명인 우수 의학교육을 위해 필수적입니다.1,2 연구에 따르면 교내 보조금과 기부 석좌 프로그램이 교육 혁신 및 교육자 성장과 같은 성과에 도움이 되는 것으로 나타났습니다.3,4 그러나 학부 및 대학원 의학교육 비용의 증가와 함께 임상 및 연구 사업에서 발생하는 수익 감소에 직면한 학부 의료 센터(AMC)가 혁신과 교육자 자체를 포함한 의학교육에 대한 자금은 여전히 상당한 위험에 처해 있습니다.2,5 등록금으로 의학교육 비용을 충당하는 경우는 거의 없기 때문에,2 우리는 기관 리더들이 지원을 옹호하도록 참여시켜야 합니다.6 
Investing in educators, educational innovation, and scholarship is essential for excellence in medical education, a core mission of all medical schools and health care.1,2 Studies demonstrate the benefit of intramural grants and endowed chair programs to outcomes like education innovation and educator growth.3,4 Yet, funding for medical education, including for innovation and the educators themselves, remains at significant risk as academic medical centers (AMCs) face declining revenue from the clinical and research enterprises along with increasing costs of undergraduate and graduate medical education.2,5 Because tuition rarely covers the costs of medical education,2 we must engage institutional leaders to advocate for support.6

교내 보조금이나 기부형 석좌와 같은 교육자 투자 프로그램(EIP)을 위한 기금은 임상 행정이나 생의학 연구 경력을 가진 리더가 할당하는 경우가 많습니다. 현재 임상 수익이 AMC의 교육 및 연구 임무에 보조금을 지급하기 때문에 이들은 경제적인 측면에 초점을 맞추는 것이 당연합니다.2,7 그러나 투자 수익률(ROI)과 같은 경제적 지표만으로는 EIP의 가치를 판단하기에는 충분하지 않습니다.7 AMC 리더와 교육자는 향후 대화와 자원 배분을 위한 더 광범위한 프레임워크가 필요합니다. 현재까지 기관 리더의 관점, 특히 재무적 ROI 이상의 가치를 지닌 요소에 대한 관점은 누락되어 있습니다. 
Funds for educator investment programs (EIPs) like intramural grants or endowed chairs are often allocated by leaders with backgrounds in clinical administration or biomedical research. They hold an understandable economic focus since clinical revenue currently subsidizes the AMC’s education and research missions.2,7 However, economic metrics, such as return on investment (ROI), are insufficient to determine the value of EIPs.7 AMC leaders and educators need a broader framework to inform future dialogue and resource allocation. To date, the perspective of institutional leaders is missing, especially regarding the factors that have value beyond financial ROI.

전통적인 재무 지표를 넘어서는 가치 요소를 탐색하기 위한 한 가지 프레임워크는 연방 최고정보위원회에서 개발한 가치 측정 방법론(VMM)입니다.8 미국 연방 정부에서 사용하는 VMM은 5가지 영역에서 가치 요소를 제시합니다.8 

  • 개인(수혜자 개인에게 혜택),
  • 재무(비용, 수익 포함),
  • 운영(시스템 운영 개선),
  • 사회적 또는 사회적(집단 또는 사회에 혜택),
  • 전략적 또는 정치적(기관의 사명, 전략적 목표, 우선순위 및 의무 발전)

한 임상 시뮬레이션 프로그램에서는 운영(재원 기간 감소) 및 전략적(환자 안전 문화) 등의 추가 영역에서 가치 요소를 식별하는 데 VMM을 사용했습니다.9 이 프레임워크는 잠재력이 있지만 아직 학술 EIP에 적용되지 않은 상태입니다. 이 글에서는 VMM 프레임워크를 사용하여 교육자 성장과 교육 혁신에 투자하는 교내 프로그램에 관한 의료 전문 교육기관 및 관련 의료 시스템 리더의 가치 요인을 탐색합니다. 
One framework for exploring value factors beyond traditional financial metrics is the value measurement methodology (VMM) developed by the Federal Chief Information Council.8 Used by the US federal government, VMM presents value factors in 5 domains:

  • individual (benefits individual recipient),
  • financial (including costs, revenues),
  • operational (improves system operations),
  • social or societal (benefits group or society), and
  • strategic or political (advances institution’s mission, strategic goals, priorities, and mandates).8 

One clinical simulation program used VMM to identify value factors in additional domains, such as operational (decreased length of stay) and strategic (patient safety culture).9 This framework has potential and has yet to be applied to academic EIPs. Using the VMM framework, we present an exploration of value factors of leaders at a health professions institution and affiliated health systems regarding intramural programs that invest in educator growth and education innovation.

연구 방법
Methods

이 질적 연구는 주제 분석10,11을 사용하여 기관 리더의 관점을 명확히 하기 위해 구성주의적 지향으로 인터뷰 데이터에서 주제를 식별했습니다. 샌프란시스코 캘리포니아 대학교의 기관윤리심의위원회는 이 연구가 일반원칙에 따라 면제되는 것으로 간주했습니다. 구두와 이메일을 통해 사전 동의를 얻었습니다. 이 연구는 질적 연구 보고 표준(SRQR) 보고 가이드라인을 따랐습니다.
This qualitative study used thematic analysis10,11 to identify themes in interview data, with a constructivist orientation, to elucidate the perspectives of institutional leaders. The institutional review board at the University of California, San Francisco, deemed this study exempt via the Common Rule. Informed consent was obtained orally and via email. This study followed the Standards for Reporting Qualitative Research (SRQR) reporting guideline.

환경 및 참여자
Setting and Participants

본 연구는 치과, 의학, 간호학, 약학 등 4개 학과와 물리치료학과가 있는 캘리포니아대학교 샌프란시스코 캠퍼스(UCSF)에서 수행되었습니다. UCSF와 3개의 부속 의료 시스템에서 55명의 리더를 이메일을 통해 초대했습니다. 충분한 이해의 깊이를 확보하기 위해 다양한 직급, UCSF 및 EIP 프로그램, 다양한 전문 분야와 규모의 부서에서 다양한 역할 경험을 가진 리더를 선정했습니다.12 처음에 응답하지 않은 응답자 31명을 대상으로 리더 역할을 충분히 대표할 수 있을 때까지 후속 인터뷰를 진행했으며, 이 중 2명은 파일럿 인터뷰 대상자로 참여했습니다. 
Our study was conducted at the University of California, San Francisco (UCSF), an academic health professions institution with 4 schools—dentistry, medicine, nursing, and pharmacy—and a physical therapy department. We invited 55 leaders via email from across UCSF and 3 affiliated health systems. We chose leaders at multiple levels of leadership with a range of experience in their roles, with UCSF and the EIP programs, and from departments of different specialties and sizes to reach a sufficient depth of understanding.12 We followed up with those who did not initially respond until we had sufficient representation of leader roles at 31 respondents, 2 of whom served as pilot interviewees.

도구
Instrument

5가지 VMM 영역(재무, 개인, 운영, 사회적 또는 사회적, 전략적 또는 정치적)8에 기반한 반구조화된 인터뷰 가이드(부록 1의 전자 부록)를 개발하여 교육에 맞게 조정했습니다. UCSF 하일레 데바스 의학교육자 아카데미(AME)3,4,13에서 개발한 두 가지 EIP, 즉 혁신 교육 기금 프로그램(IF)과 연간 15,000달러에서 30,000달러의 기부 의자 프로그램(EC)을 모범 사례로 강조했습니다. IF는 경쟁력 있는 교내 보조금을 통해 UCSF 보건 전문직 교육자들이 커리큘럼 혁신을 개발, 시범 운영 및 연구할 수 있도록 지원합니다.3 EC는 AME 회원의 경력 개발을 지원하여 기관 안팎에서 영향력을 확대하는 것을 목표로 합니다.4 우리는 각 리더에게 해당 부서 또는 단위의 IF 수혜자 및 EC 보유자 명단을 제공하여 이러한 프로그램을 설명하고 자극을 주도록 했습니다. 또한 리더들에게 가장 중요한 요소를 설명해 달라고 요청했습니다.
We developed a semistructured interview guide (eAppendix in Supplement 1) based on 5 VMM domains (financial, individual, operational, social or societal, and strategic or political),8 adapted for education. We highlighted as exemplars 2 EIPs developed by the UCSF Haile T. Debas Academy of Medical Educators (AME)3,4,13: the Innovations Funding for Education program (IF), and the Endowed Chair program (EC) at $15 000 to $30 000 per year. The IF supports UCSF health professions educators to develop, pilot, and study curricular innovations with competitive intramural grants.3 The EC supports the career development of AME members, with the aim of expanding their impact within and beyond the institution.4 We gave each leader a list of IF recipients and EC holders from their department or unit to illustrate these programs and serve as a prompt. We also asked leaders to articulate the most important factor.

zoi221604supp1_prod_1675800338.73239.pdf
0.24MB

 

두 명의 고위 리더가 파일럿 인터뷰 대상자로 참여하여 인터뷰 가이드를 다듬고, 3명의 저자 인터뷰 담당자(A.C., D.F., G.R.)를 교육하고, 일관성을 확보했습니다. 이 데이터는 분석에 포함되지 않았습니다. 나머지 29건의 인터뷰는 2019년 6월부터 9월 사이에 진행되었으며, 직접 보고하는 리더를 인터뷰한 사람은 없었습니다. 인터뷰는 12~59분(평균[SD], 35.7[8.8]분) 동안 진행되었으며, 휴대폰 음성 녹음 애플리케이션을 사용하여 직접 대면하거나 Zoom 화상 회의를 사용하여 가상으로 진행되었습니다. 모든 인터뷰 오디오 녹음은 필사되었습니다. 
Two senior leaders served as pilot interviewees to refine the interview guide, train 3 author-interviewers (A.C., D.F., G.R.), and ensure consistency. These data are not included in the analysis. The remaining 29 interviews were conducted between June and September 2019; no one interviewed a leader to whom they reported directly. Interviews lasted 12 to 59 minutes (mean [SD], 35.7 [8.8] minutes) and were conducted in-person using a mobile phone voice recording application or virtually using Zoom videoconferencing. All interview audio recordings were transcribed.

반사성
Reflexivity

이 연구는 우리의 주관과 맥락이 연구 과정에 어떤 영향을 미치는지 자의식적으로 비판, 평가, 평가하는 일련의 지속적이고 협력적이며 다각적인 관행을 사용했습니다. 저자 중 6명은 임상 교육자(A.P., A.C., G.R., D.F., G.S., H.L.)이고, 1명은 교육 연구자(A.T.), 1명은 교육 연구 동료(S.C.)입니다. 5명의 저자는 연구 이전 또는 연구 시점에 석좌교수를 역임했으며(A.C., D.F., G.R., G.S., A.P.), 3명(A.C., G.R., A.T.)은 이전에 IF 프로그램의 감독 또는 공동 감독을 맡은 적이 있습니다. 저자 중 한 명은 현재 AME의 디렉터(A.P.)이고 다른 한 명은 전 디렉터(H.L.)입니다. 세 명의 저자는 이 프로그램의 수혜자였던 적이 없습니다(S.C., A.T., H.L.). 우리는 정기적이고 성찰적인 토론에 참여하여 연구 결과를 해석할 때 서로의 가정을 고려하거나 이의를 제기할 수 있었습니다.
This study used a set of continuous, collaborative, and multifaceted practices through which we self-consciously critique, appraise, and evaluate how our subjectivity and context influence the research processes. Six authors are clinician educators (A.P., A.C., G.R., D.F., G.S., H.L.), 1 is a is an education researcher (A.T.), and 1 is an education research associate (S.C.). Five authors held endowed chairs prior to or at the time of the study (A.C., D.F., G.R., G.S., A.P.), and 3 (A.C., G.R., A.T.) had previously directed or codirected the IF program. One author is the current director of the AME (A.P.), and another a former director (H.L.). Three authors have never been a recipient of these programs (S.C., A.T., H.L.). We engaged in regular, reflective discussion, which enabled us to consider or challenge each other’s assumptions when interpreting findings.

데이터 분석
Data Analysis

귀납적 접근법에 따른 주제분석10,11을 사용하여 모든 인터뷰를 필사한 후 데이터를 분석했습니다.

  • 4명의 저자(A.P., S.C., A.T., G.S.)는 인터뷰 질문을 바탕으로 예비 추론 코드(비판단적인 방식으로 표현된 관찰 가능한 데이터)를 사용하여 코드북을 개발했으며, 이 코드북은 3개의 인터뷰 녹취록을 사용하여 반복적인 합의 형성 접근법을 통해 데이터에서 생성된 코드로 더욱 구체화되었습니다.
  • 그런 다음 4명의 저자가 나머지 트랜스크립트를 코딩했습니다. 각 녹취록은 팀원 중 2명이 코딩하고 합의를 도출하기 위해 토론을 통해 조정했습니다.
  • 5명의 저자(A.C., A.P., G.R., D.F., S.C.)가 분석된 데이터를 검토하여 중요한 주제를 종합했습니다. 분석 전반에 걸쳐 모든 팀원의 관점을 반영하여 데이터에 대한 이해를 돕고 연구팀의 관점이 아닌 리더의 목소리를 주제에 반영할 수 있도록 했습니다.
  • 코딩된 녹취록은 Dedoose 분석 소프트웨어 버전 9.0.62(사회문화 연구 컨설턴트)를 사용하여 정리했습니다.

We used thematic analysis,10,11 following an inductive approach, to analyze data once all interviews had been transcribed.

  • Four authors (A.P., S.C., A.T., G.S.) developed a codebook from preliminary, low-inference codes (observable data expressed in a nonjudgmental way) based on the interview questions, which was further refined with codes generated from the data through an iterative consensus-building approach using 3 interview transcripts.
  • These 4 authors then coded the rest of the transcripts. Each transcript was coded and reconciled by 2 of these team members through discussion to achieve consensus.
  • Five authors (A.C., A.P., G.R., D.F., S.C.) reviewed the analyzed data to synthesize overarching themes. We drew on the perspectives of all team members throughout the analysis to facilitate a shared understanding of the data and ensure that the themes represented the voices of the leaders, not the perspectives of the study team.
  • We organized coded transcripts with Dedoose analytic software version 9.0.62 (SocioCultural Research Consultants).
결과
Results

분석을 위해 29명의 리더를 인터뷰했습니다(캠퍼스 또는 대학 리더 5명[17%], 의료 시스템 리더 3명[10%], 보건 전문학교 리더 6명[21%], 부서 리더 15명[52%])(표 1). 개인, 재무, 운영, 사회적 또는 사회적, 전략적 또는 정치적 등 VMM 프레임워크 영역을 사용하여 나타난 주제를 제시합니다(표 2). EIP의 개별 수혜자에게 미치는 영향부터 시작하여 리더들이 가장 중요하다고 생각하는 가치 요소로 결론을 내립니다. 

We interviewed 29 leaders for analysis (5 [17%] campus or university leaders; 3 [10%] health systems leaders; 6 [21%] health professions school leaders; 15 [52%] department leaders) (Table 1). We present the themes that emerged using VMM framework domains: individual, financial, operational, social or societal, and strategic or political (Table 2). We begin with the impact on the individual beneficiary of EIPs and conclude with the value factors leaders identified as the most important.

개인의 가치
Individual Value

조직 리더들은 수혜자 개인의 발전, 경력, 위상 강화에 따른 혜택을 중요하게 여겼습니다. 교수진은 교육자 및 리더로서 기술과 전문성을 개발합니다. 이러한 이니셔티브는 의학교육에 참여할 수 있는 발판 역할을 하며, 특히 초기에 지원을 받으면 승진 및 리더십 역할을 포함하여 오랫동안 성공적인 커리어를 쌓을 수 있는 주니어 교수진에게 도움이 됩니다. 리더들은 교육자 투자가 개인에게 미치는 긍정적인 영향에 대해 강조했습니다. 기부형 의자나 교육 보조금을 받는다는 것은 의미 있는 일이며, 교육자들이 열정을 갖고 있는 일을 할 수 있도록 신뢰를 줍니다. 교수진에게 이러한 기금을 맡기면 아이디어를 실험하고, 자신감과 성공에 대한 열망을 얻고, 기금 기간 이후에도 더 많은 일을 할 수 있는 유연성을 확보할 수 있습니다. 리더들은 인재를 믿고 투자하면 개인을 넘어 운영, 전략 또는 정치, 사회적 또는 사회적 차원의 영향력까지 다양한 영역에서 보상을 얻을 수 있다는 점을 거듭 강조했습니다.
Organizational leaders viewed the benefits to the individual recipients as enhancing their development, career, and stature. Faculty develop skills and expertise as educators and leaders. Such initiatives serve as a launch pad for involvement in medical education, especially junior faculty who, if supported early on, can be engaged to have long successful careers, including accelerated promotion and leadership roles. Leaders underscored the affirmational impact that educator investments have on individuals. The honor of receiving an endowed chair or education grant is meaningful and gives educators credibility to do work about which they are passionate. Entrusting faculty with these funds gives them the flexibility to experiment with ideas, gain confidence and a desire to succeed, and do more beyond the duration of the funding. Leaders repeatedly highlighted that entrusting and investing in people also reaps rewards in domains beyond the individual to the operational, strategic or political, and social or societal levels of impact.

재무적 가치
Financial Value

리더들은 EIP가 실용적, 철학적 이점을 모두 가지고 있다는 점을 인정했습니다. 실용적인 측면에서, 자금 지원은 교수진에게 새로운 교육 프로젝트를 수행할 수 있는 권한과 책임을 부여합니다. 또한 교수진의 시간을 보호하고 여러 부서 또는 프로그램에 걸친 프로젝트에 실질적인 리소스를 제공합니다. 리더들은 이러한 프로그램을 매칭 기부를 통해 부서, 임상 현장 또는 기관의 추가 자원을 장려할 수 있는 수단으로 인식했습니다. 또한 리더들은 교육 기금의 더 깊은 철학적 이점을 강조하며, 수익 창출로 이어지지 않더라도 교육에 투자하는 것이 중요하다고 강조했습니다.
Leaders acknowledged that EIPs have both practical and philosophical benefits. On a practical level, funding gives faculty permission and accountability to accomplish new educational projects. It protects faculty time and provides tangible resources for projects that cross departments or programs. Leaders saw these programs as a means to encourage additional resources from departments, clinical sites or the institution through matching contributions. Leaders also highlighted a deeper philosophical benefit of educational funding, emphasizing the importance of investing in education even though the result is not revenue generation.

운영 가치
Operational Value

리더들은 교육 프로그램, 효율성, 채용 및 유지라는 세 가지 영역에서 운영상의 가치가 있다고 설명했습니다.

  • 리더들은 EIP가 교육 방법을 형성하고, 변화시키고, 개선하는 데 중요한 기여를 한다고 생각했습니다. 이들은 학습자가 최첨단 과학에 참여하고 성공적인 커리어를 준비할 수 있도록 돕는 활동에 직접적으로 기여합니다. 혁신적인 프로그램과 교육 프로젝트에 참여할 수 있는 기회는 최고의 학습자를 끌어들이고 전통적으로 의학교육에서 가시성이 부족했던 분야에 대한 노출 기회를 제공합니다. 일부 리더들은 의학 커리큘럼에 다양한 학문과 사회 문제를 도입하는 전문직 간 교육자들의 협업이 가져다주는 이점에 대해 설명했습니다. 리더들은 이러한 프로그램이 인증 기준을 충족하는 데 기여한다고 언급했습니다. 이러한 프로그램은 제한된 리소스로 인해 달성할 수 없었던 교육 제공 방식에 변화를 가져올 수 있는 기회를 창출합니다.
  • 또한 리더들은 효율성 향상을 중요한 가치로 여겼습니다. 여기에는 보다 비용 효율적인 교육 프로그램과 임상 시스템의 효율성을 높이고 낭비를 줄이기 위한 학습자 교육이 포함되었습니다.
  • 리더들은 교수진과 학습자의 채용 및 유지를 중요한 운영 가치로 강조했습니다. EIP는 최고의 교수진을 유치 및 채용하고 이들이 교육기관에 계속 머물도록 장려하는 데 도움이 됩니다. EIP는 우수한 학습자를 교육기관으로 끌어들여 향후 교수진 임용을 가능하게 하는 경우가 많습니다.

Leaders described 3 areas of operational impact that were of value: educational programs, efficiency, and recruitment and retention.

  • Leaders viewed EIPs as important contributors to shaping, transforming, and improving teaching methods. They feed directly into activities that engage learners in cutting-edge science and prepare them for successful careers. Innovative programs and opportunities to participate in education projects attract the best learners and offer exposure for disciplines traditionally lacking visibility within medical education. Some leaders described the benefit of a collaboration of interprofessional educators who bring a broader spectrum of disciplines and social issues into the medical curriculum. Leaders noted that these programs contribute to meeting accreditation standards. They create opportunities for changes in the way that education is delivered, which might not otherwise be achieved because of limited resources.
  • Leaders also saw improved efficiency as an important value. This included having more cost-effective educational programs and educating learners to enhance efficiency and decrease waste in clinical systems.
  • Leaders highlighted recruitment and retention of faculty and learners as a critical operational value. EIPs help attract and recruit the best faculty and encourage them to stay at the institution. EIPs draw high-quality learners to the institution, often enabling future faculty appointments.

사회 및 사회적 가치
Social and Societal Value

리더들은 조직을 넘어선 사회적 가치와 교수진, 학습자, 환자 등 조직 내부 커뮤니티에 대한 사회적 가치에 대해 설명했습니다.

  • 중요한 가치는 발표와 출판을 통해 일반화할 수 있고 기관 외부로 전파할 수 있는 교육 내 연구를 가능하게 하는 것이었습니다. 그들은 교수진이 환자와 사회에 도움이 되고 기관을 넘어 교육 프로그램을 향상시키는 교육 장학의 리더 또는 전문가가 되기를 원했습니다. 리더들은 기관 외부의 커뮤니티와의 관계를 구축하고 더 큰 사회적 이익에 도움이 되는 투자의 중요성을 강조했습니다. 이들은 지역, 국가, 글로벌 수준에서 영향력을 발휘하는 교육자 및 프로그램의 예를 들었습니다. 리더들은 긴급한 사회 문제를 해결하고, 의료 격차를 줄이며, 다양한 환자 집단의 치료 접근성을 개선하는 EIP의 중요성을 인식했습니다. 이들은 최상의 치료를 제공할 수 있도록 의료 서비스 제공자를 교육해야 하는 기관의 사회적 책임을 강조했습니다.

Leaders described societal value beyond the organization and social value to the organization’s internal community, including faculty, learners, and patients.

  • An important value was enabling work within education that is generalizable and disseminated beyond the institution through presentations and publications. They wanted faculty to be leaders or experts in education scholarship that benefits patients and society and enhances education programs beyond the institution. Leaders highlighted the importance of investments that build connections to the community outside the institution and benefit the greater societal good. They gave examples of educators and programs with impact ranging from local to national and global levels. Leaders recognized the importance of EIPs that address pressing societal issues, reduce health care disparities, and improve access to care for diverse patient populations. They highlighted the institution’s social responsibility to train health care providers to deliver the best care.

리더들은 이러한 이니셔티브를 통해 구축된 네트워크와 협력이 여러 부서와 학교에서 서로 배우는 교육자에게 도움이 된다는 점을 인식했습니다. 

  • 이러한 전문적인 관계는 교육자 및 학자 커뮤니티를 통해 혁신과 아이디어의 융합을 지원합니다. 이러한 커뮤니티는 후배 교육자를 멘토링하고, 학습자를 유치하며, 모범 사례를 모델링할 수 있는 포럼을 제공합니다. 이러한 연결은 개인이 교육기관 내 교육자 커뮤니티의 일원으로서 소속감을 느낄 수 있는 중요한 방법입니다. 리더들은 EIP가 환자 치료를 개선하는 교육 프로그램 개발을 지원하기를 희망한다고 설명했습니다. 그들은 교육자를 지원하면 참여도와 열정이 높아져 환자 치료로 이어진다고 느꼈습니다.

Leaders recognized that networks and collaborations created through these initiatives benefit educators who learn from each other across departments and schools.

  • These professional relationships support innovation and the confluence of ideas through a community of educators and scholars. They provide a forum to mentor junior educators, attract learners, and model best practices. These connections are important ways for individuals to experience a sense of belonging as part of a community of educators within the institution. Leaders described a desire for EIPs to support the development of educational programs that improve patient care. They felt that supporting educators lead to increased engagement and enthusiasm, which transferred to their patient care.

전략적 및 정치적 가치
Strategic and Political Value

리더들은 교육에 대한 투자가 조직 문화, 혁신, 성공에 도움이 되는 전략적, 정치적으로 중요하다고 생각했습니다. 교육에 대한 투자는 개인, 조직, 그 너머에 상징적인 영향을 미칩니다.

  • 리더들은 이러한 프로그램에 헌신하는 것이 교육과 교육적 학술활동의 중요성을 인식하는 것이라고 말했습니다.
  • 교육자의 우수성과 업적에 대한 보상은 교육자의 기관 내 가시성을 보장합니다.
  • 교육자의 업무가 중요하고 다른 임무와 동등하다는 메시지를 전달할 수 있습니다.
  • 포상은 수상자의 부서와 리더에게 명성을 부여하고, 부서의 교육 가시성을 높이며, 기관의 평판에 기여합니다.
  • 리더들은 교육적 기여를 저평가하는 문화에서 교육 교직원이 환영받고, 참여하고, 권한을 부여받는 문화로 변화시킬 수 있는 힘에 대해 설명했습니다.

Leaders deemed investing in education as strategically and politically important, which benefits organizational culture, innovation, and success. Investing in education has a symbolic impact for individuals, the organization, and beyond.

  • Leaders expressed how committing to these programs recognizes the importance of education and education scholarship.
  • Rewarding educators’ excellence and achievements ensure their visibility within the institution.
  • It sends a message that educators’ work is crucial and equal to other missions.
  • Awards bring prestige to recipients’ departments and leaders, raise departments’ education visibility, and contribute to the institution’s reputation.
  • Leaders described the power to transform culture from one that undervalues educational contributions to one where education faculty feel welcomed, engaged, and empowered.

혁신은 리더들이 교육 기관을 우수성의 최전선에 두기 위해 중요하게 생각하는 전략적 가치 요소입니다. 

  • 그들은 이러한 리소스를 통해 창의적으로 사고하고, 위험을 감수하고, 실험하고, 새로운 지식을 창출할 수 있다고 생각했습니다. 
  • 그들은 교육자들이 생각하고, 읽고, 더 깊이 파고들 수 있는 보호된 공간을 갖는 것을 중요하게 생각했습니다. 
  • 리더들은 학술 의료 기관은 교육 프로그램 때문에 훌륭한 것이지 그 반대가 아니라고 말했습니다
  • 교육 혁신, 창의성, 프로젝트, 협업은 개인과 기관 전체의 성공을 가능하게 합니다. 업무가 기관의 더 넓은 사명과 일치하는 것이 중요합니다.

Innovation is an important strategic value factor expressed by leaders to keep the institution at the forefront of excellence.

  • They felt these resources allowed for thinking creatively, taking risks, experimenting, and generating new knowledge.
  • They prized having a protected space for educators to think, read, and dive deeper.
  • Leaders expressed that an academic medical institution is great because of its educational programs, not the reverse.
  • Education innovation, creativity, projects, and collaborations enable the success of both individuals and the entire institution. It is important that the work be aligned with the institution’s wider mission.
가장 중요한 가치 요소
Most Important Value Factors

많은 리더들이 교육자 개인의 경력 및 전문성 개발과 교육 프로그램 향상과 관련된 가치 요소를 가장 중요한 것으로 꼽았습니다. 한 참가자는 "인적 요소가 가장 중요한 요소입니다... 직원에게 투자하는 것은 옳은 일이며 훌륭한 결과를 가져옵니다."라고 말했습니다.
Many leaders identified value factors that relate to enhancing both individual educator careers and professional development and the education programs as most important. One participant stated that “the human factor is the most important factor…investing in your people is the right thing to do, and it has wonderful outcomes.”

토론
Discussion

VMM 프레임워크를 사용하여 교육 기관 리더들이 교육자 및 교육 혁신에 투자할 때 중요하게 생각하는 요소를 파악했습니다. 리더들은 EIP의 가치를 인식하고 모든 영역에서 재무적 ROI 이상의 이점을 강조했으며, 적은 재정 자원으로도 상당한 영향력을 발휘할 수 있음을 보여주었습니다(그림). 이번 조사 결과는 보건 전문직 교육 교내 시드 보조금 프로그램3,14-21 및 기부 석좌 프로그램에 관한 문헌과 일치하며 리더의 관점을 반영한다는 점에서 독특합니다.4,22,23 문헌에 기술된 긍정적 영향에는 경력 성장, 개인 및 전문성 개발, 채용 및 유지(특히 교수진의 다양성 증가와 관련하여), 교육 프로그램, 협업, 교육 장학금 및 보급, 혁신 및 문화가 포함됩니다.3,4,14-23
Using the VMM framework, we identified factors that academic institutional leaders value for investing in educators and education innovation. Leaders saw value in EIPs and emphasized benefits beyond financial ROI across all domains and illustrated that small financial resources can have substantial reach (Figure). Our findings are congruent with literature about health professions education intramural seed grant programs3,14-21 and endowed chair programs and are unique in that they represent the leader perspective.4,22,23 The positive impact described in the literature includes career growth, personal, and professional development, recruitment and retention (particularly with regards to increasing the diversity of faculty), education programs, collaboration, education scholarship and dissemination, innovation, and culture.3,4,14-23

트위터의 리더들은 EIP가 개별 지원자의 경력 성장과 전문성 개발에 미치는 영향을 매우 중요하게 생각합니다. 그들은 채용 및 유지의 운영적 가치를 강조했습니다. 이는 "헌신적이고 재능 있는 직원으로 구성된 간부를 개발하는 데 필요한 시간과 자원을 투자한다."24는 볼먼과 딜의 인적 자원 조직 원칙과 일치합니다. 교육자 경력을 지원하기 위해 EIP를 설계하면 직무 만족도 및 참여도에 상당한 영향을 미칠 수 있으며, 이는 웰빙과도 연결됩니다.25-30 연구에 따르면 번아웃을 경험한 의사는 이직할 가능성이 2배 이상 높으며,30 의사를 교체하는 데 드는 비용은 의사 연봉의 2~3배로 추정됩니다.28 의료 서비스 제공자의 개발과 참여에 투자하는 것이 교체하는 것보다 비용이 적게 듭니다.28,30,31 EIP가 교수진 경력, 채용 및 유지에 미치는 긍정적인 영향을 알리면 리더들이 EIP를 위한 신규 또는 지속적인 리소스를 제공하도록 장려할 수 있습니다.
Our leaders strongly value the impact of EIPs on individual recipients’ career growth and professional development. They highlighted the operational value of recruitment and retention. This aligns with Bolman and Deal’s human resource organizational principle of “investing the time and resources necessary to develop a cadre of committed, talented employees.”24 Designing EIPs to support educator careers can have considerable impact on job satisfaction and engagement, which is linked to well-being.25-30 Studies show that physicians who experience burnout are more than twice as likely to leave,30 and estimate the cost for replacing a physician at 2 to 3 times the physician’s annual salary.28 It is less expensive to invest in the development and engagement of health care providers than to replace them.28,30,31 Communicating the positive impact of EIPs on faculty careers, recruitment, and retention may encourage leaders to provide new or ongoing resources for EIPs.

개인의 경력과 개발에 실질적인 자원을 투자하는 것은 기관 차원에서 조직의 성공에 더 큰 영향을 미칠 수 있습니다.3,4 많은 리더들이 교육 프로그램 개선과 혁신에 대한 EIP의 긍정적인 효과를 가장 중요한 가치 요소로 꼽았습니다. 이러한 운영 결과는 EIP 프로그램 평가 및 영향력 문서화에 적합합니다. 리더들은 EIP가 조직 문화에 미치는 상징적인 영향력을 높이 평가했습니다. 그들은 학습자, 교수진, 환자 커뮤니티를 위한 교육의 가치에 대한 성명을 발표하는 것이 중요하다고 설명했습니다. 이는 상징이 기관의 문화와 가치를 반영하고 영향을 미친다고 주장하는 볼먼과 딜의 상징적 프레임과 일치합니다.24
Investing tangible resources in the career and development of individuals can have further impact at the institutional level on the success of the organization.3,4 Many of our leaders described as a most important value factor the positive effect of EIPs on educational program improvement and innovation. This operational outcome lends itself well to EIP program evaluation and documenting impact. Leaders appreciated the symbolic influence of EIPs on the organizational culture. They described the importance of making a statement about the value of education for the community of learners, faculty, and patients. This aligns with Bolman and Deal symbolic frame, which argues that symbols reflect and influence institutional culture and values.24

순수한 경제적 모델은 학술 교육 미션의 가치 있는 결과를 측정하기에는 너무 협소한 도구입니다.7 보다 광범위한 프레임워크는 기관 리더와 교육자 간의 보다 효과적인 대화를 위한 토대를 형성합니다. 교육 기관에 VMM 도메인을 적용함으로써 기존의 재무적 투자 수익률을 넘어 기관 리더의 가치 변수를 더 잘 이해할 수 있었습니다. Jabbar 외7는 경제 제국주의를 "교육을 포함한 사회생활과 정책의 많은 측면에 대한 [경제의] 힘"의 근간이 되는 개념으로 제안하며, 우리가 자원 배분을 안내하는 데 사용하는 메트릭을 주도합니다. 이러한 헤게모니를 이해하고 인정함으로써 정책 입안자들은 "대안적인 프레임워크와 모델을 통해 의사 결정에 접근할 수 있습니다."7 Jabbar 외7 의 관점은 임상 수익에 의해 주도되는 현재 학술 의료 센터의 자금 조달 모델 내에서 전통적인 ROI 모델을 넘어 보다 총체적인 모델로 나아가기 위해 이와 같은 연구가 중요하다는 점을 강조하고 있습니다. 이러한 지표를 재무적 측면을 넘어선 프레임워크에서 정의하면 리소스 배치를 촉진하고 문화와 더 넓은 사회적 요인의 맥락에 맞는 언어로 옹호를 유도할 수 있습니다. 
A pure economic model is too narrow a tool to measure valued outcomes from the academic education mission.7 A broader framework forms the basis for more effective dialogue between institutional leaders and educators. Applying VMM domains to an academic institution allowed us to better understand institutional leaders’ value parameters beyond traditional financial return on investment. Jabbar et al7 proposes economic imperialism as a concept that underlies the “power [of economics] over so many facets of social life and policy—including education” and drives the metrics we use to guide resource allocation. Understanding and acknowledging this hegemony allows policymakers to “approach decision-making through alternate frameworks and models.”7 The perspective in Jabbar et al7 underscores the importance of a study such as this to move beyond a traditional ROI model to a more holistic one within the current funding models for academic medical centers, which is driven by clinical revenue. Defining these metrics in a framework beyond financial can drive resource deployment and guide advocacy with language that speaks to the culture and a wider context of social factors.

이 연구의 강점은 기관의 다양한 계층에 있는 리더들의 의견을 폭넓게 반영했다는 점입니다. 리더들이 밝혀낸 많은 가치 요소, 특히 개인의 경력, 채용 및 유지, 교육 프로그램, 장학금에 대한 혜택은 여러 기관에 걸쳐 일반화할 수 있습니다. VMM 프레임워크는 다른 기관에서도 적용하여 현지 이해관계자의 상황별 가치 요소를 탐색하는 데 사용할 수 있습니다. 예를 들어, 혁신은 기관의 핵심 사명이며, 리더들은 혁신을 조직의 성공과 연결시켰습니다. 다른 학교에서는 최적의 환자 참여 또는 공중 보건 발전이라는 핵심 사명을 명시할 수 있습니다. EIP를 각 학교의 고유한 미션에 맞춰 조정하면 이러한 프로그램을 시작하고 유지하는 데 필요한 리소스에 대한 리더십의 지원을 이끌어낼 수 있습니다. 가치 요소는 같은 기관의 리더마다 다를 수 있습니다. 이러한 가치 요소를 정의하면 EIP를 설계 및 평가하고, 이러한 프로그램의 가치와 영향에 대해 리더에게 피드백을 제공하고, 신규 또는 지속적인 자금 지원을 옹호하는 방법을 알 수 있습니다(그림). 
Our study’s strengths include a broad representation of leaders’ opinions at various echelons of our institution. Many value factors elucidated by our leaders are generalizable across institutions, particularly the benefit to individual careers, recruitment and retention, educational programs, and scholarship. The VMM framework can be adapted and used in other institutions to explore context-specific value factors from local stakeholders. For example, innovation is a core mission at this institution, and our leaders connected it to the organization’s success. Other schools may articulate core missions of optimal patient engagement or advancing public health. Aligning EIPs with each school’s unique mission can engage leadership support for resources to initiate and sustain such programs. Value factors can differ between leaders at the same institution. Defining these value factors can inform how we design and evaluate EIPs, provide feedback to leaders about the value and impact of these programs and advocate for new or sustained funding (Figure).

제한 사항
Limitations

이 연구에는 한계가 있었습니다. 인터뷰 이후 발생한 코로나바이러스 팬데믹과 사회 정의 운동으로 인한 역동적인 변화로 인해 연구 결과가 제한되었습니다. 오늘날 인터뷰를 진행한다면 리더들이 추가로 적절한 가치를 포함할지 알아보는 것도 흥미로울 것입니다. 연구자로서 우리는 연구팀의 많은 사람들이 과거에 연구비나 연구자금을 받은 적이 있다는 사실을 인지하고 참가자들의 진솔한 목소리를 보고하기 위해 최선을 다했습니다. 연구자와 참여자 간의 상호작용을 통해 공동 구축된 연구 결과를 보고할 때 이러한 관점은 데이터 해석을 향상시켰습니다. 본 연구 결과는 교육 미션을 지원하는 단일 기관을 대표하며, 많은 주제가 광범위하게 적용될 수 있지만 가치의 정의는 현지 상황에 따라 달라질 수 있습니다. 향후 연구에서는 경쟁적인 우선순위의 맥락에서 리더의 투자 결정을 탐구할 수 있습니다.
This study had limitations. Our findings are limited by the dynamic shifts due to the coronavirus pandemic and social justice movements that occurred since our interviews. It would be of interest to know if leaders would include additional pertinent values if interviewed today. As researchers, we made every effort to report the true voices of the participants, acknowledging that many on the research team have been past grant or chair recipients. In reporting findings co-constructed through the interaction between researchers and participants, these perspectives also enhanced our data interpretation. Our findings represent a single institution supportive of the education mission; while many themes may be broadly applicable, the definition of value could vary based on local contexts. Future studies might explore leaders’ investment decisions in the context of competing priorities.

결론
Conclusions

이러한 조사 결과는 보건 과학 및 의료 시스템 리더들이 직접적인 재정적 투자 수익을 넘어 다양한 영역(개인, 재정, 운영, 사회적 또는 사회적, 전략적 또는 정치적)에서 교육자 투자 프로그램에 자금을 지원하는 것의 가치를 발견했음을 시사합니다. 이들은 개별 교육자의 성장과 전문성 개발을 지원하는 유형적 자원이 조직과 커뮤니티에 강력한 이점을 창출한다고 인식했습니다. 이러한 가치는 프로그램 설계 및 평가, 리더에 대한 효과적인 피드백, 학계에서 경쟁하는 우선 순위의 맥락에서 향후 투자에 대한 옹호 등에 영향을 줄 수 있습니다. 이 접근 방식은 다른 기관에서도 리더의 상황 및 리더별 가치 요소를 파악하는 데 사용할 수 있습니다.

These findings suggest that health sciences and health system leaders found value in funding educator investment programs in multiple domains (individual, financial, operational, social or societal, and strategic or political) beyond direct financial return on investment. They perceived that tangible resources to support the growth and professional development of individual educators generate potent benefits to the organization and community. These values can inform program design and evaluation, effective feedback to leaders, and advocacy for future investments in the context of competing priorities in academic medicine. This approach can be used by other institutions to identify context and leader-specific value factors from their leaders.


JAMA Netw Open. 2023 Feb 1;6(2):e2256193. doi: 10.1001/jamanetworkopen.2022.56193.

Identifying Value Factors in Institutional Leaders' Perspectives on Investing in Health Professions Educators

Affiliations collapse

1University of California, San Francisco, San Francisco, California.

2University of California, San Francisco, School of Medicine, San Francisco, California.

PMID: 36795413

PMCID: PMC9936339

DOI: 10.1001/jamanetworkopen.2022.56193

Free PMC article

Abstract

Importance: Investing in educators, educational innovation, and scholarship is essential for excellence in health professions education and health care. Funds for education innovations and educator development remain at significant risk because they virtually never generate offsetting revenue. A broader shared framework is needed to determine the value of such investments.

Objective: To explore the value factors using the value measurement methodology domains (individual, financial, operational, social or societal, strategic or political) that health professions leaders placed on educator investment programs, including intramural grants and endowed chairs.

Design, setting, and participants: This qualitative study used semi-structured interviews with participants from an urban academic health professions institution and its affiliated systems that were conducted between June and September 2019 and were audio recorded and transcribed. Thematic analysis was used to identify themes with a constructivist orientation. Participants included 31 leaders at multiple levels of the organization (eg, deans, department chairs, and health system leaders) and with a range of experience. Individuals who did not respond initially were followed up with until a sufficient representation of leader roles was achieved.

Main outcomes and measures: Outcomes include value factors defined by the leaders for educator investment programs across the 5 value measurement methodology domains: individual, financial, operational, social or societal, and strategic or political.

Results: This study included 29 leaders (5 [17%] campus or university leaders; 3 [10%] health systems leaders; 6 [21%] health professions school leaders; 15 [52%] department leaders). They identified value factors across the 5 value measurement methods domains. Individual factors emphasized the impact on faculty career, stature, and personal and professional development. Financial factors included tangible support, the ability to attract additional resources, and the importance of these investments as a monetary input rather than output. Operational factors identified educational programs and faculty recruitment or retention. Social and societal factors showcased scholarship and dissemination benefits to the external community beyond the organization and to the internal community of faculty, learners, and patients. Strategic and political factors highlighted impact on culture and symbolism, innovation, and organizational success.

Conclusions and relevance: These findings suggest that health sciences and health system leaders find value in funding educator investment programs in multiple domains beyond direct financial return on investment. These value factors can inform program design and evaluation, effective feedback to leaders, and advocacy for future investments. This approach can be used by other institutions to identify context-specific value factors.

건강의 사회적 결정요인과 관련한 환자돌봄의 관찰가능한 프로세스 정의 (Med Educ, 2022)
Defining the observable processes of patient care related to social determinants of health
Junki Mizumoto | Toshichika Mitsuyama | Satoshi Kondo | Masashi Izumiya | Shoko Horita | Masato Eto 

 

 

1 소개
1 INTRODUCTION

건강의 사회적 결정요인(SDH)'사람들이 태어나고, 성장하고, 생활하고, 일하고, 노화하는 조건으로... 글로벌, 국가 및 지역 수준에서 돈, 권력 및 자원의 분배에 의해 형성된다'으로 정의됩니다.1 SDH는 인구 건강 결과에 상당한 기여를 합니다. 의료 전문가는 건강 결과에 영향을 미치는 사회 구조를 인식하는 능력(구조적 역량이라고 함)을 훈련하고2 환자에게 개별화된 치료를 제공하고 공중 보건을 개선해야 하지만3 환자의 사회적 요구에 항상 부응하는 것은 아닙니다.4, 5 일부 연구에 따르면 의사는 사회적 어려움을 가진 환자를 돌보는 데 장벽을 경험한다고 보고했습니다.6-11  
Social determinants of health (SDHs) are defined as ‘the conditions in which people are born, grow, live, work and age … shaped by the distribution of money, power and resources at global, national and local levels’.1 SDH makes a substantial contribution to population health outcomes. Although health care professionals are expected to train the ability to recognise social structure to influence health outcomes (as is called structural competency)2 and to deliver individualised care to patients and improve public health, 3 they do not always respond to patients' social needs.4, 5 Some studies have reported that physicians experience barriers in caring for patients with social difficulties.6-11

[사회적 불평등이 건강 상태에 미치는 영향]은 임상 수련 전문가들에 의해 종종 무시되어 왔습니다.12 예를 들어, 의학전문대학원 교육인증위원회는 레지던트 및 펠로우십 프로그램에서 수련의가 사회적 결정 요인을 식별 및 해결하고 환자의 건강 결과에 미치는 영향을 최적화하도록 준비시킬 수 있는 프로그램은 거의 없다고 보고했습니다.13 그럼에도 불구하고, SDH의 대학원 교육은 독특한 이점을 제공할 수 있을 것입니다. 레지던트는 환자의 건강 요구를 효과적으로 해결하는 기술을 개발하고 지역사회에 보다 생산적으로 참여함으로써 자기 효능감과 만족감을 경험할 수 있습니다.3 환자 건강에 영향을 미치는 사회적 결정 요인에 대해 학습하면 레지던트가 환자의 상황에 집중하고 환자-의사 관계를 강화하며 의료 팀 기반 파트너십을 구축하는 데 도움이 될 수 있습니다.13 이러한 학습은 어려운 상황에 있는 환자를 치료할 때 특히 중요합니다.3, 14 이러한 긍정적인 결과를 고려할 때 건강과 건강 관리를 개선하려면 SDH에 대한 의학 교육의 중요성을 강조하고 보다 효과적인 교육 전략을 개발하는 것이 중요해야 합니다.15 
The effects of social inequity on health conditions have often been ignored by clinical training experts.12 For example, the Accreditation Council for Graduate Medical Education reported that few residency and fellowship programmes can prepare trainees to identify and address social determinants and optimise their effects on patients' health outcomes.13 Nevertheless, postgraduate education in SDH would provide unique benefits. Residents would experience self-efficacy and satisfaction by developing skills to effectively address patients' health needs and engaging more productively with their communities.3 Learning about social determinants that affect patient health could help residents focus on patients' contexts, strengthen patient–physician relationships and build health care team-based partnerships.13 Such learning is particularly important when residents deliver care for patients in difficult situations.3, 14 Considering these positive outcomes, emphasising the importance of medical education about SDH and developing a more effective educational strategy should be important to improve health and health care.15

SDH는 의학교육과 수련의 중요한 부분이지만, 전통적인 접근 방식으로는 SDH 역량을 문서화하기 어렵습니다.16 교육과정에서 비의료 교육 주제를 제시하는 것은 종종 모호한 목표와 불명확한 평가 기준을 특징으로 합니다.17 모호한 교육과정은 수련의에게 이러한 주제가 저평가되고 임상 학습과 단절되어 있다는 인상을 주고 환자 치료와 수련의의 전문적인 성장 모두에 해를 끼칠 수 있습니다.17 또한 SDH와 관련된 커리큘럼 기준이 부족하면 레지던트가 사회적 어려움을 가진 환자에 대한 편견과 오해를 조장할 수 있습니다.18 또한 대부분의 지도전문의는 자연스럽게 생물의학적 및 기술적 기술의 습득에 높은 가치를 두고 수련의의 사회적 책임과 윤리의식 함양에 소홀히 하는 경향이 있습니다.19 
Although SDH is an important part of medical education and training, SDH competence is difficult to document using traditional approaches.16 Presentation of non-medical educational topics in the curriculum is often characterised by vague objectives and unclear assessment criteria.17 An ambiguous curriculum gives trainees the impression that these topics are undervalued and disconnected from clinical learning and may harm both patient care and the professional growth of trainees.17 Lack of curriculum criteria related to SDH may also foster residents' prejudices and misconceptions about patients with social difficulties.18 Additionally, most supervisors naturally place a high value on the acquisition of biomedical and technical skills and tend to neglect the development of social responsibility and ethics in trainees.19

전공의가 SDH 교육을 통해 무엇을 성취해야 하는지에 대한 명시적인 문서화가 중요합니다. 소외 계층에 대한 의료 서비스 제공과 같은 일부 역량은 자체적으로 위탁 가능한 전문 활동(EPA)을 수행할 가치가 있는 것으로 간주됩니다.20 예를 들어, 미국의과대학협회는 병력 청취와 관련된 EPA 하위 항목 중 하나는 SDH에 주의를 기울이는 환자 중심 면담을 수행하는 기술을 입증하는 것이라고 명시하고 있습니다.21 관찰 가능한 실습 활동(OPA)의 위탁 및 매핑은 레지던트 커리큘럼 및 평가 시스템에서 레지던트를 평가하는 데 사용할 수 있습니다.22, 23 OPA'위탁 결정을 내리기 위해 일상적인 실습에서 관찰해야 하는 학습 목표/활동의 모음'으로 정의됩니다.22 OPA는 두 가지 범주로 나뉩니다.

  • 내용 OPA는 '호중구감소성 발열에 대한 항생제 치료 선택'과 같이 각 로테이션마다 다른 특정 활동입니다.
  • 과정 OPA는 '정확하고 관련성 있는 병력 획득'과 같이 로테이션에 걸쳐 공통적으로 적용되는 활동입니다.22 과정 OPA의 개념은 모호한 역량 용어나 광범위한 EPA 대신 공동 의사 결정과 같은 특정 상황에서 의료 전문가가 따라야 하는 정확한 과정과 의료 절차를 설명하는 데 사용됩니다.24

Explicit documentation about what residents should achieve from SDH education is important. Some competencies, such as providing medical care for underserved populations, are considered worthy of their own entrustable professional activities (EPAs).20 For example, the Association of American Medical Colleges states that one of the EPA sub-items related to history taking is to demonstrate the skills to conduct patient-centred interviews that are attentive to SDH.21 Entrustment and mapping of observable practice activities (OPAs) can be used in residency curriculum and assessment systems to assess residents.22, 23 OPA is defined as ‘a collection of learning objectives/activities that must be observed in daily practice to form entrustment decisions’.22 There are two categories of OPA.

  • Content OPAs are specific activities that differ for each rotation, such as ‘choose antibiotic therapy for neutropenic fever’.
  • Process OPAs are activities that are conserved across rotations, such as ‘acquire accurate and relevant history’.22 The concept of process OPA is used to describe the precise process and medical procedures that medical professionals are expected to follow in specific situations, such as shared decision making,24 instead of using ambiguous competency terms or broad EPA.22

일차 의료 또는 포괄적인 일상 건강 관리에서 사회적 결정 요인에 대한 개입을 설명하는 여러 임상 프레임워크가 보고되었습니다. 이러한 개입은 세 가지 수준으로 나뉩니다.15, 25-28 

  • 일반적으로 개별 환자를 위한 미시적 수준
  • 의료 기관 및 지역사회를 위한 중시적 수준
  • 의료 시스템 및 행정 부서를 포함한 광범위한 인구를 위한 거시적 수준 

이러한 프레임워크는 SDH를 의료 시스템에 통합하기 위한 청사진을 제공하지만, 의사가 [실제로 환자 치료를 제공하는 과정]은 자세히 설명되어 있지 않습니다. 또한 이러한 프레임워크의 대부분은 체계적인 합의 그룹 방법을 사용하여 개발되지 않았습니다. 임상 환경에서 의사가 하는 일을 자세히 설명하지 못하면 SDH 교육의 가치가 떨어집니다. 따라서 환자의 사회적 맥락과 필요에 대한 이해를 바탕으로 환자를 돌보는 [과정 OPA]를 설명해야 합니다. 본 연구는 합의 집단 방법을 사용하여 일차 진료 환경에서 SDH와 관련된 환자 진료 [과정 OPA]를 체계화하는 것을 목표로 했습니다.
Several clinical frameworks have been reported that describe interventions for social determinants in primary health care or comprehensive day-to-day health care. These interventions are usually divided into three levels:

  • microlevel for individual patients,
  • mesolevel for medical institutions and communities and
  • macrolevel for broader populations, including health care systems and administrative divisions.1525-28 

Although such frameworks provide a blueprint for integrating SDH into health care systems, the processes by which physicians actually deliver patient care have not been described in detail. In addition, most of these frameworks have not been developed using systematic consensus group methods. Failure to describe in detail what physicians do in clinical settings reduces the value of SDH education. The process OPA for caring for patients based on an understanding of patients' social contexts and needs should therefore be explicated. This study aimed to codify the process OPA for patient care related to SDH in a primary care setting, using consensus group methods.

2 연구 방법
2 METHOD

연구 설계
Study design

일본에서 2022년 1월부터 2022년 2월까지 일차 진료 환경에서 SDH와 관련된 진료 프로세스에 대한 합의를 도출하기 위해 수정된 델파이 기법29 을 시행했습니다. 수정 델파이 연구에서는 연구자들이

  • (i) 문헌 검토를 수행하거나 다른 학술적 방법을 사용하여 작성한 초기 목록을 배포하고,
  • (ii) 전문가들로부터 의견을 추출하여 익명으로 반복적으로 피드백을 제공하며,
  • (iii) 구조화된 방식으로 합의에 도달합니다.30

수정 델파이 기법은 전문가 합의를 도출하는 데 널리 사용되며,31-33 특히 다양한 이해관계자가 있고 익명으로 독립적인 의견 제공이 필요할 때 유용합니다.34, 35 본 연구는 델파이 연구 수행 및 보고 지침(CREDES)을 따랐습니다.36 모든 라운드에서 설문지는 웹 기반 설문조사 시스템(SurveyMonkey: Survey Monkey, Inc., San Mateo, CA, USA)을 통해 제공되었습니다. 이 연구에 대한 윤리적 승인은 도쿄대학교 의학전문대학원 및 의학부 연구윤리위원회로부터 받았습니다(제2020250NI호).
A modified Delphi technique29 was implemented from January 2022 to February 2022 in Japan to reach a consensus on the processes of medical care related to SDH in a primary care setting. In a modified Delphi study, researchers

  • (i) distribute an initial list that is prepared by conducting a literature review or using other academic methods,
  • (ii) extract opinions from the experts and give them anonymous feedback iteratively and
  • (iii) reach a consensus in a structured way.30 

The modified Delphi technique is widely used to create expert consensus,31-33 especially when there are diverse stakeholders and when anonymous and independent provision of their opinions is needed.34, 35 This study followed the Guidance on Conducting and Reporting Delphi Studies (CREDES.)36 In all rounds, questionnaires were provided via a web-based survey system (SurveyMonkey: Survey Monkey, Inc., San Mateo, CA, USA). Ethical approval for this study was obtained from the research ethics committee of the University of Tokyo Graduate School of Medicine and Faculty of Medicine (No. 2020250NI).

잠재적으로 중요한 요인 목록 작성
Compilation of a list of potentially important factors

잠재적으로 중요한 단계의 예비 목록을 작성하기 위해 2021년 9월에 제1저자가 세 개의 전자 데이터베이스(MEDLINE, CINAHL, PsycINFO)에서 문헌 검색을 수행했습니다. 검색 결과 중복을 제거한 후 447개의 논문이 확인되었습니다. 논문 제목과 초록을 기준으로 선별한 결과 110편의 논문이 선정되었습니다. 전체 텍스트 심사 후 27편의 논문이 선정되었습니다.2, 3, 12, 14, 20, 37-58 이 문헌 검색의 세부 사항은 표 S1에 나와 있습니다. 이 문헌 검색을 통해 연구자들은 '구조적 역량'이라는 용어가 자주 사용된다는 것을 재발견하고 검색 공식을 수정했습니다. 의견, 리뷰, 사설, 사례 보고서, 활동 보고서 등 다양한 종류의 논문이 포함되었습니다. SDH와 관련된 환자 치료 과정을 직접적으로 언급한 논문은 거의 없었습니다. 첫 번째 저자는 논문을 종합하여 잠정적인 목록을 작성했습니다. 두 번째, 네 번째, 다섯 번째, 여섯 번째 저자는 이 목록을 비판적으로 검토 및 분석하고 반복적인 토론을 거쳐 일부 수정을 가했습니다. 목록은 커뮤니케이션, 실천, 자기 관리, 옹호라는 네 가지 구성 요소로 나뉜 13개 항목으로 구성되었습니다(표 S2 참조). 
To develop a preliminary list of potentially important steps, a literature search of three electronic databases (MEDLINE, CINAHL and PsycINFO) was conducted by the first author in September 2021. The search identified 447 articles after the removal of duplicates. The articles were screened by article titles and abstracts, and 110 articles were selected. After full-text screening, 27 articles were selected.2, 3, 12, 14, 20, 37-58 Details of this literature search are shown in Table S1. Through this literature search, researchers rediscovered that the term ‘structural competency’ was often used, and the search formula was modified. A wide range of articles was included (e.g. opinions, reviews, editorials, case reports and activity reports). Few articles directly referred to the processes of patient care related to SDH. The first author synthesised the articles and constructed a tentative list. The second, fourth, fifth and sixth authors critically reviewed and analysed the list, held iterative discussions and made some modifications. The list consisted of 13 items, which were divided into four components: communication, practice, self-management and advocacy (see Table S2).

참가자 모집
Participant recruitment

성별, 연령, 직업, 진료 환경 등 구성원의 다양성을 보장하기 위해 의도적인 샘플링 전략을 사용했습니다. 패널은 임상 전문가(의사, 간호사, 공중보건의, 사회복지사, 약사, 의료 사무원), 레지던트, 의대생, 연구자, 소외 계층 및 환자를 위한 지원 위원으로 구성되었습니다. 모집 대상을 결정한 이유는 다음과 같습니다. 

  • 첫째, SDoH와 관련된 일차 진료에는 의료 사무원을 포함한 여러 직종이 관여하고 있으며,59, 60 의사가 수행하는 과정을 다루는 연구에서도 다양한 의료 직종의 의견이 존중되어야 합니다.
  • 둘째, 의학교육의 수요를 충족하기 위해 과정에 대한 명시적인 설명이 필요하며, 의학교육의 이해당사자인 전공의, 학생, 환자의 의견을 수렴해야 합니다.
  • 셋째, SDoH와 관련된 환자 진료의 범위는 전 국민이었지만, 사회적 소외계층도 주요 이해관계자로서 본 연구에서 중요한 잠재적 의사결정권자로 고려되어야 합니다. 그러나 소외 계층을 모집하고 참여시키는 데 현실적인 어려움이 있어61, 62 연구자들은 이 연구에서 이들을 모집할 수 없었습니다. 따라서 연구자들은 공중보건간호사, 소외계층 지원위원, 사회복지사, 사회학 전문 연구자 등 소외계층의 옹호자로 간주될 수 있는 사람들에게 본 연구에 참여해 줄 것을 요청했습니다.63 연구자들은 소외계층의 의견을 물어볼 필요가 없다고 생각하지 않는다는 점을 강조해야 합니다.

A purposive sampling strategy was used to ensure member diversity, including gender, age, profession and practice setting. The panel consisted of clinical professionals (physicians, nurses, public health nurses, social workers, pharmacists and medical clerks), residents, medical students, researchers, support members for marginalised people and patients. The reasons for determining recruitment targets were as follows. First, multiple professions, including medical clerks, are involved in the primary care practice related to SDoH,59, 60 and, even in research dealing with processes undertaken by physicians, the opinions of various medical professions should be respected. Second, an explicit description of the processes was to meet the demands of medical education, and the opinions of residents, students and patients, all of whom were stakeholders in medical education, should be solicited. Third, although the scope of patient care related to SDoH was the entire population, the socially marginalised population should be considered a key stakeholder and thus an important potential decision maker in this study. However, due to practical difficulties in recruiting the marginalised population and their participation,61, 62 researchers could not recruit them in this study. Researchers thus asked public health nurses, support members for marginalised people, social workers and expert researchers in sociology, who could be considered advocators, to participate in this study.63 It should be emphasised that researchers do not believe that the opinions of marginalised people need not be asked.

우리는 SDH에 대한 깊은 이해와 상당한 실무 경험이 있는 임상 전문가, SDH에 대한 관심과 지식이 있는 레지던트 및 의대생, 만성 질환을 앓고 있고 제1저자가 근무하는 병원에 정기적으로 다니는 환자를 모집했습니다. 환자 모집 과정에서 환자의 사회적 배경과 상황은 고려하지 않았습니다. 일반 의학(일반 내과, 호스피탈리스트, 가정의학과), 소아과, 정신과, 응급의학과 의사와 간호사를 모집했는데, 이러한 전문 분야는 일반적으로 미분화 문제를 가진 환자를 진료하고 이전 연구에서 SDH를 다룰 필요가 있는 것으로 확인되었기 때문입니다.20, 39, 50, 64-67 보건의료 및 의학 교육 분야 연구자뿐만 아니라 소외된 사람들과 함께 일하는 사회학자도 모집했습니다. 소외된 사람들을 위한 지원 멤버는 비의대생 2명과 의사 및 정치인 1명으로 구성되었습니다. 이 세 명은 노숙자 및 기타 소외된 사람들을 위해 자발적으로 지원을 제공했습니다. 모든 참가자는 이 연구에 참여하기 위해 자발적으로 서면 동의를 제공했습니다. 참가자에게는 2000엔(약 15달러) 상당의 상품권으로 구성된 인센티브가 제공되었습니다. 
We recruited clinical professionals who had a deep understanding and considerable practice of SDH, residents and medical students who had an interest in and some knowledge of SDH and patients who had chronic conditions and regularly attended the hospital in which the first author worked. The social backgrounds and contexts of the patients were not considered in the recruitment process. We recruited physicians and nurses in general medicine (general internal medicine, hospitalist and family medicine), paediatrics, psychiatry and emergency medicine because these specialties usually care for patients with undifferentiated problems and have been identified as needing to address SDH in previous studies.20, 39, 50, 64-67 We recruited not only researchers in health care and medical education but also sociologists working with marginalised populations. Support members for marginalised people consisted of two non-medical students and one physician and statesperson. These three members voluntarily provided support for people experiencing homelessness and other marginalised people. All participants voluntarily provided their written consent to participate in this study. An incentive comprising a gift certificate worth JPY 2000 (approximately US$ 15) was offered to participants.

데이터 수집 및 분석
Data collection and analysis

각 라운드에서 참가자들은 5점 리커트 척도(1점: 전혀 중요하지 않다 ~ 5점: 매우 중요하다)로 각 항목을 평가하도록 요청받았습니다. 또한 참가자들은 각 항목에 대한 수정 사항을 제안하고 SDH와 관련된 의료 측면에 적합하다고 생각되는 다른 단계를 추가하도록 요청받았습니다. 참가자들의 응답은 이전 검토33 및 기타 선행 연구를 바탕으로 사전 정의된 기준에 따라 분석 및 선정되었습니다.68, 69 구체적으로 (i) 평균 점수가 4점 이상, (ii) 표준편차가 1 미만, (iii) 패널의 75% 이상이 4점 또는 5점을 준 항목은 모두 통과했습니다. 참가자들은 각 설문지를 작성하는 데 2주의 시간이 주어졌습니다. 각 마감일 며칠 전에 리마인더 이메일이 발송되었습니다. 모든 저자는 모든 라운드에서 모든 서면 의견을 읽었습니다. 첫 번째, 두 번째, 세 번째 저자는 내용 분석을 사용하여 의견을 분석했습니다.70 코딩 데이터는 다른 연구자들과의 논의를 통해 축소되고 수정되었습니다. 다음 라운드가 시작될 때 이전 라운드의 익명 결과 요약본이 전문가들에게 제공되었습니다. 우리는 모든 진술이 가능한 한 높은 수준에서 합의에 도달할 수 있는 동등한 기회를 갖도록 하기 위해 일부 진술이 이미 합의에 도달했더라도 후속 라운드에서 모든 진술을 다시 포함시키기로 결정했습니다.71 각 델파이 단계는 일본어로 진행되었습니다. 결과는 연구원과 전문 번역가가 영어로 번역했습니다. 

In each round, participants were asked to evaluate each item on a 5-point Likert scale (1: absolutely unimportant to 5: absolutely important). The participants were also asked to propose revisions for each item and add any other steps considered suitable for aspects of medical care related to SDH. Their responses were analysed and selected using a predefined standard based on a previous review33 and other previous studies.68, 69 Specifically, items that met all the following standards were passed: (i) a mean score of 4 or higher, (ii) a standard deviation of less than 1 and (iii) scores of 4 or 5 from 75% or more of the panellists. Participants were given 2 weeks to fill out each questionnaire. Reminder e-mails were sent a few days before each deadline. All authors read all written comments in every round. The first, second and third authors analysed the comments using content analysis.70 Coding data were collapsed and then modified through discussion with the other researchers. At the start of the subsequent round, a summary of anonymous results from the previous round was provided to the experts. We chose to include all statements again in the subsequent round (even though some had already reached consensus) to ensure that every statement had an equal chance of achieving consensus at the highest possible level.71 Each Delphi step was conducted in Japanese. The results were translated into English by the researchers and professional translators.

3 결과
3 RESULTS

델파이 패널에는 총 63명이 모집되었으며, 이 중 61명이 참여했습니다. 2명은 업무 부담으로 인해 참여를 거부했습니다. 인구통계학적 데이터는 표 1에 나와 있습니다. 모든 참가자가 델파이 라운드를 완료했습니다.

A total of 63 people were recruited to the Delphi panel, of which 61 participated. Two declined participation because of work pressure. The demographic data are shown in Table 1. All participants completed the Delphi rounds.

1라운드에서는 초기 목록에서 탈락한 항목이 없었습니다. 동의 수준, 평균 및 표준편차는 표 S3에 나와 있습니다. 참가자들의 의견에 대한 내용 분석을 바탕으로 12개 항목이 수정되었습니다. 분석 결과에 따라 전문가 간 팀 기반 성찰에 대한 단계가 하나 추가되었고, 이에 따라 세 번째 구성 요소인 '자기 관리'를 '유지 관리'로 변경했습니다. 또한 모든 단계가 지속적인 환자-의사 관계 내에서 다른 전문가 및 이해관계자와 협력하여 수행되어야 한다는 서문을 추가했습니다(표 S4 참조). 
In Round 1, no items were eliminated from the initial lists. The agreement levels, means and standard deviations are shown in Table S3. Twelve items were modified based on the content analysis of the participants' comments. According to the analysis, we drafted one new step about interprofessional team-based reflection and accordingly changed the third component ‘self-management’ to ‘maintenance’. We also developed a preface that stated that all steps should be conducted within an ongoing patient–physician relationship and in collaboration with other professionals and stakeholders (see Table S4).

2라운드 이후, 수정된 모든 항목이 긍정적인 합의 기준을 충족하는 것으로 확인되었습니다(표 S5 참조). 일부 사소한 수정이 이루어졌으며 모든 참가자의 승인을 받았습니다. 3라운드는 진행되지 않았습니다. 최종 버전의 목록은 14개 항목으로 구성되었으며, 이는 4개의 구성 요소로 나뉩니다(표 2).
After Round 2, all modified items were confirmed to meet the positive consensus standard (see Table S5). Some minor revisions were made and were approved by all participants. Round 3 was not performed. The final version of the list consisted of 14 items, which were divided into four components (Table 2).

4 토론
4 DISCUSSION

이 연구에서는 SDH와 관련된 환자 치료의 OPA 프로세스를 확인했습니다. 프로세스는 커뮤니케이션, 진료, 유지 관리, 옹호라는 네 가지 구성 요소로 분류되었습니다. 각 구성 요소는 서로 밀접한 관련이 있습니다. 커뮤니케이션은 좋은 진료에 기여합니다. 의사 자신과 팀원을 위한 유지 관리는 좋은 의사 소통과 진료의 기본입니다. 커뮤니케이션, 진료, 유지보수에 대한 경험은 옹호에 동기를 부여할 수 있습니다. 이 커뮤니케이션-실무-유지-옹호(CPMA) 프레임워크(그림 1 참조)는 이전에 보고된 접근 방식과 몇 가지 요소를 공유합니다. 예를 들어, 빈곤 아동 간호와 관련된 역량은 환자 간호, 대인관계 및 의사소통 기술, 전문성, 시스템 기반 실무 영역과 관련이 있습니다.39, 72 또한 간호 교육 전문가들은 환자의 기저질환을 변화시키고 건강 형평성을 증진하기 위해 비판적 돌봄 접근법과 세심한 권한 부여 간호 제공을 제안해 왔습니다.55, 73 
This study identified the process OPA for patient care related to SDH. The processes were classified into four components: communication, practice, maintenance and advocacy. Each component is closely related to each other. Communication contributes to good practice. Maintenance for physicians themselves and team members is the base of good communication and practice. Experience about communication, practice and maintenance may motivate advocacy. This Communication–Practice–Maintenance–Advocacy (CPMA) framework (see Figure 1) shares some elements with previously reported approaches. For example, the competency relevant to the care of children in poverty is related to the domains of patient care, interpersonal and communication skills, professionalism and system-based practice.39, 72 In addition, professionals in nursing education have proposed a critical caring approach and provision of sensitive empowering care to change patients' underlying conditions and promote health equity.55, 73

 

CPMA 프레임워크에는 몇 가지 독특한 장점이 있습니다. 

  • 첫째, 전문가 패널은 다양한 이해관계자로 구성됩니다. 환자, 사회학자, 소외된 사람들을 위한 지원 단체의 참여로 패널의 다양성이 증가했습니다. 
  • 둘째, 프레임워크는 환자 치료 과정을 상세하게 설명하여 의학교육에 대한 활용도를 높였습니다. 

[생물 의학 및 기술 중심의 지배적인 커리큘럼 패러다임]은 정답을 얻고 문제를 해결하는 데 중점을 둡니다. 이 패러다임에서 학습자는 SDH와 같이 확실성이 낮은 문제에 직면할 때 불편함을 느낄 수 있습니다.74, 75 문제를 해결하는 과정을 명시하면 이러한 불편함을 줄일 수 있습니다.17, 20 이 프레임워크는 잠재적으로 학습자에게 명확한 지표를 제공하고 임상 실습 및 현장 기반 평가에 반영하는 데 적용될 수 있습니다.
The CPMA framework has some unique advantages.

  • First, the expert panel consisted of various stakeholders. The participation of patients, sociologists and support members for marginalised people increased panel diversity.
  • Second, the framework clearly described in detail the patient care processes to increase its availability for medical education.

The dominant curriculum paradigm, which is centred on biomedicine and technology, focuses on obtaining correct answers and solving problems. In this paradigm, learners may feel uncomfortable when facing challenges with low levels of certainty, such as SDH.74, 75 Making explicit the process of tackling challenges could potentially reduce this discomfort.17, 20 This framework potentially provides learners with a clear indication and could be applied to reflection on their clinical practice and workplace-based assessment.

의사소통
Communication

이 구성 요소에는 환자와의 커뮤니케이션다른 전문가 및 이해관계자와의 커뮤니케이션이 모두 포함됩니다. 델파이 라운드에서는 환자에 대한 의사의 편견과 태도에 대한 의견이 나왔습니다. 예를 들어, 패널리스트들은 첫 번째 목록의 일부 항목('의사는 공동 의사 결정을 수행한다', '의사는 환자가 자신의 건강 상태에 대처할 수 있도록 적절한 설명과 지원을 제공한다')에 가부장적인 측면이 포함되어 있다고 지적했습니다. 일부 패널리스트는 의사 자신의 편견과 편견에 대한 의사의 인식이 언급되어야 한다고 제안했습니다. 이러한 편견과 태도는 항상 관찰할 수 있는 것은 아니지만, 환자 중심주의,49 다른 전문가와의 협력76 및 환자 결과를 저해할 수 있으므로76-78 의사소통 관련 교육 및 실습에서 심각하게 고려해야 합니다. 이 구성 요소는 의사소통에 대한 자세한 내용을 설명하며 효과적인 교육 및 평가에 기여할 수 있습니다. 
This component includes both communications with patients and communications with other professionals and stakeholders. The Delphi rounds revealed opinions about physicians' biases and attitudes towards patients. For example, panellists pointed out that some of the items in the first list (‘The physician performs shared decision making’ and ‘The physician provides adequate explanation and support to help the patient deal with their health condition’) included some paternalistic aspects. Some panellists suggested that physicians' awareness of their own prejudice and biases should be mentioned. Although these biases and attitudes are not always observable, they may impair patient-centredness,49 collaboration with other professionals76 and patient outcomes76-78 and should thus be seriously considered in communication-related education and practice. This component illustrates detailed content about communication and may contribute to effective education and assessment.

실습
Practice

이 구성 요소는 의사가 환자 진료를 위해 진료실에서 실제로 하는 일을 설명합니다. 목록의 각 단계는 의사가 수행하는 활동의 관점에서 설명되었습니다. 이는 의사가 특권적인 위치에 있다는 것을 의미하지는 않습니다. 델파이 라운드에서 두 가지 관련 의견이 발견되었습니다. 

  • 첫째, 사회적 요구가 충족되지 않은 환자를 진료할 때 환자 중심 진료를 고려하는 것이 특히 중요합니다. 의사는 지시적이고 억압적인 방식으로 말함으로써 그러한 환자의 권한을 박탈하는 경향이 있기 때문에 이 점을 강조해야 합니다.79-81
  • 둘째, 의사가 환자의 사회적 상황에 혼자 대처할 수 없으므로 다학제 간 협업의 중요성을 강조해야 합니다. 이러한 의견은 의료 시스템에서 일어나는 일은 환자에게 일어나는 일의 일부에 불과하다는 사실을 강조합니다.54 이러한 전문가들의 견해에 따라 이 구성 요소는 종단적이고 포괄적이며 팀 기반의 접근 방식이 필요하다고 주장합니다. 사회적 어려움을 포함한 복잡한 문제는 단기간에 해결되는 경우가 드물고 안정화가 더 적절한 목표인 경우가 많기 때문에 문제 해결을 요구하지 않습니다.82

This component illustrates what physicians actually do in the office for patient care. Each step on the list was described in terms of the activities performed by physicians. This does not imply that physicians are in a privileged position. The Delphi rounds uncovered two related opinions.

  • First, it is particularly important to consider patient-centred care when seeing patients with unmet social needs. This should be emphasised because physicians tend to disempower such patients by speaking in a directive and oppressive manner.79-81 
  • Second, a physician cannot cope alone with patients' social circumstances, and the importance of interdisciplinary collaboration should be emphasised. These opinions underlined the fact that what happens in the health care system is only part of what happens to the patient.54 Under these expert views, this component insists that a longitudinal, comprehensive and team-based approach is needed. It does not ask for problem solving because complex problems, including social difficulties, are rarely resolved over a short period, and stabilisation is often a more appropriate goal.82

유지 관리
Maintenance

이 구성 요소는 전문가의 소진을 완화하고 환자 치료의 질을 유지 및 개선하기 위해 의사가 진료실 밖에서 하는 일을 보여줍니다. 환자의 사회적 요구 해결과 관련된 소진83-85은 자가 모니터링38 및 팀 빌딩을 통해 줄일 수 있습니다.86, 87 이 구성 요소는 다른 공개된 프레임워크와 비교하여 독특하며 수련의와 임상 감독자에게 새롭고 필수적인 통찰력을 제공할 수 있습니다.
This component illustrates what physicians do outside their office to mitigate professionals' burnout and then maintain and improve the quality of patient care. Burnout associated with addressing patients' social needs83-85 can be reduced by self-monitoring38 and team building.86, 87 This component is unique compared with other published frameworks and may give novel and essential insights to trainees and clinical supervisors.

옹호
Advocacy

의료 전문가들 사이에서 윤리적 가치와 인권에 대한 존중 부족은 건강 형평성에 대한 도전이며,88 건강 옹호는 의사에게 중요한 역량으로 간주되어야 합니다.89 그러나 건강 옹호에 대한 정의가 불분명하여 이 영역을 가르치고 평가하기 어려운 경우가 있습니다.90 [미시적 수준과 거시적 수준의 사회적 책임]에 대한 강조가 증가하고 있지만,91 일부 의사들은 옹호를 [사회 시스템의 개선이 아닌 개별 환자에게 적절한 치료를 제공하는 것]으로 생각했습니다.92,93 델파이 라운드에서는 개별 의사가 실행하는 옹호의 실현 가능성에 대한 몇 가지 의문이 제기되었습니다. 예를 들어, 사회 및 지역사회 구조의 변화는 개별 의사의 범위를 벗어난 것으로 판단되었습니다. 일부 전문가들은 일부 수련의, 특히 초심자들이 시스템 차원의 문제를 지나치게 강조하는 것에 대해 거부감을 나타낼 수 있다고 주장했습니다. 의대생을 대상으로 한 이전 연구에서도 비슷한 질문이 제기되었습니다: 학생들은 의사가 개별 환자를 옹호해야 한다는 데는 반대하지 않았지만, 옹호와 관련된 어려움을 예상했습니다.94 본 연구는 의사가 임상 환경에서 실천할 수 있고 실천해야 하는 옹호의 구체적인 측면에 대한 합의를 보여주었습니다. 예를 들어, 의료 시스템 내에서 접근성을 저해하는 요인을 파악하는 것은 일차 진료에 종사하는 모든 의사가 수용할 수 있는 것으로 판단되었습니다.

A lack of respect for ethical values and human rights among health professionals is a challenge for health equity,88 and health advocacy should be considered an important competency for physicians.89 However, definitions of health advocacy are sometimes unclear, which makes it difficult to teach and assess this domain.90 Although an increasing emphasis on mesolevels and macrolevels of social accountabilities has been put,91 some physicians thought of advocacy as the delivery of appropriate care for individual patients rather than an improvement in social systems.92, 93 The Delphi rounds generated some questions about the feasibility of advocacy as practised by individual physicians. For example, changing social and community structures was judged to be beyond the scope of individual physicians. Some experts insisted that some trainees, especially novices, may express a refusal of excessive emphasis on system-level challenges. A previous study of medical students raised similar questions: Although students did not disagree that physicians should advocate for individual patients, they anticipated challenges related to advocacy.94 The present study demonstrated a consensus on the specific aspects of advocacy that physicians can and should practise in clinical settings. For example, identifying factors that impede access within their health care system was judged to be acceptable for all physicians in primary care.

이 연구에는 몇 가지 한계가 있었습니다. 

  • 첫째, 모든 저자와 참여자가 일본에 거주하고 일하고 있습니다. 일본 상황과 관련된 환자 치료 측면에만 초점을 맞추지 않았기 때문에 이러한 연구 결과를 다른 국가에 일반화할 수 있을지는 불확실합니다. 일본은 보편적 의료 서비스와 잘 발달된 사회 보장 시스템으로 유명하지만, 사회 및 건강 불평등이 여전히 존재합니다.95 예를 들어, 일본의 상대적 빈곤층 비율은 2016년에 16%로 경제협력개발기구 평균인 11%보다 높았습니다.95 한부모 가구의 비율은 2013년에 50.8%로 가장 높았습니다.96 이러한 사회 경제적 불평등은 건강 결과를 손상시킵니다.98-101 본 연구 결과는 고소득 국가에서도 SDH 해결이 중요하다는 점을 강조합니다. 또한 일본에는 다양한 인종과 민족 정체성을 가진 사람들이 살고 있지만 대다수는 같은 인종과 민족입니다. 연구자들은 패널리스트의 인종 및 민족 정체성에 대해 묻지 않았지만, 아마도 거의 모든 패널리스트가 아시아인이었을 것으로 추정됩니다. 이 결과를 다른 문화와 국가에 적용할 때는 이러한 사실에 유의해야 합니다.
  • 둘째, 각 라운드의 찬성률이 일관되게 높았다는 점입니다. 이는 임상에서 SDoH를 다루는 것이 정치적으로나 윤리적으로 옳은 것처럼 보이기 때문일 수 있으며, 참가자들은 제안된 항목을 거부하거나 거부하지 않으려는 경향이 있었습니다. 패널들의 의견에 대한 질적 분석은 상호보완적으로 작용하여 연구자들이 참가자들의 중요한 아이디어를 연구 결과에 반영하는 데 도움이 되었습니다.
  • 셋째, 돌봄의 과정을 항목화하는 데에는 불가피한 한계가 있습니다. 일련의 과정이 미리 정해져 있고 불변하며 엄격하게 따라야 한다는 것은 오해입니다. 모든 환자는 고유한 사회적 맥락을 가지고 있으며, 의사와 이해관계자는 복잡한 사회적 결정 요인을 시급히 해결하기 위해 비판적이고 창의적인 사고의 중요성을 인식해야 합니다.
  • 40 넷째, 이 연구는 치료 과정에서 관찰 가능한 단계만을 조사했습니다. SDH를 충분히 다루기 위해서는 전문가가 자신의 편견을 인식하는 능력과 같이 관찰할 수 없는 진료의 측면을 조사할 필요가 있습니다.40, 42, 45, 50, 77
  • 다섯째, 이 연구에 참여한 환자 수가 적었습니다. 다른 전문가와 동등한 수준의 연구 참여에 대한 부담감이 환자 참여에 장애가 될 수 있습니다. 환자의 참여와 파트너십이 촉진되어야 하며,102, 103 새로 개발된 프레임워크는 보다 현실적이고 실질적인 방식으로 환자의 의견을 수렴하는 추가 연구를 통해 개선되어야 합니다.104 특히 소외된 환자들의 의견을 반영할 수 있는 잘 설계된 연구가 필요합니다.105, 106

This study had several limitations.

  • First, all authors and participants live and work in Japan. Although we did not focus only on aspects of patient care specific to the Japanese situation, it is uncertain whether these findings can be generalised to other countries. Although Japan is known for its universal health care and well-developed social security systems, social and health inequities still exist.95 For example, the proportion of people with relative poverty in Japan was 16% in 2016, higher than that of the Organization for Economic Co-operation and Development average of 11%.95 This proportion among single-parent households was 50.8% in 2013,96 which was the highest.97 These socio-economic inequities impair health outcomes.98-101 The present findings highlight the importance of addressing SDH even in high-income countries. In addition, although people with various racial and ethnic identities live in Japan, the majority are of the same race and ethnicity. Researchers did not ask about panellists' racial and ethnic identities, but, presumably, almost all of the panellists were Asian. These facts should be noted when adapting the result to other cultures and countries.
  • Second, the approval rates of each round were consistently high. It may be that addressing SDoH in clinical practice sounds politically and ethically correct, and participants tended to avoid denying and refusing proposed items. Qualitative analysis of the panellists' comments functioned complementarily, and it helped researchers to reflect participants' important ideas to the result.
  • Third, there are inevitable limitations in itemising the process of care. It is a misapprehension that a series of processes is predetermined and immutable and should be strictly followed. Every patient has a unique social context, and physicians and stakeholders must be aware of the importance of critical and creative thinking to urgently address complex social determinants.40 
  • Fourth, this study examined only observable steps in the processes of care. To fully address SDH, it is necessary to investigate unobservable aspects of care, such as professionals' ability to recognise their own biases.4042455077 
  • Fifth, the number of patients who participated in this study was a few. The burden of participating in research, which was equivalent to other professionals, might be a barrier to patient participation. Patient involvement and partnerships should be promoted,102103 and this newly developed framework should be refined by further research that gathers patient feedback in a more realistic and practical manner.104 In particular, well-designed research to reflect the opinions of marginalised patients is needed.105106

저희는 OPA 과정을 설명하는 것이 SDH에 대한 대학원 교육을 개선하는 데 중요한 단계라고 생각합니다. 이러한 연구 결과는 몇 가지 임상적 함의를 가지고 있습니다. 

  • 첫째, 건강 상태 불평등의 근원은 부와 권력의 불평등한 분배에 있지만, 많은 개입이 불평등 문제를 개인화하여 개인의 행동을 주요 해결책으로 삼고 있습니다. 이러한 접근 방식은 '판타지적 패러다임'이라는 비판을 받아야 합니다.107 이 연구는 의사가 개별 환자와 관련된 SDH를 해결하는 방법에 중점을 두었지만, 우리는 이러한 문제의 상위 원인을 이해하는 것이 중요하다는 것을 인식하고 있습니다. SDH의 상류 측면을 고려하는 것은 맥락화된 진료에 필수적입니다.3
  • 둘째, CPMA 프레임워크는 일차 진료에 종사하는 모든 의사의 기본 요구 사항을 설명합니다. 전문 진료를 위해서는 보다 전문화된 역량이 필요합니다.108-111 의사는 특히 소외된 인구에게 진료를 제공할 때 이 목록 이상의 것을 목표로 해야 합니다.112-114

We believe that describing process OPA is an important step towards improving postgraduate education about SDH. These findings have several clinical implications.

  • First, although the root of health status inequity is in the unequal distribution of wealth and power, many interventions personalise the problem of inequality and target individual behaviour as the main solution. This approach should be criticised as a ‘fantasy paradigm’.107 Although this study focused on how physicians address SDH related to individual patients, we recognise the importance of understanding upstream causes of these problems. A consideration of the upstream aspects of SDH is essential to contextualised care.3 
  • Second, the CPMA framework describes basic requirements for all physicians in primary care. More specialised competency is needed for expert practice.108-111 Physicians should aim beyond the list, especially when delivering care to marginalised populations.112-114

OPA는 수련의 평가 계획에 적절하게 매핑되어 있다면 평가할 수 있습니다.22 따라서 CPMA 프레임워크를 사용하는 평가 방법은 다른 맥락에서 개발 및 검증되어야 합니다. 이 프레임워크 실행의 교육 효과는 환자가 보고한 결과를 사용하여 평가해야 합니다. 또한, 이 프레임워크가 공유된 의사결정 및 환자 중심주의를 포함한 다른 환자 치료 영역과의 연관성을 확인해야 합니다.
OPA can be assessed if they are appropriately mapped in a trainee assessment plan.22 Thus, assessment methods using the CPMA framework should be developed and validated in other contexts. The educational effectiveness of implementing this framework should be evaluated using patient-reported outcomes. Additionally, the association of this framework with other realms of patient care, including shared decision making and patient-centredness, should be confirmed.

5 결론 5 CONCLUSION

수정된 델파이 기법을 사용하여 SDH와 관련된 환자 치료를 위한 프로세스 OPA를 설명했습니다. CPMA 프레임워크는 SDH 관련 의료 교육을 개선하는 데 도움이 될 수 있습니다. 다른 맥락에서 추가 검증, 구현 및 탐색을 위해서는 추가 연구가 필요합니다.
Using the modified Delphi technique, the process OPA for patient care related to SDH was described. The CPMA framework could help to improve medical education in SDH. Further research is needed for further validation, implementation and exploration in other contexts.

 


Med Educ. 2023 Jan;57(1):57-65. doi: 10.1111/medu.14915. Epub 2022 Aug 23.

Defining the observable processes of patient care related to social determinants of health

Affiliations collapse

1Department of Medical Education Studies, International Research Center for Medical Education, Graduate School of Medicine, The University of Tokyo, Tokyo, Japan.

PMID: 35953461

DOI: 10.1111/medu.14915

Abstract

Introduction: An understanding of social determinants of health (SDH) and patients' social circumstances is recommended to deliver contextualised care. However, the processes of patient care related to SDH in clinical settings have not been described in detail. Observable practice activities (OPAs) are a collection of learning objectives and activities that must be observed in daily practice and can be used to describe the precise processes for professionals to follow in specific situations (process OPA.) METHODS: We used a modified Delphi technique to generate expert consensus about the process OPA for patient care related to SDH in primary care settings. To reflect the opinions of various stakeholders, the expert panel comprised clinical professionals (physicians, nurses, public health nurses, social workers, pharmacists and medical clerks), residents, medical students, researchers (medical education, health care, sociology of marginalised people), support members for marginalised people and patients. The Delphi rounds were conducted online. In Round 1, a list of potentially important steps in the processes of care was distributed to panellists. The list was modified, and one new step was added. In Round 2, all steps were acknowledged with few modifications.

Results: Of 63 experts recruited, 61 participated, and all participants completed the Delphi rounds. A total of 14 observable steps were identified, which were divided into four components: communication, practice, maintenance and advocacy. The importance of ongoing patient-physician relationships and collaboration with professionals and stakeholders was emphasised for the whole process of care.

Discussion: This study presents the consensus of a variety of experts on the process OPA for patient care related to SDHs. Further research is warranted to investigate how this Communication-Practice-Maintenance-Advocacy framework could affect medical education, quality of patient care, and patient outcomes.

의학교육이 인공지능을 만나는 곳: 테크놀로지가 돌봄을 할 수 있을까? (Med Educ, 2020)
Where medical education meets artificial intelligence: ‘Does technology care?’
Anneke G. van der Niet1,2 | Alan Bleakley3

핵심 메시지
key message

인공지능은 의학의 모습을 바꿀 것이지만 '해결책주의'의 증상을 겪을 수 있습니다. 인공 지능이 돌봄의 관계가 치료의 일부가 될 수 있는 '수작업' 의학을 대체해서는 안 됩니다.
Artificial intelligence will change the face of medicine, but may suffer from symptoms of ‘solutionism.’ Artifical intelligence must not displace 'hands on' medicine, where caring relationship can be part of the cure.

기계는 공감과 연민과 같은 인간의 자질이 부족하기 때문에 환자는 인간 의사가 상담을 주도하고 있다는 인식을 가져야 합니다. 또한 불신에 가려진 기술인 AI를 환자들이 즉각적으로 신뢰하는 것은 기대할 수 없습니다.1
Machines lack human qualities such as empathy and compassion, and therefore, patients must perceive that consultations are being led by human doctors. Furthermore, patients cannot be expected to immediately trust AI; a technology shrouded by mistrust.1

1 소개
1 INTRODUCTION

16세기 베살리우스의 해부학 텍스트 이후 '기계로서의 신체'는 의학의 중심적이고 환원적인 은유가 되었으며, 특히 환자를 객관화하는 데 기여했습니다.2 기술화된 의학의 시대에 접어들면서 인공지능(AI)은 진단과 치료를 통해 구체화되는 의학을 쉽게 강화할 수 있게 되었습니다. 이러한 도구적 접근 방식은 신자유주의적 효율성의 가치에 기반한 '관리형' 의료와 일맥상통합니다. 점점 더 환원적이고 기술화되고 알고리즘적 사고의 지배를 받는 의학 교육은 특히 기계적 시뮬레이션을 통한 학습에서 두드러지게 나타납니다. 한편, 교육학은 인격 형성과 같은 커리큘럼의 더 넓은 가치를 무시한 채 겉으로 드러난 역량과 그 평가에 집중함으로써 '축소 포장'되고 있습니다. 
Since Vesalius’ 16th century anatomy texts, ‘the body as machine’ has become a central, reductive metaphor in medicine, serving in particular to objectify patients.2 As we enter the age of technologised medicine, artificial intelligence (AI) readily reinforces a medicine that disembodies through both diagnosis and treatment. Such an instrumental approach chimes with a ‘managed’ medicine grounded in values of neoliberal efficiency. There is a knock-on effect for a medical education that has become increasingly reductive, technologised and subject to algorithmic thinking, particularly evident in learning through mechanical simulation. Pedagogy, meanwhile, is ‘shrink-wrapped’ through focus on bare competencies and their assessment, ignoring wider value in the curriculum such as character formation.

최근(2019년 10월) 런던에서 열린 한 컨퍼런스에서는 글로벌 '업계 리더'들이 모여 최근 AI의 혁신에 대해 논의했습니다. '모범 사례' 중에는 영국 국민보건서비스(NHS)의 AI 활용 사례도 있었습니다.3 이 사례는 'AI는 복잡한 문제를 해결할 수 있는 잠재력을 가지고 있으며, 이를 통해 더 많은 시간을 확보하고 생산성 수준을 높일 수 있다'는 문제 해결의 틀을 가지고 있었습니다. 이 문장의 후반부는 환자 치료라는 '느린' 의학의 가치보다 생산성, 도구성, 효율성을 우선시하는 신자유주의 윤리를 반영합니다.4 후자는 '차가운' 기술 중심 의료보다 '따뜻한' 감각 기반 임상을 우선시하는 '체화된' 의학에 대한 관심 증가로 대표됩니다.5 이 글에서는 비감각적 AI 문제 해결의 순간과 '느린' 체화된 의학 사이에 생산적인 대화를 생성하는 방법을 질문함으로써 의학에서 AI를 문제화합니다. 도구주의와 수동적이고 포괄적인 승인을 넘어서는 AI의 잠재적 영향력을 탐구함으로써 의학교육의 '솔루션주의'에 관한 논의에 기여합니다. 
A recent (October 2019) conference in London brought together global ‘industry leaders’ to discuss recent innovations in AI. Amongst the ‘Best Practice Examples’ was AI use in the UK National Health Service (NHS).3 This was framed as a problem-solving exercise, where ‘AI has the potential to solve complex problems and in doing so, freeing up more time and increasing levels of productivity.’ The second half of the sentence reflects a neoliberal ethic, where output, instrumentality and efficiency are prioritised above the values of the ‘slow’ medicine of patient care.4 The latter is exemplified by a growth of interest in ‘embodied’ medicine that places a ‘warm’ sense-based clinical practice before a ‘cold’ technology-driven health care.5 In this article, we problematise AI in medicine by asking how we can generate productive dialogue between the instants of non-sensory AI problem-solving and ‘slow’ embodied medicine. By exploring the potential impact of AI beyond instrumentalism and passive, all-encompassing approval, we contribute to discussions concerning ‘solutionism’ in medical education.

1960년대 후반, 존 매카시는 '생각할 수 있는 기계'를 설명하기 위해 '인공 지능'이라는 용어를 만들었습니다.6 1980년대 중반에 AI는 의학 담론에 들어왔지만,7,8 비교적 최근에야 의학교육에 관여하기 시작했습니다.9 21세기 AI 시대는 컴퓨터 사용을 통해 증가하는 정보의 양을 관리하는 데 중점을 두었던 20세기 중후반의 정보화 시대를 넘어섰습니다. 이제 우리는 정보의 지능적 활용에 초점을 맞추고 있습니다.10 AI가 의학에 광범위하게 도입됨에 따라 '대규모 데이터 세트, 머신러닝 및 로봇과 관련된 지능 도구'11를, 인간 능력의 확장뿐만 아니라, 잠재적 자율 에이전트로서 배우기 위한 의학교육의 심각한 '리부팅'이 필요합니다. 이는 행위자 네트워크 이론(ANT)12과 같은 객체 지향 교육법과 '객체 지향 온톨로지'13와 같은 철학적 프레임워크에 기반하여 기술 자체를 객관화하지 않고 개념적으로 정교한 미래의 의료 교육을 형성할 것을 제안합니다. 이러한 새로운 지평은 이미 과부하로 인한 잠재적 소진에 시달리고 있는 수련의들에게 더 큰 부담을 주는 부담이 아니라 기회로 받아들여져야 합니다.14 우리는 AI의 포스트휴먼 요구와 체화된 반성적 의학 및 의학교육의 기초가 되는 시대를 초월한 민감한 인간적 지원을 결합해야 합니다.15 그러면 의학 교육은 두 가지 과제에 직면하게 됩니다. 

  • (a) 체화된 의학을 새로운 기술과 대화하게 하고
  • (b) ANT와 같은 객체 의도성을 수용하는 개념적 모델에 익숙해지는 

In the late 1960s, John McCarthy coined the term ‘artificial intelligence’ to describe ‘machines that can think.’6 By the mid-1980s, AI had entered medical discourse,7, 8 but only relatively recently has it engaged medical education, albeit mainly descriptively rather than critically.9 The AI age of the 21st century has eclipsed the previous age of information of the mid- to late 20th century that focused on management of increasing quantities of information through the use of computers. Now we are focused upon the application of information as intelligent use.10 The widespread introduction of AI into medicine requires a serious ‘reboot’ of medical education to learn about ‘intelligence tools involving large datasets, and machine learning and robots,’11 not just as extensions of human capacities but as potentially autonomous agents. This suggests drawing on object-oriented pedagogies such as actor-network theory (ANT)12 and on philosophical frameworks such as ‘object-oriented ontology13 to shape a conceptually sophisticated medical education for the future, where technologies themselves are not objectified. Such a new horizon must be framed as an opportunity, rather than a burden placing increasing pressure on trainee doctors already suffering from potential burnout through overload.14 We must blend the post-human demands of AI with timeless sensitive human support that is the basis of an embodied, reflexive medicine and medical education.15 Medical pedagogies then face a twin challenge:

  • (a) bringing embodied medicine into dialogue with new technologies, and
  • (b) gaining familiarity with conceptual models that embrace object intentionality such as ANT.

2 의료 분야의 AI
2 AI IN MEDICINE

의학적 의사 결정에서 인간의 사고를 모방하는 컴퓨터 기반 시스템은 1970년대 초 내과 진단 도구로 설계된 INTERNIST-I와 스탠포드 대학에서 개발한 감염을 일으키는 박테리아를 식별하고 체중에 따라 적절한 항생제 치료법을 추천할 수 있는 AI 시스템인 MYCIN이 등장한 이래로 사용되어 왔습니다. 환자가 증상을 컴퓨터에 설명하면 컴퓨터가 진단을 내립니다. 1970년대 중반부터 피츠버그 대학의 해리 피플은 진단 능력이 뛰어나기로 유명한 내과 의사 잭 마이어스 박사를 인터뷰하고 데이터를 수집했습니다. 10년이 넘는 기간 동안 피플은 혈액 매개 감염성 박테리아에 국한된 마이신의 범위를 확장하여 1000가지 질병을 진단할 수 있는 '의료 전문가 시스템' 또는 AI의 한 형태인 캐듀서스(CADUCEUS)를 개발했습니다. 현재 암 치료에는 최소 800여 가지의 약물이 사용되고 있으며, 이러한 치료법에 대한 연구 증거의 양은 방대합니다. 
Computer-based systems that emulate human thinking in medical decision-making have been around since the early 1970s, when INTERNIST-I was designed as a diagnostic tool in internal medicine, paralleled by MYCIN (developed at Stanford University) as an AI system that could both identify bacteria causing infections and recommend suitable antibiotic treatment with dosage keyed to body weight. The patient describes symptoms to the computer, which in turn offers a diagnosis. Starting in the mid-1970s, Harry Pople at the University of Pittsburgh interviewed and gathered data from Dr Jack Meyers, an experienced internal medicine doctor renowned for his diagnostic acumen. Over a period of more than 10 years, Pople expanded on the narrow focus of MYCIN on blood-borne infectious bacteria, to develop CADUCEUS, a ‘medical expert system’ or form of AI that could diagnose 1000 diseases. As a current example, the spectrum of current cancer treatments involves at least 800 different drugs and the volume of research evidence behind such treatments is enormous.

AI는 망원경과 현미경을 동시에 적용하여 대규모 데이터베이스를 활용하여 스마트하고 개인화된 정밀 치료를 가능하게 하며, Microsoft가 진행 중인 '프로젝트 하노버'가 그 예입니다.16 이러한 전문가 시스템은 저장된 지식(사실과 규칙 모두)을 기반으로 작동하여 추론을 실현하고, 알려진 사실에 규칙을 적용하여 새로운 사실과 설명까지 추론합니다. 대량의 데이터를 개인의 편견과 편견의 오염 없이 빠르게 처리할 수 있습니다.17 그러나 여기에는 인공지능의 내부 모순이 있습니다. 

  • '폐쇄적' 연산이기 때문에 '객관적'으로만 작동할 수 있으며, 입력에서 출력까지 경로를 오염시키는 역설적인 요소가 존재하지 않는다는 것입니다. 

물론 이러한 선형 시스템은 인간의 피드백과 재설계를 통해 조정할 수 있습니다. 반면 인간은 여러 개의 경쟁적인 피드백 루프가 있는 '개방형' 시스템(복잡하고 역동적이며 적응적)이며, 많은 인지적 편향과 정서적 오염을 보입니다. 그러나 불확실성에서 비롯된 이러한 색채가 바로 의료 상담에 의미를 부여하거나 진심을 다할 수 있는 이유입니다. 연구에 따르면 환자들은 의사가 세심하고 공감적으로 경청할 뿐만 아니라 적절하게 접촉하는 '따뜻한' 의술을 원합니다.5 또한 은유의 광범위한 사용은 이러한 의료적 만남에 언어적 '신체'를 더하는 반면,2 아직까지 AI는 은유를 잠재적 오염으로 간주합니다.18 이는 우리가 AI 시대로 더 깊이 들어가면서 의학 교육도 감각을 교육하고 임상 업무에서 은유를 능숙하게 사용하는 데 중점을 두어야 한다는 것을 시사합니다.
AI allows simultaneous application of a telescope and microscope, harnessing large databases for smart, personalised precision treatments, exemplified in Microsoft's ongoing ‘Project Hanover.’16 Such expert systems work on the basis of stored knowledge (both facts and rules) realising inferences, where rules are applied to known facts to deduce new facts and even explanations. Large quantities of data can be dealt with rapidly without contamination from personal bias and prejudice.17 Here, however, is AI’s internal contradiction:

  • it can only function ‘objectively’ because it is a ‘closed’ operation; there are no paradoxical elements contaminating the route from input to output.

Such a linear system can of course be adapted through human feedback and redesign. Humans, by contrast, are ‘open’ systems (complex, dynamic and adaptive) with multiple, competing feedback loops, and show many cognitive biases and emotional contaminations. But it is precisely in such colourations arising from uncertainty that medical consultations may gain meaning, or are given heart. Research shows that patients want an embodied ‘warm’ medicine where doctors not only listen attentively and empathically, but also touch appropriately.5 Further, the widespread use of metaphor adds linguistic ‘body’ to such medical encounters,2 whereas AI, as yet, spurns metaphor as potential contamination.18 This suggests that as we move deeper into the AI age, medical education must also focus on educating the senses and on fluency in use of metaphor in clinical work.

우리는 AI의 사용 증가가 '해결책주의'의 형태로 의학과 의학교육에 가져올 수 있는 오만함을 인식해야 하며, 이는 교육적, 윤리적 함의를 모두 가져올 수 있습니다. 최근 파이낸셜 타임즈에서 유전자 편집 기술인 크리스퍼 유전자가위(CRISPR)의 공동 발견자인 제니퍼 두드나와의 인터뷰는 똑똑한 AI '해결책주의'가 어떻게 다양하고 복잡한 문제를 가릴 수 있는지 보여줍니다.19

  • 첫째, 과학적 오만: 크리스퍼 유전자 편집 기술은 복잡성과 잠재적 위험을 가리는 은유를 통해 설명됩니다. 이 기술은 '유전자를 자르거나' '가위처럼 자를 수 있는' '강력한 기술'입니다.
  • 둘째, 교육적 도구주의: 위 크리스퍼의 은유에 담긴 공학적 사고방식을 반영하듯, 의학교육은 복잡한 학습을 평가 목적으로 관찰 가능한 도구적 결과로 축소하는 경향이 있는 '역량' 운동의 게으른 언어와 관행에 제대로 도전하지 못했습니다. 이는 '가능성 학습'과 혁신이 '안정을 위한 학습'이라는 안전지대로의 후퇴로 대체되는, 의도적이고 상상력이 풍부한 역량 교육학을 잠재적으로 속일 수 있습니다.

We should be aware of the creeping hubris that increasing use of AI may bring to medicine and medical education in the form of ‘solutionism,’ introducing both pedagogical and ethical implications. A recent interview in the Financial Times with Jennifer Doudna, the co-discoverer of clustered regularly interspaced short palindromic repeats (CRISPR), the gene-editing technology, shows how smart AI ‘solutionism’ can mask a range of complex issues.19 

  • First, scientific hubris: Crispr gene-editing technology is described through metaphors that mask complexity and potential dangers. It is a ‘powerful technology’ that can ‘snip out genes’ or ‘cut like scissors.’
  • Second, pedagogical instrumentalism: mirroring the engineering mentality of Crispr's metaphors above, medical education has simply failed to properly challenge the lazy language and practices of the ‘competence’ movement that tend to reduce complex learning to instrumental, observable outcomes for assessment purposes. This potentially cheats a purposeful and imaginative pedagogy of capabilities, where ‘possibility learning and innovation are replaced by a retreat to the safe ground of ‘learning for stability.’

 

  • 셋째, 매우 중요한 것은 인공지능이 중심이 되는 의학 및 관련 의학교육의 미래에 대한 윤리적 함의가 복잡하고 긴급하며, 종종 상황에 따라 달라진다는 점입니다. 예를 들어, 중국 과학자 허젠쿠이는 크리스퍼 편집 쌍둥이의 탄생을 발표했을 때 이 위험한 행동으로 인해 3년의 징역형과 거액의 벌금을 선고받고 과학자 경력을 끝내야 했습니다. 물론 이것은 수사학적으로 사용된 극단적인 예입니다. 하지만 예를 들어 의학교육 시뮬레이션에서 SimMan과 같은 복잡한 마네킹이 널리 사용되고 있다고 가정하면, 이러한 시뮬레이션 작업이 남성 성별 모델에 국한된다는 윤리적 수수께끼는 어떻게 설명할 수 있을까요?
  • Third, and extremely important, the ethical implications of an AI-heavy medicine and associated medical education future are complex and emergent, often context-specific. For example, when the Chinese scientist He Jiankui announced the birth of Crispr-edited twins, this rogue action led to a jail sentence of 3 years, a hefty fine and the end of a scientific career. This of course is an extreme example used rhetorically. Yet, if we tone this down to, for example, the widespread use of complex manikins such as SimMan in medical education simulations, what do we make of the ethical conundrum that such simulation work is limited to a male-gendered model?

3 인간이 컴퓨터가 되는 순간
3 WHERE THE HUMAN BECOMES COMPUTER

일반적으로 우리는 컴퓨터를 인간처럼 만들기 위해 노력하지만, 어떤 사람들은 인간이 컴퓨터처럼 되기를 원합니다. 에드워드 프렌켈의 사랑과 수학: 숨겨진 현실의 핵심은 1980년대에 수학 가정교사이자 멘토였던 야코프 이사에비치가 모스크바 병원의 선임 비뇨기과 의사들과 함께 일하면서 그들이 진단 및 의사 결정 전문 지식을 합리화하도록 도왔던 이야기를 담고 있습니다.20 컴퓨터의 도움 없이 이사에비치는 의사들이 구두로 사고 전략을 연습하면 이사에비치가 이를 간소화하여 의사들에게 다시 반영하여 고려하도록 하는 'thinking aloud' 연구를 사용했습니다. Isaevich는 의사들의 진단 의사 결정 과정 중 85%~90%가 중복적이며, 나머지 핵심적인 10%~15%는 의사 결정 트리로 명확하게(명료화 및 매핑) 만들 수 있다는 결론을 내렸습니다. 인간의 직관, 순수한 알고리즘 논리, 발명이 겹치는 영역에서 일하면서 Isaevich는 AI를 임상 업무에 적용하기 위한 템플릿을 공식화했지만, 기계가 개입하지 않았다는 점을 제외하면 말이죠. 오히려 논리학자 자신이 기계가 되었습니다. 
Where we generally strive to make computers more like humans, some would like humans to be more like computers. Edward Frenkel's Love and Math: The Heart of Hidden Reality tells the story of how his mathematics tutor and mentor Yakov Isaevich had been working in the 1980s with senior urologists in a Moscow hospital, helping them to rationalise their diagnostic and wider decision-making expertise.20 Without the help of computers, Isaevich employed ‘thinking aloud’ research, where doctors would verbally rehearse thinking strategies and Isaevich would streamline these, mirroring them back to the doctors for consideration. Isaevich concluded that 85%-90% of the doctors’ diagnostic decision-making processes were redundant, and the remaining key 10%-15% could be made plain (articulated and mapped) as decision trees. Working in the territory of overlap between human intuition, pure algorithmic logic and invention, Isaevich formulated a template for the application of AI to clinical work, except that, again, no machine was involved. Rather, the logician was himself becoming the machine.

이사예비치는 학부 마지막 학기였던 프렌켈에게 이사예비치의 의료 추론 프로젝트에 함께 일해 달라고 요청했습니다. 그는 프렌켈을 신장 이식 시 발생하는 문제에 대한 박사 학위 논문을 위해 데이터를 수집하고 있던 젊은 비뇨기과 의사 세르게이 아루티유니안과 짝을 이루게 했습니다. 환자의 면역 체계가 이식된 신장을 거부하는 경우, 의사와 외과의는 그 자리에서 신장을 제거할지 아니면 사투를 벌일지 결정해야 합니다. 신장을 유지하면 환자가 사망할 수도 있지만, 제거하면 환자는 다시 원점으로 돌아가 또 다른 기증을 받아야 합니다. 아루투니안은 초음파 판독을 통해 수집한 환자 집단의 정량적 데이터를 바탕으로 40가지의 다양한 매개변수(의사의 의사 결정은 보조 기술에 의해 보강되기도 하고 압도되기도 함)를 바탕으로 이 딜레마에 대한 직관적이지 않고 논리적인 해답을 찾고자 했습니다. 우아하고 최소한의 의사 결정을 위한 객관적인 기준이 마련될 수 있을까요? 
Isaevich invited Frenkel, now in his final undergraduate year, to work with Isaevich's medical reasoning project. He paired Frenkel with a young urologist, Sergei Arutyunyan, who was collecting data for a PhD thesis on problems encountered in transplanting kidneys. Where patients’ immune systems reject transplanted kidneys, doctors and surgeons must make on-the-spot decisions about removing the kidney or fighting for it. Keeping the kidney could lead to the patient's death, but removal puts the patient back to square one, needing another donation. Arutyunyan wanted a logical, and not intuitive, answer to this dilemma based on quantitative data from patient populations gathered through ultrasound readings yielding a wieldy 40 different parameters (the doctor's decision making then both augmented and overwhelmed by the supplemental technology). Could a set of objective criteria be established for elegant, minimalist decision making?

흥미롭게도 프렌켈이 이룬 의료 AI의 도약은 성과가 있었습니다. 순수하게 수학자로 교육을 받은 프렌켈은 의사 아루투니안과 역할을 바꿔, 구현된 기계로서 이 좁은 전문 영역에서 의사의 의사 결정 전문성을 모방하고 심지어 향상시킬 수 있는지 확인했습니다. 프렌켈은 아루투니안이 복잡한 데이터를 가지고 있는 270명의 환자 각각이 고유한 사례를 제시했지만, 환자들에게도 주요 공통 속성과 문제가 있음을 발견했습니다. 그는 30명의 환자 데이터를 무작위로 선택하고 패턴을 찾아 이식 신장이 거부반응을 일으킬 가능성이 있는 경우 의사 결정을 내릴 때 일반화할 수 있는 일련의 규칙을 조합하려고 했습니다. 
Interestingly, the medicine AI leap that Frenkel made was performative. Although educated purely as a mathematician, he reversed roles with the doctor Arutyunyan to see if, as an embodied machine, he could imitate, even enhance, the doctor's decision-making expertise in this narrowly defined area of expertise. Although each of 270 patients for whom Arutyunyan had complex data presented a unique case, Frenkel saw that patients also displayed key common attributes and issues. He tried to piece together a generalisable set of rules for decision making in the case of potential rejection of transplanted kidneys by randomly choosing data for 30 patients and looking for patterns.

이를 통해 그는 아루티유니안이 모델링한 진단 추론 프로세스에 들어가 일반적인 규칙을 이해했습니다. 프렌켈은 의사 의자에 앉았을 뿐만 아니라 환자의 기록을 참조할 때도 아루티유니안이 환자에 대한 질문을 하면 프렌켈이 파일을 참조하여 대답하는 방식으로 환자의 마음속으로 들어갔습니다. 이런 식으로 프렌켈은 퍼즐을 완성하는 것이 아니라 풀어나가는 것처럼 질문하는 패턴을 고안해냈습니다. 그는 실제로 서너 가지의 핵심적인 임상 해부학적 질문만으로 진단에 도달할 수 있으며, 이를 의사 결정 트리로 쉽게 매핑할 수 있다는 사실에 주목했습니다. 프로세스 규칙을 학습한 프렌켈 역시 정확한 진단을 내릴 수 있었고, 말콤 글래드웰이 제안한 집중적인 연습을 통해 전문 지식을 습득하는 '1만 시간'의 법칙에 도전할 수 있었습니다.21 프렌켈은 지능형 기계처럼 불확실성을 수용해야 하는 실시간 의료 상담이나 주요 의료-수술 팀 기반 의사 결정의 '뜨거운' 실제 상황과는 완전히 분리된 작동 규칙을 '차갑게' 학습한 상태였습니다. 프렌켈은 이전에는 초음파 진단에서 확인되지 않았던 '간단한 알고리즘'을 발견했는데, 이 알고리즘은 소수를 제외한 모든 경우에 적합했습니다. 
Thus, he got inside the diagnostic reasoning process that Arutyunyan modelled and made sense of its generic rules. Frenkel not only sat in the doctor's chair but also in his mind, when consulting patients’ records as Arutyunyan asked him questions about these patients that Frenkel would then answer by consulting their files. In this way, as he describes it himself, Frenkel conceived a pattern of questioning as if he were undoing, rather than completing, a jigsaw. He noted that Arutyunyan could in fact arrive at a diagnosis after only three or four key clinical-anatomical questions and that these could be readily mapped as a decision tree. Having learned the process rules, Frenkel too could make a correct diagnosis, challenging the ‘10 000 hours’ rule of acquiring expertise through focused practice as suggested by Malcolm Gladwell.21 Frenkel had, as intelligent machines do, learned ‘cold’ the rules of operation, completely divorced from the ‘hot’ flesh-and-blood contexts of live medical consultations, or key medico-surgical team-based decisions, that must embrace uncertainty. Frenkel had discovered a ‘simple algorithm’ at play (that fitted all but a small number of cases) that, he claimed, had previously not been identified in ultrasound diagnostics.

컴퓨터의 도움 없이 의사의 역할을 수행한 프렌켈은 '하루 동안의 의사'이자 '기계 지능'이 되었으며, AI형 인지가 의사의 업무를 보완하는 것이 아니라 대체하는 것이라는 프레임을 만들었습니다. 하지만 기술 프레임워크에 의해 실제 의술이 강화되는 것이 아니라 대체되는 것을 환영해야 할까요? 체화된 의학과 체화되지 않은 인공지능 사이의 모순된 공간에서 점점 더 많은 일을 하게 될 미래 의사들의 정체성은 어떤 식으로 형성될까요? 다시 말하지만, 

  • 인간의 전문성은 모호성을 용인하도록 설계된 개방적이고 역동적이며 복잡한 시스템의 산물이지만,
  • 전문가 시스템은 폐쇄적이고 피드백을 통해 반복되며 모호성을 용납하지 않는 선형적입니다. 

전문가 시스템은 선형적이고 계층적인 명령과 통제 프로세스를 따르기 때문에 권위적일 수밖에 없으며, 동적인 문제에 대해 안정적인 해결책을 제시하기도 합니다.
Acting in the role of the computer but without the help of a computer, Frenkel became ‘doctor for a day’ and ‘machine intelligent,’ framing AI-type cognition as not just supplementing but supplanting the doctor's work. But do we welcome this displacement, rather than enhancement, of hands-on medicine by a technological framework? In what sense might this shape doctors’ identities in the future as they increasingly work in a contradictory space between embodied medicine and disembodied AI? Again,

  • although human expertise is the product of an open, dynamic, complex system designed to tolerate ambiguity,
  • expert systems are linear: closed; looped through feedback, and intolerant of ambiguity.

They are inevitably authoritarian as they conform to linear, hierarchical processes of command and control, also suggesting stable solutions to problems that are dynamic.

4 계산의 지배력 증가: 의료 도구에서 디지털 도구까지
4 THE INCREASING DOMINANCE OF CALCULATION: FROM MEDICAL TOOLS TO DIGITAL TOOLS

육체가 있고 구체화된 실습 의학과 육체가 없고 구체화되지 않은 기계가 주도하는 의학을 분리하는 것은 기술의 가치를 깎아내리기 위한 수사학적 장치(거짓 반대)로 볼 수 있습니다. 결국 기술은 인간의 감각, 존재론적 대상13, 인간의 '기계적 집합체'의 연장이 아닐까요?"22, 23 우리가 우려하는 것은 기계 시뮬레이션과 같은 기술이 인간 행위자로부터 분리되어 거품 속에서 작동하고, 실습이 실습 검사를 대체하기 시작할 수 있다는 점입니다. 과연 기술에 대한 매혹과 그에 따른 과잉 검사는 의학계에서 치료해야 할 증상일까요?"24 이 분야에 대해 간략히 살펴봅시다.
Our separation of a fleshly, embodied hands-on medicine and a fleshless, disembodied machine-led medicine could be seen as a rhetorical device (a false opposition) to discount the value of technologies. After all, are technologies not extensions of the human senses, ontological objects13 and humans’ ‘machinic assemblages?’22, 23 Our fear is that technologies, like mechanical simulations, can become disengaged from human actors to operate in a bubble, where hands-off testing begins to be a substitute for hands-on examinations. Indeed, is fascination with technology and consequent over-testing a symptom in medicine that should be treated?24 Let us briefly review the field.

로봇 수술과 같은 새로운 형태의 '하드' 의료 기술25과 앞서 논의한 유전자 편집에 사용되는 크리스퍼 염기서열과 같은 '소프트' 의료 기술19, 26은 계속해서 빠른 속도로 의료 행위를 재편하고 있습니다.27 컴퓨터화된 기기는 의료 서비스 제공에서 인간을 빠르게 대체하고 있습니다. 마취는 이제 의사가 마취를 유도할 때를 제외하고는 환자의 육체를 거의 보지 않을 정도로 기계에 의존하고 있습니다. 또한 한때 인간 기술자가 수행하던 다양한 실험실 테스트가 컴퓨터화된 기계에 의해 일상적으로 수행되고 있으며,28 일부 수술 절차에서는 로봇과 같은 자동화 기술이 외과의의 모습을 보완했다가 대체하기도 합니다. 
New forms of ‘hard’ medical technologies such as robotic surgery25 and ‘soft’ ones such as Crispr sequences used to edit genes, discussed earlier,19, 26 continue to reshape medical practice at an accelerating pace.27 Computerised devices are rapidly replacing human beings in the delivery of medical services. Anaesthesia is now so machine dependent that practitioners rarely look at the patient in the flesh, other than at induction. In addition, a wide variety of laboratory tests that were once conducted by human technicians have long been routinely performed by computerised machines,28 and forms of automation such as robotics have in some surgical procedures come to augment and then displace the figure of the surgeon.

실제로 최근 출시된 'GP at Hand' 앱과 같은 자가 진단 AI가 성공을 거두면 의사들은 일자리를 잃게 될 것입니다.29 이 기술은 챗봇 형태의 AI를 사용하여 사용자가 질문 메시지에 따라 증상을 입력하면 확률을 통해 감별 진단을 내리고 다음 단계를 제안합니다. 신약 개발과 같은 임상시험에 준하는 규제와 절차가 없기 때문에 이러한 종류의 애플리케이션에 대한 윤리적 문제가 해결되지 않은 채로 남아있습니다. 또한 데이터 보호, 제3자 개입, 잘못된 코딩(백인 인종에 대한 알고리즘 편향성 포함)도 이러한 개발로 인해 발생하는 문제입니다. 진단을 지원하기 위해 기계와 컴퓨터를 사용한다는 것은 행동, 감각 및 상호 작용을 모두 숫자 또는 픽셀 단위의 기술 형식에 맞게 계산해야 한다는 것을 의미합니다. 결과적으로 건강과 신원 모두 점점 더 계산적인 관점에서 접근하게 될 것이며, 20년 전 굿맨의 예측이 현실화되고 있습니다: '의료 전문직의 미래는 계산적이다."28 이러한 도구주의와 효율성은 건강을 '관리'하는 앞서 설명한 신자유주의적 가치 체계로 우리를 되돌려 놓습니다. 또한 계산적 프레임워크는 의료 교육학을 형성할 것이며, 이러한 교육학은 말 그대로 도구적 교육이 될 것이며, 인도적이고 윤리적 치료를 배양하는 전통적인 병상 참여에서 요구되는 다양한 감각적 능력보다는 주로 학생들의 시각적 기반 '판독'에 대한 피드백을 요구하게 될 것입니다. 
Indeed, doctors themselves will be threatened with redundancy should self-diagnosing AI such as the recent app ‘GP at Hand’ prove successful!29 This technology uses AI in the form of a chatbot, where users write their symptoms according to question prompts, resulting in a differential diagnosis with probabilities and suggested next steps. Without regulation and the equivalent of clinical trials (as you would expect for, say, a new pharmaceutical) this leaves ethical questions unanswered for this kind of application. Further, data protection, third party intervention and poor coding (including algorithmic bias towards white ethnic groups) are all problems arising from such developments. Employing machines and computers to assist in diagnosing means that actions, sensations and interactions all have to be calculated in order to fit the technological form, which is in digits or pixels. Consequently, both health and identities are increasingly framed in computational terms, realising Goodman's prediction from two decades ago that: ‘the future of the health professions is computational.’28 Such instrumentalism and efficiency returns us to the neoliberal values framework described earlier, where health is ‘managed.’ Further, calculative frameworks will shape medical pedagogy, such pedagogy in turn becoming literally instrumental, requiring mostly feedback on students’ optically based ‘readouts’ rather than a range of sensory capabilities such as those required in traditional bedside engagement that cultivates humane, ethical care.

5 중재 및 의미 만들기
5 MEDIATION AND MEANING MAKING

의료에 새로운 기술이 도입될 때마다 의사는 고유한 맥락에서 이를 어떻게 이해할지 고민해야 합니다.30 의료 기술을 매개하는 것은 '확장된 인지'를 가진 '확장된 인간'으로 초점을 이동시키고 있습니다.31 매개체는 변형하고 번역할 뿐만 아니라 전달해야 할 요소의 의미를 잠재적으로 왜곡하고 수정합니다.32 기술 혁신이 의사의 감각을 확장하고 조절하는 의학에서 감각 비율, 지각 유형 또는 지각 패턴을 변경하는 등의 효과는 광범위하게 조사되지 않았습니다. 의료 정보 기술이 의료 과실을 줄여 환자 안전을 향상시킬 것이라는 약속은 아직 완전히 실현되지 않았습니다.33, 34
Whenever new technologies are introduced into health care, doctors have to figure out how to make sense of them in unique contexts,30 where mediating medical technologies are shifting the focus to the ‘extended human’ with ‘extended cognition.’31 Mediators transform and translate, but also potentially distort and modify the meaning of the elements they are supposed to carry.32 In medicine, where technological innovations extend and modulate the doctor's senses, the effects, such as altering sense ratios, type of perception or patterns of perception, have not been extensively investigated. The promise that health information technologies will increase patient safety by decreasing medical error has yet to be fully realised.33, 34

AI는 또한 자아의 기술을 구성하여 이를 사용하는 의사의 정체성과 성격을 적극적으로 형성하며,35 여기서 정체성은 수행 능력, 즉 자신이 하는 것입니다.36 로봇 수술이 대표적인 예로, 최악의 시나리오는 개복 수술을 수행할 줄 아는 외과의가 감소하는 것입니다.36 기술 사용의 변화로 인해 혈관 방사선 전문의가 혈관 외과 전문의의 내부자가 되는 등 새로운 기술의 도입에는 다양한 내부자 및 외부자 역학 관계가 수반됩니다.37 [새로운 기술의 도입]은 [새로운 언어의 채택]과 누가 (하위) 전문 분야에 포함되거나 제외되는지에 대한 [새로운 관할권 주장]으로 이어져 정체성에 혼란을 야기합니다. 
AI also constitutes technologies of the self, actively shaping the identities and characters of the doctors that make use of them,35 where identity is performance, something that one does.36 A key example is robotic surgery, where a worst-case scenario is a decline in surgeons knowing how to perform an open surgery.36 A variety of insider and outsider dynamics come with the uptake of new technologies, such as vascular radiologists becoming insiders to the vascular surgeons due to a change in use of technology.37 The introduction of new technologies leads to the adoption of new languages and new jurisdictional claims about who is included in or excluded from the (sub)specialty, confounding identities.

그러나 기술은 사용 방식에 따라 그 의미와 궁극적인 기능이 달라지는 [관계적인 것]으로 볼 수 있습니다.38 의료 서비스에서 컴퓨팅 시스템의 효과는 정확성과 성능뿐만 아니라 사용자의 수용도, 사용 맥락, 임상적 만남을 포함한 사회적 및 직업적 상호 작용에 대한 결과도 분석해야 합니다.39 기술이 종종 의사와 환자 간의 거리를 넓히기 때문입니다. 환자 침대 옆에 있는 컴퓨터 주위에 모여 있는 의사나 환자의 증상보다는 검사 결과 등 환자에 대한 정보를 얻기 위해 컴퓨터 화면을 바라보는 가정의의 많은 예처럼,40 기술은 임상의의 시선을 환자의 신체에서 기기로 향하게 할 수 있습니다. 아브라함 베르게스와 동료들은 '차트 선반 주변과 방사선실에서 모든 전문의가 모여 환자에 대해 논의하던 중요한 사회적 의식의 상실'을 통해 EMR 구축의 의도하지 않은 결과를 지적합니다.41 
However, technologies can be seen as relational: only getting their meanings and ultimate functions in the ways that they are put to use.38 The effects of computing systems in health care are subject to analysis not only of accuracy and performance but also of acceptance by users, of the context of use, and of consequences for social and professional interaction, including the clinical encounter, as technologies often distance doctors from their patients.39 Technologies can direct the clinician's gaze away from the patient's body towards the device,40 like the many examples of doctors gathering around a computer next to the bed of the patient, or the family practitioner who looks at the computer screen for information about the patient, such as test results, rather than at the patient's presenting symptoms. Abraham Verghese and colleagues point to the unintended consequences of establishing the electronic medical record through ‘loss of important social rituals … around the chart rack and in the radiology suite, where all specialties converged to discuss patients.41

6 선물 상자를 열었지만 블랙박스만 발견하다
6 OPENING THE GIFT BOX ONLY TO FIND A BLACK BOX

인공지능은 우리에게 거절할 수 없는 선물을 선사할까요? 대부분의 기술 혁신은 의사와 환자 사이를 매개하지만, AI는 그 이상의 역할을 합니다. 존 매카시의 '생각할 수 있는 기계'에서처럼 예측을 합니다.6 대부분의 AI 시스템에서는 입력과 출력이 모두 소프트웨어 프로그램에 제공됩니다. 전문가가 AI 프로그램이 올바른 연결을 할 수 있도록 지원하는 훈련 기간 동안 AI가 입력과 출력 사이의 패턴을 인식한 후에는 입력만으로 소프트웨어가 개발된 알고리즘에 따라 출력을 제공할 수 있습니다. 예를 들어, 피부과에서 피부암을 인식하는 연구에서 다양한 디지털 사진과 해당 질병 라벨을 각각 입력과 출력으로 제공했습니다.42 훈련 기간이 지나자 AI 소프트웨어는 새로운 피부암 이미지를 기반으로 특정 질병을 진단할 수 있었습니다. 이제 AI는 질병을 인식하는 방법을 '학습'하여 숙련된 안과 의사의 망막 이미지 진단이 AI보다 정확도가 떨어지는 등 숙련된 의사와 동일하거나 더 높은 수준까지 매우 빠르고 정확하게 진단할 수 있게 되었습니다.43 이러한 형태의 AI인 머신러닝은 영상의학, 병리학 및 종양학에서 광범위하게 적용됩니다.44 임상적으로 관련된 출력 데이터를 생성하려면 AI는 정의된 샘플 데이터베이스에 대한 이전 학습 기록의 맥락에서 처리된 많은 환자의 입력 데이터가 필요합니다.45 이러한 유형의 AI는 의사 결정을 내리는 것으로 가정된 특정 신경 아키텍처를 수학적으로 표현한 것이므로 인공 신경망(ANN)을 생성한다고도 합니다. 이러한 유형의 AI는 중환자실과 정신과에서 환자의 특정 위험이나 약물 부작용을 예측하는 데 사용되었습니다. 
Surely AI offers a gift we cannot refuse? Although most technical innovations mediate between the doctor and patient, AI does something more than that: it predicts, as in John McCarthy's ‘machines that can think.’6 In most AI systems, both input and output are provided to a software program. After AI has recognised a pattern between input and output during a training period in which the expert assists the AI program in making the right connection, only the input will be sufficient for the software to provide an output based on its developed algorithm. For example, in a study on recognising skin cancer in dermatology, many different digital pictures and their corresponding disease labels were provided as input and output, respectively.42 After a training period, the AI software was able to diagnose the specific disease based on a new image of skin cancer. AI has now ‘learned’ to recognise diseases, and is able to do so very fast and accurately, up to the same, or an even higher, level as an experienced doctor, such as experienced ophthalmologists’ diagnoses of retinal images being less accurate than those of their AI counterpart.43 This form of AI, machine learning, is applied widely in radiology, pathology and oncology.44 To produce clinically relevant output data, AI needs input data from many patients, processed in the context of previous training history on a defined sample database.45 As this type of AI is a mathematical representation of the particular neural architecture assumed to make decisions, it is often said to produce an artificial neural network (ANN). This type of AI has been used to predict certain risks or adverse reactions to drugs for patients in both intensive care and psychiatry.

이 모든 것은 문제가 없는 것처럼 보입니다. 하지만 몇 가지 한계가 있습니다. 앞서 언급했듯이 AI가 사용하는 데이터 세트의 기본 입력은 지금까지 숫자 또는 디지털 픽셀 단위인 AI 소프트웨어 프로그램의 언어에 맞아야 합니다. 즉, 의사는 이 언어를 준수하는 입력으로 기억상실증의 측면만 제공할 수 있으므로 AI의 일부가 될 수 있는 항목에 제한이 있습니다. 다른 측면은 무시되거나(예: 터치) 컴퓨터의 '생각' 형태에 맞게 조작(예: 텍스트)됩니다. 의사가 물리적으로 경험하는 행동과 감각은 이제 계산되어야 합니다. 
This all seems to be non-problematic. However, there are several limitations: as mentioned earlier, the basic input for the datasets that AI uses should fit the language of the AI software program, which is so far in digits or digital pixels. This means that a doctor can only provide those aspects of the anamnesis as input that complies with this language, thus placing a limitation on what can become part of AI. Other aspects are either ignored (eg, touch) or manipulated (eg, text) in such a way that they fit to the form of ‘thought’ of a computer. Actions and sensations that are experienced physically by a doctor now have to be calculated.

결국 의료 교육도 계산화될 것입니다. 이에 대한 글은 이미 벽에 써져 있습니다. 혁신적인 사회물질적, 사회문화적 모델이 의학교육에 도입된 후 초기의 흥분 이후, 역량 운동과 '위탁 가능한 전문 활동'(EPA)은 의학교육에 대한 새로운 믿음을 대표하며, [비판적 학습은 예측 가능한 결과의 '고지'가 아니라 불확실성, 독특성, 가치 갈등(과정으로서의)의 '늪지대'에서 발생한다]는 도널드 쇤의 주요 통찰을 버린 것처럼 보입니다.46 
In turn, perhaps medical pedagogies too will become calculated. The writing is on the wall for this. After an initial flush of excitement after innovative sociomaterial and sociocultural models were introduced to medical education, the competence movement and ‘entrustable professional activities’ (EPAs) represent the new faith in medical education, seemingly abandoning the major insight of Donald Schön, that critical learning occurs in the ‘swampy lowland’ of uncertainty, uniqueness and value conflict (as process), rather than the ‘high ground’ of predictable outcomes.46

이런 식으로 전문가와 환자의 신체, 다른 의료진, 도구와의 임상적 관계는 파편화되고 기술적인 형태로 재구성됩니다. 이는 질병의 복잡성을 줄일 수 있지만, 의사의 전문성을 제약하는 측면이 더 큽니다. 이러한 전문성은 상당 부분 암묵적이거나 암묵적(예: '눈 깜빡임' 패턴 인식 또는 유형 1 추론 사용)이며, 구현된 많은 기술은 모델링에 접근하기 어렵습니다.47 이는 비뇨기과 의사들이 '소리 내어 생각하기'를 통해 대부분 은밀한 의사 결정 과정을 언어화(따라서 합리화)하도록 도왔던 프렌켈의 도전이기도 했습니다. 또한 AI가 의학의 모호성을 줄이기 위한 수단으로 여겨질 수 있지만, 전문성을 통해 발달한 직관이나 '인지 무의식'48의 중요성을 거부함으로써 불확실성을 증가시키는 의도하지 않은 결과를 초래할 수 있습니다. 다시 말하지만, 이러한 중재는 환자의 생생한 경험에 대한 현상학적 '파악'인 구체화된 실천의 잠재적 손실을 수반합니다.15 입력 경계와 관련된 또 다른 한계는 '가치 정렬'로, 이는 AI가 인간의 목표와 가치에 따라 행동하도록 학습할 수 있는 방법을 다룹니다. 옳고 그름을 구분하기 위해서는 일련의 가치가 필요한 경우가 많습니다. 숙련된 의사는 특정 환자의 목표에 따라 어떤 치료법을 다른 치료법보다 더 중요하게 생각해야 할 때를 알고 있습니다. 이러한 유형의 목표는 아직까지 AI에서 설정할 수 없지만, 설령 가능하다고 하더라도 시스템이 어떻게 이해하고 도덕화할 수 있을지는 알 수 없습니다. 
In this way, the clinical relation of the expert to patients’ bodies, to other practitioners and to tools becomes fragmented and reconstructed in technological forms. This might reduce the complexity of a disease, but more, it constrains the expertise of a doctor. Such expertise is to a great extent implicit or tacit (eg, in employing ‘blink-of-an-eye’ pattern recognition or Type 1 reasoning) and many embodied techniques are inaccessible to modelling.47 This was also the challenge of Frenkel, who helped the urologists to verbalise (and hence rationalise) their mostly covert decision-making process by making them ‘think aloud,’ a contested way of trying to get inside doctors’ minds. In addition, where AI might be seen as a means to reduce ambiguity in medicine, it can have unintended consequences of increasing uncertainty by refusing the importance of intuition or the ‘cognitive unconscious48 developed through expertise. Again, the mediation thus comes with a potential loss of embodied practice, the phenomenological ‘grasp’ of the patient's lived experience.15 Another limitation related to the input boundary is ‘value alignment’; this deals with how AI can be taught to act in accordance with human goals and values. Often, a set of values is necessary to tell right from wrong. An experienced doctor knows when to value one treatment over another, given the goal for this particular patient. These types of goals cannot be set in AI so far, but even when this will be possible, it is unknown how (and how we know that) the system understands and can moralise.

AI의 출력이 뉘앙스의 여지가 없는 참/거짓 유형의 목표로 제한되는 경우, 이는 제어와 관련된 또 다른 한계에 직면하게 됩니다. 앞서 언급했듯이 지금까지 AI에는 피드백이 불가능했습니다. 소프트웨어가 실행되면 목표를 변경할 수 없습니다. 루프가 닫힙니다. 또한 어떤 경로가 사용되어 특정 결과가 나왔는지, 신경망의 특정 계층 사이에 왜 특정 가중치가 있었는지 역추적하기가 어렵습니다. 의료 현장에서 의사들은 환자에 대한 진단을 확인하기 위해 서로에게 자주 물어보며 일반적인 규칙에 맞지 않을 수 있는 특이한 사례의 구체적인 세부 사항을 추론할 여지를 제공하지만, 컴퓨터에게 특정 진단에 도달한 방법과 이유를 설명해 달라고 요청하기는 어렵습니다. 가장 중요한 것은, 환자와의 의도적인 상호 작용이 없으며, 정보로 축소된다는 점입니다. 성능의 정확성과 효율성(선물 상자)에도 불구하고 AI는 종종 블랙박스와 같습니다. 
Where the output of AI is limited to a true/false type of goal, with no room for nuance, this brings us to another limitation, which deals with control. As noted earlier, there is until now no feedback possible in AI. Once the software is running, goals cannot be changed. The loop is closed. Moreover, it is difficult to trace back what pathways have been used that led to a certain outcome, or why there was a particular weight between certain layers of the neural network. In medical practice, doctors frequently ask each other to verify a diagnosis for a patient, giving room for reasoning on the specific details of the idiosyncratic case that might not fit the general rules, but it is difficult to ask a computer to explain how and why it came to a particular diagnosis. Most importantly, there is no intentional interaction with the patient, who is also reduced to information. Despite the accuracy and efficiency of the performance (a gift box), AI is often a black box.

이 블랙박스를 열어야 합니다. 그렇지 않으면 AI는 의사가 이해하지 못하거나 동의하지 않거나 잘못된 '솔루션'을 제공하여 의사를 온갖 종류의 어려운 상황에 처하게 할 수 있으며, 이는 다시 책임과 대행에 관한 윤리적 문제를 제기할 수 있습니다. 이미 크리스퍼 기술을 이용한 유전자 치료제에 대해 언급했으며, 최근 간세포에 개입하면 '나쁜' 콜레스테롤 생성 수준을 바꿀 수 있다는 연구 결과가 발표되었습니다. 하지만 심장마비 가능성을 제어할 수 있다는 이 약속이 구조적 불평등을 통한 사회적 불공평이라는 더 큰 건강 문제에 의해 가로채질 수 있을까요? 새로운 기술의 혜택은 누구에게 돌아갈까요? 인공지능에 대한 입력은 병원에 내원하는 환자에 국한되기 때문에 인공지능의 결과물은 같은 사람으로 제한되어 잠재적으로 불평등을 강화할 수 있습니다. 최악의 경우, 미디어에 의해 심하게 악용되는 AI는 '폭주하는 대상'49이 되어 치료법이 증상으로 바뀔 수도 있습니다. 
This black box needs to be opened. If not, AI can place doctors in all kinds of difficult situations, providing ‘solutions’ that they do not understand or agree with, or that are erroneous, again raising ethical questions concerning responsibility and agency. We have already mentioned gene therapeutics using Crispr technologies, and recent developments show that intervention in liver cells can alter levels of production of ‘bad’ cholesterol. However, could this promise to control the potential for heart attacks be hijacked by the greater health problem of social injustice through structural inequality? Who will benefit from the new technologies? The input for AI goes as far as the patients who present themselves at the clinic, thus limiting the output of AI to those same people, potentially reinforcing inequality. In a worst-case scenario, heavily exploited by the media, AI may become a ‘runaway object,’49 rogue or autonomous, turning a cure into a symptom.

7 치료를 위한 교육인가, 관리를 위한 교육인가?
7 EDUCATING TO CURE OR TO CARE?

AI는 의사의 업무를 변화시키고 직업적 정체성을 재구성함으로써 의료 서비스에 영향을 미칠 것입니다. 그러나 컴퓨터가 진단과 치료 측면에서 무엇을 할 수 있더라도 임상적 직관, 비언어적 지각, 공감과 같은 임상 치료의 복잡한 측면을 모두 충족시킬 수는 없으며, 이는 좋은 임상 치료와 전문직업성의 핵심입니다.50 따라서 '학습자가 환자와의 연결 능력을 희생하지 않고 이러한 새로운 기술을 습득하도록 돕는 것이 차세대 의사 교육의 큰 과제 중 하나가 될 것입니다'40. 
AI will impact health care by changing doctors’ tasks and reshaping professional identities. However, no matter what a computer can do in terms of diagnosing and treating, it cannot fulfil the full range of complex aspects of clinical care, such as clinical intuition, ineffable perceptions and empathy, that are key to good clinical care and professionalism.50 Therefore, ‘helping our learners to master these new technologies without sacrificing their ability to connect with their patients will undoubtedly be one of the great challenges of educating the next generation of physicians.’40

의학교육은 의료 분야에서 [AI 시스템에 대한 비판적 검토]를 커리큘럼 과정과 강의 계획서 내용에 어떻게 가장 잘 통합할 것인가라는 교육학적 과제에 직면해 있습니다. 커리큘럼 자체가 알고리즘화되는 도구주의의 함정에 빠질 위험이 있습니다. 학습자는 이러한 기술이 의료 문제에 대한 인식을 중재하고 틀을 짜는 데 사용될 것이라는 점을 이해해야 합니다. [AI의 원리에 대한 교육 측면]에서 공식화는 학습자가 [자신의 감각적 경험과 관련하여], 그리고 [일상적인 의료 행위의 복잡하고 불확실한 현실에서 추론과 관련하여] AI 결과를 이해하는 데 도움이 될 것입니다. 이는 또한 전문가적 통제력을 회복하는 방법이기도 한데, AI를 사용한 작업을 규제함으로써 전문가적 정체성의 왜곡을 방지할 수 있습니다.38

Medical education faces a pedagogical challenge: how best to incorporate the critical examination of AI systems in health care as curriculum process and syllabus content. The danger is falling into the trap of instrumentalism, where the curriculum itself becomes algorithmic. Learners have to understand that these technologies will come to mediate and frame their perceptions of medical problems. Formalisation in terms of education on the principles of AI will help learners to make sense of AI output in relation to their own sensory experiences and in relation to reasoning in the complex and uncertain reality of everyday medical practice. It is also a way of regaining professional control, where regulating work with AI sets out to inoculate against the distortion of professional identity.38 

대부분의 AI 세부 사항은 의사의 전문적 한계를 넘어서는 것이지만, 테크놀로지는 [계산적 사고 방식을 넘어서는 의료 전문가로서의 부분]에 주의를 다시 집중할 기회를 제공하여 학습자가 독특한 환자와의 관계를 맺도록 요구합니다.

  • 의대생은 의료 행위가 배타적이고 명확하게 과학적, 통계적 또는 절차적이지 않으며, 따라서 지금까지는 계산으로 추적할 수 없다는 것을 배워야 합니다.
  • 마찬가지로 많은 임상적 의사 결정에는 사회적, 개인적, 윤리적, 심리적, 재정적, 가족적, 법적 및 기타 요소가 포함되어 있어 전적으로 의학적인 것만 있는 것은 아닙니다.

, 의학에는 본질적으로 예술성과 인간성이 내재되어 있으며, 이는 정의상 AI가 복제할 수 없는 가소성입니다. 의대생을 위한 AI 교육 개발은 '신체 교육학'을 배우는 것과 함께 진행되어야 합니다.15 
Although most AI detail will be beyond doctors’ professional limits, technologies offer an opportunity to refocus attention on the part of being a health care professional that is beyond calculative modes of thought, requiring learners to make a connection with a unique patient.

  • Medical students should be taught that the practice of medicine is not exclusively and clearly scientific, statistical or procedural, and hence is not, so far, computationally tractable.
  • Likewise, many clinical decisions are not exclusively medical, where they have social, personal, ethical, psychological, financial, familial, legal and other components.

In other words, medicine has an intrinsic artistry and humanity that by definition is plastic and cannot be replicated by AI. Developments in AI education for medical students must walk hand-in-hand with learning ‘body pedagogics.’15

8 결론
8 CONCLUSIONS

우리는 체화된 의학을 질문자로 삼아 의학 분야에서 AI의 성장을 단순히 나열하고 축하하기보다는 문제를 제기하여 비관론보다는 주의caution의 분위기를 조성했습니다. 우리는 또한 무비판적이고 반성적이지 않은 '해결책주의'에 대해서도 문제를 제기했습니다. 우리는 의학교육이 학부 과정에서 AI에 대한 조기 학습에 대한 요구에 느리게 대응해 왔으며, 특히 윤리적 의학을 실천하는 데 있어 AI와의 교류가 미래 의사의 정체성 구성에 핵심 요소가 될 것이라는 점에 주목합니다. 미래의 의사는 AI가 알지 못하는 것을 알아야 하며, 의사는 진단의 정확성을 넘어 환자에게 무엇을 계속 제공할 수 있을까요?"51 AI의 뼈대에 살을 붙여 봅시다.

Using embodied medicine as interrogator, we have problematised, rather than simply catalogued and celebrated, the growth of AI in medicine to set a tone of caution rather than pessimism. We have also problematised uncritical and unreflexive ‘solutionism.’ We note that medical education has been slow to respond to the demands of early learning about AI in undergraduate medicine, and that engaging with AI will become a key factor in the identity constructions of doctors of the future, particularly in the exercise of an ethical medicine. What will the physician of the future need to know that AI does not, and what will doctors continue to offer patients beyond diagnostic precision?51 Let us put flesh on the bones of AI.

 


 

Med Educ. 2021 Jan;55(1):30-36. doi: 10.1111/medu.14131. Epub 2020 Mar 30.

Where medical education meets artificial intelligence: 'Does technology care?'

Affiliations collapse

Affiliations

1Faculty of Health, Medicine and Life Sciences, School of Health Professions Education, Maastricht University, Maastricht, the Netherlands.

2Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, Utrecht, the Netherlands.

3Plymouth University Peninsula School of Medicine, Plymouth, UK.

PMID: 32078175

DOI: 10.1111/medu.14131

Abstract

'COLD' TECHNOLOGIES AND 'WARM' HANDS-ON MEDICINE NEED TO WALK HAND-IN-HAND: Technologies, such as deep learning artificial intelligence (AI), promise benign solutions to thorny, complex problems; but this view is misguided. Though AI has revolutionised aspects of technical medicine, it has brought in its wake practical, conceptual, pedagogical and ethical conundrums. For example, widespread adoption of technologies threatens to shift emphasis from 'hands-on' embodied clinical work to disembodied 'technology enhanced' fuzzy scenarios muddying ethical responsibilities. Where AI can offer a powerful sharpening of diagnostic accuracy and treatment options, 'cold' technologies and 'warm' hands-on medicine need to walk hand-in-hand. This presents a pedagogical challenge grounded in historical precedent: in the wake of Vesalian anatomy introducing the dominant metaphor of 'body as machine,' a medicine of qualities was devalued through the rise of instrumental scientific medicine. The AI age in medicine promises to redouble the machine metaphor, reducing complex patient experiences to linear problem-solving interventions promising 'solutionism.' As an instrumental intervention, AI can objectify patients, frustrating the benefits of dialogue, as patients' complex and often unpredictable fleshly experiences of illness are recalculated in solution-focused computational terms. SUSPICIONS ABOUT SOLUTIONS: The rate of change in numbers and sophistication of new technologies is daunting; they include surgical robotics, implants, computer programming and genetic interventions such as clustered regularly interspaced short palindromic repeats (CRISPR). Contributing to the focus of this issue on 'solutionism,' we explore how AI is often promoted as an all-encompassing answer to complex problems, including the pedagogical, where learning 'hands-on' bedside medicine has proven benefits beyond the technical. Where AI and embodied medicine have differing epistemological, ontological and axiological roots, we must not imagine that they will readily walk hand-in-hand down the aisle towards a happy marriage. Their union will be fractious, requiring lifelong guidance provided by a perceptive medical education suspicious of 'smart' solutions to complex problems.

 

보건의료전문직이 인공지능-기반 도구를 사용하기 위한 역량(Acad Med, 2023)
Competencies for the Use of Artificial Intelligence–Based Tools by Health Care Professionals
Regina G. Russell, PhD, MA, MEd, Laurie Lovett Novak, PhD, Mehool Patel, MD, Kim V. Garvey, PhD, MS, MLIS, Kelly Jean Thomas Craig, PhD, Gretchen P. Jackson, MD, PhD, Don Moore, PhD, and Bonnie M. Miller, MD, MMHC 

 

인공 지능(AI)은 머신 러닝, 딥 러닝, 자연어 처리, 신경망을 활용하는 알고리즘을 포함하여 인간의 지능을 모방하는 컴퓨터 과학 기술을 말합니다. 의료 분야에서 AI 알고리즘은 이해할 수 없을 정도로 크고 복잡한 데이터 세트를 임상적 결정을 내릴 수 있는 정보로 변환할 수 있습니다. 1 이러한 기술은 점점 더 정교해지고 있으며, 위험 점수, 이미지 해석, 의료 기록 요약과 같이 AI로 생성된 결과물은 이미 환자 치료에 직접적인 영향을 미치는 데 사용되고 있습니다. 2,3 AI 기반 도구에는 이러한 유형의 계산 기술을 운영 체제에 통합하는 모든 기기와 소프트웨어가 포함됩니다.  
Artificial intelligence (AI) refers to computer science techniques that mimic human intelligence, including algorithms that leverage machine learning, deep learning, natural language processing, and neural networks. In health care, AI algorithms can transform incomprehensibly large and complex data sets into information that can guide clinical decisions. 1 These technologies have become increasingly sophisticated, and AI-generated outputs, such as risk scores, image interpretation, and health record summarization, are already being used to directly influence patient care. 2,3 AI-based tools include all instruments and software that incorporate these types of computational technologies into their operating systems.

의료 서비스에서 AI 기반 도구를 구현하면서 도구 개발의 편향성, 후속 건강 개선의 분배 정의 등 다양한 실용적, 윤리적 문제가 제기되고 있습니다. 4-7 이러한 도구의 도입이 증가함에 따라 임상의는 이러한 도구의 유익한 효과와 해로운 효과를 모두 경험하게 될 것이며, 따라서 AI의 영향을 받는 의료 환경에서 효과적이고 효율적이며 안전하며 공평하게 진료하기 위한 기본 역량을 습득해야 합니다. 그러나 이러한 역량의 성격은 아직 명확히 밝혀지지 않았습니다.
The implementation of AI-based tools in health care has given rise to a variety of practical and ethical concerns, including bias in tool development and justice in the distribution of subsequent health improvements. 4–7 As adoption increases, clinicians will encounter both beneficial and harmful effects of these tools and will therefore need to acquire baseline competencies for practicing effectively, efficiently, safely, and equitably in AI-influenced health care environments. However, the nature of such competencies has not yet been articulated.

6개 의과대학에서 제공하는 프로그램을 검토한 Paranjape 등 8은 고급 수학, 의료 데이터 세트, AI 기반 도구의 임상 적용 등 다양한 AI 관련 주제를 다루는 학습 활동을 설명했습니다. Lee 등 9은 학부 의과 교육에서 AI에 대한 범위 검토를 수행하여 명확한 학문적 구조와 특정 역량이 필요하다는 결론을 내렸습니다. 최근 논평에서 Garvey 등10은 임상 현장에서 새로운 AI 기반 도구를 테스트하고 구현할 때 구조화된 학습 프로그램이 거의 수반되지 않는다고 지적하며 역량에 대한 요구를 반영했습니다. 임상 정보학에 대한 역량이 발표되었지만,11 AI 기반 도구의 계산 복잡성과 진료 및 결과에 대한 잠재적 영향력을 고려할 때 안전하고 윤리적인 배포를 보장하기 위한 구체적인 역량이 필요합니다. 
In a review of programs offered at 6 medical schools, Paranjape et al 8 described learning activities that covered a range of AI-related topics, including advanced mathematics, health care data sets, and clinical applications of AI-based tools. Lee et al 9 conducted a scoping review of AI in undergraduate medical education and concluded that a clear academic structure and specified competencies were needed. In a recent commentary, Garvey et al 10 noted that structured learning programs rarely accompanied the testing and implementation of new AI-based tools in clinical workplaces and they echoed the call for competencies. Although competencies have been published for clinical informatics, 11 the computational complexity of AI-based tools and their potential influence on practice and outcomes necessitate specific competencies aimed at ensuring safe and ethical deployment.

역량 기반 교육은 이 연구를 위한 개념적 틀을 제공합니다. 12-14 역량 기반 교육의 기본 가정은 특정 지식, 기술 및 태도(역량)가 전문직 실무에 필요하며 이러한 역량 기대치는 전문직 환경의 핵심 측면을 분석하여 학습자를 위해 개발될 수 있다는 것입니다. 기대치가 정의되면 학습자는 개별화된 학습 경로를 따라 해당 역량을 습득할 수 있습니다. 12-14 본 연구의 목표는

  • (1) 의료 전문가를 위한 AI 관련 임상 역량을 정의하고,
  • (2) 향후 의료 AI 기술의 사용과 감독 및 관리에 필요한 조직의 책임을 탐구하는 것입니다.

Competency-based education provides a conceptual framework for this study. 12–14 The basic assumption of competency-based education is that specific knowledge, skills, and attitudes (competencies) are necessary for professional practice and that these competency expectations can be developed for learners by analyzing core aspects of their professional environments. Once expectations are defined, learners can acquire the competencies along individualized learning pathways. 12–14 Our study’s goals were

  • (1) to define AI-related clinical competencies for health care professionals and
  • (2) to explore future uses of health care AI technologies and the organizational responsibilities required for oversight and management.

방법
Method

2020년 12월부터 2021년 7월까지 의료 전문가가 AI 기반 도구를 효과적으로 사용하는 데 필요한 역량에 대해 주제별 전문가(SME)를 인터뷰하는 정성적 연구를 수행했습니다. 15-19그림 1은 연구 프로세스를 간략하게 보여줍니다.
From December 2020 through July 2021, we conducted a qualitative study in which we interviewed subject matter experts (SMEs) about the competencies health care professionals need to work effectively with AI-based tools. 15–19Figure 1 outlines the study process.

다학제 연구팀은 밴더빌트 대학교 의과대학의 교수진 5명(R.G.R., L.L.N., K.V.G., D.M., B.M.M.)과 AI 관련 연구 및 분석에 전문성을 갖춘 IBM 왓슨 헬스의 직원 3명(M.P., K.J.T.C., G.P.J.)으로 구성되었습니다. 이 팀에는 의사 3명(M.P., G.P.J., B.M.M.), 의료 AI 전문 인류학자(L.L.N.), 기술 업계 과학 책임자이자 전 의학교육자(K.J.T.C.), 교육 기술 전문가(K.V.G.), 직장 및 역량 기반 학습에 전문성을 갖춘 의학교육 학자 2명(D.M. 및 R.G.R.)이 포함되었습니다. 과도한 영향을 피하기 위해 IBM 왓슨 헬스의 팀원은 인터뷰에 참여하지 않았습니다. 두 연구팀은 연구의 다른 모든 측면에 대해 협력했습니다. 밴더빌트 대학교 기관윤리심의위원회는 이 프로젝트가 최소한의 위험을 초래한다고 판단하여 면제 대상으로 지정했습니다. 
The multidisciplinary research team consisted of 5 faculty members (R.G.R., L.L.N., K.V.G., D.M., and B.M.M.) from Vanderbilt University School of Medicine and 3 members (M.P., K.J.T.C., and G.P.J.) from IBM Watson Health who have expertise in AI-related research and analytics. The team included 3 physicians (M.P., G.P.J., and B.M.M.), an anthropologist specializing in health care AI (L.L.N.), a science officer in the technology industry and former medical educator (K.J.T.C.), a specialist in educational technology (K.V.G.), and 2 medical education scholars with expertise in workplace and competency-based learning (D.M. and R.G.R.). To avoid undue influence, team members from IBM Watson Health did not participate in the interviews. The combined team collaborated on all other aspects of the study. The Vanderbilt University Institutional Review Board determined that the project posed minimal risk and designated it exempt.

참여자
Participants

문헌 검토, 전문가 접촉, 눈덩이 기법(참가자에게 다른 내용전문가(SME)을 추천하도록 요청)을 통해 의료 환경에서 AI 기반 도구를 사용하는 내용전문가(SME)을 파악하고 이 연구에 참여하도록 초대했습니다. 19 연구에 참여하려면 미국 보건 전문직 교육 또는 보건 정보학 분야에서 일해야 했습니다. 이러한 의도적인 표본 추출 과정을 통해 다양한 직업을 대표할 수 있도록 했으며 윤리 및 형평성에 대한 특별한 전문성을 가진 사람들을 포함했습니다. 내용전문가(SME)은 연구 참여에 대한 사례금을 받았습니다. 
We identified SMEs in the use of AI-based tools in health care settings through literature review, professional contacts, and snowball techniques (asking participants to recommend other SMEs) and invited them to participate in this study. 19 To be eligible for the study, individuals needed to work in U.S. health professions education or health informatics. This purposive sampling process ensured representation from multiple professions and included those with special expertise in ethics and equity. The SMEs received an honorarium for their participation in the study.

데이터 수집
Data collection

반구조화된 인터뷰가 데이터의 주요 원천이었지만, 인구통계학적 설문지와 역량 초안에 대한 피드백을 요청하는 양식에서도 데이터를 수집했습니다. 저자들의 사전 논평에서 확인된 내용은 인터뷰 가이드 개발에 도움이 되었습니다. 10 반구조화된 인터뷰 프로토콜에는 연구 목표에 대한 설명이 포함된 소개 스크립트와 인터뷰에 대한 공유된 이해를 확립하기 위한 의료 분야에서의 AI에 대한 실무적 정의가 포함되었습니다. 인터뷰는 "의료 전문가가 임상 치료에서 AI 기반 도구를 사용하려면 어떤 역량이 필요한가?"라는 질문에 답하기 위해 설계되었습니다. 또한 인터뷰에서는 의료 AI 기술의 향후 활용과 이에 대한 조직의 감독 및 관리 책임에 대해서도 논의했습니다. 질문의 질과 유용성을 평가하기 위해 두 차례의 파일럿 인터뷰가 실시되었습니다. 초기 인터뷰에 대한 팀의 검토를 거쳐 다양성, 포용성, 건강 형평성과 관련된 질문이 추가되었습니다. 15번의 인터뷰 끝에 데이터 포화, 즉 내용전문가(SME)이 공유하는 아이디어의 반복에 도달했습니다. 20 최종 인터뷰 가이드는 부록 디지털 부록 1에서 확인할 수 있습니다.
Semistructured interviews were the primary source of data, although we also collected data from a demographic questionnaire and a form that solicited feedback on drafts of the competencies. Articles identified from the authors’ prior commentary informed development of the interview guide. 10 The semistructured interview protocol included an introductory script with a description of the study objectives and a working definition of AI in health care that was intended to establish a shared understanding for the interviews. Interviews were designed to answer the question, “What competencies do health care professionals need for use of AI-based tools in clinical care?” Interviews also explored future uses of health care AI technologies and organizational responsibilities for their oversight and management. Two pilot interviews were conducted to assess quality and utility of the questions. After team reflection on the initial interviews, questions were added related to diversity, inclusion, and health equity. Data saturation, or repetition of the ideas being shared by the SMEs, was reached after 15 interviews. 20 The final interview guide is available in Supplemental Digital Appendix 1 (at https://links.lww.com/ACADMED/B334).

인터뷰는 2021년 1월부터 4월까지 화상회의를 통해 진행되었으며, 참가자의 동의를 얻어 오디오를 녹음하고 필사했습니다. 인터뷰는 45분에서 60분 사이에 진행되었습니다. 인터뷰는 밴더빌트 대학교 의과 대학 팀원들이 가용성에 따라 분산되어 진행되었으며, 각 팀당 1명이 주 인터뷰어로 지정되었습니다. 보조 면접관은 일관성을 보장하기 위해 1명을 제외한 모든 면접에 참여했으며 선택한 지점에서 후속 질문을 던졌습니다. 일관성을 유지하기 위해 연구팀 전체가 매주 만나 인터뷰 내용을 보고하고, 녹취록을 검토하고, 인터뷰 가이드를 수정했습니다.
Interviews were conducted via videoconference between January and April 2021, audio-recorded with consent of the participants, and transcribed. The interviews lasted between 45 and 60 minutes. Interviews were distributed across Vanderbilt University School of Medicine team members based on availability, with 1 person designated as the primary interviewer for each. A secondary interviewer participated in all but 1 interview to ensure consistency and pose follow-up questions at selected points. To further ensure consistency, the entire research team met weekly to debrief interviews, review transcripts, and refine the interview guide.

내용전문가(SME)은 인터뷰 후 학위, 직책, 직업, 전문 분야, 나이, 경력 연수, 성별, 인종 또는 민족 등의 인구통계학적 요소를 제공했습니다. 2021년 7월, 내용전문가(SME)은 각 역량 및 하위 역량 초안의 구체적인 문구가 포함된 양식을 받았습니다. 응답자들은 각각에 대해 공개 텍스트 의견과 편집 제안을 제공하도록 요청받았으며 요약 의견을 제공할 기회를 가졌습니다. 인구통계학적 설문지와 역량 피드백 양식은 안전한 온라인 데이터 수집 도구인 REDCap(Research Electronic Data Capture)을 통해 관리되었습니다. 21 
The SMEs provided the following demographic elements after the interview: degree(s), title(s), profession, fields of expertise, age, years of professional experience, gender, and race or ethnicity. In July 2021, the SMEs received a form that contained the specific wording of each draft competency and subcompetency. Respondents were asked to provide open text comments and editorial suggestions on each and had the opportunity to provide summary comments. The demographic questionnaire and competency feedback form were administered through REDCap (Research Electronic Data Capture), a secure online data-capturing tool. 21

데이터 관리 및 분석
Data management and analysis

인터뷰 녹취록은 정성적 데이터 분석 도구인 Dedoose로 가져오기 전에 비식별화되었습니다. 22 코딩은 연구팀 전체에 분산되어 2명의 연구원이 각 인터뷰마다 코딩을 담당했습니다. 코딩에는 녹취록에서 텍스트 섹션을 발췌하고 각 발췌에 1개 이상의 코드를 할당하는 작업이 포함되었습니다. 연구팀은 연역적 코딩귀납적 코딩 방법을 모두 사용했습니다. 23

  • 연역적 코딩에서는 의료 전문직 전반의 영역을 통합한 기존 역량 분류법을 적용했습니다. 24
  • 귀납적 코딩에서는 선험적 코딩 체계 없이 데이터를 검토하여 기존 프레임워크에 포함되지 않은 새로운 아이디어를 식별할 수 있었습니다.

이전에 개발된 역량 프레임워크를 사용하면 코더 간 개념의 초기 조율을 지원하는 한편, 개방형 코딩의 유연성을 통해 예상치 못한 인사이트를 발견할 수 있었습니다.
Interview transcripts were deidentified before importing into the qualitative data analysis tool Dedoose. 22 Coding was distributed across the research team, with 2 members coding each interview. This coding involved excerpting sections of text from transcripts and assigning 1 or more codes to each excerpt. The team used both deductive and inductive coding methods. 23 

  • In deductive coding, an existing competency taxonomy that integrated domains across health professions was applied. 24 
  • In inductive coding, data were reviewed without an a priori coding scheme, allowing identification of new ideas not included in existing frameworks.

The use of a previously developed competency framework supported initial alignment of concepts across coders, whereas the flexibility of open coding allowed discovery of unexpected insights.

주제별 분석은 코딩 과정에서 떠오른 아이디어를 탐색하고 요약하는 데 사용되었습니다. 15,19,23 코딩된 데이터에 대한 반복적인 팀 토론을 통해 6개의 광범위한 개념적 주제가 확인되었습니다. 이러한 주제는 인터뷰 전반에 걸쳐 반복적으로 나타났으며, 15개의 녹취록을 분석한 결과 귀납적 주제 포화도에 도달했습니다. 20,25
Thematic analysis was used to explore and summarize ideas that surfaced during the coding process. 15,19,23 Six broad conceptual themes were identified through iterative team discussion of the coded data. These themes appeared repeatedly throughout interviews, and after analysis of 15 transcripts, inductive thematic saturation was reached. 20,25

전문 분야에 따라 각 개념적 주제에 1명의 연구원이 배정되어 역량 진술로 번역했습니다. 연구자들은 주제에 매핑된 발췌문을 사용하여 향후 커리큘럼 개발에 필요한 구체성을 제공하는 하위 역량도 만들었습니다. 두 번째 팀원이 각 초안을 검토하고 수정 사항을 제안하도록 배정되었습니다. 그런 다음 팀 전체가 작성된 목록을 검토하고 유사한 하위 역량을 중복 제거하고 합의에 도달할 때까지 문구와 프레임워크를 반복적으로 수정했습니다. 이 합의 문서는 후속 양식을 작성하는 데 사용되었으며, 역량 진술에 대한 피드백을 요청하기 위해 내용전문가(SME)에 이메일을 통해 배포되었습니다.
On the basis of areas of expertise, 1 researcher was assigned to each conceptual theme for translation into competency statements. Using excerpts that mapped to the themes, researchers also created subcompetencies, which provided the specificity needed for future curriculum development. A second team member was assigned to review each draft and suggest revisions. The entire team then reviewed the compiled list, deduplicated similar subcompetencies, and iterated the wording and framework until consensus was reached. This consensus document was used to construct the follow-up form, which was distributed via email to the SMEs to solicit feedback on competency statements.

내용전문가(SME)의 피드백을 수집한 후, 각 역량과 그 하위 역량에 대해 1차 및 2차 검토자가 배정되었습니다. 1차 검토자는 제출된 답변을 변경 없음, 문구 수정, 삭제 또는 새로운 하위 역량 추가 등의 루브릭을 사용하여 코드화했습니다. 그런 다음 1차 검토자는 코드화된 답변을 사용하여 필요에 따라 진술을 수정했습니다. 2차 검토자는 성문화된 내용을 확인하고 수정된 내용이 정확하고 완전한지 평가했습니다. 수정된 역량 목록은 전체 연구팀의 합의 프로세스를 통해 검토 및 최종 확정되었습니다.
After collection of feedback from the SMEs, primary and secondary reviewers were assigned to each competency and its subcompetencies. The primary reviewer codified the submitted responses using the following rubric: no change, edit wording, delete, or add new subcompetency. The primary reviewer then used the codified responses to revise statements as needed. The secondary reviewer confirmed codification and evaluated the revisions for accuracy and completeness. The revised competency list was reviewed and finalized using a consensus process across the combined research team.

매주 회의를 통해 팀원들의 다양한 관점과 경력 경험이 데이터 해석에 미치는 영향에 대한 지속적인 성찰을 장려했습니다. 성적표 코딩, 성명서 작성, 내용전문가(SME) 피드백을 통한 수정 작업을 함께 진행함으로써 팀 내 참여가 더욱 활발해져 요약 결과의 신뢰도가 높아졌습니다. 25
Weekly meetings encouraged ongoing reflection on the impact of team members’ diverse perspectives and career experiences on data interpretation. Mixed pairings for transcript coding, statement writing, and revisions from the SME feedback encouraged deeper engagement within the team, increasing the credibility of the summary findings. 25

결과
Results

남성 10명, 여성 5명을 포함하여 15명의 내용전문가(SME)을 인터뷰했습니다. 내용전문가(SME)의 연령은 34세부터 73세까지 다양했으며, 3명은 소수 인종 또는 소수 민족 출신이었습니다. 내용전문가(SME)을 대표하는 임상 전문직은 의학(n = 9), 간호(n = 2), 약학(n = 1, 표 1) 등이었습니다. 비임상 학자 3명은 윤리(n = 1), 비즈니스 및 교육(n = 1), 사회 의학(n = 1)에 대한 전문성을 추가했습니다. 여러 역할을 맡고 있다고 답한 내용전문가(SME)이 많았지만,

  • 주로 의료 AI, 생의학 정보학 및/또는 직장에서의 AI의 윤리적 적용에 대한 전문성으로 11명이 선정되었고,
  • 의료 전문직 교육에 대한 전문성으로 2명이,
  • 의료 전문직 교육과 의료 AI에 대한 이중 전문성으로 2명이 선정되었습니다.

Fifteen SMEs were interviewed, including 10 men and 5 women. The SMEs ranged in age from 34 to 73 years, and 3 were from racial or ethnic minority groups. Clinical professions represented by the SMEs included medicine (n = 9), nursing (n = 2), and pharmacy (n = 1; Table 1). Three additional nonclinical scholars added expertise in ethics (n = 1), business and education (n = 1), and social medicine (n = 1). Although many SMEs reported multiple roles,

  • 11 were selected primarily for their expertise in health care AI, biomedical informatics, and/or the ethical application of AI in the workplace;
  • 2 for their expertise in health professions education; and
  • 2 because of dual expertise in health professions education and health care AI.

주요 주제 및 선정 의견
Key themes and selected comments

연구원들은 코딩된 발췌문을 반복적으로 분석하여 6개의 개념적 주제를 식별했습니다. 이러한 주제는 초기 전문가 간 역량 프레임워크에 의해 연역적으로 형성되었으며, 연구팀의 다양한 전문성을 반영한 오픈 코딩을 통해 귀납적으로 형성되었습니다. 주제와 대표 의견은 아래에 요약되어 있습니다. 의견은 길이와 명확성을 위해 최소한의 편집을 거쳤습니다. 
Researchers identified 6 conceptual themes based on iterative analysis of coded excerpts. These themes were shaped deductively by the initial interprofessional competency framework as well as inductively by open coding, which reflected the diverse expertise of the research team. Themes and representative comments are outlined below. Comments have been minimally edited for length and clarity.

기초 지식의 필요성.
The need for foundational knowledge.

내용전문가(SME)들은 임상 치료에 활용되는 AI 도구의 유형, 목적, 데이터의 품질, 관련 분야의 개발 기여도에 대한 기초 지식이 필요하다고 설명했습니다. 내용전문가(SME)들은 임상의가 이러한 분야의 전문가가 될 필요는 없지만 일반적인 정보학 역량과 데이터 입력, 생성된 출력, 알고리즘의 특성 등 AI의 구성 요소에 대한 높은 수준의 이해가 있어야 한다고 강조했습니다.
The SMEs described a need for foundational knowledge about the types of AI tools deployed in clinical care, their purposes, the qualities of data, and the contributions of related fields to their development. The SMEs stressed that clinicians need not become experts in any of these disciplines but instead should gain general informatics competency and a high-level understanding of the components of AI, including data inputs, generated outputs, and the nature of algorithms.

모든 의료 전문가를 위한 기본적인 정보학 역량 교육부터 시작해야 합니다. 어떤 기술을 사용하고 있나요? 어떻게 효율적으로 사용하나요? 데이터란 무엇인가요? 데이터를 어떻게 수집하나요? 왜 정확한 데이터가 필요한가?... 그런 다음 다음과 같은 더 복잡한 내용으로 넘어갑니다: AI란 무엇인가요? (내용전문가(SME) 4)
You have to start with fundamental informatics competency education for all our health professionals. What are the technologies you use? How do you use them efficiently? What’s data? How do you collect data? Why do we need accurate data?… Then, we move up to more complex stuff like: What is AI? (SME 4)

어떤 사람들은 이것을 새로운 형태의 문해력이라고 설명합니다.
Some described this as a new form of literacy.

더 많은 의료 서비스가 기술을 통해 제공될 수밖에 없기 때문에, 그 편안함, 그것은 수리력도 아니고 문해력도 아닙니다. 기술 리터러시와 같은 것이죠. (내용전문가(SME) 3)
We have no choice that more health care is going to be funneled through technology, so that comfort, it’s not quite numeracy and it’s not quite literacy. It’s like tech literacy. (SME 3)

내용전문가(SME)들은 또한 특정 도구를 특정 환자 또는 환자 집단에 안전하게 적용하기 위해서는 해당 도구에 대해 "알아야 할 것을 알아야 한다"고 설명했습니다.
The SMEs also described a need to “know what you need to know” about a specific tool to safely apply it to any given patient or population of patients.

우리는 약리학에 대해 충분히 배워서 누군가 와서 "이 마법의 콩이 있어요"라고 말하면 "그래요, 약리학을 해봤어요. 어떻게 작용하죠?" 그러면 그들은 "그냥 마술이에요."라고 대답할 것입니다. 환자에게 마법의 콩을 주기 전에 그 마법에 대해 조금 더 알고 싶을 것입니다. AI는 사람들에게 판매되는 방식이 마법의 콩과 비슷하기 때문에 우리 모두가 올바른 질문을 할 수 있기를 바랍니다. 그 뒤에 있는 기술은 무엇인가요? 어떻게 테스트되었나요? 외부 검증을 거쳤나요? 어떤 종류의 데이터 세트인가요? (내용전문가(SME) 6)
We learn enough about pharmacology that if someone came in and said, “I have these magic beans” and you said “Okay, I’ve had pharmacology. How do they work?” And they said “Well, it’s just magic.” You’d probably want to know a little more about the magic before you would give it to your patients. AI is kind of like magic beans the way it gets sold to people, and I want all of us to be able to ask the right questions. What’s the technology behind it? How has it been tested? Did you do external validation? What kind of data sets? (SME 6)

마지막으로, 내용전문가(SME)들은 일부 기초 지식은 기존 교육 과정에서 이미 다루어졌을 수 있지만 AI의 맥락에서 명시적으로 적용해야 할 필요가 있다는 점을 인정했습니다.
Finally, the SMEs acknowledged that some foundational knowledge may already be covered in existing courses but would need to be applied explicitly in the context of AI.

이 부분은 이미 어느 정도 교육을 받고 있다고 생각합니다. 민감도, 특이성, 정확도 측정에 대한 기본적인 개념만 "AI가 모든 것을 해결해 줄 거야"라고 말하는 경향이 있기 때문입니다. 하지만 완벽한 방법은 없습니다. (내용전문가(SME) 14)
This is something I think they already get some training on. Just the basic notion of sensitivity, specificity, the accuracy measures because there’s a tendency to say, “Ooh, you know, AI is going to solve everything.” But no particular method is perfect. (SME 14)

윤리적, 법적, 규제적, 사회적, 경제적, 정치적 문제.
Ethical, legal, regulatory, social, economic, and political issues.

많은 내용전문가(SME)은 공정성을 보장하기 위한 신중한 조치 없이 AI 기반 도구가 구현될 경우 건강 불평등을 악화시킬 가능성에 대해 논의했습니다. 잠재적인 부정적 결과에 대해 충분한 주의를 기울이기 전에 형사 사법, 교육, 주택 분야에 AI 기반 기술이 도입되었을 때 발생하는 문제점을 언급하기도 했습니다. 이러한 우려를 바탕으로 내용전문가(SME)은 모든 임상의가 AI 기반 도구가 건강 격차와 의료 격차를 좁힐지 넓힐지를 결정할 사회적, 윤리적, 법적, 규제적 문제를 이해해야 한다고 설명했습니다. 
Many of the SMEs discussed the potential for AI-based tools to worsen health inequities if implemented without deliberate action to ensure fairness. Several cited problems occurred when AI-based technologies were implemented in criminal justice, education, and housing before sufficient attention was given to potential negative consequences. On the basis of these concerns, the SMEs described a need for all clinicians to understand the social, ethical, legal, and regulatory issues that will determine whether AI-based tools will narrow or widen health disparities and health care gaps.

우리는 형사 사법, 교육, 주택 등 다양한 분야에서 인공지능이 형평성과 포용성을 무시할 때 어떤 일이 벌어지는지 보아왔습니다. 우리는 이러한 분야에서 인공지능을 사용해서는 안 되며, 의료 분야에서도 인공지능을 사용해서는 안 됩니다. (내용전문가(SME) 15) 
We have seen what AI does when it ignores equity and inclusion in criminal justice, in education, in housing, in you name it. We should not have done it in those fields, and we certainly cannot do it in health care. (SME 15)

내용전문가(SME)들은 이해도를 넘어 임상의가 AI 도구의 윤리적 배포와 적절한 평가를 보장하기 위해 개인적이고 공동의 책임감을 개발해야 할 필요성을 설명했습니다. 
Beyond understanding, the SMEs described a need for clinicians to develop a sense of personal and shared responsibility for ensuring ethical deployment and appropriate evaluation of AI tools.

많은 임상의가 "나에게는 책임이 있고 그 책임의 일부는 목소리를 내는 것입니다. 저는 이 문제에 대해 방관자가 되지 않을 것입니다." 라고 말하기보다는 시스템이 저절로 작동하도록 내버려 둡니다.  (내용전문가(SME) 15) 
Many clinicians allow the system to run itself rather than saying, “I have a responsibility and part of that is to speak up. I’m not going to be a bystander on this one.” (SME 15)

내용전문가(SME)들은 AI 기반 도구가 제공하는 지원과 관계없이 임상의가 환자 치료의 모든 측면과 임상 결과에 대한 일차적인 전문적, 법적 책임을 여전히 져야 한다고 한목소리로 말했습니다. 
The SMEs uniformly stated that clinicians should still hold primary professional and legal responsibility for all aspects of patient care and for clinical outcomes, regardless of the support provided by AI-based tools.

학생들에게 필요한 윤리적 부분은 의사가 그 과정에서 전문적인 의무를 포기하지 않는다는 개념이라고 생각합니다. (내용전문가(SME) 10) 
I think the ethical piece that the students would need is this notion of the physician not abdicating their professional duty in the process. (SME 10)

몇몇은 형평성과 격차 악화 가능성에 대한 우려를 신랄하게 표현하고 공평한 배치를 방해할 수 있는 정치적, 구조적 요인에 대해 설명했습니다. 
Several expressed concerns for equity and the potential for worsening disparities in poignant terms and described the political and structural factors that could impede equitable deployment.

그렇다면 기술에 접근할 수 없거나 교육, 언어 또는 기타 장벽으로 인해 뒤처진 사람들은 어떤 상황에 놓이게 될까요? AI 혁명은 그 범주에 속하는 사람들을 어디에 남겨둘까요? 그리고 환급 구조와 전체 시스템 측면에서 의료 서비스 모델은 어디에서 장애를 일으키기 시작할까요? AI가 본질적으로 격차를 줄이는 데 도움이 되기보다는 오히려 악화시킬 수 있는 부분은 어디일까요? 저는 그것이 앞으로 문제가 될 수 있다고 생각하기 때문입니다. (내용전문가(SME) 14)  
So where does it leave those who don’t have access to technology or who are left behind because of educational, language, or other barriers? Where does an AI revolution leave people who fall in that bucket? And where does the health care model start to impede, in terms of the way reimbursements are structured, the entire system? Where might it introduce gaps that AI essentially exacerbates rather than helps close? Because I can see that being a problem down the line. (SME 14)

임상의의 역할과 책임, 그리고 임상 진료의 본질.
Clinician roles and responsibilities and the nature of the clinical encounter.

내용전문가(SME)들은 AI 기반 도구의 큰 잠재력은 임상의가 다양한 소스에서 파생된 방대한 양의 데이터를 관리할 수 있도록 지원함으로써 임상 진료를 개선하는 데 있다는 데 동의했습니다. 
The SMEs agreed that the great promise of AI-based tools is to improve clinical care by helping clinicians manage massive amounts of data derived from diverse sources.

과거 데이터, 현재 데이터. 환자의 환경에서 무슨 일이 일어나고 있는지 진료 시점에 이 모든 것을 종합하려고 합니다....... 증상. 치료법. 우리를 돕기 위해 이러한 도구가 필요합니다. 우리의 마음은 무한하지 않습니다. 우리는 산만합니다. 우리는 피곤합니다. 우리는 무슨 일이 일어나고 있는지 따라잡을 수 없습니다. 올바른 결정을 내리는 데 도움을 줄 수 있는 도구가 필요합니다. (내용전문가(SME) 4) 
Historic data, current data. Trying to piece all that together at the point of care with whatever’s going on in the environment of the patient…. The symptomatology. Their treatment. We need these tools to help us. Our minds are not infinite. We are distracted. We get tired. We can’t keep up with what’s going on. We need these tools to be able to provide us the support to help make good decisions. (SME 4)

내용전문가(SME)들은 AI 기반 도구의 결과물이 임상적 의사 결정을 보강하고 지원하는 데 사용되어야 하며, 임상의가 개별 환자를 돌볼 때 이러한 결과물을 적용하는 데 판단력을 발휘해야 하며, 이상적으로는 공유 의사 결정 모델을 사용해야 한다고 강조했습니다. 
The SMEs stressed that the outputs of AI-based tools should be used to augment and support clinical decision-making and that clinicians should exercise judgment in applying those outputs in caring for individual patients, ideally using shared decision-making models.

르네 라넥은 튜브를 통해 들을 수 있는 청진기를 사용하면 더 잘 들을 수 있다는 사실을 발견했고, 이는 결국 현대 청진기로 발전했습니다. 마찬가지로 AI 도구는 진단 또는 치료 능력을 향상시키는 청진기처럼 보아야 하며, 귀를 대체해서는 안 됩니다. (내용전문가(SME) 7) 
René Laënnec discovered that you could hear better with a tube you could listen through that was eventually refined to the modern stethoscope. Similarly, AI tools should be viewed like a stethoscope that increases your diagnostic or therapeutic ability but shouldn’t replace what goes on between your ears. (SME 7)
의사의 역할은 해당 지침이나 프레젠테이션을 환자의 맥락에서 해석한 다음 공유된 의사 결정 프로세스를 통해 환자가 현재 처한 상황에서 환자의 요구를 가장 잘 충족할 수 있는 치료 계획을 생성하는 것이며, 일부 상충 관계가 있을 수 있음을 이해해야 합니다. (내용전문가(SME) 2) 
The physician’s job is to interpret that guideline or presentation in the context of their patient and then through a shared decision-making process generate a care plan that’s going to best meet the needs of the patient where they are at that time, understanding that there may be some tradeoffs. (SME 2)

내용전문가(SME)들은 환자의 치료에 영향을 미칠 수 있는 AI 기반 도구의 결과를 환자에게 설명할 때 향상된 커뮤니케이션 기술이 필요하다고 설명했지만, 이러한 도구가 임상의와 환자 관계의 인간적인 측면을 향상시킬 수 있는 기회를 제공할 수 있다고 믿었습니다. 
The SMEs described a need for enhanced communication skills in explaining to patients the outputs of AI-based tools that might influence their care but also believed that these tools may present an opportunity to enhance humanistic aspects of clinician–patient relationships.

의사 결정에 따른 인지적 부하를 일부 상쇄하고 이를 돕기 위해 이 도구를 통합함으로써 이득을 얻을 수 있다고 한다면, 이러한 인지적 부하를 인문학적 상호 작용에 다시 주입할 수 있을까요? (내용전문가(SME) 10) 
If we’re saying we’re going to benefit from offsetting some of the cognitive load of the decision making and incorporating this tool to help us do that, can we reinfuse that cognitive load in the humanistic interactions? (SME 10)

내용전문가(SME)들은 대체가 아닌 보강을 강조했지만, 임상 치료의 일부 측면이 결국 AI 기반 기능으로 대체될 수 있다는 점을 인정했습니다. 
Although the SMEs emphasized augmentation and not replacement, they acknowledged that some aspects of clinical care might eventually be replaced by AI-based capabilities.

기술이 여러분이 하는 일의 일부를 대체할 것이라는 점을 이해해야 합니다. 그리고 그거 아세요? 괜찮습니다. 여러분의 스킬 세트에는 여전히 중요한 다른 측면이 있습니다. 하지만 기술이나 시스템의 다른 부분으로 더 잘할 수 있는 일에 매달리는 것은 어떨까요? 환자와 시스템 모두에게 중요합니다. AI가 이를 더욱 강화할 것이라고 생각합니다. (내용전문가(SME) 2) 
You need to understand that technology is going to replace some of the stuff you do. And guess what? That is okay. There are other aspects of your skill set that are going to remain really important. But hanging onto stuff that can be done better by either technology or other points in the system? It’s important for both the patients and the system. AI, I think, is going to push that. (SME 2)

마지막으로, 내용전문가(SME)들은 임상의가 임상적으로 적절한 경우 환자가 생성한 데이터를 통합해야 한다고 생각했습니다. 
Finally, the SMEs believed that clinicians should incorporate patient-generated data when clinically appropriate.

엄청나게 가치 있는 정보죠? 예를 들어 진료실에서 혈압을 한 번 측정하는 것보다 훨씬 낫죠. 이제 환자의 일상 생활과 건강 상태에 대한 훨씬 더 정확한 데이터를 얻을 수 있습니다. 따라서 우리는 이를 활용해야 합니다. 엄청난 잠재력을 제공합니다. 사람들을 너무 단편적으로 만나기 때문에 우리가 할 수 있는 모든 일을 할 수 없습니다. 하지만 이를 어떻게 처리하고 최대한 활용할 수 있을지 함께 노력해야 합니다. (내용전문가(SME) 10) 
It’s incredibly valuable information, right? It’s so much better than the one shot of, for instance, a blood pressure measurement when they’re in the office. Now you’ve got much more accurate data about their day-to-day life and how things work. And so, we need to lean into it. It offers tremendous potential. It’s all the stuff that we can’t do because we see folks so episodically. But we will need to work together to figure out how to process it and make the most of it. (SME 10)

팀 역학 및 워크플로에 미치는 영향.
The impact on team dynamics and workflows.

몇몇 내용전문가(SME)은 AI 기반 도구가 기존 워크플로우에 미칠 영향에 대해 논의하고 워크플로우 변경 사항을 명시적으로 처리하는 구현 프로세스를 권장했습니다. 
Several SMEs discussed the impact that AI-based tools would have on established workflows and recommended implementation processes that explicitly address workflow changes.

"우리의 워크플로는 무엇이며, 이 기술은 어디에 적합fit한가?"라는 질문을 던질 수 있어야 합니다. 그것이 위험 모델을 제시하는 것이든, 약물을 전달하는 로봇이든, 정맥주사 펌프이든 상관없습니다. 이 새로운 도우미를 수용하기 위해 워크플로에 어떤 변화가 필요할까요? (내용전문가(SME) 1) 
We should be able to ask the question, “What is our workflow and where does this technology fit into that?” And that’s whether it’s the presentation of a risk model or it’s the robot that delivers medications or the IV [intravenous] pump. How will this fit in, and what do we need to change about our workflows, if anything, to accommodate this new helper? (SME 1)

내용전문가(SME)들은 이러한 워크플로우의 중단이 전문가 간 팀과 팀원 간의 관계에 영향을 미칠 수 있다는 점을 인정했습니다. 
The SMEs acknowledged that these disrupted workflows could have an impact on interprofessional teams and the relationships between team members.

또한 전문가 간 팀은 팀 내에서 증강 지능이 할 수 있는 일과 할 수 없는 일에 대해 공유된 이해를 가져야 합니다. 특히 진행성 암과 같은 복잡한 치료 문제와 관련하여 중요한 부분이라고 생각할 수 있습니다. (내용전문가(SME) 2) 
The interprofessional team would also have to have a shared understanding of what that augmented intelligence can and can’t do within that team aspect. But you could imagine that being an important part, particularly around complex care issues like advanced cancers. (SME 2)

데이터 세트의 편향성 및 대표성에 대한 우려.
Concerns about bias and representativeness of data sets.

내용전문가(SME)들은 개인적 편향, 조직적 편향, 시스템적 편향 등 여러 수준에서, 그리고 도구 제작 과정의 여러 지점에서 편향에 대한 우려를 논의했습니다. 특히 내용전문가(SME)들은 데이터 세트 학습과 검증에 사용되는 모집단의 특성에 따라 AI 기반 도구에 내장될 수 있는 편향성에 대해 설명할 때 우려를 표명했습니다.
The SMEs discussed concerns for bias at multiple levels, including personal, organizational, and systemic bias, and at multiple points in the tool creation process. Concerns were particularly strong when SMEs described the bias that can be built into AI-based tools based on the characteristics of populations used for training data sets and for validation.

문제는 이러한 도구를 사용하는 사람들이 도구의 기반이 되는 데이터에 사용된 개별 연구에 대해 알지 못한다는 것입니다. 예를 들어 70세 미만은 임상시험에 참여하지 않거나 어린이는 임상시험에 참여하지 않는 등의 배제 기준이 있었을 수 있습니다. 모든 임상시험에 해당되는 것은 아니지만, 이상값에 대한 예측의 기본 통계적 근거가 편향될 수 있을 정도로 많은 수의 임상시험에 해당될 수 있습니다. 따라서 일반적인 AI 도구에서 경계 조건이 무엇인지 모른다면 도구를 적용하는 것은 매우 위험합니다. (내용전문가(SME) 7) 
The problem there is that the people that are using those tools don’t know about the individual studies that went into the data underlying the tool. There may have been exclusion criteria for, like, people under 70 weren’t admitted into the trial, or children weren’t admitted into the trial, and so on. And that may not have been true of all of the trials, but enough of them that it would bias the underlying statistical basis for predictions for outliers. So, it’s very dangerous to apply a tool, in general AI tools, if you don’t know what their boundary conditions are. (SME 7)

데이터의 대표성은 특히 과소대표된 환자 및 인구 집단에 특정 AI 도구를 적용하는 결정을 내릴 때 매우 중요합니다.
Representativeness of data becomes especially critical in decisions about applying specific AI tools to underrepresented patients and populations.

얼굴 인식 및 치안 분야에서는 소수 민족의 얼굴에 대해 학습되지 않은 도구가 판매되어 근처에 있지도 않은 사람을 부적절하게 용의자로 분류하는 등의 문제가 발생하기도 했습니다. 의료 분야에도 이와 유사한 사례가 있습니다. (내용전문가(SME) 14) 
And we’ve seen issues with that in facial recognition and policing space where tools have been sold that were not trained on minority faces and then inappropriately flagged people as being suspects for things where they were not even anywhere near the vicinity. There’s an analog in medicine to that as well. (SME 14)

내용전문가(SME)들은 편견을 완화하기 위해 커뮤니티 참여와 권한 부여 접근 방식을 권장했습니다.
The SMEs recommended a community engagement and empowerment approach to mitigate bias.

환자와 보호자를 테이블로 끌어들이는 것은 점점 더 넓어지고 있습니다. 그리고 네, 그렇게 하면 속도가 느려집니다. 저도 동의합니다. 하지만 모두를 위한 진전을 이루려면 모든 사람을 테이블에 참여시켜 발언권을 부여하고 영향력을 행사할 수 있는 권한을 부여해야 합니다. (내용전문가(SME) 15) 
Bringing in patients and their caregivers to the table is broadening and widening. And, yes, that slows things down. I agree. But if you wanted to find progress for everyone, then we need to empower and have everyone at the table and give them a power to speak and influence and shape. (SME 15)

지속적인 전문성 개발.
Continuing professional development.

내용전문가(SME)들은 변화의 속도가 빠르기 때문에 보건 전문직 교육 단계 전반에 걸쳐 이니셔티브가 필요하다고 말하며, 후배 학습자가 상급자보다 더 발전할 수 있다고 언급했습니다. 
The SMEs stated that the rapid pace of change created an imperative for initiatives across the phases of health professions education and noted that junior learners might be more advanced than their supervisors.

이러한 상황에서 공동 제작 및 공동 학습이 도움이 될 수 있다고 생각하며, 교수진이 학생 및 레지던트와 함께 자신들이 접하지 못했던 분야에 대해 배울 수 있는 공간을 마련할 수 있습니다. (내용전문가(SME) 2)
This is where I think co-production and co-learning can be helpful, creating space for faculty to kind of learn with their students and the residents in areas that they themselves may not have been exposed to. (SME 2)

의료 전문가를 위한 AI 관련 역량
AI-related competencies for health care professionals

인터뷰 분석, 주제 파악, 전문가 피드백 통합, 연구팀 전체의 반복적인 합의 도출을 통해 최종적으로 의료 전문가를 위한 6개의 AI 관련 역량 영역 진술과 25개의 하위 역량이 공식화되고 구체화되었습니다. 의료 전문가 역량 분류체계 24가 초기 코딩을 용이하게 했지만, 일부 AI 관련 역량은 이미 확립된 여러 영역에 매핑되는 것을 발견했으며 최종 프레임워크를 구성할 때 이 분류체계에 국한하지 않았습니다. 
Ultimately, 6 AI-related competency domain statements and 25 subcompetencies for health care professionals were formulated and refined from analysis of the interviews, identification of themes, incorporation of expert feedback, and iterative consensus development across the study team. Although the health professions competency taxonomy 24 facilitated initial coding, we found that several of the AI-related competencies mapped to multiple established domains and did not limit ourselves to this taxonomy in constructing our final framework.

역량 영역 진술은 다음과 같습니다:

  • (1) AI에 대한 기본 지식: AI가 무엇인지 설명하고 의료 분야에 적용하는 방법을 설명한다.
  • (2) AI의 사회적, 윤리적 의미: 사회, 경제, 정치 시스템이 AI 기반 도구에 미치는 영향과 이러한 관계가 정의, 형평성, 윤리에 미치는 영향을 설명한다.
  • (3) AI로 강화된 임상 진료: 환자 중심 치료 계획을 수립할 때 다양한 정보원을 통합하는 AI 강화 임상 진료 수행,
  • (4) AI 기반 도구의 증거 기반 평가를 수행: 환자 및 집단에 대한 치료를 제공할 때 AI 기반 도구와 그 기반 데이터 세트의 품질, 정확성, 안전성, 맥락적 적절성, 편향성을 평가하고,
  • (5) AI 기반 도구의 워크플로 분석: AI 기반 도구 구현으로 인한 팀, 역할, 책임, 워크플로 변화를 분석하고 이에 적응하며,
  • (6) AI 기반 도구 관련 실무 기반 학습 및 개선: 의료 서비스에서 AI 도구 사용과 관련된 지속적인 전문 개발 및 실무 기반 개선 활동에 참여합니다.

The competency domain statements are as follows:

  • (1) basic knowledge of AI: explain what AI is and describe its health care applications;
  • (2) social and ethical implications of AI: explain how social, economic, and political systems influence AI-based tools and how these relationships impact justice, equity, and ethics;
  • (3) AI-enhanced clinical encounters: carry out AI-enhanced clinical encounters that integrate diverse sources of information in creating patient-centered care plans;
  • (4) evidence-based evaluation of AI-based tools: evaluate the quality, accuracy, safety, contextual appropriateness, and biases of AI-based tools and their underlying data sets in providing care to patients and populations;
  • (5) workflow analysis for AI-based tools: analyze and adapt to changes in teams, roles, responsibilities, and workflows resulting from implementation of AI-based tools; and
  • (6) practice-based learning and improvement regarding AI-based tools: participate in continuing professional development and practice-based improvement activities related to use of AI tools in health care.


그림 2는 역량 영역을 보여줍니다. 목록 1은 AI 관련 역량과 그 하위 역량을 제시합니다. 전문직별 번역의 예로, 부록 디지털 부록 2에는 이러한 역량이 의학전문대학원 교육 인증위원회의 핵심 역량 영역 26과 대략적인 난이도에 매핑된 목록이 나와 있습니다.

Figure 2 depicts the competency domains. List 1 presents the AI-related competencies along with their subcompetencies. As an example of profession-specific translation, Supplemental Digital Appendix 2 (at https://links.lww.com/ACADMED/B334) lists these competencies mapped to the Accreditation Council for Graduate Medical Education core competency domains 26 and approximate level of difficulty.

토론
Discussion

AI 기반 도구의 임상적 사용에 대한 내용전문가(SME)의 일반적인 열의에도 불구하고 거의 모든 인터뷰에 신중함이 스며들어 있었습니다. 사람에 대한 부적절한 분류 및 라벨링, 왜곡된 데이터 세트에 기반한 예측 모델, 기존의 불평등을 의도치 않게 유지하는 강력한 기술 등 잠재적인 위험은 AI 개발의 역사 전반에 걸쳐 목록화되어 왔습니다. 4,27 인공지능의 역사에 대한 최근의 비판적 검토에 따르면 "인공지능은 인공적이지도 지능적이지도 않다"고 주장합니다. 4 AI 기반 도구의 출력은 사용 가능한 데이터에 의해 결정되므로 시스템의 "지능"은 인간의 결정에서 비롯된 기본 데이터 소스의 구조, 가정 및 범위에 의해 결정됩니다. 따라서 강력한 윤리적 지향과 형평성에 대한 헌신이 모든 AI 관련 역량의 핵심이 되어야 합니다. 
Despite the SMEs’ general enthusiasm for clinical use of AI-based tools, a sense of caution permeated nearly all interviews. Potential dangers have been cataloged across the history of AI development, including inappropriate categorization and labeling of people, predictive models based on skewed data sets, and powerful technologies that inadvertently maintained existing inequalities. 4,27 A recent critical review of the history of AI claims that “artificial intelligence is neither artificial nor intelligent.” 4 Outputs from AI-based tools are determined by available data, and therefore system “intelligence” is determined by the structure, assumptions, and scope of underlying data sources, which originate from human decisions. Thus, a strong ethical orientation and commitment to equity should lie at the heart of all AI-related competencies.

수많은 저자와 전문 기관은 AI 관련 지식, 기술 및 태도를 전달하기 위한 체계적인 접근 방식과 이러한 교육 및 학습을 안내할 역량 목록을 개발할 것을 촉구해 왔습니다. 7-10,28-36 의료37 및 간호 교육에서 정보학을 위한 역량이 개발되었으며,38 이들 중 일부는 AI와 관련이 있지만 임상 치료에서 강력한 AI 기반 도구를 사용하는 데 필요한 구체성이 부족합니다. 저자들은 보건 전문직 커리큘럼에 AI 관련 주제를 포함할 것을 권고하고10 다른 저자들은 모범 프로그램을 인용했지만,17,18 역량 목록이 공식화되지는 않았습니다. 
Numerous authors and professional organizations have called for systematic approaches to imparting AI-related knowledge, skills, and attitudes and for the development of a list of competencies that would guide this teaching and learning. 7–10,28–36 Competencies have been developed for informatics in medical 37 and nursing education, 38 and although some of these are relevant to AI, they lack the specificity needed for the use of powerful AI-based tools in clinical care. Although authors have made recommendations for AI-related topics to be included in health professions curricula 10 and others have cited exemplar programs, 17,18 no list of competencies had been formulated.

이 연구는 반구조화된 전문가 인터뷰의 주제별 분석을 통해 의료 전문가를 위한 AI 관련 임상 역량 목록을 생성함으로써 이러한 격차를 해결합니다. 임상 진료를 제공하는 모든 사람이 결국 AI 기반 도구와 상호작용하게 되므로 전문가 간 접근 방식을 의도적으로 선택했습니다. 연구 결과에 따르면 의료 전문직 간에 상당한 연관성이 있으며, 이는 전문직 간 학습자가 협력하여 역량을 개발할 수 있는 잠재적 기회를 나타냅니다. 
This study addresses that gap by generating a list of AI-related clinical competencies for health care professionals through thematic analysis of semistructured expert interviews. An interprofessional approach was deliberately chosen because all who provide clinical care will eventually interface with AI-based tools. Findings suggest significant alignment across health professions, indicating potential opportunities for interprofessional learners to develop competencies collaboratively.

이 연구는 개인의 역량에 초점을 맞추었지만, 내용전문가(SME)은 유능한 개인이 AI 기반 도구와 관련된 구조, 산출물 및 결과를 평가하고 모니터링할 수 있는 'AI 역량 조직' 내에서 기능해야 한다고 강조했습니다. 이러한 조직은 안전과 공정성에 중점을 둔 규제 시스템의 지원이 필요합니다. 최근 의약품 및 의료 기기 개발과 유사한 평가 프로세스가 AI 기반 도구에 대해 설명되었으며,39 다른 사람들은 새로운 실험실 테스트에 대해 유사한 프로세스를 제안했습니다. 40 그러나 두 모델 모두 현지 조직과 규제 기관 간의 책임(및 그에 따른 역량) 배분은 아직 명확히 밝혀지지 않았습니다. 
Although this study focused on individual competencies, the SMEs emphasized that competent individuals need to function within “AI-capable organizations,” which can evaluate and monitor the structure, outputs, and outcomes associated with AI-based tools. These organizations in turn need support from regulatory systems focused on safety and fairness. An evaluation process similar to that for drug and medical device development was recently described for AI-based tools, 39 whereas others have suggested a process similar to that for new laboratory tests. 40 However, in both models, the distribution of responsibilities (and thus capabilities) between local organizations and regulatory bodies remains to be clarified.

직접적인 질문 없이도 투명성은 유능한 임상의를 지원하는 데 필요한 조직과 시스템의 필수적인 특성으로 떠올랐습니다. 조직 차원에서 개방적이고 잘 소통되는 프로세스를 통해 환자와 임상의가 임베디드 툴에 대해 인지할 수 있습니다. 또한 투명성은 도구가 강력한 경쟁적 이해관계가 아닌 환자와 대중의 이익을 위해 기능할 수 있도록 보장하는 안전장치 역할을 합니다. 이러한 투명성을 확보하려면 모든 관련 이해관계자가 의사결정 테이블에 참여해야 하며 새로운 도구를 구현할 때 표준화된 접근 방식을 사용해야 합니다. 5 
Even without direct questioning, transparency emerged as an essential characteristic of the organizations and systems needed to support competent clinicians. Open and well-communicated processes at the organizational level can ensure that patients and clinicians are aware of embedded tools. Transparency also serves as a safeguard to ensure that tools will function primarily for the benefit of patients and populations and not for powerful competing interests. This transparency requires that all relevant stakeholders have seats at decision-making tables and that standardized approaches are used for implementing new tools. 5

개인과 조직의 역량은 시스템 전반의 투명성을 요구하며, 규제 표준은 AI 기반 도구의 구성 방식, 답변하도록 설계된 질문, 학습 및 검증에 사용되는 모집단에 대한 명확한 라벨링을 요구합니다. 1,41-43 복잡한 신경망과 진화하는 알고리즘을 갖춘 도구는 조사하기 어렵고 추가적인 지속적인 감독이 필요합니다. 39,43 
Individual and organizational competency demands system-wide transparency, with regulatory standards requiring clear labeling about how AI-based tools are constructed, the questions they are engineered to answer, and the populations used for training and validation. 1,41–43 Tools with complex neural networks and evolving algorithms will be difficult to interrogate and will require additional ongoing oversight. 39,43

내용전문가(SME)들은 또한 필요한 조직 및 규제 프로세스를 수립하고 감독할 AI 기반 도구의 구현과 관련된 분야에 대한 고급 교육을 받은 새로운 의료 전문가 집단cadre의 필요성에 대해서도 논의했습니다. 학위 프로그램, 펠로우십, 자격증 프로그램 등 이러한 인력의 성격과 교육에 대해서도 추가적인 정의가 필요합니다.
The SMEs also discussed the need for a new cadre of health care professionals with advanced training in areas related to the implementation of AI-based tools who would establish and oversee the needed organizational and regulatory processes. The nature of this workforce and its training, whether through degree programs, fellowships, or certificate programs, also needs further definition.

마지막으로, 역량 기반 교육은 개인이 역량을 습득하는 다양한 방법이 있다고 가정하므로 이 보고서에서는 교수 및 학습 접근 방식에 대한 구체적인 권장 사항을 제시하지 않습니다. 그러나 역량 목록(목록 1)은 학습자의 역량 습득을 돕기 위한 이니셔티브를 설계, 실행 및 결과를 측정하고자 하는 교육자에게 청사진을 제공합니다. 개별 임상의가 이미 보유하고 있을 수 있는 AI 관련 역량은 학습자 수준에 따라 다를 수 있으므로 교육 연속체의 뚜렷한 단계에 역량을 할당하지 않았습니다. 대신, 부록 디지털 부록 2에서는 어떤 역량을 초급으로 간주할 수 있는지, 어떤 역량을 고급으로 간주할 수 있는지 제안합니다. 
Finally, competency-based education assumes there are diverse ways in which individuals gain competencies; thus, this report does not make specific recommendations for teaching and learning approaches. However, the competency list (List 1) provides a blueprint for educators who wish to design, implement, and measure results of initiatives aiming to help learners acquire them. The AI-related competencies that individual clinicians might already possess are likely to vary across and within learner levels; therefore, we did not assign competencies to distinct phases of the educational continuum. Instead, Supplemental Digital Appendix 2 (at https://links.lww.com/ACADMED/B334) suggests which competencies could be considered entry level and which could be considered more advanced.

acadmed_2022_08_31_miller_acadmed-d-21-02684_sdc1.pdf
0.23MB

Competencies for the Use of Artificial Intelligence (AI)–Based Tools by Health Care Professionals

1. Basic Knowledge of AI: Explain what artificial intelligence is and describe its health care applications.

  • a. Identify the range of health-related AI applications.
  • b. Describe contributions from the disciplines of data science, computer science, and informatics to the development of health care AI tools. 
  • c. Summarize the factors that influence the quality of data and explain how they impact the outputs of AI-based applications. 
  • d. Explain how different approaches to data visualization can affect interpretation of the outputs of AI-based tools and the subsequent actions that might be taken. 
  • e. Describe the statistical properties of AI-based tools and explain how they should be used in interpreting outputs.

2. Social and Ethical Implications of AI: Explain how social, economic, and political systems influence AI-based tools and how these relationships impact justice, equity, and ethics.

  • a. Acknowledge personal responsibility for fairness and equity in the use of AI-based tools in health care.
  • b. Describe how system-level factors and regulatory structures influence the implementation of AI-based tools in health care.
  • c. Identify and evaluate how personal and structural biases can impact health data and the outputs of AI-based tools.
  • d. Recognize the potential for use of AI-based tools to reduce or exacerbate health disparities and participate in debiasing activities to mitigate negative impacts.
  • e. Appraise the ethical issues for clinicians, patients, and populations raised by various design, implementation, and use scenarios involving AI.

3. AI-Enhanced Clinical Encounters: Carry out AI-enhanced clinical encounters that integrate diverse sources of information in creating patient-centered care plans.

  • a. Recognize that clinicians are responsible for all patient care decisions, including those that involve support from AI-based tools, and exercise judgment in applying AI-generated recommendations.
  • b. Discern a patient’s information needs, preferences, numeracy, and health literacy levels regarding the use of AI-based tools in their care.
  • c. Explain to patients the concepts of risk and uncertainty as they relate to the outputs of AI-based tools and describe practical implications for their care.
  • d. Integrate information derived from multiple AI and non-AI sources in patient-centered decision-making processes that result in personalized care plans.
  • e. Demonstrate comfort and humility in caring for data-empowered patients and incorporate patient-reported data and outcomes in developing care plans.
  • f. Apply methods of data visualization to facilitate patient understanding of AI-derived data, with sensitivity to possible differential impacts related to race, ethnicity, sex, gender, and social determinants of health.
  • g. Describe how AI-based tools can be used to enhance access and quality of care in remote and underserved settings. 

4. Evidence-Based Evaluation of AI-Based Tools: Evaluate the quality, accuracy, safety, contextual appropriateness, and biases of AI-based tools and their underlying datasets in providing care to patients and populations.

  • a. Access critical information about specific AI-based tools before applying them to patient care, including sources and representativeness of training data, algorithm performance for the question being asked, and how they were validated. 
  • b. Describe how the scope and quality of data sets used in development of AI tools influence their applicability to specific patients and populations.
  • c. Identify potential biases in the design of an AI-based tool, and the implications of those biases for patient care and population health. 
  • d. Collaborate with patients, caregivers, informaticians, and others in the ongoing monitoring of AI-based applications and communicate feedback through established organizational channels.

5. Workflow Analysis for AI-Based Tools: Analyze and adapt to changes in teams, roles, responsibilities, and workflows resulting from implementation of AI-based tools.

  • a. Participate collaboratively in team-based discussions that analyze changing roles, responsibilities, and workflows associated with the adoption of novel AI-based tools and help implement necessary changes.
  • b. Effectively use AI-based tools to facilitate critical communications between all members of health care teams.
  • c. Recognize data and informatics professionals as valuable members of health care teams and collaborate with them in the design of AI tools that address clinical problems.
  • d. Contribute to micro- and macro-system decision-making processes regarding which AI-based tools should augment and which should replace parts of current health care practices.

6. Practice-Based Learning and Improvement Regarding AI-Based Tools: Participate in continuing professional development and practice-based improvement activities related to use of AI tools in health care

 

목록1. 의료 전문가가 인공지능(AI) 기반 도구를 사용하기 위한 역량
List 1 
Competencies for the Use of Artificial Intelligence (AI)–Based Tools by Health Care Professionals

  • 1. 인공지능에 대한 기본 지식: 인공지능이 무엇인지 설명하고 인공지능의 의료 분야 적용 사례를 설명합니다.
    1. Basic Knowledge of AI: Explain what artificial intelligence is and describe its health care applications.
  • 2. AI의 사회적, 윤리적 의미: 사회, 경제, 정치 시스템이 AI 기반 도구에 어떤 영향을 미치는지, 그리고 이러한 관계가 정의, 형평성, 윤리에 어떤 영향을 미치는지 설명합니다.
    2. Social and Ethical Implications of AI: Explain how social, economic, and political systems influence AI-based tools and how these relationships impact justice, equity, and ethics.
  • 3. AI로 강화된 임상적 만남: 환자 중심의 치료 계획을 수립할 때 다양한 정보 소스를 통합하는 AI로 강화된 임상 진료를 수행합니다.
    3. AI-Enhanced Clinical Encounters: Carry out AI-enhanced clinical encounters that integrate diverse sources of information in creating patient-centered care plans.
  • 4. AI 기반 도구의 증거 기반 평가: 환자와 집단에게 치료를 제공할 때 AI 기반 도구와 그 기본 데이터 세트의 품질, 정확성, 안전성, 맥락적 적절성, 편향성을 평가합니다.
    4. Evidence-Based Evaluation of AI-Based Tools: Evaluate the quality, accuracy, safety, contextual appropriateness, and biases of AI-based tools and their underlying datasets in providing care to patients and populations.
  • 5. AI 기반 도구에 대한 워크플로 분석: AI 기반 도구의 구현으로 인한 팀, 역할, 책임, 워크플로우의 변화를 분석하고 이에 적응합니다.
    5. Workflow Analysis for AI-Based Tools: Analyze and adapt to changes in teams, roles, responsibilities, and workflows resulting from implementation of AI-based tools.
  • 6. AI 기반 도구에 대한 실습 기반 학습 및 개선: 의료 서비스에서 AI 도구 사용과 관련된 지속적인 전문성 개발 및 실무 기반 개선 활동에 참여합니다.
    6. Practice-Based Learning and Improvement Regarding AI-Based Tools: Participate in continuing professional development and practice-based improvement activities related to use of AI tools in health care.

제한 사항
Limitations

다양한 전문 분야를 가진 내용전문가(SME)을 의도적으로 선정했지만, 15건의 인터뷰는 일부 관점에 대한 표본을 제공했습니다. 예를 들어, 사회복지사, 병원 관리자 또는 AI 기반 도구와 상호 작용할 기타 의료 전문가는 표본에 포함되지 않았습니다. 또한 이 연구는 미국에서 일하는 전문가로 제한되었으며, 해외 전문가들은 다른 견해를 추가할 가능성이 높습니다. 향후 연구에서는 더 많은 전문가 그룹을 참여시켜 다양한 치료 환경, 팀 역할, 전문 분야에 따라 어떤 역량 요소가 가장 중요한지 알아볼 수 있습니다. 또한 이 연구는 환자 중심의 임상적 AI 사용에만 초점을 맞추었으며 비즈니스 프로세스나 시스템 수준의 운영을 용이하게 하는 도구는 포함하지 않았습니다. 마지막으로, 빠른 변화의 속도를 고려할 때 이러한 역량은 적응이 필요하며 이 목록은 자주 재검토되어야 한다는 점을 인정합니다. 
Although we purposively selected the SMEs for their diverse areas of expertise, 15 interviews provided a small sampling of perspectives. For example, the sample did not include social workers, hospital administrators, or other health care professionals who will interact with AI-based tools. In addition, the study was restricted to experts working in the United States, and international experts would likely add different views. Future research could engage a wider group of experts to determine which competency elements are most critical for different care settings, team roles, and specialty contexts. Furthermore, this study focused only on patient-centered, clinical uses of AI and did not include tools that facilitate business processes or system-level operations. Finally, given the rapid pace of change, we acknowledge that the competencies will need to adapt and that this list should be frequently revisited.

결론
Conclusions

전문가 인터뷰와 주제별 분석을 사용한 이 정성적 연구에서는 의료 전문가가 AI 기반 도구를 사용하는 데 필요한 6개의 역량 영역과 25개의 하위 역량을 확인했습니다. 역량 선언문과 하위 역량은 향후 교육 단계에 걸쳐 의료 전문가를 위한 교육 및 학습 프로그램을 안내하는 데 사용할 수 있습니다. AI 기반 도구의 잠재적 이점을 극대화하고 잠재적 해악을 줄이려면 윤리적으로 유능한 개인과 조직을 개발하는 것이 중요합니다. 
This qualitative study using expert interviews and thematic analysis identified 6 competency domains and 25 subcompetencies needed for the use of AI-based tools by health care professionals. The competency statements and subcompetencies can be used to guide future teaching and learning programs for health care professionals across the phases of education. The development of ethically competent individuals and organizations is critical if the potential benefits of AI-based tools are to be maximized and their potential harms diminished.

 


Acad Med. 2023 Mar 1;98(3):348-356. doi: 10.1097/ACM.0000000000004963. Epub 2022 Sep 6.

Competencies for the Use of Artificial Intelligence-Based Tools by Health Care Professionals

Affiliations collapse

1R.G. Russell is director of learning system outcomes, Office of Undergraduate Medical Education, and assistant professor of medical education and administration, Vanderbilt University School of Medicine, Nashville Tennessee; ORCID: https://orcid.org/0000-0002-5540-7073 .

2L.L. Novak is director, Center of Excellence in Applied Artificial Intelligence, Vanderbilt University Medical Center, and associate professor of biomedical informatics, Vanderbilt University School of Medicine, Nashville, Tennessee; ORCID: https://orcid.org/0000-0002-0415-4301 .

3M. Patel is associate chief health officer and chief medical officer of provider analytics, IBM Watson Health, Cambridge, Massachusetts, and clinical professor, Northeast Ohio Medical University, Rootstown, Ohio.

4K.V. Garvey is research instructor in anesthesiology, Vanderbilt University School of Medicine, and director of operations, Center for Advanced Mobile Healthcare Learning, Vanderbilt University Medical Center, Nashville, Tennessee; ORCID: https://orcid.org/0000-0002-2427-0182 .

5K.J.T. Craig is lead director, Clinical Evidence Development, Aetna Medical Affairs, CVS Health. At the time this work was completed, the author was deputy chief science officer of evidence-based practice, Center for AI, Research, and Evaluation, IBM Watson Health, Cambridge, Massachusetts; ORCID: https://orcid.org/0000-0002-9954-2795 .

6G.P. Jackson is vice president and scientific medical officer, Intuitive Surgical, Sunnyvale, California, and associate professor of surgery, pediatrics, and biomedical informatics, Vanderbilt University School of Medicine, Nashville, Tennessee. At the beginning of this work, the author was vice president and chief science officer, IBM Watson Health, Cambridge, Massachusetts; ORCID: https://orcid.org/0000-0002-3242-8058 .

7D. Moore is emeritus professor of medical education and administration, Vanderbilt University School of Medicine, Nashville, Tennessee.

8B.M. Miller is professor of medical education and administration, Vanderbilt University School of Medicine, and director, Center for Advanced Mobile Healthcare Learning, Vanderbilt University Medical Center, Nashville, Tennessee; ORCID: https://orcid.org/0000-0002-7333-3389 .

PMID: 36731054

DOI: 10.1097/ACM.0000000000004963

Abstract

Purpose: The expanded use of clinical tools that incorporate artificial intelligence (AI) methods has generated calls for specific competencies for effective and ethical use. This qualitative study used expert interviews to define AI-related clinical competencies for health care professionals.

Method: In 2021, a multidisciplinary team interviewed 15 experts in the use of AI-based tools in health care settings about the clinical competencies health care professionals need to work effectively with such tools. Transcripts of the semistructured interviews were coded and thematically analyzed. Draft competency statements were developed and provided to the experts for feedback. The competencies were finalized using a consensus process across the research team.

Results: Six competency domain statements and 25 subcompetencies were formulated from the thematic analysis. The competency domain statements are: (1) basic knowledge of AI: explain what AI is and describe its health care applications; (2) social and ethical implications of AI: explain how social, economic, and political systems influence AI-based tools and how these relationships impact justice, equity, and ethics; (3) AI-enhanced clinical encounters: carry out AI-enhanced clinical encounters that integrate diverse sources of information in creating patient-centered care plans; (4) evidence-based evaluation of AI-based tools: evaluate the quality, accuracy, safety, contextual appropriateness, and biases of AI-based tools and their underlying data sets in providing care to patients and populations; (5) workflow analysis for AI-based tools: analyze and adapt to changes in teams, roles, responsibilities, and workflows resulting from implementation of AI-based tools; and (6) practice-based learning and improvement regarding AI-based tools: participate in continuing professional development and practice-based improvement activities related to use of AI tools in health care.

Conclusions: The 6 clinical competencies identified can be used to guide future teaching and learning programs to maximize the potential benefits of AI-based tools and diminish potential harms.

의학교육에서 효과적인 질문을 위한 12가지 팁(Med Teach, 2020)
12 tips for effective questioning in medical education
Stacey Pylman and Amy Ward

 

소개
Introduction

질문하기는 의학교육자에게 일반적이고 필수적인 교육 관행입니다. 우리는 의학교육자라는 용어를 모든 의학교육 환경을 포함하여 미래의 임상의를 준비하는 모든 사람을 지칭하는 데 사용합니다(Branch et al. 1997). 질문은 철학자 소크라테스의 교육적 전통에 뿌리를 두고 있으며(Neenan 2009; Kost and Chen 2015), 교육자가 학습의 발판을 마련하고 비판적 사고 능력의 개발을 장려하기 위해 사용하는 강력한 전략입니다(Smith 1977; Costa 1985; Garside 1996; Ritchhart 외. 2011). 질문을 효과적으로 실행하면 학습자의 참여도, 집중도, 콘텐츠 이해도에서 긍정적인 결과를 이끌어낼 수 있습니다(Cho et al. 2012). 학습자와 의학교육자 모두 질문이 '학습, 논리적 사고, 자신의 결정에 대한 방어, 빠른 기억, 자기 평가, 동료와의 원활한 의사소통'을 촉진하는 데 유용하다고 생각합니다(Kost and Chen 2015; Goebel et al. 2019). 의학교육자는 임상 환경에서 학습자의 독립성을 향한 진전을 촉진하기 위해 이해, 임상 기술 및 추론을 평가하기 위해 질문도 사용합니다(Bowen 2006). 대부분의 사람들은 질문이 의학교육의 주요 교육 활동 중 하나라는 데 동의합니다(Amin and Khoo 2003; Lo and Regehr 2017). 그러나 의학교육에서 질문 기법을 효과적으로 사용하는 방법에 대한 문헌은 제한적입니다. 
Asking questions is a common and necessary teaching practice for medical educators. We use the term medical educator to refer to anyone preparing future clinicians inclusive of all medical education settings (Branch et al. 1997). Questioning is rooted in the educational tradition of the philosopher Socrates (Neenan 2009; Kost and Chen 2015) and is powerful strategy educators use to scaffold learning and encourage the development of critical thinking skills (Smith 1977; Costa 1985; Garside 1996; Ritchhart et al. 2011). When effectively executed, questioning can elicit positive outcomes in learner participation, concentration, and understanding of content (Cho et al. 2012). Learners and medical educators alike believe questioning is useful to ‘promote learning, logical thinking, defending one’s decisions, quick recall, self-assessment, and communicating well with one’s peers’ (Kost and Chen 2015; Goebel et al. 2019). Questioning is also used by medical educators to assess understanding, clinical skills, and reasoning in order to promote learner progress toward independence in the clinical setting (Bowen 2006). Most agree that questioning is one of the primary educational activities in medical education (Amin and Khoo 2003; Lo and Regehr 2017). However, the literature on how to effectively use the questioning technique in medical education is limited.

비효과적인 질문이 어떤 것인지에 대한 문헌은 상당수 있지만(Detsky 2009, Kost and Chen 2015, Mavis 외. 2014, Stoddard and O'Dell 2016), 어떤 종류의 질문을 해야 하는지, 의학교육자가 학습자의 굴욕감을 예방하는 방법과 같은 효과적인 질문에 대한 문헌은 거의 없습니다(Kost and Chen 2015). 질문하기의 개념은 간단한 것처럼 보이지만, 많은 의학교육자들은 공식적인 교육학 교육을 받지 못했습니다.

  • 종종 의학교육자들은 학습자 시절에 질문을 받았던 것과 같은 방식으로 질문을 하는데, 이것이 항상 가장 효과적인 것은 아닙니다(Kost and Chen 2015).
  • 또한 많은 의학교육자들은 질문이 학습자의 과도한 불안을 증가시킨다고 잘못 생각하고 있습니다(Amin and Khoo 2003).
  • 모든 질문이 동일한 결과를 얻거나 학습자에게 동일한 수준의 사고와 반응을 유도하는 것은 아닙니다(Sachdeva 1996; Wink 1993).

다음 12가지 팁은 의학교육자가 강의실 또는 임상 환경에서 질문을 계획하고, 질문하고, 분석할 때 목적에 맞게 효과적으로 질문하는 데 도움이 될 것입니다.  
There is significant literature about what ineffective questioning looks like (Detsky 2009; Kost and Chen 2015; Mavis et al. 2014; Stoddard and O’Dell 2016), but little about effective questioning, such as what kind of questions should be asked and how medical educators can prevent learner humiliation (Kost and Chen 2015). Although the concept of asking questions seems a simple practice, many medical educators lack formal pedagogical training. Often medical educators ask questions in the same way questions were asked of them when they were learners, which is not always the most effective (Kost and Chen 2015). Additionally, many medical educators mistakenly assume questioning increases undue anxiety for learners (Amin and Khoo 2003). Not all questioning achieves the same results or prompts learners to the same level of thinking and response (Sachdeva 1996; Wink 1993). The following twelve tips will help medical educators be purposeful and effective as they plan, ask, and analyze questions in classroom or clinical settings.

팁 1: 학습의 발판이 되는 질문하기
Tip 1: Ask questions to scaffold learning

질문은 학습자가 알고 있는 내용을 평가하는 방법으로 가장 흔히 생각되며, 이는 효과적인 방법이지만, 질문은 학습자가 이해하도록 비계를 세우는 데에도 사용할 수 있습니다. 스캐폴딩교육자가 학습자에게 더 유능한 다른 사람의 지원 없이 자신의 능력을 넘어서는 목표를 달성할 수 있도록 지원하는 것을 말합니다(Van de Pol 외. 2010). 스캐폴딩은 의학교육자가 학습자가 직면한 지식 격차를 파악한 다음 학습자가 더 발전된 이해를 향해 나아갈 수 있도록 지원을 제공할 수 있어야 하는 역동적인 과정입니다(Van de Pol 외. 2010). 이를 잘 수행하려면 의학교육자는 학습 목표학습자의 현재 지식 수준이라는 두 가지 아이디어를 염두에 두어야 합니다. 의료 교육자가 학습의 시작점과 끝점을 파악한 후에는 효과적인 비계 역할을 하는 질문을 설계할 수 있습니다. 임상 환경에서 학습의 발판을 마련하기 위한 질문은 학습자와 의료 교육자 간의 주고받기를 통해 이루어지는 경우가 많습니다. 보다 공식적인 수업 환경을 이끄는 의료 교육자는 계획 시간을 활용하여 훌륭한 스캐폴딩 질문을 설계할 수 있습니다. 의료 교육자는 질문을 설계할 때 다음 사항을 고려하는 것이 중요합니다:
Questioning is most often thought of as a way to assess what learners know, and it is effective in doing so; however, questioning can also be used to scaffold learners toward understanding. Scaffolding refers to the support educators offer learners to help them accomplish goals that are beyond their capacity without the support of a more capable other (Van de Pol et al. 2010). Scaffolding is a dynamic process that requires the medical educator to be able to identify the knowledge gap faced by a learner, and then offer support for the learner as they move toward a more advanced understanding (Van de Pol et al. 2010). To do this well, the medical educator must have two ideas in mind: the learning objective, and the learner’s current level of knowledge. Once the medical educator has identified the starting and endpoints of learning they can design questions that serve as effective scaffolding. Asking questions to scaffold learning in the clinical setting is often done at the moment through a give and take between the learner and medical educator. Medical educators leading more formal class settings can use planning time to design great scaffolding questions. It is important for the medical educator to think through the following when designing questions:

  • 학습자가 이 주제에 대해 이미 무엇을 알고 있는가?
  • 학습자가 오늘 떠날 때 이 주제에 대해 무엇을 알기를 원하는가?
  • 학습자가 생각하고 지식의 공백을 메우는 데 도움이 되는 질문은 무엇일까요?
  • What do my learners already know about this topic?
  • What do I want my learners to know about this topic when they leave today?
  • What questions can I ask that will help learners think and fill in knowledge gaps?

요령은 학습자가 이미 알고 있는 내용에 접근하는 데 도움이 되는 질문을 하고, 거기에서 추가 질문을 가이드로 삼아 구축하는 것입니다. 예를 들어, 다음 세그먼트에서 의학 교육자(ME)는 질문을 사용하여 학습자가 세균성 중이염이 어떻게 발생하는지 이해할 수 있도록 발판을 마련합니다:
The trick is to ask questions that help learners access what they do know and build from there with further questions as guides. For example, in the following segment the medical educator (ME) uses questions to scaffold the learner to understand how bacterial middle ear infections occur:

  • ME - 이 미생물은 호흡기에서 흔한데 왜 모든 사람이 중이염에 걸리지 않나요?
  • 학습자 - 모르겠어요...
  • 나 - 생각해 보세요, 왜 우리 몸에 유스타키오관이 있을까요?
  • 학습자 - 귀를 배출하기 위해서요.
  • 나 - 네, 중이를 배출하기 위해서요. 하지만 유스타키오관이 체액과 병원균으로 막히면 이 공간은 어떻게 될까요?

  • ME – These microbiotas are common in the respiratory tract so why doesn’t everyone get an ear infection?
  • Learner – I don’t know…
  • ME – So think about it, why do we have a eustachian tube?
  • Learner – To drain the ear.
  • ME – Yes to drain the middle ear. But what happens to this space once the eustachian tube gets blocked with fluid and pathogens?

이 토론을 시작할 때 학습자는 질문에 대답할 수 없습니다. 그러나 의학교육자는 다음 단계의 학습을 지원하기 위해 사고를 발판으로 삼습니다.
At the start of that discussion, the learner is unable to answer the question asked. However, the medical educator scaffolds thinking to support learning to this next level.

강의실에서 스캐폴딩을 할 때 질문을 설계하는 의학교육자는 스스로에게 질문해야 합니다:
When scaffolding in the classroom setting, medical educators designing questions should ask themselves:

  • 질문이 단순히 흥미를 유발하는가? 아니면 질문을 통해 주요 학습 목표에 도달할 수 있을까요? (위긴스와 윌버 2015)
  • 질문이 이상하거나, 직관적이지 않거나, 쉽게 오해할 수 있는 내용을 다루고 있습니까? 아니면 평범하고 비교적 분명한 답이 있는 예측 가능한 질문입니까? (위긴스 및 윌버 2015)
  • Is the question merely engaging? Or will asking it lead toward the main learning objective? (Wiggins and Wilbur 2015)
  • Does the question get at what’s odd, counterintuitive, or easily misunderstood? Or is it a predictable question with mundane and relatively obvious answers? (Wiggins and Wilbur 2015)

의학 교육자가 학습자가 무엇을 알고 있고 무엇을 알아야 하는지 명확하게 이해하면 학습자가 지식 격차를 해소하는 데 도움이 되는 질문을 할 수 있습니다.
When medical educators have a clear understanding of what learners know and what they need to know, they can ask questions to help learners bridge knowledge gaps.

팁 2: 학습에서 개방형 및 폐쇄형 질문의 역할 이해하기
Tip 2: Understand the roles of open and closed questions in learning

다양한 범주의 질문을 이해하고 활용하면 의료 교육자가 특정 학습 목표를 의도적으로 타겟팅할 수 있습니다. 일반적으로 사용되는 범주 구분 중 하나는 개방형 질문과 폐쇄형 질문의 구분입니다.

  • 개방형 질문은 제기된 질문에 대해 다양한 방식으로 사고할 수 있도록 합니다. 개방형 질문에서는 학습자가 어떤 사실을 어떻게 또는 왜 알고 있는지 종합하고 옹호하도록 요구하는 경우가 있습니다(Webb 2009; Ritchhart 외. 2011).
  • 폐쇄형 질문은 정답이 명확하며 학습자가 사실을 회상하는 데 참여해야 하는 경우가 많습니다(Webb 2009; Ritchhart 외. 2011).

Understanding and utilizing different categories of questions allows the medical educator to purposefully target specific learning goals. One commonly used categorical distinction is that of open-ended and closed-ended questions.

  • Open-ended questions allow for multiple ways of thinking about the question posed. Sometimes open-ended questions ask learners to synthesize and defend how or why they know something to be true (Webb 2009; Ritchhart et al. 2011).
  • Closed-ended questions have a clear answer, often requiring learners to engage in the recall of facts (Webb 2009; Ritchhart et al. 2011).

의학교육자는 교육에 두 가지 유형의 질문을 모두 사용해야 합니다. 각 질문 유형은 목적에 따라 학습자가 서로 다른 목표를 달성하는 데 잠재적으로 도움이 됩니다. 예를 들어,

  • 학습자가 심전도를 읽는 것이 목표인 경우 의학교육자는 '이 심전도는 무엇을 보여주나요?"라는 폐쇄형 질문을 할 수 있습니다. 이 폐쇄형 예에서 학습자는 심전도를 읽는 방법을 학습하는 데 있어 중요한 첫 단계인 심전도의 부분을 식별해야 합니다.
  • 그러나 학습자가 심전도를 기반으로 응급 상황과 비응급 상황을 구분하려는 경우 의료 교육자는 '심전도를 보면 이 환자가 응급 상황이라고 생각하십니까, 왜 그렇다고/그렇지 않다고 생각하나요?'와 같은 개방형 질문을 사용할 수 있습니다.

Medical educators should employ both question types in their teaching. Each serves a purpose and each potentially helps learners meet different objectives.

  • For example, if the goal is for learners to read an electrocardiogram, the medical educator might ask the closed-ended question, ‘What can you identify on this electrocardiogram?’ In the closed-ended example, the learner needs to identify the parts of an electrocardiogram, an important first step in learning how to read an electrocardiogram.
  • However, if learners are attempting to differentiate between emergent and non-emergent situations based on an electrocardiogram, the medical educator might use an open-ended question such as, ‘Looking at the electrocardiogram, do you think this patient is emergent, why or why not?’

개방형 질문에서는 학습자가 자신의 생각을 설명하기 위해 증거를 사용하여 선택해야 합니다. 개방형 질문의 또 다른 예로는 '이 사례와 관련된 클리닉에서 경험한 사람이 있나요...'와 같이 정답이 없는 질문이 있습니다. 팁 3번에서 설명한 대로 개방형 질문과 폐쇄형 질문을 더 세분화하여 다양한 수준으로 만들 수 있습니다. 
In the open-ended question, learners have to make a choice using evidence to explain their thinking. Another example of an open-ended question would be a question with no certain answer such as, ‘Has anyone had an experience from the clinic that relates to this case…?’ Open-ended and close-ended questions can be further refined into additional levels, as explained in tip number three.

팁 3: 모든 수준의 질문 사용
Tip 3: Use all levels of questioning

학습자가 단순히 기본적인 사실을 기억하는 것 이상으로 나아가도록 하기 위해 블룸의 분류법(Krathwohl and Anderson 2009)과 같은 질문 프레임워크가 개발되었습니다. 블룸의 분류법은 학습자가 기본 사실을 기억하는 것에서 개념을 종합하는 데 필요한 복잡한 사고로 이동하여 개념에 대한 지식을 개발하는 방법에 대한 사고 구조를 제공합니다(Webb 외. 2013). 블룸의 분류법을 의학교육과 연결하여 Barrett 등(2017)은 블룸의 분류법에 따라 대학원 의학교육의 수술 맥락에서 질문을 연구했습니다. 이스트 버지니아 의과대학(EVMS 2019)에서도 블룸의 분류법을 사용하여 의학교육에서 질문을 안내했습니다. 표 1은 이들의 모든 작업을 각색한 버전으로 결합한 것입니다.

In an effort to push learners beyond simply recalling basic facts, questioning frameworks such as Bloom’s Taxonomy (Krathwohl and Anderson 2009) were developed. Bloom’s Taxonomy provides a structure for thinking about how learners develop knowledge of a concept moving from recalling basic facts towards the complex thinking needed to synthesize concepts (Webb et al. 2013). Connecting Bloom’s Taxonomy with medical education, Barrett et al. (2017) researched questioning in graduate medical education surgical contexts according to Bloom’s Taxonomy. The East Virginia Medical School (EVMS 2019) also used Bloom’s Taxonomy to guide questioning in medical education. Table 1 combines all of their work in an adapted version.

위의 표에는 의료 교육자가 학습자가 블룸의 분류 체계의 단계를 통과하는 데 필요한 비계의 양과 유형을 조정할 때 질문이 어떻게 변화할 수 있는지에 대한 예가 나와 있습니다. 모든 수준의 질문은 다양한 시기에 다양하게 던져야 하며, 어떤 수준의 질문이 학습자가 목표에 도달하는 데 가장 도움이 될지 결정하는 것은 교육자의 몫입니다. 그러나 학습자가 지식이 풍부해지고 능숙해짐에 따라 질문은 고차원적 사고의 질문으로 전환되어야 합니다(Phillips and Duke 2001).
In the table above, we provide examples of how questioning might shift as a medical educator adjusts the amount and type of scaffolding learners need to help move them through the levels of Bloom’s Taxonomy. All levels of questions need to be asked at various times, and it is up to the educator to decide which level of questioning will best support learners in reaching the objective. However, as the learner becomes more knowledgeable and proficient, the questions should move toward higher-order thinking questions (Phillips and Duke 2001).

팁 4: 임상 추론의 모델로서 질문하기
Tip 4: Ask questions as a model of clinical reasoning

임상의는 증상을 조사하고 감별 진단을 위해 노력할 때 환자에게 자주 질문을 합니다. 이러한 기술은 의과대학에서 사례를 통해 생각하는 의사로서 스스로에게 던질 수 있는 질문이나 환자에게 던질 수 있는 질문을 배우면서 시작됩니다. 의료 교육자의 역할 중 하나는 의료 학습자를 위한 전문적인 행동과 사고 방식을 모델링하는 것입니다(Rencic 2011). 임상의는 특히 환자를 진단하고 치료할 때 고유한 사고 패턴을 사용하며, 이를 흔히 '임상적 추론'이라고 합니다(Eva 2005). 이러한 사고 패턴은 학습자가 이러한 유형의 사고를 내면화하고 독립적으로 실행할 수 있을 때까지 학습자에게 명시적으로 가르치고 연습해야 합니다(Myrick and Yonge 2002; Tharp and Gallimore 1991). 질문은 이 과정을 지원하는 데 사용될 수 있습니다. 의료 교육자가 환자 사례를 본 후 학습자에게 구체적인 질문을 할 때, 이는 종종 암묵적으로 학습자의 사고를 유도하는 것입니다. 또한 이러한 방식으로 질문하면 학습자는 자신이 알고 있는 것을 말로 표현하는 연습을 할 수 있습니다(Lo and Regehr 2017). 그러나 의료 교육자가 소리 내어 생각함으로써 자신의 생각을 명확하게 표현하면 초보 학습자에게 특히 유용할 수 있습니다(Ritchhart 외. 2011). 학습자가 의료 교육자가 왜 그 질문을 선택했는지 이해하면 임상의의 전문적인 독백을 들여다볼 수 있는 창이 열립니다. 
Clinicians frequently ask questions of patients as they investigate symptoms and work toward a differential diagnosis. This learned skill begins in medical school by learning questions to ask oneself as a physician thinking through a case, or questions to ask the patient. Part of the role of a medical educator is to model professional behaviors and ways of thinking for medical learners (Rencic 2011). Clinicians engage in unique patterns of thinking, especially when diagnosing and treating a patient – often referred to as ‘clinical reasoning’ (Eva 2005). These patterns of thinking must be explicitly taught to and practiced by learners until the learners are able to internalize this type of thinking and practice it independently (Myrick and Yonge 2002; Tharp and Gallimore 1991). Questions can be used to support this process. When medical educators ask learners specific questions after seeing a patient case, they are often implicitly walking learners through their thinking. Additionally, questioning in this manner encourages learners to practice verbalizing what they know (Lo and Regehr 2017) However, when a medical educator makes their thinking explicit by thinking aloud, this can be especially beneficial for novice learners (Ritchhart et al. 2011). When a learner understands why the medical educator has chosen to ask that question, it allows the learners a window into the professional monologue of the clinician.

예를 들어, 의료 교육자는 호흡 곤란을 겪고 있는 환자의 병상을 떠난 후 학습자에게 '엑스레이를 찍어야 할까요?"라고 물어볼 수 있습니다. 이 질문을 할 때 의료 교육자는 학습자가 엑스레이를 찍어야 한다는 것을 암시하는 징후를 찾아내도록 도와줌으로써 학습자의 사고력을 키우려고 할 수 있습니다. 의대 1학년 학생의 경우, 의학교육자는 '환자의 신체 검사 중에 왼쪽 하부 폐에서 딱딱한 소리가 나는 것을 발견했습니다'와 같은 질문을 통해 학습자의 사고 과정을 보다 분명하게 만들 수 있습니다. 이 소리의 원인을 더 잘 이해하기 위해 엑스레이를 찍어야 하는지 궁금해졌습니다. 이 소리의 원인이 무엇이라고 생각하시나요? 이 예에서 의료 교육자는 학습자가 결론을 도출하도록 하는 질문을 하는 대신, 학습자의 사고를 모델링하고 그 과정에서 학습자가 함께 참여하도록 요청하고 있습니다. 따라서 의학교육자는 질문을 사용하여 임상적 추론의 발달을 모델링하고, 발판을 마련합니다. 
For example, a medical educator might ask learners after leaving the bedside of a patient experiencing difficulty breathing, ‘should we order an X-ray?’ In asking that question, the medical educator may be trying to scaffold learners’ thinking by helping them hone in on what the learner should notice that would suggest an X-ray should be ordered. For an early first-year medical student, the medical educator might make their thinking process more evident by phrasing the question like, ‘I noticed crackling in the left lower lung during the physical exam of our patient. This made me wonder if we should order an X-ray to better understand what was causing that sound. What do you think could be causing the sound?’ In this example, rather than asking a question where the learner is asked to draw their conclusions, the medical educator is instead modeling their thinking and asking the learner to engage along with them in the process. Thus, the medical educator is using questioning to model and scaffold the development of clinical reasoning.

팁 5: 심리적 안전감을 조성하여 '찌르지 말고' 질문하기
Tip 5: Probe, don’t ‘prod,’ by creating psychological safety

학습자에게 고차원적인 사고를 요구하는 도전적인 질문을 할 때는 주의를 기울여야 합니다. 질문(즉, 직접적인 질문)을 받는 의료 학습자는 종종 자신이 도전받는 것이 아니라 'pimped'가 되고 있다고 느낍니다(Lo and Regehr 2017). 다른 연구자들과 마찬가지로 저희도 'pimped'라는 용어와 그 용어가 내포하는 젠더적 의미에 문제를 제기합니다(Martin and Wells 2014; Nagarur et al. 2019). 앞으로는 이를 '프로딩prodding'으로 지칭하며, 의료계에서도 이를 따르기를 권장합니다. 프로딩은 질문자가 모욕감을 주거나 불쾌감을 주거나 지식의 격차를 악의적으로 드러내려는 의도를 내포합니다. 진정한 소크라테스 방식과 prodding는 교육자가 학습자에게 질문을 던지는 교환이라는 점에서 그 차이가 항상 명확하지는 않습니다(Goebel 외. 2019; Kost와 Chen 2015). 소크라테스식 교수법과 프로딩의 차이점은 질문자의 의도와 학습자의 인식에 있습니다(Kost and Chen 2015; Stoddard and O'Dell 2016). 교육 환경이 학습자가 소크라테스식 교수법을 인식하는 데 도움이 되지 않는다면 선의의 의대 교육자의 프로빙 질문조차도 프로딩으로 인식될 수 있습니다(Stoddard and O'Dell 2016). 가장 효과적인 프로빙 질문도 때때로 학습자의 지식의 한계를 넘어설 수 있으므로 심리적 안전이 확립되지 않은 경우 프로빙으로 인식될 수 있습니다(Stoddard and O'Dell 2016). 
Caution must be used when asking learners challenging questions that demand higher-order thinking. Medical learners who are probed with questions (i.e. direct questioning) often feel they are being ‘pimped’ instead of challenged (Lo and Regehr 2017). Like others, we take issue with the term ‘pimping’ and the gendered connotations it implies (Martin and Wells 2014; Nagarur et al. 2019). We will from here on refer to it as ‘prodding’ and we encourage the medical community to follow suit. Prodding implies the intent of the questioner to humiliate, cause discomfort, or malignantly illuminate knowledge gaps. The distinction between the genuine Socratic method and prodding is not always obvious since the essential transaction of both is an exchange in which the educator poses questions to learners (Goebel et al. 2019; Kost and Chen 2015). The differentiation between Socratic teaching and prodding lies in the intent of the questioner and the perception of the learner (Kost and Chen 2015; Stoddard and O’Dell 2016). If the educational environment is not conducive to learners recognizing Socratic teaching, then even a well-intentioned medical educator’s probing questions may be perceived as prodding (Stoddard and O’Dell 2016). The most effective probing questions will occasionally exceed the bounds of learners’ knowledge and may thus be perceived as prodding if psychological safety has not been established (Stoddard and O’Dell 2016).

심리적 안정감을 조성하기 위한 권장 사항(Mavis 외. 2014; Stoddard 및 O'Dell 2016).
Recommendations for creating psychological safety (Mavis et al. 2014; Stoddard and O’Dell 2016).

  • 조사 방법과 조사 이유를 미리 설명하고 학습자에게 조사가 어떤 느낌인지 상기시킴으로써 존중하는 분위기를 조성합니다.
  • 학습자의 적절한 수준에서 질문을 던집니다.
  • 프로빙을 할 때는 질문을 던진 후 침묵 또는 '대기 시간'을 주어 학습자가 질문을 고려하거나 자신의 지식을 되돌아보거나 큰 소리로 생각할 수 있는 기회를 제공함으로써 답변할 수 있는 충분한 시간을 갖도록 합니다.
  • 학습자가 '모르겠다'고 말하거나 다른 사람에게 도움을 요청할 수 있도록 허용합니다.
  • 학습자가 '모릅니다'라고 말할 때 프롬프트를 제공하여 학습자가 알고 있는 내용을 말할 수 있도록 돕고, 학습자가 알고 있는 내용을 연결할 수 있도록 돕습니다.
  • 학습자가 어려움을 겪는 동안 언어적 및 비언어적 커뮤니케이션을 사용하여 학습자에게 지지를 표시합니다(예: 고개 끄덕이기, 미소 짓기, 언어적 격려).
  • 부적절한 수행을 간과하거나 무시하지 말고, 굴욕감이나 분노를 유발하지 않으면서 지식이나 기술의 격차를 명확히 파악하는 자비로운 방식으로 교정합니다(가능하면 비공개로 수행).
  • Create an atmosphere of respect by explaining early on how they will probe, the reason for probing, and reminding learners of what probing will sound like.
  • Ask questions are asked at an appropriate level for the learners.
  • When probing, give sufficient time for answers by allowing silence or ‘wait time’ after posing questions, during which learners have the opportunity to consider the question, reflect on their knowledge, or think aloud.
  • Allow learners to say ‘I don’t know’ or ask for help from others.
  • Prompt when learners ‘don’t know’ to help learners voice what they do know and help them build connections to what is known.
  • Use verbal and non-verbal communication to show support to learners while they are challenged (e.g. nodding, smiling, verbal encouragement).
  • Do not overlook or ignore inadequate performance; however, correct in a compassionate manner that clearly identifies gaps in knowledge or skill without causing humiliation or resentment (often done privately if possible).

학습자의 사고를 조사하는 것은 올바르게 적용하면 의학교육자와 학습자에게 매우 매력적이고 보람이 크지만, 시간과 노력이 많이 들기 때문에 가장 중요한 개념에 신중하게 사용해야 하며 심리적 안전이 보장되는 환경에서 올바르게 실행해야 합니다(Stoddard and O'Dell 2016).
When correctly applied, probing learners’ thinking is very engaging and highly rewarding for medical educators and learners; however, it is time- and effort-intensive, so it should be used judiciously on the most important concepts, and it must be executed correctly in an environment of psychological safety (Stoddard and O’Dell 2016).

팁 6: 시작-응답-평가(I-R-E) 패턴에서 벗어나기
Tip 6: Break away from the Initiate-Respond-Evaluate (I-R-E) pattern

교육자와 학습자 간의 전통적인 질문과 답변 패턴은 I-R-E 패턴을 따릅니다: 시작(교육자), 응답(학습자), 평가(교육자)(Almasi 1996). 다음은 IRE 패턴의 예입니다: 
Traditional question and answer patterns between educators and learners follow an I-R-E pattern: Initiate (educator), Respond (learner), Evaluate (educator) (Almasi 1996). The following is an example of the IRE pattern:

  • 의학 교육자: 토마스, 이 환자의 고혈압을 어떻게 치료해야 할까요?
  • 학습자: 저는 환자에게 몇 가지 생활 습관 변화에 대해 이야기하는 것부터 시작하겠습니다.
  • 의료 교육자: 좋아요, 맞습니다.
  • Medical educator: Thomas, how do we treat hypertension in this patient?
  • Learner: I would start by talking to the patient about some lifestyle changes.
  • Medical educator: Good, that’s right.

의학 교육자의 반응이 긍정적이고 안심이 되는 경우가 많음에도 불구하고, 의학 교육자의 반응은 더 이상의 사고와 토론을 종결시킵니다(Wink 1993). 학습자가 비판적으로 사고하고 고차원적인 수준에서 반응하도록 장려하기 위해 의학교육자는 '세 번째 회전 반응'을 사용하여 추가적으로 질문하고 조사하여 학습을 더 발전시킬 수 있는 다른 학습자의 반응을 유도해야 합니다(Ford-Connors and Robertson 2017). 학습을 늘리기 위해 세 번째 턴 반응을 사용하는 방법에는 다음이 포함됩니다:
Despite the fact that the medical educator’s response is often positive and reassuring when the medical educator responds it ends further thinking and discussion (Wink 1993). In order to encourage learners to think critically and respond at higher-order levels, medical educators need to use the ‘third-turn response’ to prompt and probe further and invite other learner responses which can further learning (Ford-Connors and Robertson 2017). Ways to use the third-turn response to increase learning include:

  • 학습자에게 평가를 공동으로 구성하도록 요청합니다.
    • 예를 들어, '현재로서는 왜 그것이 좋은 선택인지 설명해 줄 수 있나요?
  • Asking learners to co-construct the evaluation. For example, ‘Can someone explain why that is a good option at the moment?’
  • 추가 설명 요청하기
    • 예를 들어, '어떤 상황에서 약학적 개입을 할 수 있는지 설명해 줄 수 있나요?'
  • Asking for a further explanation. For example, ‘Can someone explain under what circumstances we might turn to a pharmaceutical intervention?’
  • 학습자에게 동의하거나 동의하지 않는지 묻기
    • 예를 들어, '이에 대해 이의를 제기할 사람이 있나요? 우리가 고려해야 할 다른 사항이 있나요?'(Wink 1993)
  • Asking learners to agree or disagree. For example, ‘Would someone like to challenge that? Are there other things we should consider?’ (Wink 1993)
  • 임상적 추론을 명시적으로 제시하여 학습자의 대답에 추가로 쌓아나가기
    • 예를 들어, '일반적인 지침에 따라 생활 습관 변화에 대한 대화로 시작해야 하는 것은 맞습니다. 하지만 이 경우에는 이 사람의 배경에 섭식 장애가 있다는 것을 알고 있고, 그 장애를 악화시킬까 봐 걱정됩니다. 그래서 대신...'로 시작할 수 있습니다.
  • Building on the learner’s response by making clinical reasoning explicit. For example, ‘You are correct that based on the usual guidelines we should start with a conversation about lifestyle changes. However, in this case, I know that this person has an eating disorder in their background, and I would worry about exacerbating that. So, instead I might start with…’

각각의 경우에 세 번째 차례 응답을 사용하여 대화와 사고를 멈추지 않고 학습자 토론 과정을 계속하는 것이 목표입니다(Ford-Connors and Robertson 2017).
In each case, the goal is to use the third-turn response to continue the process of learner discussion rather than stop the conversation and thinking (Ford-Connors and Robertson 2017).

팁 7: 학습자가 질문에 답하기 전에 짝을 지어 토론하도록 허용하기
Tip 7: Allow learners to discuss in pairs before answering questions

교실 환경에서 학습자가 답변하기 전에 중요한 질문에 대해 서로 토론할 수 있도록 허용하면 이해가 더 깊어집니다. 토론은 협력적이고 능동적인 학습의 한 형태입니다. Kuh 등(2011)은 학습자가 토론에 적극적으로 참여할 때 단순히 듣기만 할 때보다 더 많은 것을 배운다는 사실을 발견했습니다. 짝을 지어 작업하면 학습자가 참여를 피하는 것이 사실상 불가능하므로 각자가 생각하고, 배우고, 수업의 집단 학습에 추가하는 데 책임을 져야 합니다. Smith 등(2009)의 연구 결과에 따르면 토론 그룹에 속한 학습자 중 원래 정답을 알고 있는 사람이 없더라도 동료 토론을 통해 이해도를 높일 수 있다고 합니다. 또한 협동 학습은 학습자가 자신의 이해에 대해 큰 소리로 생각하고, 오해나 지식의 격차를 파악하고, 다른 사람에게 정보를 가르칠 수 있는 기회를 제공함으로써 지식 유지와 심도 있는 이해에 도움이 되는 것으로 알려져 있습니다(Smith 외. 2009).
In the classroom setting, when learners are allowed to discuss important questions with each other before answering, understanding is deepened. Discussion is one form of cooperative, active learning. Kuh et al. (2011) found when learners are actively participating in the discussion, they learn more than when they merely listen. Working in pairs makes it virtually impossible for learners to avoid participating, thus making each person accountable for thinking, learning and adding to the collective learning of the class. Results from Smith et al. (2009) indicate that peer discussion enhances the understanding, even when none of the learners in a discussion group originally knows the correct answer. Additionally, cooperative learning helps learners think out loud about their understanding, identify misconceptions or gaps in knowledge and give learners an opportunity to teach the information to another person – known to help knowledge retention and deeper understanding (Smith et al. 2009).

분명히 학습자가 답변하기 전에 모든 질문에 대해 짝을 지어 토론할 수 있는 시간은 충분하지 않습니다. 짝을 지어 토론하기에 좋은 질문은 다음을 기반으로 출제됩니다.

  • (a) 학습자가 자주 오해하는 어려운 개념,
  • (b) 학습자가 숙달해야 하는 주요 학습 목표,
  • (c) 정답과 오답이 명확하지 않은 문제

Clearly, there is not enough time to allow learners to discuss all questions in pairs before answering. Questions that are good for paired discussions are asked based on

  • (a) difficult concepts about which learners often have misconceptions,
  • (b) main learning objectives that learners need to master, and
  • (c) issues that do not have a clear right or wrong answer.

팁 8: 질문 후 학습자에게 생각할 시간 주기
Tip 8: Give learners think-time after asking a question

의료 학습자는 결국 스스로 생각하고 빠르게 반응할 수 있어야 하지만, 개념이 새로운 경우 학습자에게는 생각할 시간이 필요합니다(Rowe 1986). 생각 시간을 통해 학습자는 자신의 답을 충분히 생각하고 설명할 수 있습니다. Rowe(1986)는 교육자와 모든 학습자가 적절한 정보 처리 작업을 완료하고, 감정을 소화하고, 구두 반응과 행동을 개발할 수 있도록 교육자와 모든 학습자가 방해받지 않고 침묵하는 뚜렷한 시간으로 정의되는 '생각 시간'의 개념을 구축했습니다. 교육자가 학습자에게 질문을 할 때 학습자는 일반적으로 약 1초 동안 응답을 기다립니다(Rowe 1986). 교육자는 종종 힌트, 자신의 답변 또는 더 빠르게 반응하는 학습자의 답변으로 침묵을 채우고 싶은 유혹을 받습니다(Cho et al. 2012). 한 연구(Tobin 1987)에 따르면 질문을 한 후 최소 3초 이상 기다리는 것이 영어 학습자에게 상당한 이점을 가져다준다고 합니다:
Medical learners need to eventually be able to think and respond quickly on their feet, but when concepts are new, learners need think time (Rowe 1986). Think time allows learners to think through and explain their answers. Rowe (1986) constructed the concept of ‘think-time,’ defined as a distinct period of uninterrupted silence by the educator and all learners so that they both can complete appropriate information processing tasks, digest feelings, and develop oral responses and actions. When educators ask learners questions, they typically wait about one second for a response (Rowe 1986). Educators are often tempted to fill the silence with hints, their own answers, or a response from a quicker responding learner (Cho et al. 2012). A review of research (Tobin 1987) found that waiting at least 3seconds after asking questions, and more time for English language learners, had significant benefits:

  • 학습자의 더 긴 반응
  • 더 많은 학습자 담화
  • 대체 반응의 증가
  • 학습자 반응의 복잡성 및 인지 수준 증가
  • 학습자가 주도하는 담화 증가
  • 더 많은 학습자 간 상호 작용
  • 학습자의 혼란 감소
  • 자신감 향상
  • 더 높은 성취도
  • Longer learner responses
  • More learner discourse
  • Increase in alternative responses
  • Increase in the complexity and cognitive level of learner responses
  • More learner-initiated discourse
  • More learner to learner interactions
  • A decrease in learner confusion
  • More confidence
  • Higher achievement

의학교육자는 질문을 한 후 조용히 셋까지 세어 더 긴 생각의 시간을 제공하도록 도전해야 합니다. 처음에는 침묵이 어색하게 느껴질 수 있지만 의학교육자와 학습자는 침묵에 익숙해질 것입니다. 의학교육자가 의도적으로 생각할 시간을 준다고 말하면 어색함을 어느 정도 완화할 수 있습니다.
Medical educators should challenge themselves to provide longer think-time by silently counting to three after asking questions. The silence may feel awkward at first, but medical educators and learners will become used to it. When medical educators state that they are purposely giving think-time, some awkwardness can be alleviated.

팁 9: 질문을 중단해야 할 때 알기
Tip 9: Know when to stop questioning

스캐폴딩 또는 생각 시간을 제공하더라도 학습자가 제기된 질문에 대답하지 못하는 경우가 있습니다. 학습자는 일부 질문에 답할 수 있지만 모든 질문에 답하지 못할 수도 있습니다. 학습자가 몇 가지 질문에 답할 수 있지만 더 깊이 들어가면 학습자의 능력을 넘어서는 도전을 받게 될 수도 있습니다. 매우 어려운 문제이거나 준비가 되지 않아서 학급 전체가 질문에 답하지 못할 수도 있습니다. 학습자가 질문에 답할 수 없는 경우 의료 교육자에게는 몇 가지 옵션이 있습니다:
There will be times when learners are unable to answer questions posed, even when provided with scaffolding or think-time. A learner might be able to answer some of the questions but not all. A learner might be able to answer a few questions, but when probed further they are challenged beyond capacity. The entire class might not be able to answer a question posed because it is very challenging, or they are unprepared. When questions become unanswerable for learners, medical educators have a few options:

  • 다른 학습자에게 답변을 도와달라고 요청합니다.
  • 학습자에게 조사하여 다음 수업 세션에서 다시 보고하도록 요청합니다(McKibbon 및 Marks 2001).
  • 교수자가 직접 개입하여 답의 일부 설명하기
  • 교수자가 직접 개입하여 답의 전체 설명하기
  • Ask another learner to help answer
  • Ask learners to research it and report back in the next class session (McKibbon and Marks 2001)
  • Step in and explain part of the answer
  • Step in and explain the entire answer

이러한 옵션은 모두 허용되며 상황에 따라 사용해야 합니다. 

  • 예를 들어 학습자가 준비한 결과 답을 알아야 하고 이것이 주요 학습 목표인 경우에는 다른 학습자에게 답을 도와달라고 요청하거나 학습자 그룹에 답을 조사하고 다시 보고하도록 할당하는 것이 더 합리적입니다. 
  • 질문이 학습자를 해당 세션의 학습 목표에서 벗어나게 하는 경우, 학습자에게 해당 질문을 조사하여 보고하도록 요청하거나 학습자가 직접 나서서 답을 설명하는 것이 좋습니다. 
  • 의료 교육자가 부분적으로 설명한 후 학습자가 답을 이해하거나 기억하는 것 같으면 의료 교육자는 학습자에게 나머지 부분을 설명할 기회를 주어야 합니다. 

All of these options are acceptable, and their use should depend on the situation. For example,

  • if the learners should know the answer as a result of their preparation and it is the main learning objective, then it makes more sense to ask another learner to help answer or assign the group of learners to research the answer and report back.
  • If the question is moving learners beyond the learning objectives for that session, it makes sense to ask learners to research it and report back or step in and explain the answer.
  • If a learner appears to understand or remember the answer after a partial explanation from the medical educator, then the medical educator should allow the learner a chance to explain the rest.

의학교육자는 학습자 또는 학습자에게 무슨 일이 일어나고 있는지 주의를 기울이고 답을 알 수 없는 질문에 적절히 대응하는 방법을 조정하는 것이 중요합니다(Koole and Elbers 2014). 의학교육자는 항상 학습자가 준비하지 않았다고 가정해서는 안 되며, 의학교육자가 항상 모든 질문에 개입하여 답변해서는 안 됩니다. 그러나 학습자가 좌절감에 도달했을 때 인지 부하를 완화하기 위해 개입하여 답을 완전히 설명하는 것이 적절할 때가 있습니다(Kirschner 외. 2006). 의료 교육자가 개입하여 설명하는 경우, 학습자가 이해했는지 확인하고 학습자에게 요약하도록 요청하는 등 후속 조치를 취해야 합니다(Van de Pol 외. 2010).

It is important for the medical educator to pay attention to what is happening with the learner or learners and adjust how they respond to unanswerable questions accordingly (Koole and Elbers 2014). Medical educators should not always assume learners didn’t prepare, and medical educators should not always step in and answer the entire question. However, there are times when learners have reached frustration and it is apropos to step in and explain the answer in full to alleviate cognitive load (Kirschner et al. 2006). When medical educators do step in and explain, they need to follow-up by checking for learner understanding and asking the learner to summarize (Van de Pol et al. 2010).

팁 10: 학습자의 반응 듣기
Tip 10: Listen to learner responses

의료 교육자는 학습자의 사고를 돕고 학습의 발판을 마련하기 위해 질문을 할 뿐만 아니라, 학습자의 지식을 평가하고 다음에 무엇을 가르쳐야 할지 결정하기 위해 질문을 합니다(Amin and Khoo 2003; Pylman and Ward 2020). 이러한 작업을 수행하려면 의료 교육자는 학습자의 말에 귀를 기울여야 합니다. 가르치는 순간에는 많은 단어가 사용됩니다. 의학교육자는 향후 교육을 위해 학습자의 반응에 집중하는 방법을 배워야 합니다(Sherin and van Es 2005). 전문 교육자는 학습자의 자료 이해도를 점진적으로 평가하는 체크포인트를 두는 경향이 있습니다(Leinhardt et al. 1991). 의료 교육자는 학습자의 이해를 돕기 위해 다음 질문을 만들기 위해 학습자의 반응에 주의를 기울여야 합니다. 학습자의 반응을 경청하는 것이 당연한 것처럼 보일 수도 있지만, 실제로 의학교육자는 학습자의 정서적 요구와 콘텐츠 요구 모두에 대응하기 위해 경청해야 합니다. 효과적인 교육자는 학습자의 반응에 주의를 기울이며 다음과 같은 점에 주목합니다:
Medical educators ask questions to help learners think, to scaffold their learning, but also to assess learners’ knowledge and decide what needs to be taught next (Amin and Khoo 2003; Pylman and Ward 2020). Accomplishing any of those tasks requires that medical educators listen to their learners. During a teaching moment, many words are spoken. Medical educators must learn how to focus on learner responses for the purpose of future instruction (Sherin and van Es 2005). Expert educators tend to have check-points where they assess learners’ progressive understanding of the material (Leinhardt et al. 1991). Medical educators need to pay attention to how learners respond in order to craft the next question to further their understanding. While listening to learner responses may appear obvious to some, in practice medical educators need to listen in order to respond to both learners’ emotional needs and their content needs. Effective educators pay attention to learner responses noting:

  • 학습자가 개념을 이해했습니까? 어떻게 알 수 있나요?
  • 학습자의 반응은 어느 정도의 이해도를 보여줬나요? 
    • 학습자가 기본 사항만 알고 있었나요, 피상적인 수준의 이해를 보였나요, 아니면 개념을 완전히 이해했나요? 다른 학습자에게 추가를 요청해야 합니까?
  • 오해가 있었나요? 학습자에게 응답에 대해 질문하도록 요청해야 합니까?
  • 학습자에게 이 질문이 너무 어려웠나요? 더 나은 스캐폴드를 위해 어떻게 세분화할 수 있을까요?
  • 학습자의 반응에 따라 수업을 되돌아가야 할까요(다시 가르칠까요), 아니면 앞으로 나아가야 할까요?
  • 학습자가 자신의 반응을 정당화할 수 있었나요(이유를 설명할 수 있었나요)? 교육자는 학습자가 잘못된 이유로 올바른 응답을 하지 않았는지 확인해야 합니다.
  • Did the learner understand the concept? How do I know?
  • What level of understanding did the learner response reveal? Did the learner just know the basics, display a surface level understanding, or provide a full understanding of the concept? Should I call on another learner to add?
  • Was there a misconception? Should I call on learners to question the response?
  • Was this question too difficult for the learner? How can I break it down to better scaffold?
  • Based on learner responses, should I go back (reteach) or forward in my teaching?
  • Was the learner able to justify their response (explain why)? Educators need to make sure learners didn’t give the right response for the wrong reasons.

의료 교육자는 단순히 정답을 찾는 것이 아니라 학습자가 무엇을 말하고 있는지, 그리고 그 응답을 향후 교육에 어떻게 활용할 수 있는지에 대해 학습자의 응답에 귀를 기울여야 합니다.
When medical educators really listen to learner responses, not simply for the correct answer but for what learners may be saying and how responses can be leveraged for further teaching moments.

팁 11: 서면 응답을 위한 질문하기
Tip 11: Ask questions for a written response

질문을 할 때 의료 교육자는 보통 한 명 또는 소수의 학습자로부터 답변을 듣고 모든 학습자가 무슨 생각을 하고 있는지 궁금해합니다. 서면 응답을 요청하는 질문은 강의실 환경에서 의료 교육자가 모든 학습자의 응답을 수집하는 데 도움이 됩니다. 이러한 방식으로 의료 교육자는 개별 학습자 또는 학급 전체의 이해도를 측정하거나 사고 과정을 확인할 수 있습니다.
When asking questions, medical educators usually hear a response from one or a few learners and are left wondering what all learners were thinking. Asking questions for written response helps medical educators in the classroom setting collect responses from all learners. In this way, medical educators are able to gauge understanding or see the thought processes of individual learners or the class as a whole.

서면 응답을 위한 질문은 빠른 쓰기, 퇴실 슬립 또는 디지털 대화를 통해 수행할 수 있습니다.
Asking questions for a written response can be done through quick writes, exit slips, or digital conversations.

  • 빠른 쓰기: 의료 교육자가 질문을 던지고 학습자에게 1~2분 동안 사려 깊은 답변을 작성하도록 요청합니다.
  • 출구 슬립: 수업 세션이 끝날 때 의료 교육자는 '오늘 무엇을 명확히 알게 되었습니까?' '여전히 혼란스러운 점은 무엇입니까?' '오늘 그룹 내 다른 사람에게서 배운 것은 무엇입니까?' 등의 1~2가지 질문을 던집니다. 학습자는 자신의 답변을 작성하여 수업에서 나갈 때 제출합니다.
  • 디지털 대화: 의료 교육자는 의료 교육자 및/또는 학습자가 서로 질문을 하고 답변을 작성하는 공유 온라인 문서를 생성합니다. 이 문서는 수업 세션 전, 도중 및 후에 추가할 수 있습니다.
  • Quick writes: The medical educator poses a question and asks learners to take 1–2 minutes writing their own thoughtful responses.
  • Exit slips: At the end of a class session the medical educator asks 1–2 questions such as ‘What was clarified for you today?’ ‘What still confuses you?’ ‘What is something you learned from another person in your group today?’ Learners write their own answers and hand them in as they leave class.
  • Digital conversations: The medical educator creates a shared online document where the medical educator and/or learners pose questions and write responses to each other. This document can be added to before, during, and after class sessions.

이러한 각 예에서 의학교육자는 평가나 채점을 위한 것이 아니라 학습자의 사고를 평가하여, 향후 교육을 안내하는 데 사용하기 위해 잘 발달된 질문을 하고 있으며, 이를 형성 평가라고 합니다(Heritage 2007). 의학교육자는 학습자에게 답안 작성의 목적이 무엇인지, 그리고 답안 작성은 평가 목적이 아니라 향후 교육을 계획하기 위한 것임을 명확히 알려야 합니다. 응답은 의학교육자가 개별 학습자에 대해 알고자 하는지 또는 수업 전체를 측정하고자 하는지에 따라 익명으로 작성할 수도 있고 그렇지 않을 수도 있습니다.
In each of these examples, the medical educator is asking well-developed questions not to evaluate or grade, but to assess learner thinking and use it to guide further instruction – this is formative assessment (Heritage 2007). Medical educators need to make it clear to learners the purpose of the written responses, and that the writings are not for evaluative purposes but for the purpose of planning future instruction. Responses may or may not be anonymous depending on whether the medical educator desires to know about individual learners or gauge the class as a whole.

팁 12: 질문 분석하기
Tip 12: Analyze the questions asked

교육자는 자신이 어떤 유형의 질문을 하는지, 생각할 시간을 충분히 주는지 또는 I-R-E 패턴의 질문을 사용하고 있는지 알지 못하는 경우가 많습니다(Sherin and van Es 2005). 비디오 또는 오디오 녹화 강의 세션은 의학교육자가 질문을 포함하여 강의의 여러 측면을 분석하는 데 도움이 될 수 있습니다. 질문 분석에는 세션을 녹화하고, 녹음을 듣고, 의학교육자가 질문한 내용을 적고, 다음에 따라 분석하는 과정이 포함됩니다:
It is common for educators to be unaware of the types of questions they ask, whether they give enough think-time, or whether they are using the I-R-E pattern of questioning (Sherin and van Es 2005). Video or audio recording classroom sessions can help medical educators analyze many aspects of their teaching including their questioning. Analyzing questioning involves recording a session, listening to the recording, writing down the questions the medical educator asked, and analyzing according to:

  • 블룸의 분류법 - 질문의 수준
  • IRE 패턴 사용
  • 프로빙 또는 찌르기
  • 학습자 응답의 품질
  • 생각 시간 사용
  • Bloom’s Taxonomy – levels of questioning
  • Use of the IRE pattern
  • Probing or prodding
  • Quality of learner responses
  • Use of think-time

의료 교육자가 임상 환경에서 가르치는 경우, 녹화는 옵션이 아닐 수 있습니다. 대신, 의료 교육자는 학습자에게 의료 교육자가 질문하는 방식에 대한 피드백을 제공하도록 요청하거나 동료에게 특정 질문 기법을 관찰하고 메모하도록 요청할 수 있습니다.
If medical educators are teaching in the clinical setting, the recording might not be an option. Instead, medical educators can ask learners to give feedback on the way medical educators are asking questions, or they can ask a colleague to observe and take note of specific questioning techniques.

결론
Conclusion

질문은 의학교육에서 항상 일반적인 교수 전략입니다. 질문을 잘못 사용하면 학습자가 소외감을 느끼고 학습할 수 있는 위치에 있지 않다고 느끼거나 더 심한 경우 위협이나 굴욕감을 느낄 수 있습니다. 질문을 효과적으로 사용하면 강력한 학습 도구가 됩니다. 따라서 의료 교육자는 질문 기법을 더욱 발전시켜야 합니다. 여기에 제공된 팁을 사용하여 효과적으로 질문하면 의학교육자가 어려움을 겪고 있는 학습자를 지원하고, 고급 학습자에게 도전하고, 학습자의 이해를 평가하고, 학습자의 참여와 참여의 균형을 맞추는 데 도움이 될 것입니다. 의학교육자가 효과적인 질문을 더 많이 할수록 의학교육은 오늘날 의료 분야의 도전에 맞설 준비가 된 사려 깊은 임상의를 양성할 수 있는 가능성을 높일 수 있습니다.Questioning will always be a common teaching strategy in medical education. When used incorrectly questioning can leave learners feeling singled out and not in a position to learn, or worse – threatened or humiliated. When used effectively, questioning is a powerful learning tool. Therefore, medical educators deserve further development around questioning techniques. Effective questioning using the tips provided here will help medical educators support struggling learners, challenge advanced learners, assess learner understanding, and balance learner participation and engagement. The more targeted medical educators can be in asking effective questions, the more likely medical education can develop thoughtful clinicians ready to take on the challenges of today’s medical field.

 


 

Med Teach. 2020 Dec;42(12):1330-1336. doi: 10.1080/0142159X.2020.1749583. Epub 2020 Apr 16.

12 tips for effective questioning in medical education

Affiliations collapse

Affiliation

1Office of Medical Education Research and Development, College of Human Medicine, Michigan State University, East Lansing, MI, USA.

PMID: 32297833

DOI: 10.1080/0142159X.2020.1749583

Abstract

Questioning is one of the most frequently used and powerful teaching strategies across levels and settings in medical education. Although the concept of asking questions may seem like a simple practice, many medical educators lack pedagogical training. When effectively executed, questioning can elicit positive outcomes in learner participation, concentration, and understanding of content. When used incorrectly, questioning can leave learners feeling singled out and not in a position to learn, or worse - threatened or humiliated. There is a lot of literature in medical education about what ineffective questioning looks like, but little about how to enact effective questioning, such as what kind of questions should be asked and how to design those questions to improve learning. The following twelve tips will help medical educators be purposeful and effective as they plan, ask, and analyze questions in classroom or clinical settings.

Keywords: Methods; clinical; general; lectures/large group; small group teaching.

전공의 수련과정에서 직접관찰의 현상학: 밀러의 'does'레벨은 관찰가능한가? (Med Educ, 2022)
A phenomenology of direct observation in residency: Is Miller's ‘does’ level observable?
Chris B. T. Rietmeijer1 | Suzanne C. M. van Esch2 | Annette H. Blankenstein1 | Henriëtte E. van der Horst1 | Mario Veen3 | Fedde Scheele4 | Pim W. Teunissen5 

 

 

1 서론
1 INTRODUCTION

직접 관찰(DO)은 역량 기반 의학교육(CBME)의 초석이며, 형성적 및 총체적 목적을 가진 작업장 기반 평가(WBA)의 핵심입니다.1, 2 그러나 졸업후의학교육(PGME)에서 DO의 활용은 저조합니다.1, 3, 4 PGME에서 DO에 관한 문헌은 불분명한 이해관계, 평가에 대한 두려움, 환자와의 상호작용의 어려움, 관찰을 요청하거나 제공하는 것과 상충되는 자율성과 효율성에 대한 기대 등 이러한 저조한 활용에 대한 충분한 설명을 제공합니다.3-12 한 가지 중요한 반복되는 발견은 '관찰자 효과'입니다. Ladonna와 동료들이 발견한 바와 같이, 관찰하는 레지던트들은 마치 '공연을 하는 것'처럼 느껴져 환자에게 덜 자연스럽게 행동하고 '평소'의 진료 방식을 '교과서적인' 접근 방식으로 바꾼다고 합니다. 이러한 '진정성 없는 행동'에 대한 피드백은 레지던트들에게 유용한 것으로 간주되지 않았습니다.6 
Direct observation (DO) is a cornerstone of competency-based medical education (CBME); it is at the heart of workplace-based assessment (WBA) with its formative and summative purposes.1, 2 Yet the uptake of DO in postgraduate medical education (PGME) is poor.1, 3, 4 The literature on DO in PGME provides ample explanations for this poor uptake, such as unclear stakes, fear of assessment, difficulties in interacting with patients and expectations concerning both autonomy and efficiency that conflict with asking for, or offering, observation.3-12 One important recurring finding is the ‘observer effect’. As Ladonna and colleagues found, observed residents felt as if they were ‘staging a performance’; they behaved less naturally towards patients and ‘they exchanged their ‘usual’ practice for a ‘textbook’ approach’. Feedback on this ‘inauthentic behaviour’ was not considered useful by these residents.6

진정성 없는 행동이 '임상 역량을 평가하기 위한 밀러 피라미드의 최상단에 있는 '한다'에 대한 평가'로서 DO의 가치에 심각한 위협이 된다는 것은 의심의 여지가 없습니다.1, 13 '한다' 수준의 평가는 학습자가 임상 상황에서 독립적으로 기능할 수 있는 능력을 평가하는 것입니다.13 코건과 동료들은 DO에 대한 지침에서 '학습자가 관찰을 받을 때 실제와 다른 의료행위를 할 수 있다(예: 환자 병력을 기록할 때 전자 건강 기록을 입력하지 않거나 더 집중적인 검사가 적절한데도 종합적인 신체 검사를 하는 경우)'는 점을 인식했습니다. 그러나 저자들은 '관찰자는 학습자가 실제 업무 행동에 대한 피드백을 받을 수 있도록 학습자가 '평소에 하던 대로' 하도록 격려해야 한다'고 말하며 이 문제에 반박합니다.1 
There can be no doubt that inauthentic behaviour is a serious threat to the value of DO as ‘an assessment of “does” at the top of Miller's pyramid for assessing clinical competence’.1, 13 Assessment of the ‘does’ level is an assessment of the learner's ability to function independently in clinical situations.13 In their guidelines on DO, Kogan and colleagues recognised that ‘learners may default to inauthentic practice when being observed (e.g., not typing in the electronic health record when taking a patient history or doing a comprehensive physical exam when a more focused exam is appropriate)’. But the authors counter this problem by stating that ‘observers should encourage learners to “do what they would normally do” so that learners can receive feedback on their actual work behaviour’.1

동일한 지침에 따르면, 이를 더 쉽게 하기 위해 감독자는 물리적으로 그 상황에 있지만, 예를 들어 환자의 시야에서 벗어난 곳에 앉는 등 가능한 한 적게 존재해야 합니다.1 
To make this easier, according to the same guidelines, supervisors, while physically being in the situation, should be as little present as possible, for example, by sitting out of the patient's line of sight.1

그러나 이러한 슈퍼바이저의 역할은 일반 진료(GP) 교육에서 환자의 진료 상황 경험에 대한 최근 현상학적 연구와 상충되는 것으로 보입니다.14 환자들은 여러 가지 이유로 슈퍼바이저가 어느 정도 대화에 참여하기를 원했고, 실제로 그렇게 했습니다. 그 이유 중 하나는 환자들이 [후배의 접근에 대한 선배의 승인]이 필요했기 때문입니다. 이를 심각하게 받아들이면 선배의사가 [가능한 한 대화에 벗어나있는 것]이 아니라, 반대로 [신중하게 대화에 참여]하는 방향으로의 변화를 의미할 수 있습니다.14 
This take on the supervisor's role, however, seems to conflict with our recent phenomenological research on patients' experiences in DO situations in general practice (GP) training.14 Patients, for several reasons, needed—and indeed caused—supervisors to participate in the conversation to some extent. One of those reasons was that patients needed the senior's approval of the junior's approach. Taking this seriously would imply a movement in the opposite direction, where a supervisor judiciously participates in the activity, rather than keeping out of it as much as possible.14

그러나 이러한 대조적인 통찰은 단 하나의 연구, 즉 환자의 관점에 대한 연구로만 뒷받침됩니다.14 중요한 것은 현상학적 접근법을 통해 환자의 관점에 대한 이해를 얻었으며, 즉 환자가 진료 상황에서 본질적으로 경험하는 것의 규칙성을 조사했다는 점입니다. 빈과 시안치올로가 조언했듯이, 의학교육에서 지속적인 문제(대부분의 교육 상황에서 DO가 부족하다는 점 등)에 직면할 때 우리는 '속도를 늦춰야 할 때 속도를 늦출 수 있도록 하여, 연구 대상과 더 직접적으로 소통하고, 우리의 가정을 드러내고, 새로운 각도에서 성가신 문제를 해결하도록 돕는' 철학적 접근 방식을 취해야 합니다.15 현상학은 그러한 접근 방식이며, 이를 통해 DO 상황에서 환자가 필요로 하는 것의학교육이 DO를 개념화하는 방식 사이의 불일치를 볼 수 있게 해줍니다.1, 14 DO 상황에서 레지던트(및 수퍼바이저)의 경험에 대한 유사한 현상학적 이해가 부족하며, 모든 관점에서 DO 상황에 대한 보다 완전한 이해를 바탕으로 제기된 질문에 대한 답을 찾기 위해 필요합니다. 따라서 우리는 현상학적 접근 방식을 따라 다음을 조사했습니다. 
This contrasting insight, however, is supported by only one study, on one perspective, that is, that of patients.14 Importantly, we obtained our understanding of the patients' perspective by following a phenomenological approach, meaning that we investigated regularities in what patients essentially experienced in DO situations. As Veen and Cianciolo advised, when facing persistent problems in medical education (such as the lack of DO in most training contexts), we should take a philosophical approach that ‘empowers us to slow down when we should, thereby engaging us more directly with our subjects of study, revealing our assumptions, and helping us address vexing problems from a new angle’.15 Phenomenology is such an approach; it enabled us to see the discrepancy between what patients needed in DO situations and how medical education conceptualises DO.1, 14 A similar phenomenological understanding of residents'(and supervisors') experiences in DO situations is lacking and needed to find answers to the questions that have arisen, based on a more complete understanding of DO situations from all perspectives. We, therefore, followed a phenomenological approach to investigate

수퍼바이저가 직접 환자를 관찰하는 동안 레지던트들이 환자와 함께 일할 때 어떤 규칙적인 경험을 했는지 조사했습니다.
the regularities in how residents essentially experienced working with a patient while a supervisor was physically present, observing them.

앞서 언급한 바와 같이, DO는 WBA의 핵심입니다.1, 2 그러나 현상학적 접근은 예를 들어 목적 등의 측면에서 현상을 미리 정의하지 않고 조사하는 것을 의미합니다.16 따라서 우리는 DO의 목적이나 다른 정의에 관계없이 위의 연구 질문에 정의된 대로 DO 상황을 조사했습니다.
As indicated, DO is central to WBA.1, 2 A phenomenological approach, however, implicates investigating phenomena without pre-defining them, in terms of their purposes for example.16 We therefore investigated DO situations as defined in the research question above, regardless of the purposes or other definitions of DO.

2 방법
2 METHODS

현상학적 접근
Phenomenological approach

우리는 네덜란드의 한 GP 수련 센터에서 현상학적 인터뷰 연구를 수행했습니다. 
We performed a phenomenological interview study in one Dutch GP training centre.

의학교육 문헌에서는 종종 해석적(또는 해석학적) 현상학서술적(또는 초월적) 현상학을 구분한다.17, 18 그러나 Rietmeijer와 Veen은 저자들이 특정 학파에 가입하기보다는 현상학을 어떻게 이해하고, 연구에서 현상학의 원칙을 어떻게 적용했는지 명확히 할 것을 제안했다.16 이제 이러한 원칙과 우리가 사용한 방법을 설명한다. 
Medical education literature often distinguishes interpretive (or hermeneutic) phenomenology from descriptive (or transcendental) phenomenology.17, 18 However, Rietmeijer and Veen proposed that, rather than subscribing to a specific school, authors should make clear how they understand phenomenology and how they applied principles of phenomenology in their study.16 We now describe these principles and the methods we used.

2.1.1 사전 반성적 경험의 공통 구조
2.1.1 Common structures in pre-reflective experience

우리는 전공의들이 DO 상황에서 무엇을 경험했는지, 즉 이러한 상황에 대해 반성하기 전에 '반성 전 경험'을 조사했습니다. 이 반성 전 경험을 조사하는 것은 달성할 수 없는 이상이지만, 우리의 목표는 참가자들의 반성, 아이디어 및 의견을 통해 반성 전 경험의 공통 구조에 대해 알아보는 것이었습니다. 이러한 공통 구조규칙성 또는 불변 구조 또는 경험의 본질이라고도 불립니다.16, 19, 20
We investigated what residents experienced in DO situations before they had reflected on these situations: the ‘pre-reflective experience’. Although investigating this pre-reflective experience is an unattainable ideal, our goal was to learn what participants' reflections, ideas and opinions revealed about the common structures of this pre-reflective experience. These common structures are also called regularities, or invariant structures, or essences of the experience.16, 19, 20

2.1.2 이론에 얽매이지 않는 개방적 연구, 브래킷 사용
2.1.2 Open, theory-free; bracketing

우리는 의도적으로 참가자의 역할, 방법 또는 목표와 같은 DO 상황에 대한 이론 없이 이 연구를 시작했습니다. 우리는 상황 자체가 전공의들에게 어떻게 발생하는지에 초점을 맞췄습니다.16 현상학적 원칙에 따라 이러한 개방적인 접근 방식을 통해 현상을 미리 정의하고 관심 대상을 좁혔더라면 발견하지 못했을 현상의 측면을 볼 수 있었습니다.16 
We deliberately started this study without a theory on DO situations, for example, in terms of participants' roles, methods or goals. We focused on how the situation in itself occurred to residents.16 In line with phenomenological principles, this open approach enabled us to see aspects of the phenomenon that would remain unnoticed had we pre-defined it and narrowed our object of interest.16

이러한 개방성을 얻기 위해 우리는 조사 대상에 대한 '자연스러운 태도'를 '괄호'(=보류)해야 했습니다.16, 19-22

  • 자연스러운 태도에서는21 레지던트, 환자, 감독자, DO 상황 사이의 관계에 대한 가정을 당연한 것으로 간주하게 됩니다. 즉, 인터뷰를 시작하기 전에 이미 교육 이벤트와 같은 DO 상황을 미리 정의하고 모든 참가자에게 특정한 역할을 부여합니다.
  • 반면에 현상학적 태도를 취하는 경우, 우리는 자신과 다른 참가자, 상황 사이의 관계에 대한 참가자의 경험에 정확히 관심을 갖습니다.16, 19-22

In order to attain this openness, we had to ‘bracket’ (= suspend) our ‘natural attitude’ towards our object of investigation.16, 19-22 

  • With a natural attitude,21 we would take our assumptions about relations between the resident, the patient, the supervisor and the DO situation for granted. In other words, before starting the interviews, we would already have predefined DO situations, for instance, as a teaching event, with particular roles for all the participants.
  • With a phenomenological attitude, by contrast, we are precisely interested in participants' experiences of these relationships between themselves, the other participants and the situation.1619-22

[가정에 대한 반성적 성찰]은 모든 질적 연구에서 일반적이지만, 현상학에서 괄호치기는 그보다 더 나아가, 현상에 대한 시야를 좁힐 수 있는 이론적, 개념적 아이디어를 중단하는 것을 의미합니다

  • 이는 종종 후설의 '사물 그 자체로'라는 말과 함께 언급됩니다.21 따라서 연구자는 자신의 자연스러운 태도와 인터뷰 대상자의 자연스러운 태도를 끊임없이 의식해야 합니다.22
  • 따라서 괄호치기는 인터뷰와 분석 과정에서 똑같이 중요했습니다. 이를 위해서는 DO에 대한 의견과 이론을 끊임없이 중단하고 전공의들이 DO 상황을 어떻게 경험했는지, 그리고 이러한 경험의 공통된 구조가 무엇인지에 대해 그들이 밝힌 내용으로 되돌아가는 것이 수반되었습니다.22

Although reflexivity on one's assumptions is common in all qualitative research, in phenomenology, bracketing goes further than that and means suspending theoretical and conceptual ideas that may narrow one's sight of the phenomenon.

  • This is often referred to with Husserl's dictum ‘to the things themselves’.21 Researchers must, therefore, constantly be aware of both their own natural attitude and the natural attitude of the interviewees.22 Bracketing was, consequently, equally important during the interviews and the analysis of them.
  • This entailed constantly suspending opinions and theories about DO that arose and bringing them back to what they revealed about how residents experienced DO situations and what were the common structures of this experience.22

인터뷰를 시작하기 전에 CBTR과 SCMvE는 각각 DO 상황에서 레지던트가 된 경험에서 중요한 측면이라고 생각하는 것에 대한 에세이를 썼습니다. 이 에세이에서 그들은 DO에 대한 이전 연구 결과를 포함하여 DO 상황에 대해 당연하게 여기는 자신의 자연스러운 태도에 대해서도 성찰했습니다.10, 11, 14 이후 그들은 이 에세이에 대해 서로 인터뷰하고, 인터뷰 및 분석 기간 동안 반성적 일기와 추가 메모 작성의 시작으로 이러한 성찰을 활용했습니다. 이 연습을 통해 밝혀진 한 가지 예로, 두 연구자 모두 후배 의사는 선배 의사로부터 배워야 하며, 그 역할을 DO가 해야 한다고 확신하고 있는 것으로 나타났습니다. 이것이 그럴듯해 보이지만, 이 의견 및 다른 의견/이론(예: 서론에서 설명한 바와 같이)을 의도적으로 중단함으로써, 그들은 인터뷰 과정에서 레지던트들의 사전 성찰적 경험에 기여하는 DO 상황의 다른 측면도 보기 위해 더 민감해지려고 노력했습니다
Before starting the interviews, CBTR and SCMvE each wrote an essay on what they thought were important aspects of the experience of being the resident in a DO situation. In these essays, they also reflected on their natural attitude, what they tend to take for granted regarding DO situations, including findings from their previous research on DO.10, 11, 14 They subsequently interviewed one another about these essays and used these reflections as the start of a reflexive diary and further memo writing throughout the interview and analysis period. As one example of what this exercise revealed, it appeared that both researchers were convinced that a junior doctor must learn from a senior doctor, with DO playing a role. However plausible this seems, by deliberately suspending this and other opinions/theories (e.g., as described in the introduction), they tried to become more sensitive in their interviews to see also other aspects of DO situations that contributed to residents' pre-reflective experience.

컨텍스트
Context

저희는 네덜란드 서부에서 인터뷰를 진행했습니다. 네덜란드의 GP 수련은 역량 기반의 3년 수련 프로그램으로, 레지던트는 첫해와 마지막 해를 GP에서 보내며 교대로 근무하는 두 명의 GP 트레이너의 감독 하에 일합니다. 레지던트는 주간 퇴원 프로그램을 위해 매주 1일씩 수련 기관을 방문합니다. 수퍼바이저와 레지던트는 수련 기관에서 환자 진료 중에 의사 또는 관찰자가 되어 번갈아 가며 정기적인 양방향 DO 세션에 참여하도록 점점 더 많이 권장하고 있습니다. 인터뷰 당시 실제로 이러한 조언을 받아들이는 비율은 점점 증가하고 있었지만 여전히 적당한 수준이었습니다. 저자는 해당 수련 기관에서 근무하지 않았으며 인터뷰에 응한 레지던트들과는 아무런 관계가 없습니다. 
We performed our interviews in the western part of the Netherlands. Dutch GP training is a competency-based, 3-year training programme; residents spend their first and final years in GP, working under the nearby supervision of one—sometimes two alternating—GP trainers. Residents visit their academic training institute 1 day each week for their day release programme. Supervisors and residents are increasingly encouraged by the training institute to engage in regular bi-directional DO sessions, taking turns being the doctor or the observer, during patient care. The take-up of this advice in practice, at the time of our interviews, was growing but still moderate. The authors did not work at the training institute and had no relationship with the residents interviewed.

데이터 수집
Data collection

2021년에는 무작위로 선정된 총 30명의 1~3년차 레지던트에게 DO 상황에서의 경험에 대한 인터뷰에 응해달라는 이메일 초대장을 보냈습니다. 수락한 레지던트는 SCMvE 또는 CBTR에서 인터뷰를 진행했습니다. 인터뷰는 코로나19 팬데믹으로 인해 물리적 만남이 제한되었기 때문에 화상 통화로 진행되었습니다. 인터뷰는 '환자를 진료하는 동안 상사가 방에 함께 있어 환자를 관찰했던 상황에 대해 말씀해 주시겠습니까'라는 첫 질문 외에는 미리 정해진 질문이 없다는 점에서 비구조화22되었습니다. 그러나 경험의 방식을 이해하려는 우리의 목표는 우리가 사용한 질문 유형에 영향을 미쳤습니다: 저희는 반 마넨을 따라 의도적으로 '경험의 실존적 요소'를 찾았습니다. 반 마넨은 사람들이 다음에서 경험한다고 주장합니다.20

  • 신체(예: 무엇을 느끼고 무엇을 하는지),
  • 시간(예: 언제 무슨 일이 일어나고 시간이 얼마나 빨리 가는지),
  • 장소(예: 누가 어디에 앉고 가구의 위치),
  • 관계(예: 환자와의 친숙함 및 수련 관계의 질)

경험의 방법을 파악하기 위해 이러한 경험의 실존적 요소에 따라 구체적인 DO 상황에서 어떤 일이 일어났는지 매우 사실적으로 질문했습니다.20, 22 인터뷰 시간은 60분에서 75분까지 다양했습니다. 
In 2021, we sent an email invitation to a total of 30 first- and third-year residents, randomly chosen, to be interviewed about their experiences in DO situations. Those who accepted were interviewed by either SCMvE or CBTR. The interviews took place via video calls because physical encounters were restricted because of the Covid 19 pandemic. The interviews were unstructured22 in the sense that there were no pre-fixed questions other than the opening question: ‘Can you tell me about a situation in which your supervisor was present in the room, observing you while you were working with a patient?’ However, our aim of understanding the how of the experience did influence the type of questions that we used: We followed van Manen by deliberately looking for his ‘existential elements of experience’. Van Manen claims that people experience things

  • in their body (e.g., what they feel and what they do),
  • in time (e.g., what happens when and how fast the time goes),
  • in place (e.g., who sits where and position of furniture) and
  • in relationship (e.g., familiarity with the patient and quality of the training relationship).20 

To get to the how of the experience, we asked quite factually what happened in specific DO situations, guided by these existential elements of experience.20, 22 The interviews varied in length from 60 to 75 min.

분석
Analysis

인터뷰는 비디오로 녹화되었습니다. CBTR과 SCMvE는 먼저 비디오 녹화를 개별적으로 분석하여 이 인터뷰가 주제에 대해 말하는 내용을 한두 개의 문구(즉, 'sententious phrases'[반 마넨])로 캡처하여 전체적으로 분석했습니다.20 그런 다음 인터뷰를 필사하고 익명화했으며, CBTR은 이 필사본을 반 마넨의 다양한 렌즈인 [살아있는 신체, 살아있는 공간, 살아있는 시간, 살아있는 관계]를 통해 네 차례에 걸쳐 코딩하여 분석했습니다.20 이 네 가지 렌즈 사용은 경험의 이러한 모든 측면에 더 민감하게 반응하도록 만들었습니다. 이는 분석에서 중요한 단계였습니다. 그러나 이 작업의 목적은 경험을 네 가지 범주로 설명하는 것이 아니라 경험에 대한 보다 완전한 그림을 얻는 것이었습니다.20 따라서 결과 섹션에서는 이러한 실존적 요소에 대해 보고하지 않고 반복적이거나 공통적인 구조로 사전 반영 경험을 세분화할 것입니다.16, 19, 20
The interviews were videotaped. Both CBTR and SCMvE first—separately—analysed the video recordings holistically by capturing in one or two phrases what this interview told them about our topic (i.e., ‘sententious phrases’ [van Manen]).20 They then transcribed and anonymised the interviews, and CBTR analysed these transcriptions in four rounds of coding through Van Manen's different lenses of lived body, lived space, lived time and lived relationship.20 Using these four lenses made us more sensitive to all these aspects of the experience. It was an important step in the analysis. The aim of this, however, was to gain a more complete picture of the experience, not to describe the experience in four categories.20 Therefore, in the results section, we will not report on these existential elements but will break down the pre-reflective experience in recurrent, or common, structures.16, 19, 20

CBTR은 사전 반성적 경험의 특정 공통 구조(예: '평가자로서의 수퍼바이저에 대한 전공의의 인식')에 대해 드러나는 것으로 보이는 코드를 해석하여 코드를 그룹화했습니다. 그는 '상상적 변형' 과정을 통해 이러한 공통 구조를 파악했습니다.

  • 상상적 변형이란 이 구조가 없더라도 같은 경험이 될 수 있는지 스스로에게 물어보는 것을 의미합니다. 대답이 '아니오'면 공통 구조로 간주했습니다.16, 20

CBTR grouped the codes by interpreting what they seemed to reveal about specific common structures of the pre-reflective experience (e.g., ‘residents' awareness of the supervisor as an assessor’). He determined these common structures through a process of ‘imaginative variation’.

  • Imaginative variation means asking oneself if the experience would still be the same experience without this structure. If the answer was no, it was a common structure.1620

CBTR은 이 과정에서 반성적인 메모를 작성했습니다. 그런 다음 이 모든 자료를 SCMvE에게 보냈고, SCMvE는 직접 녹취록을 읽고 코드, 코드 그룹, 메모에 댓글을 달고 더 많은 코드와 메모를 추가했습니다. SCMvE와 CBTR은 각 인터뷰가 끝난 후 화상 통화를 통해 결과를 논의했습니다. 3번과 6번의 인터뷰가 끝난 후 PWT는 이들과 함께 회의에 참여하여 지금까지의 분석 결과를 검토했습니다. 분석에 대한 추가 검토는 MV, AHB, HEvdH, FS를 포함한 전체 팀과 두 차례의 회의에서 이루어졌으며, 이들은 코드, 코드 그룹 및 메모의 예와 분석 시스템에 대해 의견을 제시했습니다. 
CBTR wrote reflexive memos during this process. He then sent all this material to SCMvE who read the transcripts herself, commented on codes, code groups and memos, and added more codes and memos. SCMvE and CBTR discussed their findings during video calls, after each interview. After three and six interviews, PWT joined them in a meeting to review the analyses thus far. A further review of the analyses took place in two meetings with the whole team, including MV, AHB, HEvdH and FS, who commented on examples of codes, code groups and memos and on the system of analysis.

3 결과
3 RESULTS

총 6명의 레지던트를 인터뷰했는데, 그 중 5명은 수련 1년차 후반에, 1명은 수련 3년차에 있었습니다. 모든 레지던트는 전체 상담 과정에서 지도의의 관찰을 받은 경험이 있었습니다. 이러한 DO 세션은 형성적인 목적으로 진행되었습니다. 우리가 들은 대부분의 이야기는 이러한 예정된 DO 상황에 관한 것이었지만, 일부는 환자상담 중에 조언을 구하기 위해 슈퍼바이저를 호출하여 발생한 부가적ad hoc 관찰에 관한 것이었습니다. 우리는 인터뷰를 분석하여 전공의의 사전 성찰적 경험의 공통 구조에 대해 밝혀진 내용을 해석했습니다. 다음 단락에서는 이러한 공통 구조에 대해 보고합니다.
We interviewed a total of six residents, five of them in the second half of their first year, and one in her third year of the training. All residents had experience with being observed by their supervisor throughout a whole consultation. These DO sessions were intended to be formative. Most accounts we heard were about these scheduled DO situations, but some were about ad hoc observations when the supervisor was called in for advice during a consultation. We analysed the interviews by interpreting what they revealed about common structures of residents' pre-reflective experiences. We report on these common structures in the following paragraphs.

[첫 번째이자 명백한 공통 구조]는 DO 상황에서 레지던트들은 환자와 수퍼바이저가 함께 있는 방에 있는 경험을 했다는 것입니다. 레지던트들은 자신과 환자, 수퍼바이저 간의 언어적 및 비언어적 상호작용은 물론 수퍼바이저와 환자 간의 상호작용도 경험했습니다:
A first and obvious common structure was that in DO situations residents experienced being in a room with a patient and with a supervisor. Residents experienced verbal and non-verbal interactions between themselves and the patient and the supervisor, as well as interactions between the supervisor and the patient:

  • R2: 그래서 좀 더 열심히 해야겠다는 생각이 들어요, 거의 '안녕하세요!!' (손을 흔들며) [...] 환자가 수퍼바이저에게 계속 말을 걸면 [...] 그러면 '내가 이 대화를 해야 하는데, 이렇게 하면 잘 안 되겠구나'라는 생각이 들어요.
    R2: So, then I feel that I have to work a bit harder, I'm almost doing like ‘hallo!!’ (waves her hand) […] if the patient keeps talking to the supervisor […] Then I think: I was supposed to do this conversation, but this way I'm not quite succeeding.

레지던트들의 사전 성찰 경험에서 [두 번째로 공통적인 구조]수퍼바이저와 환자 모두에게 인상을 남기면서 수퍼바이저의 관찰을 받는 경험을 했다는 것이었습니다:
A second common structure in residents' pre-reflective experiences was that they experienced being observed by their supervisor while making an impression on both the supervisor and the patient:

  • R6: 네, 하지만 수련 중이라는 것과 환자가 당신에 대한 의견을 형성하고 슈퍼바이저가 당신에 대한 의견을 형성한다는 것을 매우 의식하고 있죠 [...].
    R6: Well, but yes, you are very conscious of being in training and that, um, the patient forms an opinion of you, and that the supervisor forms an opinion of you […].

수퍼바이저에게 어떤 인상을 남겼는지에 대한 인식은 레지던트들이 자신의 성취에 대해 자부심을 가질 수 있게 합니다:
Awareness of the impression they made on supervisors could make residents proud of their accomplishments:

  • R1: 그리고 나서 [...] 이게 잘 되고 있구나, 이게 잘 되고 있구나, 이게 잘 되고 있구나, 그리고 은근히 이렇게 생각했죠, 아, 잘 되고 있구나, 슈퍼바이저가 여기 와서 (이걸) 봐주셔서 다행이다.
    R1: And then I thought, yeah, […] this is going well, this is going well, this is going well, and I secretly thought like, oh, this is going nicely and I'm glad that my supervisor is here (and sees it).

이러한 관찰에 대한 인식은 전공의들이 관찰되지 않는 환자면담과 비교해보면, 관찰받는 때에 더 불안감을 느끼고 심지어 장애를 느낄 수도 있습니다:
This awareness of being observed could also make residents feel insecure and even handicapped compared with a not-observed consultation:

  • R6: 글쎄요, 관찰을 받을 때는 그것이 무엇인지, 무엇을 해야 하는지 잘 모르는 경우가 많고, 평상시에는 무언가를 생각하거나 무언가를 만들어내지만, 감독관이 저를 관찰하고 있으면 잘못된 말을 할까 봐 두렵습니다.
    R6: Well, um, I feel that when I'm being observed I know less often what it is or what I have to do; and, normally, I would think of something, or make something up, but if my supervisor is observing me, I'm afraid that I'll say the wrong things.

레지던트들이 환자와 진단 및 치료 계획에 대해 논의할 때 불안감과 장애를 느끼는 것이 가장 두드러졌습니다:
Feeling insecure and handicapped was most prominent when residents discussed the diagnosis and care plan with the patient:

  • R2: [...] 진단과 치료 방법에 대해 완전히 확신하지 못하면 [...] 감독관이 끼어들어 전혀 하지 않겠다고 말할지도 모르기 때문에 [...] 그렇게 해야 하는 것이 매우 어색했습니다.
    R2: […] concerning the diagnosis and how to handle this, if I am not entirely certain, I can't be very firm in saying we're going to do this […] because perhaps the supervisor will interrupt and say that we're not going to do this at all […] I found that very awkward to have to do.

불안감 및/또는 장애를 느끼는 것은 레지던트들이 환자와 상호작용하는 개인적인 방식과도 관련이 있을 수 있습니다:
Feeling insecure and/or handicapped could also relate to residents' personal way of interacting with patients:

  • R2: [...] 제가 환자와 소통하는 방식 등 여러 가지 측면에 대해 상사가 인정해줄지 궁금합니다 [...] 저는 상당히 친근하고 가능하면 격식을 차리지 않는데, 상사도 그런 점을 인정해줬으면 좋겠어요 [...].
    R2: […] that I wonder if my supervisor approves […] that can concern multiple aspects, such as how I communicate with patients, I'm quite approachable and not so formal if possible, and then I hope that she will appreciate that too […].

[또 다른 공통적인 경험 구조]로, 레지던트들은 관찰 수퍼바이저를 선배 동료이자 잠재적 조력자로 경험했습니다. 이로 인해 레지던트들은 독립적인 노동자로서 수퍼바이저에게 주는 인상에 해가 되더라도 최적의 환자 치료를 위해 수퍼바이저의 의견을 묻게 될 수 있습니다:
As another common structure of the experience, residents experienced their observing supervisor as a senior colleague and potential helper. This could lead residents to ask the supervisor's opinion, for the sake of optimal patient care, even if this was to the detriment of the impression they made on their supervisor as an independent worker:

  • R3: 특히 치료 계획은 최소한 확인을 받고 싶어요. 특히 전문가가 제 옆에 앉아있는데 환자가 최적의 치료보다 못한 치료를 받는 것은 원치 않아요 [...] 저는 항상 갈등을 겪습니다: 이것은 관찰이기 때문에 전문가가 없는 것처럼 행동해야 한다. 하지만 어쨌든 저는 그와 상담합니다 [...].
    R3: Especially the care plan, I want to have that checked at least. I don't want the patient to get less than optimal treatment when the expert, notably, was sitting beside me […] I always have that conflict: this is an observation so I should act as if he wasn't there. But then I consult him anyway […].

또한 레지던트들은 자신의 지도의가 환자에게 친숙한 GP인 경우가 많았습니다. 이것 역시 레지던트가 대화에 감독관을 참여시킬 수 있습니다:
Also, residents often experienced their supervisor as the patient's familiar GP. This, too, could make residents engage their supervisor in the conversation:

  • R4: [...] 환자가 (슈퍼바이저와 상의할 때) 그걸 좋아하는 것 같아요 [...] 자기 담당 GP가 동의한다는 것을 알기 때문이죠.
    R4: […] I think that the patient likes that (when I discuss things with my supervisor) […] because she sees that her own GP agrees.

[또 다른 공통된 구조]는 레지던트들이 진료실에서 슈퍼바이저의 위치에 대해 경험하는 것이었습니다:
Another common structure was the residents' experience of the position of the supervisor in the room:

  • R3: 네, 그녀가 제 뒤에 조금 더 옆으로 앉았다면 도움이 되었을 거예요 [...] 이제 저는 그녀가 우리 사이에 [...] 거의 중재자처럼 [...] 앉았다는 것을 깨달았습니다.
    R3: Yes, it would have helped if she had sat more to the side, a bit behind me […] Now I realise that she sat right between us […] almost like a mediator […].
  • R5: 그는 정말 말 그대로 기댈 수 있는 사람이고 저를 지지해주는 사람이기 때문에 그가 제 뒤에 앉지 않고 옆으로 더 멀리 앉았다면 아마도 (책임자라는) 느낌을 줄 수 있었을 것입니다.
    R5: Well he is really quite literally someone to lean on, someone who supports me, so if he did not sit behind me but to the side and further away, that would perhaps give me the feeling (of being in charge).

놀랍게도 수퍼바이저의 존재로 인한 방해에도 불구하고 레지던트들은 종종 관찰 상황을 독립적으로 일하는 방법을 보여주기 위한 초대 또는 과제로 경험했습니다:
Strikingly, despite the disturbances resulting from the presence of the supervisor, residents often did experience the observation situation as an invitation, or assignment, to show how they work independently:

  • R1: [...] 이것은 (환자와 그의 아들과) 꽤 좋은 3자 대화였습니다; 제 수퍼바이저는 옆에 앉아서 개입하지 않았고, 어..., 그냥 관찰했고, 어... 저는 모든 것을 혼자서했습니다 [...].
    R1: […] This was quite a good three-way conversation (with a patient and his son); my supervisor sat to the side, and he did not intervene, he, uh …, he just observed, and uh… I did it all by myself […].

마치 환자와 혼자 있는 것처럼 독립적으로 일하려고 하면 많은 좌절감을 느낄 수 있습니다:
Trying to work independently, as if they were alone with the patient, could cause many frustrations:

  • R5: 제가 약간 수렁에 빠지거나 개요를 잃을 때 ..., 그가 거기에 없으면 스스로 회복할 수 있지만 [...], 분명히 저는 상사가있을 때 대부분 회복하지 못합니다.
    R5: When I get bogged down a bit, or lose the overview …, if he were not there I would recover myself, […] but, apparently, I mostly don't manage to recover when my supervisor is present.
  • R1: 그리고 그것은 ..., 당신은 최고의 자아가 아니며, 최적의 기능을 발휘하지 못하고 [...] 당신이 원하지만, 그것은 역설입니다.
    R1: and that's …, then you're not your best self, you're not functioning optimally […] while you do wish you did, that's a paradox.

반면, 일부 레지던트들은 독립적으로 일하는 방식을 보여주기 위한 과제로서 DO 상황을 경험하지 않았을 때를 설명하면서, 그 상황을 [상사와 함께 일하고 서로를 관찰하면서 배울 수 있는 기회]로 해석하기도 했습니다:
By contrast, some residents provided accounts of times when they did not experience DO situations as an assignment to show how they work independently; they could also interpret the situation as an opportunity to work and learn together with their supervisor, observing each other, which they valued:

  • R2: [...] 저는 그 상황을 협력적인 상담으로 바꾸고 싶었습니다 [...] 서로를 보완하고 [...] 어떻게 할 것인지에 대해 스파링을하고, 음, 네, 그게 재미 있다고 생각했습니다 [...].
    R2: […] I was inclined to turn the situation into a collaborative consultation […] I like that, complementing each other […] sparring about what would you do, and um, yeah, I thought that was fun […].

흥미롭게도 DO 상황에 대한 이러한 해석은 대부분 자발적으로 발생했으며 감독자와 사전에 합의하지 않았습니다.
Interestingly, this interpretation of the DO situation mostly arose spontaneously and was not agreed upon in advance with the supervisor.

[마지막으로 공통적 경험 구조]로, 레지던트들은 이전 경험을 바탕으로 수퍼바이저와 기존 관계를 맺고 있었으며, 이는 DO 상황을 경험하는 방식에 영향을 미쳤습니다:
As a last common structure of experience, residents had a pre-existing relationship with their supervisor based on previous experiences, which influenced how they experienced the DO situation:

  • R2: 저는 이 수퍼바이저와 매우 잘 지낼 수 있었고, 서로 신뢰하는 관계였기 때문에 수퍼바이저의 관찰을 받는 것이 신경 쓰이지 않았어요.
    R2: I could get along very well with this supervisor, we had a trusted relationship, so I didn't mind being observed by him.

4 토론
4 DISCUSSION

일반적으로 DO에 대한 이해를 증진하기 위해, 특히 '관찰자 효과', '진정성 있는 행동', 밀러의 'DO' 수준 및 DO 상황에서 수퍼바이저의 참여와 관련하여 [레지던트들이 DO 상황에서 전-성찰적으로 경험한 것의 규칙성]을 조사했습니다. 그 결과, 관찰하는 수퍼바이저가 관찰하지 않는 상담과 비교했을 때 레지던트의 경험과 행동이 어떻게 크게 달라졌는지 밝혀냈습니다. 다음 단락에서 이에 대해 자세히 설명하겠습니다.
In order to advance our understanding of DO in general, and specifically concerning the ‘observer effect’, ‘authentic behaviour’, Miller's ‘does’ level and the participation of supervisors in DO situations, we investigated regularities in what residents pre-reflectively experienced in DO situations. Our results illuminate how an observing supervisor substantially changed the experience of residents and their behaviour, compared with unobserved consultations. We will elaborate on this in the following paragraphs.

Ladonna와 동료들은 레지던트들이 DO 하에서 '진실하지 않게' 행동한다고 보고했으며,6 따라서 관찰되지 않을 때, 즉 독립적으로 일할 때 어떻게 행동할지 보여주지 못한다는 사실을 발견했습니다.1, 13 이 저자들은 관찰자 효과(관찰 및 평가받는다고 느낄 때 다르게 행동하는 것을 말함)가 이에 대한 책임이 있다고 주장했습니다. 그러나 이러한 관찰자 효과는 레지던트에게 평소처럼 행동하도록 격려하고1 더 나은 DO 조건을 조성함으로써 극복할 수 있는 것으로 간주되는 경우가 많습니다.4, 8, 23 이러한 조건에는 다음이 포함됩니다.4, 8, 23, 24 

  • 종단적이고 신뢰할 수 있는 훈련 관계,
  • 전용 시간이 있는 반복적인 DO 세션,
  • 감독자가 의도적으로 한쪽에 앉아 환자와 눈을 마주치는 등 접촉을 피하는 등 레지던트의 자율성을 촉진하는 조치

연구 결과에 따르면 이러한 예방 조치는 실제로 산만한 상호작용을 줄이고 레지던트가 더 편안하게 느끼도록 도울 수 있음을 확인시켜 줍니다. 그러나 더 편안하고 산만함을 덜 느낀다는 것은 감독자가 없을 때처럼 '진정성 있게' 또는 독립적으로 일할 수 있다는 것과는 다릅니다.
Ladonna and colleagues found that residents reported behaving ‘inauthentically’ under DO,6 thus not showing how they would work when not observed, that is, independently.1, 13 These authors held the observer effect responsible for this, which refers to acting differently when feeling observed and assessed. This observer effect, however, is often regarded as something that can be overcome, by encouraging residents to behave as they would normally do,1 and by creating better DO conditions.4, 8, 23 Such conditions comprise

  • longitudinal, trusted, training relationships,
  • recurring DO sessions with dedicated time and
  • measures to promote residents' autonomy such as supervisors avoiding contact, including eye contact, with the patient by deliberately sitting to one side.482324 

Our results confirm that these precautions may indeed help reduce distracting interactions and make residents feel more at ease. However, feeling more at ease and less distracted is not the same as being able to work ‘authentically’, or independently, as one would when the supervisor is not there.

우리는 수퍼바이저의 존재로 인해 발생하는 [관찰자 효과]가 이전에 이해했던 것보다 훨씬 더 중요하기 때문에 독립적으로 일할 수 없다는 것을 발견했습니다. 관찰되는 상황에 놓임으로써, 수퍼바이저는 레지던트와 환자가 경험하고 느끼고 행동하는 것에 불가피한 영향을 미치면서, 그 상황을 여러 가지 방식으로 변화시켰기 때문입니다.

  • 한 가지 예로, 레지던트들은 관찰되지 않는 상황에서는 그렇게 하지 않았을지라도 최적의 환자 치료와 편안함을 위해 수퍼바이저를 대화에 참여시키고 싶은 유혹을 받는다는 사실을 발견했습니다. 이러한 레지던트들은 선배가 병실에 있을 때 자신의 전문성을 활용하지 않는 것이 부자연스럽게 느껴졌습니다.
  • 환자와 슈퍼바이저(주로 담당 GP)가 친숙하다는 점도 레지던트들이 슈퍼바이저를 참여시키는 또 다른 이유였습니다. DO 상황에서 환자의 경험에 대한 이전 연구14에 따르면 환자들도 같은 이유로 슈퍼바이저의 연륜 및/또는 환자와의 친숙함 때문에 슈퍼바이저를 대화에 끌어들이는 것으로 나타났습니다.

We found that the observer effect that is caused by the presence of the supervisor did not allow for working independently because this effect was much more material than was previously understood: By being in the situation that was observed, the supervisor changed that situation in numerous ways with an inevitable impact on what the resident and patient experienced, felt and did.

  • As one example, we found that residents were tempted to engage their supervisor in the conversation for the sake of optimal patient care and comfort, even if they would not have done so in an unobserved situation. For these residents, when the senior was in the room, it felt unnatural not to make use of their expertise.
  • The familiarity of the patient with the supervisor, often their GP, was an additional reason for residents to engage their supervisor. Our previous study of patients' experiences14 in DO situations indicated that patients also drew supervisors into the conversation, for the same reasons: the supervisor's seniority and/or familiarity with the patient.

우리는 관찰자 효과는 레지던트가 관찰되고 평가받는다고 느끼는 것뿐만 아니라, 감독하는 GP의 존재가 관찰 대상 상황을 심오한 방식으로 변화시킨다는 결론을 내립니다. 따라서 레지던트가 독립적으로 일하는 방식을 관찰하는 것으로 정의된 Miller의 '한다' 수준을 관찰하는 것은1,13 불가능해 보입니다. 
We conclude that the observer effect is not just about residents feeling observed and assessed; the presence of a supervising GP changes the situations to be observed in profound ways. Therefore, observing Miller's ‘does’ level, defined as observing how a resident works independently,1, 13 seems impossible.

우리는 레지던트들이 종종 이 문제로 어려움을 겪는다는 것을 발견했습니다: 레지던트들은 수퍼바이저나 프로그램의 기대에 따라 [독립적으로 일하는 모습을 보여줘야 한다]고 생각하면서도, 동시에 그것이 [불가능]하고, 심지어 [좋은 환자 치료를 위해서도 바람직하지 않다]는 것을 경험했습니다. 기대에 부응하고 가능한 한 독립적으로 일하려고 노력하면서 이에 대처한 레지던트들은 많은 장애와 좌절을 경험했다고 보고했습니다. 이전 연구에서도 DO가 레지던트, 환자, 감독자 세 참여자 모두에게 불편한 상황과 어색함을 초래하는 경우가 많다고 강조합니다.3, 6, 9-11, 25 이러한 원인 중 하나는 한편으로는 감독자의 참여에 대한 환자와 레지던트의 요구와 다른 한편으로는 DO 지침에 따라 감독자를 대화에서 배제하려는 감독자 및 레지던트의 시도1에서 찾을 수 있다고 이 문헌에 덧붙여 설명합니다.
We found that residents often struggled with this: They experienced the expectations of the supervisor, or the programme, as needing to show how they work independently, while they simultaneously experienced that this was impossible, and even undesirable in the interests of good patient care. Residents who coped with this by complying with the expectations and trying to work as independently as possible reported many impediments and frustrations. Previous research also highlights that DO often brings about uncomfortable situations and awkwardness for all three participants: residents, patients and supervisors.3, 6, 9-11, 25 We add to this literature that one of the causes for this may be found in the discrepancy between patients' and residents' needs for the participation of the supervisor on the one hand, and the DO guidelines-driven supervisors' and residents' attempts to keep the supervisor out of the conversation,1 on the other.

실무적 시사점
Practical implications

이 연구는 평가에 관한 연구는 아니지만, 연구 결과는 WBA에 영향을 미칠 수 있습니다.26, 27

  • 레지던트 역량 평가에 필요한 데이터를 어떻게 수집할 수 있을까요?
  • 레지던트가 독립적으로 유능한 GP 또는 전문의가 되고 있는지 어떻게 확신할 수 있을까요?
  • PGME는 어떻게 사회적 책임에 부응하고 있나요?
  • 어느 정도 거리를 두고 객관적으로 레지던트를 평가해야 하지 않을까요?

이러한 질문은 의학교육에서 일반적이고 타당한 질문이며, CBME 시대에는 더욱 그렇습니다.26
Although this study was not about assessment, our findings may have implications for WBA.26, 27 

  • How do we collect the data we need for assessing our residents' competence?
  • How can we be certain that a resident is becoming an independently competent GP or medical specialist?
  • How does PGME live up to its societal accountability?
  • Must we not assess residents with a certain degree of distance and objectivity?

These are common and valid questions in medical education; more so in the CBME era.26

그러나 우리의 연구 결과는 독립적 역량에 대한 이러한 거리두기 관찰의 실현 가능성에 의문을 제기합니다. 앞서 살펴본 바와 같이, 관찰자가 관찰하는 것은 적어도 부분적으로는 관찰자의 존재에 의해 영향을 받습니다. 자연 과학에서는 인공물이라고 할 수 있습니다. 이는 레지던트의 성과에 대해 '객관적'이고 거리를 두고 판단하고자 하는 욕구가 실제로는 자연과학자, 즉 (포스트)실증주의적 태도를 드러낸다는 것을 알 수 있게 해줍니다.28 그러나 사회구성주의에서 이러한 [인공물]은 [레지던트의 학습 궤적에 대한 의미]에 관한 [대화의 유효한 출발점]이 될 수 있습니다.28 이는 대화의 중요성이 점점 강조되는 피드백에 관한 문헌과 공명합니다.29-31 우리의 결과는 이러한 대화가 우리가 본 것은 그 자체로는 의미가 없음을 확고히 확립하는 것으로 시작해야 한다는 것을 시사합니다. 연구 결과에서 종종 반영된 것처럼, 모호하지 않고 명확하게 이 점을 밝히면 DO 상황의 긴장을 완화할 수 있습니다.
Our results, however, question the feasibility of this distanced observing of independent competence. As shown, what observers observe is, at least in part, caused by their presence. In the natural sciences, we would speak of artefacts. This helps us see that the wish for an ‘objective’, distanced, judgement of a resident's performance actually reveals a natural scientist, that is, a (post-) positivist attitude.28 In social constructivism, however, these artefacts can be valid starting points for a dialogue concerning their meaning for the resident's learning trajectory.28 This resonates with the literature on feedback, in which the importance of dialogue is increasingly emphasised.29-31 Our results suggest that this dialogue should start with firmly establishing that what we have seen has no meaning in itself. Being clear about this instead of ambiguous, as was often reflected in our results, may relieve tensions in DO situations.

위의 내용을 다시 밀러의 피라미드에 대입하면,13 우리는 절대로 'shows how'를 수준 이상은 볼 수 없으며, 'does'라는 단계는 [우리가 관찰한 것]과 [다른 출처에서 추론한 것]을 기반으로 구축하는 구조입니다. 이러한 다른 출처와 관련하여 점점 더 많은 지식이 민족지학 및 현상학에서 파생된 전공의의 역량 진전을 평가하는 새로운 보완적인 방법을 뒷받침합니다.32-34
When we translate the above again to Miller's pyramid,13 we will never see more than the ‘shows how’ level, which makes the ‘does’ level a construction that we build upon what we have observed and what we infer from other sources. Concerning these other sources, a growing body of knowledge supports new complementary ways of assessing the residents' progress in competence, derived from, for example, ethnography and phenomenology.32-34

마지막으로, 우리의 연구 결과는 레지던트와 수퍼바이저가 환자와 같은 공간에 있는 목적과 진행 방법에 관한 대화를 개선할 수 있음을 시사합니다. 유사한 GP 수련 환경에서 수행된 최근 연구에 따르면 레지던트와 수퍼바이저는 이에 대해 거의 논의하지 않는 것으로 확인되었습니다.10-12 이러한 대화의 중요한 요소는 DO 상황이 양방향적이고 전경이 아닌 경우 학습에 가장 효과적인 것으로 보입니다.10, 11 따라서 레지던트와 수퍼바이저는 서로 관찰하고, 관찰 정보를 수집, 공유 및 함께 해석하면서 함께 일하고 배우기 위해 DO 상황을 사용하는 것을 고려해야 합니다. 
As a last practical implication, our results suggest that residents and supervisors could improve their dialogue concerning the purpose of their being in the same room with a patient and how to proceed. Recent research in similar GP training settings confirms that residents and supervisors hardly discuss this.10-12 An important factor for this dialogue appears to be that DO situations seem to work best for learning when DO is bi-directional and not foregrounded.10, 11 Residents and supervisors should therefore consider using DO situations to work and learn together while observing each other, collecting, sharing and together interpreting observational information along the way.

향후 연구에 대한 시사점
Implications for future research

레지던트와 환자를 대상으로 한 현상학적 연구는 PGME에서 DO를 개념화하는 데 기여했으며, 수퍼바이저가 관찰하는 상황에 필연적으로 참여해야 함을 강조했습니다. 이 과정에서 수퍼바이저의 DO 상황에 대한 경험에 대한 현상학적 조사는 아직 중요한 부분이 누락되었습니다. 
Our phenomenological research amongst residents and patients has contributed to the conceptualisation of DO in PGME, highlighting the inevitable participation of supervisors in the situations they observe. In this, a phenomenological investigation of supervisors' experiences in DO situations is yet an important missing piece.

이 연구의 주요 기여는 관찰자 효과를 단순히 불안을 유발하는 것이 아니라 상황을 실질적으로 변화시키는 것으로 새롭게 개념화했다는 점입니다. 이러한 이해를 확인 및/또는 개선하기 위해서는 다른 맥락에서의 추가 연구가 필요합니다.35
The main contribution of this work to the literature is the new conceptualisation of the observer effect, not just as anxiety-provoking but as a material alteration of the situation. Further research in other contexts is needed to confirm and/or improve this understanding.35

또한, 함께 일하고 배우는 세션에서 얻은 정보가 전공의에 대한 요약 평가에 가장 적합한 정보를 제공할 수 있는 방법에 대한 더 많은 연구가 필요합니다. 
Moreover, we need more research on how information obtained from working and learning together sessions can best inform summative assessments of residents.

제한 사항
Limitations

본 연구는 네덜란드의 한 GP 수련 센터에서 수행되었기 때문에 다른 상황에 적용하는 데 한계가 있습니다. 언급해야 할 중요한 맥락적 요소는 GP 수련 과정에서 환자가 레지던트보다 레지던트의 지도의인 자신의 GP를 더 잘 알고 있다는 점입니다. 이 사실은 우리의 연구 결과 중 하나에 기여했습니다: 레지던트들은 자신의 슈퍼바이저를 환자에게 친숙한 GP로 경험했습니다. 이렇기에 레지던트가 수퍼바이저를 대화에 참여하도록 유도할 수 있습니다. 수퍼바이저를 참여시키는 또 다른 이유인 연공서열은 아마도 대부분의 의료 전문직의 교육 맥락에 적용될 것입니다. 
We conducted our research in one Dutch GP training centre, limiting its transferability to other contexts. An important contextual factor to mention is that patients, in GP training, usually know their own GP, who is the resident's supervisor, better than they know the resident. This fact contributed to one of our findings: Residents experienced the presence of their supervisor as the patient's familiar GP. This could encourage them to engage the supervisor in the conversation. The other reason to engage the supervisor, their seniority, will probably apply to education contexts in most health professions.

두 번째 한계는 이 연구가 네덜란드의 한 일반의 교육 센터라는 한 가지 맥락에서만 이루어진 소규모 인터뷰 연구라는 점입니다. 그러나 현상학적 연구에서는 소수의 참여자만으로도 의미 있는 결과를 얻을 수 있는 경우가 많습니다. 반 마넨의 말처럼: '모든 현상학적 주제는 언제나 다시 다루어지고 원래 의미의 차원과 의미의 측면을 탐구할 수 있다'.20 또한 귀납적으로 얻은 이론의 타당성은 그 자체의 양적 토대가 아니라 다양한 맥락에서의 유용성에 의해 결정되며,35 이는 더 많은 연구가 필요합니다.
A second limitation is that this is a small interview study in only one context: a GP training centre in the Netherlands. In phenomenological research, however, small numbers of participants often suffice to attain meaningful, though not exhaustive, results. As van Manen puts it: ‘Every phenomenological topic can always be taken up again and explored for dimensions of original meaning and aspects of meaningfulness’.20 Also, the validity of inductively obtained theory is not determined by its quantitative underpinning per se but by its usefulness in different contexts,35 which needs to be determined further.

5 결론
5 CONCLUSION

우리의 연구 결과는 '관찰자 효과'가 이전에 이해되었던 것보다 훨씬 더 중요하다는 것을 보여줍니다. 결과적으로, 밀러의 'Does' 수준에서 마치 수퍼바이저가 없는 것처럼 전공의의 '진짜' 행동을 관찰하는 것은 이론적으로나 실제적으로 불가능하며, 오해의 소지가 있는 개념으로 보입니다:

  • 수퍼바이저가 그곳에 있는데도, 마치 수퍼바이저가 그곳에 없는 것처럼 일하도록 하는 불가능한 일을 전공의에게 강요하나, 보고된 모든 관련 문제와 고통을 포함하여 상황을 실질적으로 변화시켰기 때문에 오해의 소지가 있으며,
  • 수퍼바이저가 상황에 참여하지 않도록 하여 잠재적으로 환자와 전공의의 요구를 무시하게 만들었기 때문에 오해의 소지가 있으며,
  • 전공의과 수퍼바이저가 교육과 학습의 기회를 낭비하게 만들었다는 점에서 오해의 소지가 있습니다.

Our results indicate that the ‘observer effect’ is much more material than was previously understood. Consequently, observing residents' ‘authentic’ behaviour at Miller's ‘does’ level, as if the supervisor was not there, seems—theoretically and practically—impossible and a misleading concept: misleading because it invited residents to do the impossible:

  • to work as if the supervisor was not there while he/she was there and substantially changed the situation, with all the reported associated problems and distress;
  • misleading also because it made supervisors try to avoid participating in the situation, thereby potentially neglecting patients' and residents' needs; and
  • misleading, finally, because it made residents and supervisors waste opportunities for educating and learning.

이번 연구 결과와 이전 연구 결과를 바탕으로, 레지던트와 수퍼바이저가 한 공간에서 함께 환자 진료에 참여하는 경우, 일방향 DO는 함께 일하고 배우는 세션에서 양방향 DO로 대체하는 것이 더 낫다고 제안합니다.
Based on our results and previous findings, we suggest that when a resident and a supervisor are together in one room, engaged in patient care, one-way DO is better replaced by bi-directional DO in working-and-learning-together sessions.

 


Med Educ. 2023 Mar;57(3):272-279. doi: 10.1111/medu.15004. Epub 2022 Dec 23.

A phenomenology of direct observation in residency: Is Miller's 'does' level observable?

Affiliations collapse

Affiliations

1Department of General Practice, Amsterdam UMC, location Vrije Universiteit Amsterdam, Amsterdam, The Netherlands.

2Department of General Practice, Amsterdam UMC, location University of Amsterdam, Amsterdam, The Netherlands.

3Department of General Practice, Erasmus Medical Center, Rotterdam, The Netherlands.

4School of Medical Sciences, Athena Institute for Transdisciplinary Research, Amsterdam UMC, location Vrije Universiteit Amsterdam, Amsterdam, The Netherlands.

5School of Health Professions Education, Maastricht University, Maastricht, The Netherlands.

PMID: 36515981

PMCID: PMC10107098

DOI: 10.1111/medu.15004

Abstract

Introduction: Guidelines on direct observation (DO) present DO as an assessment of Miller's 'does' level, that is, the learner's ability to function independently in clinical situations. The literature, however, indicates that residents may behave 'inauthentically' when observed. To minimise this 'observer effect', learners are encouraged to 'do what they would normally do' so that they can receive feedback on their actual work behaviour. Recent phenomenological research on patients' experiences with DO challenges this approach; patients needed-and caused-some participation of the observing supervisor. Although guidelines advise supervisors to minimise their presence, we are poorly informed on how some deliberate supervisor participation affects residents' experience in DO situations. Therefore, we investigated what residents essentially experienced in DO situations.

Methods: We performed an interpretive phenomenological interview study, including six general practice (GP) residents. We collected and analysed our data, using the four phenomenological lenses of lived body, lived space, lived time and lived relationship. We grouped our open codes by interpreting what they revealed about common structures of residents' pre-reflective experiences.

Results: Residents experienced the observing supervisor not just as an observer or assessor. They also experienced them as both a senior colleague and as the patient's familiar GP, which led to many additional interactions. When residents tried to act as if the supervisor was not there, they could feel insecure and handicapped because the supervisor was there, changing the situation.

Discussion: Our results indicate that the 'observer effect' is much more material than was previously understood. Consequently, observing residents' 'authentic' behaviour at Miller's 'does' level, as if the supervisor was not there, seems impossible and a misleading concept: misleading, because it may frustrate residents and cause supervisors to neglect patients' and residents' needs in DO situations. We suggest that one-way DO is better replaced by bi-directional DO in working-and-learning-together sessions.

평가 실무의 개편에서 복잡성 탐색: 비판적 실재주의자 관점(Adv Health Sci Educ Theory Pract. 2021)
Exploring complexities in the reform of assessment practice: a critical realist perspective
Chris Roberts1 · Priya Khanna1 · Andrew Stuart Lane1 · Peter Reimann2 · Lambert Schuwirth3

 

소개
Introduction

학습에 대한 전통적인 평가를 학습을 위한, 그리고 학습으로서의 평가를 포괄하는 프로그램으로 개혁할 때 몇 가지 문제가 있을 수 있습니다. 총괄적 고부담 의사 결정을 강조하는 평가 시스템에 대한 전통적 접근 방식은 복잡한 인간의 기술과 행동을 숫자로 변환하는 데 한계가 있다는 지적을 받아왔습니다(Cate & Regehr, 2019; Driessen et al., 2005; Epstein & Hundert, 2002; Hodges, 2006, 2013; Schuwirth & Van der Vleuten, 2006). 이러한 한계를 부분적으로 해결하기 위해 [학습을 위한 평가] 그리고 [학습으로서의 평가]를 지원하는 프로그램 평가와 같은 대안적 접근 방식이 개발되었습니다. 프로그램적 평가는 의도적으로 선택된 여러 평가 과제를 사용하여 각각 일정 기간 동안 결합된 의미 있는 데이터를 생성함으로써 다양한 역량 결과 영역에서 학습자의 진행 상황에 대한 풍부하고 종적인 삼각형 정보의 흐름을 생성할 수 있습니다(Van der Vleuten & Schuwirth, 2005). 학습자에 대한 정보는 수집(종단적), 대조(삼각 측량) 및 연결(정보의 질)되어 해당 정보의 의미에 대한 의사 결정이 평가자의 전문적 판단과 일치하도록 합니다. 이를 통해 프로그래밍 방식 평가는 학습을 위한 평가, 학습으로서의 평가, 학습에 대한 평가 간의 균형을 촉진합니다. (Dannefer, 2013; Driessen 외., 2012; Khanna 외., 2021). 
There can be several problems when reforming traditional assessment of learning to a program which encompasses assessment for and as learning. Traditional approaches to assessment systems which emphasise summative high-stakes decision making, have been challenged for their limitations in converting complex human skills and behaviours to numbers (Cate & Regehr, 2019; Driessen et al., 2005; Epstein & Hundert, 2002; Hodges, 2006, 2013; Schuwirth & Van der Vleuten, 2006). Partly to address these limitations, alternate approaches such as programmatic assessment have been developed that support assessment for and as learning. By using purposefully selected multiple assessment tasks, each generating its own meaningful data combined over a period of time, programmatic assessment allows for the creation of an abundant and longitudinal flow of triangulated information about a learner's progress in various competency outcome areas (Van der Vleuten & Schuwirth, 2005). Information about learners is collected (longitudinally), collated (triangulation), and connected (quality of information), so that decision making about the implications of that information is aligned with the professional judgement of assessors. In doing this, programmatic assessment promotes a balance between assessment for, as and of learning. (Dannefer, 2013; Driessen et al., 2012; Khanna et al., 2021).

개혁을 도입하고 보고할 때 교육자들이 직면하는 주요 과제이자 이 백서의 초점은 의과대학이 학습을 위한 평가 및 학습으로서의 평가 원칙에 동의하는 경우에도 기존의 전통적인 관행을 개혁하는 것이 어렵다는 것입니다(Driessen 등, 2012; Roberts 등, 2014; Dannefer, 2013; Torre 등, 2021). 여기에는 보건 전문가 교육 결과에 관심이 있는 이해관계자들에게 의미 있는 방식으로 변화를 촉진하고 평가하고자 하는 사람들에게 통찰력을 제공할 수 있는 복잡한 이유가 있습니다. 한 가지 이유는 다음 두 그룹 간의 의견 불일치일 수 있습니다. 

  • 학습자의 성과에 대한 기존의 '객관적', 구조화된 정량화를 평가에 유지하려는 사람들(Swanson & Roberts, 2016)과
  • 임상 역량에 대한 새로운 개념을 고려한 체계적인 접근 방식을 장려하는 사람들(Driessen et al., 2012; Schuwirth & Van der Vleuten, 2011)

When introducing and reporting reforms, a major challenge for educators, and the focus of this paper, is that medical schools have found it difficult to reform pre-existing traditional practices, even where there is agreement with the principles behind assessment for and as learning, (Driessen et al., 2012; Roberts et al., 2014; Dannefer, 2013; Torre et al., 2021). There are complex reasons for this which may provide insights for those wishing to promote and evaluate change in a way that is meaningful to stakeholders interested in health professional education outcomes. One reason might be disagreement between

  • those who wish to retain the traditional ‘objective’, structured quantification of learners’ performances to assessment (Swanson & Roberts, 2016) and
  • those promoting a systemic approach that take into account newer conceptions of clinical competence (Driessen et al., 2012; Schuwirth & Van der Vleuten, 2011).

이 백서에서 논의한 바와 같이, 의료 전문가 교육 내에서 평가 문화를 크게 변화시키는 데 중요한 요소를 조명할 수 있는 방법론을 개발하는 데 있어 교육자를 안내할 수 있는 출판된 문헌은 거의 없습니다. 이 백서에서는 이론에 기반한 접근 방식을 취하여 평가 실무에서 주요 개혁을 실행하고 평가하려는 사람들을 지원하기 위한 몇 가지 핵심 원칙과 방법론적 접근 방식을 제안합니다. 
As we discuss in this paper, there is little published literature to guide educators in developing methodologies that might illuminate significant factors in making major changes to the culture of assessment within healthcare professional education. In this paper, we take a theoretically informed approach to propose some key principles and a methodological approach to support those considering implementing and evaluating major reform in assessment practice.

표면적으로는 학습 평가에 대한 전통적인 접근 방식과 평가의 의사 결정 및 학습 기능을 통합하는 새로운 접근 방식(Schut et al., 2021)의 차이점은 방법론적인 것으로 보입니다. 이는 부분적으로는 평가 관행의 기본 프레임워크와 품질을 확립하는 데 사용되는 타당성 제안의 차이 때문일 수 있습니다.

  • 구조화된 임상 평가의 객관성이 고부담 평가의 특징이 된 이후(Newble & Swanson, 1988), [전통적인 평가 접근법]은 의과대학 평가 관행을 안내하는 데 큰 영향을 미쳐 왔습니다.
  • 이 방식을 옹호하는 사람들은 공정성 문제를 해결하는 동시에 측정 오류를 줄임으로써 학생의 임상 역량에 대한 고품질 시험의 타당성과 신뢰성을 확립할 수 있다고 주장합니다. (Clauser 외, 2006; Norcini 외, 2018)
  • [학습의 평가]는 미래의 성과를 예측할 수 있다고 가정하는 [선형 동적 과정]으로 생각됩니다.
  • '공정성'은 평등의 관점에서 다루어지며, 모든 응시자는 동일한 표준화되고 구조화된 테스트 프로세스를 적용받습니다. (AERA, 2014)
  • 검증된 평가 도구는 상황, 문화 또는 관련된 사람에 관계없이 일반화하거나 다른 환경으로 이전할 수 있는 것으로 간주됩니다. 

On the surface, the differences between traditional approaches to the assessment of learning and newer approaches that integrate the decision-making and learning functions of assessment (Schut et al., 2021) appear to be methodological. In part this may be due to differences in the validity propositions used to establish the underlying frameworks and quality of assessment practices.

  • Since the objectivity of structured clinical assessments became the hallmark of high stakes assessment (Newble & Swanson, 1988) traditional approaches to assessment have been highly influential in guiding medical school assessment practice. Advocates claim that the validity and reliability of high quality testing of the clinical competence of students is established by reducing errors of measurement, at the same time as addressing fairness. (Clauser et al., 2006; Norcini et al., 2018)
  • The assessment of learning is thought to be a linear dynamic process which is assumed to be predictable of future performance.
  • ‘Fairness’ is addressed from a perspective of equality; all candidates are subject to the same standardised and structured testing processes. (AERA, 2014)
  • Validated assessment tools are assumed to be generalisable or transferable to other settings regardless of context, culture, or persons involved.

반면에 프로그램적 평가를 지지하는 사람들은 [복잡한 비선형 시스템 접근 방식]을 취하여 임상 역량 개념에 더 잘 부합하는 평가 접근 방식을 강조합니다.

  • 이러한 복잡성 관점에서는 역량이 다면적인 것으로 간주되며 단일 수치 결과로 표현하기 어렵습니다. 대신 역량은 문제 공간의 경계를 넘나드는 민첩한 문제 해결에서 나타난다는 점을 인정합니다. 따라서 전문성, 팀워크, 전문가 간 협업과 같은 복잡하고 실용적인 기술을 포함하고 통합하도록 설계되었습니다(Khanna et al., 2021; Schuwirth & Van der Vleuten, 2020; Van der Vleuten et al., 2018).
  • 습을 위한 그리고 학습으로서의 평가는 평가 프로그램, 평가자 및 학생 간의 긴밀한 상호 작용을 가정합니다. (Schuwirth & Van der Vleuten, 2011, 2019; Van der Vleuten 외., 2012)
  • 타당도는 전체 프로그램 내의 특정 목적에 따라 다양한 평가 도구를 신중하고 맞춤식으로 조합하는 것을 기반으로 합니다. (Schuwirth & Van der Vleuten, 2012).
  • 공정성형평성의 관점에서 다루어지며, 모든 학습자가 동일한 품질의 평가를 받지만 그 과정은 맞춤형으로 이루어집니다. (Schuwirth & Ash, 2013) 검증된 평가 프로그램은 상황, 문화 및 관련 인원을 고려하여 다른 환경에 맞게 조정할 수 있습니다. (Valentine 외., 2021).

Advocates for programmatic assessment, on the other hand, take a complex non-linear systems approach, emphasising an approach to assessment that better aligns with concepts of clinical competence.

  • In this complexity perspective, competence is seen as multifaceted and not amenable to being expressed by a single numerical outcome. Instead it accepts that competence manifests itself in agile problem solving withing the boundaries of a problem space. As such it is designed to include and integrate complex and practical skills such as professionalism, team work and interprofessional collaboration (Khanna et al., 2021; Schuwirth & Van der Vleuten, 2020; Van der Vleuten et al., 2018).
  • Assessment for and as learning assumes a close interaction between the assessment program, the assessors, and the students. (Schuwirth & Van der Vleuten, 2011, 2019; Van der Vleuten et al., 2012)
  • Validity is based on a careful and tailored combination of various assessment instruments depending on the specific purposes within the overall program. (Schuwirth & Van der Vleuten, 2012).
  • Fairness is addressed from a perspective of equity; all learners receive the same quality of assessment, but the processes are tailored. (Schuwirth & Ash, 2013)
  • Validated programs of assessment can be adapted to other settings taking account of context, culture, and the people involved. (Valentine et al., 2021).

고전적 평가 관행과 프로그램적 평가 관행을 옹호하는 사람들 사이의 의견 차이는 종종 [철학적 수준에서 평가와 학습 간의 관계에 대한 확고한 신념과 서로 다른 구성]에 있습니다(Torre et al., 2020).

  • 전통적인 평가 관행[학습하는 사람]과 [평가받는 사람]의 경험을 분리하는 경험주의적 인식론에 기반을 두고 있으며(Adie 외, 2018), 따라서 학습자는 '측정의 대상'이 됩니다.
  • 학습을 위한 그리고 학습으로서의 평가를 옹호하는 접근 방식에서 평가 프로그램의 설계는 학습 및 커리큘럼 품질 보증을 최적화하기 위한 것입니다. (Van der Vleuten 외., 2015) 평가는 학습자가 평가와 학습을 경험하는 동안 자신의 행동을 선택할 수 있는 능력을 포함하여 학습자의 주체성을 촉진하는 쌍방향 대화입니다(Meeuwissen 외., 2019; Schut 외., 2018, 2020).
  • 전통적인 접근 방식에서는 평가가 학생에게 수행되는 반면,
    프로그램적 평가에서는 학생과 함께 평가가 수행됩니다.

The differences in opinion between advocates of classical and programmatic assessment practices often also lies in firmly held beliefs and differing constructions of the relationship between assessment and learning at a philosophical level (Torre et al., 2020).

  • Traditional assessment practices are based on an empiricist epistemology that separates the experience of the person-as-learning from the person-as-being assessed (Adie et al., 2018) and therefore learners become an ‘object of measurement’.
  • In approaches that advocate assessment for and as learning, the design of an assessment program is intended to optimise learning and curriculum quality-assurance. (Van der Vleuten et al., 2015Assessment is a two-way dialogue promoting learners’ agency including their ability to make choices in their actions during their experience of assessment and learning (Meeuwissen et al., 2019; Schut et al., 2018, 2020).
  • In the traditional approach, assessment is done to the students while
    in programmatic assessment, assessment
     is done with the students.

이러한 대조는 평가 분야의 평가 및 연구에 시사점을 주는데, 연구자들은 평가 실행과 같은 복잡한 사회 현상을 고려한 이론적 관점과 이를 뒷받침하는 연구 패러다임 및 이론을 설명할 필요가 있습니다(Roberts et al., 2020).

  • [학생을 대상으로] 평가가 이루어지는 연구에서는 일반적으로 맥락, 문화, 사람 및 이들 간의 상호 작용과 같은 요소와 무관하게 "무엇이 효과가 있는가"라는 질문을 던집니다. 이는 평가 도구의 고유한 가치와 타당성을 파악하고자 합니다(Cronbach & Meehl, 1955; Downing, 2003; Messick, 1995).
  • [학생과 함께] 평가를 실시하는 경우 "무엇이 효과가 있는가"라는 질문만으로는 충분하지 않습니다. 대신 "무엇이, 누구에게, 어떤 맥락에서, 그리고 가장 중요한 것은 왜 효과가 있는가"와 같은 보다 복잡한 질문이 더 적절합니다.

This contrast has implications for evaluation and research in the field of assessment, where researchers need to explain the theoretical lens through which they have considered complex social phenomena like assessment practice, including the underpinning research paradigm and theory (Roberts et al., 2020).

  • Research where assessment is done to the students typically asks the question “what works” independently of the factors such as context, culture, and people and the interactions between them. It seeks to tease out the unique value and validity of the assessment instrument (Cronbach & Meehl, 1955; Downing, 2003; Messick, 1995).
  • Where assessment is done with the students, the question of “what works” is not enough. Instead, more complex questions such as “what works, for whom, in which context and, most importantly, why” are more appropriate.

이론적 프레임워크
Theoretical framework

이러한 질문은 이러한 현상에 대한 비판적 현실주의적 평가의 전형적인 예입니다(Ellaway 외., 2020).

  • 비판적 현실주의 관점에서 평가는 특정 사회 구조 내에서 목적과 기능을 정의할 수 있는 관행으로 볼 수 있습니다(Delandshere, 2001).
  • 평가 관행을 뒷받침하는 구조를 탐구하는 것은 바스카의 현실주의적 분석을 통해 도움이 될 수 있습니다. (Bhaskar, 1978; Bhaskar 외., 2018; Groff, 2004)
    • 경험적(경험),
    • 실제적(사건),
    • 실재적(인과 구조 및 메커니즘) 수준을 구분한 
  • 인과적 구조와 메커니즘은 실재적이며 그것이 생성하는 사건의 패턴과 구별되며, 이는 다시 그것이 포착되는 경험과는 구별됩니다.
  • 현실에 대한 바스카의 광범위한 설명의 주요 응용 분야 중 하나는 의료 환경과 같은 복잡한 시스템에서 현실주의 프로그램 평가의 현대적 접근 방식으로 이어졌습니다. 현실주의 평가는 주로 개입의 효과 여부가 아니라 '무엇이, 누구에게, 어떤 상황에서, 어떻게 효과가 있는가'라는 질문을 해결하기 위해 맥락-기제-결과 구성을 이해하는 데 중점을 둡니다(Dalkin 외., 2015).
  • 현실주의적 접근법은 '생성적' 인과관계 이론에 기반합니다. 즉, 상황을 변화시키는 것은 프로그램이 아니라, 프로그램에 노출되었을 때, 주어진 메커니즘을 활성화하고 변화를 일으키는 것은 [맥락에 내재된 사람]입니다(Pawson et al., 1997).

Such questions are typical examples of a critical realist evaluation of such a phenomenon (Ellaway et al., 2020).

  • From a critical realist perspective, assessment can be viewed as a practice whose purpose and functions can be defined within particular social structures (Delandshere, 2001).
  • Exploring the structures that underpin assessment practice can profit from Bhaskar’s realist analysis of the distinction between the level of
    • the empirical (experiences),
    • the actual (events) and
    • the real (causal structures and mechanisms). (Bhaskar, 1978; Bhaskar et al., 2018; Groff, 2004)
  • Causal structures and mechanisms are real and distinct from the pattern of events they generate, which in turn are distinct from the experiences in which they are captured.
  • One of the key applications of Bhaskar’s extensive account of reality has led to the contemporary approach of realist program evaluation in complex systems such as healthcare settings. Realist evaluation is primarily concerned with understanding the context-mechanism-outcome configuration to address the question, 'what works, for whom, under what circumstances, and how' as opposed to whether an intervention works or not (Dalkin et al., 2015).
  • The realist approach is based on a ‘generative’ theory of causality: it is not programs that make things change, it is people, embedded in their context who, when exposed to programs, do something to activate both given mechanisms and change (Pawson et al., 1997).

학생의 학습 및 평가에 대한 구조적 요인의 영향을 이해하는 것뿐만 아니라 개별 학생의 주체성과 학습에서 선택할 수 있는 개인적 권한을 어떻게 행사할 수 있는지를 고려하는 것이 중요합니다(Kahn, 2013). 교사도 마찬가지입니다. 아처는 [사회 구조]와 [인간의 주체성] 간의 상호작용을 설명하기 위해 이러한 프레임워크를 제공합니다(Archer, 2000; 2003). [학습자의 주체성]은 기본적인 자아 감각에서 시작하여 프로그램에 참여하는 과정을 통해 발전하며 미래의 실무자가 되기 위한 기본 요소입니다. 에이전시를 설명하려면 [시간에 따른 구조와 에이전시 간의 복잡한 상호작용]을 포함해야 합니다(Archer, 2000, 2002; Priestley 외., 2015). Archer의 '형태 발생적' 접근법(Archer, 1982)에 따르면 학습자가 사회적, 실제적, 자연적 세계와 상호작용할 때 행위자성의 변화가 나타납니다. 개인이 교육 프로그램에 대한 누적된 경험에 반응하고 물질적, 사회적 조건이 변화함에 따라 행위자성의 변화는 긍정적이거나 부정적일 수 있습니다. 아처의 이론은 고등 교육 분야(Ashwin, 2008; Case, 2015; Leibowitz 외, 2012), 전문직 간 보건직업 교육(Hutchings 외, 2013), 의학교육에서 교육자가 소그룹 교육을 통해 전문성을 증진하는 방법을 조사하기 위해 광범위하게 사용되어 왔습니다(Kahn, 2013).
As well as understanding the influences of structural factors in student learning and assessment, it is important to consider the agency of individual students and how they might exercise their personal powers to make choices in their learning (Kahn, 2013). Similarly with teachers. Archer provides such a framework that seeks to account of the interplay between social structure and human agency (Archer, 2000; 2003). Learner agency develops from a basic sense of self through a process of engagement with programs and is fundamental to becoming a future practitioner. Explanation of agency must include the complex interplay between structure and agency over time (Archer, 2000, 2002; Priestley et al., 2015). In Archer’s ‘‘morphogenetic’’ approach (Archer, 1982) changes in agency emerge as a learner interacts with the social, practical and natural worlds. Change in agency may be both positive and negative as individuals respond to their cumulative experience of the educational program and as their material and social conditions evolve. Archer’s theory has been used extensively in the higher educational field (Ashwin, 2008; Case, 2015; Leibowitz et al., 2012), in interprofessional health professions education (Hutchings et al., 2013) and in medical education in order to investigate how educators develop their practice to promote professionalism through small group teaching (Kahn, 2013).

우리는 프로그램적 평가를 도입하는 복잡한 개입이 여러 환경에서 작동하여 무엇이, 어떻게, 누구를 위해, 어떤 맥락에서 작동하는지에 대한 유용한 통찰력을 제공할 수 있는 방법을 고려하고자 했습니다. 우리의 비판적 현실주의 접근법은 다음의 상호 작용을 탐구함으로써 평가 시스템 개혁에 영향을 미치는 조건의 복잡성을 해결하고자 합니다. 

  • 평가 구조(학생에게 수행되는 평가를 제한하거나 가능하게 하는 규칙 및 리소스)와
  • 학습자 주체성(학생과 교직원이 평가가 어떻게 학생과 함께 수행되었는지에 대한 의견을 제시하는 곳)

We wished to consider the ways in which the complex intervention of introducing programmatic assessment might work across multiple settings to provide useful insights of what works, how it works, for whom, and in what contexts. Our critical realist approach seeks to address the complexity of conditions that influence reforming of assessment systems by exploring the interplay of

  • assessment structures (rules and resources that constrain or enable assessment that is done to students) and
  • learner agency (where students and staff have input into how assessment is done with them).

우리의 중심 질문은 "기존 평가가 학습에 대한 전통적인 총괄 평가였던 의학 또는 보건 과학 프로그램에서 프로그램 평가 관행의 번성을 뒷받침하는 몇 가지 중요한 요소는 무엇인가?"였습니다.
Our central framing question was “What are some critical factors that underpin the flourishing of programmatic assessment practices in a medical or health science program where pre-existing assessment has been traditional summative assessments of learning?”.

방법
Methods

구조와 기관의 문제로서 평가 개혁에 관한 프로젝트의 맥락을 제공하기 위해, 우리는 기존 평가 시스템을 개혁하고 평가를 계획하는 한 의과대학의 전형적이지만 익명으로 처리된 비넷을 개발했습니다. 이 그림은 저자들의 평가 개혁에 대한 국제적 경험을 바탕으로 작성되었습니다. (표 1 참조). 
In order to provide a context for our project on assessment reform as a problem of structure and agency, we developed a typical but anonymised vignette of a medical school undergoing reform in its traditional assessment system, and planning an evaluation. The vignette is based on the authors’ international experience of assessment reform. (See Table 1).

평가 관행과 기관 간의 관계에 대한 우리의 프레임 질문을 다루기 위해, 우리는 헌신적인 개인의 반사성의 특징적인 측면, 즉 프로그램 평가의 측면을 구현할 때 궁극적인 관심사의 본질과 순서를 파악하기 위해 Archer가 개발한 이론에 대한 비판적 현실주의적 접근 방식을 수정했습니다(Archer, 2000, 2007). 처음 두 저자는 다른 저자들과 개별적인 개방형 질문 토론을 진행하면서 평가와 학습에 대한 서로 다르지만 광범위한 이론적, 경험적 경험들 사이에서 반사성을 장려했습니다.

  • 토론은 첫째, 프로그램적 평가를 도입하고 유지하는 데 있어 근본적인 문제를 규명하는 것으로 시작되었습니다.
  • 둘째, 잠재적인 기본 메커니즘에 대한 통찰력을 제공한 프로그램 평가의 주요 논문에 대한 해석을 논의하고,
  • 셋째, 세 개의 다른 기관에서 프로그램 평가의 구현을 설계하고 평가한 저자의 경험을 반성했습니다.

각 회의에 대한 자세한 메모가 작성되었습니다. 첫 번째 저자들은 새로운 주제를 도출하고 이를 수집 및 정리하여 추가 논의를 위해 그룹에 다시 제시했습니다. 이러한 반복적인 과정을 통해 합의에 도달하고 관련 문헌과 저자들의 전문가적 입장을 모두 담아낸 주제를 더욱 구체화할 수 있었습니다. 이 연구에서는 기관 윤리를 요구하지 않았습니다.  
In order to address our framing question about the relationship between assessment practices and agency, we modified a critical realist approach to theory developed by Archer in order to tease out characteristic aspects of reflexivity in committed individuals, about the nature and order of their ultimate concerns in implementing aspects of programmatic assessment (Archer, 2000, 2007). The first two authors held extended individual open-ended questions discussions with the other authors, encouraging reflexivity amongst our differing but extensive theoretical and empirical experiences of assessment and learning.

  • The discussions began with first, elucidating the fundamental issues in introducing and sustaining programmatic assessment.
  • Second, discussing interpretations of key papers in programmatic assessment that gave insight into potential underlying mechanisms, and
  • third reflecting on the author experience of designing and evaluating implementations of programmatic assessment across three different institutions.

Detailed notes were made of each of the meetings. The first authors drew out emergent themes which were captured and collated and presented back to the group for further discussion. Through this iterative process, we were able to reach a consensus and further refine the themes that captured both the pertaining literature and the expert positions of the authors. We did not require institutional ethics for this study.

연구 결과
Findings

프레이밍 리얼리즘 질문과 관련하여, 우리는 비판적 리얼리즘 관점에서 의료 또는 보건 과학 프로그램에서 새로운 평가 관행을 구현하는 맥락에서 네 가지 주제를 도출했습니다. 이는 다음과 같습니다:
In relation to our framing realist question, we drew out four themes in the context of implementing new assessment practices in a medical or health science program from a critical realist perspective. These were:

  • 1. 학습자의 적극적 참여와 주체성 활성화.
  • 2. 논증적 타당성에 중점을 둔 타당성 프레임워크 구축.
  • 3. 통합된 프레임워크 내에서 의도적이고 다양한 평가 과제를 설계합니다.
  • 4. [평가 규칙의 구조]와 [학습자 주체성] 간의 관계를 더 잘 이해할 수 있도록 공유된 내러티브를 일관되게 사용합니다.
  1. 1.Enabling positive learner engagement and agency.
  2. 2.Establishing validity frameworks that are oriented on argumentative validity.
  3. 3.Designing purposeful and eclectic assessment tasks within an integrated framework.
  4. 4.Developing consistent use of a shared narrative to provide a better understanding of the relationship between the structure of assessment rules and learner agency.

학습자의 긍정적인 참여 및 자율성 지원
Enabling positive learner engagement and agency

이 주제에서는 평가 관행을 변경할 때 학습자의 자율성을 향상하기 위해 무엇이 어떻게 작용할 수 있는지에 대한 관점에서 구조와 자율성 간의 상호 작용에 대해 설명합니다. [구조]는 [행동을 가능하게 하거나 제약할 수 있는 규칙 및 리소스]로 구성되며, 일부 구조적 특성은 다른 구조적 특성보다 변경하기가 더 어렵습니다. (Archer, 1982) 평가에서 학습자 주체성은 학습자를 선택을 하는 행위자로 인정하며, 학습자의 행동은 예상된 방식과 예상치 못한 방식으로 평가 관행을 형성합니다. (Adie 외., 2018).
This theme describes the interplay between structure and agency from the perspective of how and what might work to enhance learner agency when changing assessment practice. Structure consists of the rules and resources that may enable or constrain action, with some structural properties being more challenging to change than others. (Archer, 1982) Learner agency in assessment acknowledges learners as actors who make choices, and whose actions shape assessment practices in both anticipated and unexpected ways. (Adie et al., 2018).

[행위자성에 영향을 미치기 위해 평가를 재설계하는 것]은 교육 시스템 내의 모든 평가, 특히 널리 사용되는 고부담 총점 평가의 배경을 고려하여 고려해야 합니다. (Cilliers 외., 2010; Harrison 외., 2015, 2017). 개혁을 실행할 때 교수진이 새로운 평가 접근법을 통해 학습자가 참여할 수 있는 기회를 수용하면 구조와 학습자 주체 사이의 역학 관계가 바뀔 것입니다(Adie 외., 2018). 
Re-designing assessment to influence agency has to be considered against the backdrop of all assessment within the educational system, particularly the prevailing high-stakes summative assessments. (Cilliers et al., 2010; Harrison et al., 2015, 2017). In implementing reform, the dynamic between structure and learner agency will change where faculty embraces opportunities for learners to engage (Adie et al., 2018) with new assessment approaches.

[교사-학습자 관계의 특성] 또한 학습자 에이전시의 중요한 특징입니다. (Schut 외., 2021), 특히 [교사의 평가 신념](Meeuwissen 외., 2019)에 영향을 미칩니다. 교사가 커리큘럼 변화에 가져올 수 있는 신념, 가치 및 속성은 교사가 주체성을 달성할 수 있는 정도에 영향을 미칩니다(Priestley 외., 2012). 객관식(MCQ)(Haggerty, 1918)과 OSCE(Harden & Gleeson, 1979)가 지배적인 평가 관행 문화에서 교사가 사회화되어 온 방식은 보다 미묘한 평가 이론을 개발하는 데 있어 교사의 주체성을 제약할 수 있습니다(Vosniadou, 2014; Vosniadou 외., 2001). 교수개발은 교사의 주체성을 개발하고 학습 문화에 프로그램 평가에 대한 공유된 이해를 포함시키는 데 중요합니다. (Bok et al., 2016; Driessen et al., 2012; Schut et al., 2021) 
The nature of teacher-learner relationships is also an important feature of learner agency. (Schut et al., 2021) and in particular the assessment beliefs of teachers (Meeuwissen et al., 2019). The beliefs, values and attributes that teachers can bring to curriculum change impacts the extent to which they are able to achieve agency (Priestley et al., 2012). The ways in which teachers have been socialised in a culture of assessment practice dominated by multiple choice questions (MCQ) (Haggerty, 1918) and the OSCE, (Harden & Gleeson, 1979) can constrain their agency in developing a more nuanced theory of assessment (Vosniadou, 2014; Vosniadou et al., 2001). Faculty development is important in developing teacher agency and embedding a shared understanding of programmatic assessment in the learning culture. (Bok et al., 2016; Driessen et al., 2012; Schut et al., 2021)

그러나 학생은 평가 및 학습 참여를 유도하고 지원하는 다양한 상황적 요인을 경험할 가능성이 높습니다(Billett, 2001).

  • Heeneman 등(2015)은 관찰 및 피드백을 위한 작업장의 특성에 따라 학습자의 주체성에 미치는 영향이 달라질 수 있다고 지적합니다.
  • 학습자가 이해하지 못하거나, 학습의 이득에 비해 너무 복잡하거나, 업무량이 너무 많아 학습자의 주체성에 도전하는 평가 과제는 학습에 부정적인 영향을 미칠 수 있습니다(Cilliers et al., 2010).
  • 학습자의 주체성은 학습자가 평가 과제를 부담스러운 것으로 인식하는 경우에도 방해받을 수 있지만, 예를 들어 교사가 안전하고 신뢰할 수 있는 평가 관계에서 학습자의 주체성을 촉진함으로써 강화될 수 있습니다(Schut et al., 2020).

However, students are likely to experience differing situational factors that invite and support their engagement in assessment and learning (Billett, 2001).

  • Heeneman et al., 2015 also note the differences in characteristics of the workplace for observation and feedback, leading to differing impacts on learner agency.
  • Assessment tasks that challenge learners’ sense of agency by virtue of not being understood or too complex or too great a workload for the gain in learning may have a negative impact on learning (Cilliers et al., 2010).
  • Learner agency may be also hindered by learners' perceptions of assessment tasks as high stakes but can be empowered, for example, by teachers promoting learner agency in safe and trusting assessment relationships (Schut et al., 2020).

그럼에도 불구하고 학습자가 평가 설계에 적극적으로 참여할 수 있는 권리에는 책임이 수반됩니다. 학습자는 체계적인 평가의 표준과 규범, 특정 측정 방법, 학습 목표 및 달성해야 할 역량 간의 관계를 존중해야 합니다. 프로그램 평가의 원칙에 대한 학생 오리엔테이션과 개별 학생 및 학생 단체와의 지속적인 커뮤니케이션을 통해 상당한 학습자 개발이 필요합니다. 

Notwithstanding, the right for learners to be actively involved in assessment design comes with responsibilities. Learners must respect the standards and norms of systematic assessment and the relation between specific measurement methods, learning goals and the competencies to be achieved. Significant learner development will be needed with student orientation to the principles of programmatic assessment and ongoing communication with the individual students and the student body.

[학습 멘토 시스템]과 같이 학습자 행위자성을 지원할 수 있는 평가 구조에 새로운 리소스를 재할당하거나 제공하는 것도 고려해야 합니다(Schut et al., 2018). 전통적인 평가 실행 구조는 대규모 객관적 구조화 임상 시험(OSCE)과 서면 문제 은행을 실행하는 데 상당한 리소스를 필요로 합니다. 프로그래밍 방식의 평가는 맞춤형 소프트웨어인 e포트폴리오에서 보다 의미 있고 확실한 평가를 수집하고 대조하는 데 리소스가 필요할 수 있습니다(Driessen et al., 2007).
Consideration needs to be given to reallocating or providing new resources to those assessment structures which might support learner agency such as a learning mentor system (Schut et al., 2018). Traditional assessment practice structures require significant resource in running large-scale Objective Structured Clinical Examinations (OSCEs) and written question banks. Programmatic assessment might require resources to collect and collate more meaningful and authentic assessments in customized software; the ePortfolio (Driessen et al., 2007).

요약하면, 학습자 주도성을 촉진하기 위한 전략에는 [학습자와 함께] 평가가 이루어질 수 있도록 [평가 구조를 변경하는 것]이 포함됩니다. 변화의 주요 장벽은 교사와 학생이 기존의 평가 구조에서 벗어나 프로그램 평가에 유리한 조건을 제공하는 새로운 구조로 전환하는 것일 수 있습니다. 교수자가 평가 관행을 변경하는 경우, 학생은 학습 기회를 극대화하기 위해 적극적으로 참여할 책임이 있습니다. 학습자의 주체성을 촉진하기 위해 평가에 대한 리소스를 재할당해야 할 수도 있습니다. 이 시나리오에서 평가 관행의 구조를 변경하면 학습자의 주체성에 복잡한 영향을 미칠 수 있습니다. 동시에 학습자의 주체성을 촉진하려면 평가 구조의 변화가 필요합니다.
In summary, strategies to promote learner agency involve changing assessment structures so that assessment is done with learners. Major barriers to change may be the socialization of teachers and students away from pre-existing structures of assessment into newer structures that provide favourable conditions for programmatic assessment. Where faculty does change assessment practice, students have responsibilities to be active in maximizing their learning opportunities. Resources for assessment may need to be reallocated to promote learner agency. In our scenario, changing the structure of assessment practice will have complex impacts on learners’ agency. At the same time, promoting learner agency would require a change in the structure of assessment.

타당도 프레임워크 구축
Establishing validity frameworks

이 주제에서는 논증 기반 프레임워크가 제공하는 구조가 교수자, 학생 및 인증 기관과 같은 외부 이해관계자 모두에게 프로그램식 평가의 타당성 주장을 명확하게 하는 방법을 설명합니다.
This theme describes how the structure provided by an argument-based framework might give clarity around the validity claims of programmatic assessment for both faculty, students, and external stakeholders such as accreditation bodies.

프로그래밍 방식 평가는 [학습을 위한 평가]를 촉진합니다. 학습의 평가에서는, 원하는 학습 결과의 범위를 나타내는 통합된 작업 모음에 대한 결정이 내려집니다(Van der Vleuten & Schuwirth, 2005) 전통적인 평가 관행에서는 학습에 대한 검증된 총합 평가를 준비한다는 측면에서, ['형성'이라는 용어]가 [학습을 위한 평가]를 의미하는 다양한 평가를 사용하는 경향이 있습니다. 이론적 관점에서 이러한 가정을 비판하는 광범위한 문헌이 있습니다. (Bennett, 2011).
Programmatic assessment promotes assessment for learning. For assessment of learning, decisions are made on an integrated collection of work representing a range of desired learning outcomes (Van der Vleuten & Schuwirth, 2005) Traditional assessment practice tends to use a variety of assessments where the term ‘formative” implies assessment for learning, in preparation for a validated summative assessment of learning. There is an extensive literature critiquing this assumption from a theoretical perspective. (Bennett, 2011).

교수자가 형성적 평가를 프로그램적 평가로 인식하는 경우에도, 학생은 평가가 진급progression 결정에 사용되는 경우 형성적 평가를 총괄적 평가로 인식합니다(Heeneman 외., 2015). 학습에 대한 평가에서 형성적/총괄적 평가가 무엇인지에 대한 이러한 혼란을 피하기 위해, 우리는 프로그램식 평가를 뒷받침할 수 있는 타당성에 대한 논증 기반 관점으로 전환합니다.
Students perceive formative assessment as summative if it is used for any progression decisions (Heeneman et al., 2015) even where faculty perceive the assessment as programmatic. To avoid this confusion about what is formative/summative in the assessment for and of learning, we turn to an argument-based view of validity that can underpin programmatic assessment.

케인의 프레임워크는 프로그램적 평가 관행에 대한 검증 프로세스와 의도적으로 선택된 증거 기반 평가 과제의 타당성에 신뢰성을 부여하는 데 중요한 역할을 할 수 있습니다. (Kane, 2013; Schuwirth & Van der Vleuten, 2012).
Kane’s framework can play a significant role in lending credibility to the validation process for programmatic assessment practice, and the validity of purposely chosen and evidence-based assessment tasks. (Kane, 2013; Schuwirth & Van der Vleuten, 2012).

타당도는 [시험의 속성]이 아니라 [시험 점수의 해석 및 사용 제안의 속성]입니다.

  • 해석과 사용에 있어 타당성이 있고 적절한 증거에 의해 뒷받침되는 높은 타당도를 가진 것으로 간주됩니다.
  • 해석과 사용에 있어 적절하게 뒷받침되지 않거나 증거에 의해 모순되는 것은 타당도가 낮은 것으로 간주됩니다. (Kane, 2013) 

Validity is not a property of a test, but a property of the proposed interpretations and uses of test scores.

  • Interpretations and uses that make sense, and are supported by appropriate evidence, are considered to have high validity.
  • Those that are not adequately supported or are contradicted by evidence, are taken to have low validity. (Kane, 2013)

모든 이해관계자의 초점은 점수/측정/정보가 [의도한 결론을 도출하는 데 유효하게 사용될 수 있다]는 명제를 뒷받침할 수 있는 충분한 증거와 논리적 논거가 있는지에 있습니다. Kane의 주장에는 특정한 철학적 또는 이론적 제약이 없습니다(Schuwirth & Van der Vleuten, 2012). 타당성 논증을 구축하기 위한 케인의 조언에는 테스트 프로세스의 품질에 대한 구체적인 가설이나 테스트에 대한 구체적인 기법이 포함되어 있지 않습니다. 그는 평가 대상 구조의 본질에 대한 명확한 이론적 가정이 중요하며, 이해 관계자를 설득하기 위해서는 잘 논증된 일련의 추론이 필요하다고 조언합니다(Kane, 2013). 평가 프로그램에 사회적 관심사에 대한 고려가 포함된 경우, '사회적 명령으로서의 타당성'이 평가 전략에 포함될 수 있습니다(Marceau et al., 2018; St-Onge et al., 2017). 케인 프레임워크의 네 가지 추론 유형, 즉 채점, 일반화, 외삽, 결정 규칙 [평가 실무에서 일상적으로 사용할 수 있는 이론적 발판]을 제공합니다. 관찰에서 점수, 목표 영역으로, 마지막으로 구성으로 이어지는 논증적 추론은 [추론의 계층]을 추가하여 타당성을 향상시킵니다. 평가 프로그램의 타당성을 입증하는 데에도 동일한 원칙을 사용할 수 있습니다.

The focus for all stakeholders is on whether there is sufficient evidence and a logical argument to support the proposition that the scores/measures/information it yields can be validly used to draw the conclusions it was designed to support. There are no specific philosophical or theoretical constraints on Kane’s argument, (Schuwirth & Van der Vleuten, 2012). Kane’s advice for building the validity argument does not include specific hypotheses about the quality of test processes or specific techniques for testing them. He advises that a clear theoretical assumption about the nature of the construct being assessed is important and that a well-argued series of inference is needed to convince stakeholders (Kane, 2013). Where the program of assessments includes consideration of societal concerns, then ‘validity as a social imperative’ can be embedded in the assessment strategy (Marceau et al., 2018; St-Onge et al., 2017). The four types of inferences in Kane’s framework, namely, scoring, generalisations, extrapolation, and decision rules, provide theoretical scaffolding for their everyday use in assessment practice. Argumentative inferences from an observation to a score, to target domain, and finally to construct, add layers of inferences to improve validity. The same principles can be used to establish the validity of programs of assessments.

평가 설계자는 평가의 맥락과 목적에 따라 주장arguments의 성격이 추론에 가장 적합한 이유를 명확하고 일관성 있게, 그리고 그럴듯하게 방어해야 할 책임이 있습니다. 논증 기반 타당도 프레임워크를 전체 평가 프로그램으로 확장하려면 학습을 위한 평가의 [이중적 성격]을 인식해야 합니다(Van der Vleuten & Schuwirth, 2005).

  • 목표 역량을 평가하는 평가 프로그램의 경우, 채점 과정과 시험관 등급 생성에 일관된 원칙을 적용하여 평가 형식을 만들거나 조정할 때 주의를 기울여야 합니다. 이러한 평가는 단순히 완료/미완료, 만족/불만족 또는 등급을 매길 수 있습니다.
  • 의도적으로 설계된 과제 모음에서 기대되는 성과는 목표 역량을 평가하는 데 사용될 수 있는 과제의 '유니버스'에 걸쳐 일반화할 수 있어야 합니다.
  • 의미 있는 정보 삼각 측량을 통해 선택된 과제 모음과 그 평가 또는 서술적 피드백에서 외삽하는 것은 목표 역량 또는 다른 맥락에서의 향후 역량 성과를 측정하는 것이어야 합니다.
  • 선택된 과제의 수집을 기반으로 한 의사 결정 프로세스의 견고성이 갖는 함의를 고려해야 합니다. 교수진 개발은
    • 첫째, 개별 평가를 수행하고 피드백을 제공하는 방법에 대한 교사의 우려를 해결해야 합니다.
    • 둘째, 의사 결정 과정, 진행 규칙, 적절한 교정 및 추가 평가 규칙에 대한 이해와 자신감의 수준을 높입니다.

The onus is with the assessment designer to clearly, coherently, and plausibly defend why the nature of the arguments are best suited for the inferences according to the context and purpose of the assessment. Extending an argument-based validity framework into the whole program of assessments also needs to recognise the dual nature of the assessment for learning (Van der Vleuten & Schuwirth, 2005).

  • For programs of assessment to assess targeted competencies, care needs to be taken in creating or adapting assessment formats, by applying consistent principles on scoring processes and the generation of examiner ratings. These may be simple completed/not completed, satisfactory/unsatisfactory or graded.
  • The expected performance across the purposively designed collection of tasks should be generalisable across the ‘universe’ of tasks that could have been used to assess the target competencies. 
  • Extrapolating from the collection of selected tasks and their ratings or narrative feedback through meaningful triangulation of information should be a measure of the target competencies, or future performance of competencies in a different context.
  • Consideration needs to be given to the implications of the robustness of the decision-making process based on the collection of selected tasks. Faculty development will need to address teachers’ concerns about
    • first; how to undertake individual assessment and give feedback.
    • Second, increase the level of understanding and confidence in the decision-making processes, progression rules, and were appropriate remediation and further assessment rules.

케인은 평가 프로그램의 타당성 주장에 대한 엄격한 평가는 해석 사용 논증에서 [가장 약한 가정]이나 [고부담 결정을 뒷받침하는 가정]을 식별하고 우선순위를 정하는 것에서 시작해야 한다고 주장합니다. (Kane, 2013) 연구 결과는 평가 구조를 최적화하고, 교수자 개발을 개선하고, 학생의 학습 개발을 개선하는 데 다시 피드백될 수 있습니다.
Kane argues rigorous evaluation of the claims to validity of a program of assessments, should start by identifying and prioritising the weakest assumptions in the interpretation use argument or those that underpin high stakes decisions. (Kane, 2013) Findings can be fed back into optimising the structures of assessment, refining faculty development, and learning development of students.

요약하면, 논증 기반 타당도 프레임워크는 학생이 프로그램 평가를 위해 제시한 특정 정보 모음이 해당 학생의 진도 결정을 알리는 데 충분한 타당도를 가질 수 있는 방법과 이유를 명확하게 제공함으로써 교수자 및 학습자를 포함한 다양한 이해관계자에게 효과적일 수 있습니다.
In summary, an argument-based validity framework can work for a range of stakeholders including faculty and learners by providing clarity in how and why a particular collection of information presented by a student for programmatic assessment can have sufficient validity to inform progression decisions about that student.

목적이 있고 다양한 과제 설계
Designing purposeful and eclectic tasks

이 주제에서는 평가 시스템 구조와 학습자 기관 간의 교차점이 평가에 대한 다양한 접근 방식에 대한 일부 교수진의 순진함으로 인해 어떻게 영향을 받을 수 있는지에 대해 설명합니다.

  • 첫째, 순진한 경험주의: 학습자에 대한 지식의 기초를 형성하는 절대적인 사전 이론적 관찰 및 경험이 있다는 믿음입니다.
  • 둘째, 순진한 절충주의: 순전히 경험주의적이고 환원주의적인 접근 방식으로 평가 과제를 설계하거나 프로그램 내 관습 및 관행에 기반한 형식을 사용할 때 발생할 수 있습니다(Bhaskar 외., 2018).

This theme describes how the intersection between the structure of an assessment system and learner agency can be impacted by the naivety of some faculty in differing approaches to assessment.

  • First, naïve empiricism: the belief that there are some absolute pre-theoretical observations and experiences which form the basis of knowledge about a learner.
  • Second, naïve eclecticism (Bhaskar et al., 2018) which can arise when assessment tasks are designed on a purely empiricist and reductionist approach, or using formats based on custom and practice within the program.

좀 더 자세히 설명하자면, 학습자의 수행에 대한 '관찰'에 초점을 맞추면 평가 설계자가 순진한 경험주의, 즉 평가자의 평가가 오류 없이 절대적인 시험 점수를 제공할 수 있다는 생각으로 이어질 수 있습니다. 점수에 대한 지나친 강조는 학습자의 주체성을 해칠 수 있습니다. 숫자는 학습자의 성과에 대한 다양한 판단과 후속 학습을 개선하기 위한 피드백의 동화를 왜곡할 수 있습니다. 평가와 학습의 이론이 서로 어떻게 영향을 미치는지에 대한 최근의 검토에서 베어드 외, 2017은 평가의 이론적 근거는 불가지론적이라고 제안합니다. 
To illustrate further, a focus on ‘observation’ of learner performances can lead to naïve empiricism by the assessment designer; the notion that an assessor’s rating can provide a person’s test score that is absolute and free from error. An overemphasis on scores can be detrimental to learner agency. Numbers can skew the assimilation of multiple judgements of a learner’s performances and the attendant feedback aimed at refining subsequent learning. In a recent review of how theories of assessment and learning influence each other, Baird et al., 2017 suggest that the theoretical basis of assessment is agnostic.

학습을 위한 평가와 학습을 위한 평가는 평가의 서로 다른 기능을 제시하는 서로 다른 방식으로 간주됩니다. 이러한 주장은 형성 평가와 총괄 평가의 맥락에서 Norcini 외., 2018에 의해 제기되었습니다. 평가에 대한 이러한 이론적 접근의 의도하지 않은 결과는 평가 설계자가 정책, 커리큘럼 및 측정 의무를 충족하는 과정에서 발생하는 실용적 및 측정 고려 사항이 우세하다는 것입니다. 순진한 절충주의의 의도하지 않은 결과는 이론적 근거에 기반한 가정 없이, 오직 실용적으로 평가 과제를 선택할 수 있으며, 이러한 과제의 집합으로서 학습자의 다양한 역량에 대한 학습자의 진도에 대한 정보의 종적 흐름을 제공하지 못할 수 있다는 것입니다(Van der Vleuten & Schuwirth, 2005).
Assessment for learning and of learning are seen as differing ways of presenting the different functions of assessment. This argument is taken up in the context of formative and summative assessment by Norcini et al., 2018. An unintended consequence of this atheoretical approach to assessment is the dominance of practical and measurement considerations arising from assessment designers meeting policy, curriculum, and measurement imperatives. An unintended consequence of naive eclecticism is that assessment tasks may be chosen pragmatically without theoretically informed assumptions, and which may, as a collection of tasks, fail to provide for longitudinal flow of information about a learner's progress in various competencies (Van der Vleuten & Schuwirth, 2005).

[평가 관행을 혼합하는 것][이론]과 [실행의 실용성] 사이의 균형을 맞추기 위해 서로 반대되는 것처럼 보이는 평가 관행 사이의 긴장을 해결하는 한 가지 접근 방식입니다. 일반적으로 일부 학습 결과에는 전통적인 총괄 평가 요소를, 다른 학습 결과에는 프로그래밍 방식의 평가를 사용합니다. 안타깝게도 혼합형 평가 관행은 다양한 평가 과제의 중요성을 해석하는 타당성(Cook & Beckman, 2006)과 신뢰성(Driessen et al., 2012)을 둘러싸고 학습자와 교수진의 마음속에 긴장을 조성합니다. 가능한 경우 혼합 평가 관행을 피하는 것이 좋습니다. 
Hybridising assessment practice is one approach to resolving the tension between seemingly opposing assessment practices to balance theory with the pragmatics of implementation. Typically, by having an element of traditional summative assessment for some learning outcomes and programmatic assessment for others. Unfortunately, hybridised assessment practices create tensions in the minds of learners and faculty, around the validity (Cook & Beckman, 2006) and trustworthiness (Driessen et al., 2012) of interpreting the importance of various assessment tasks. It is beneficial to avoid hybrid assessment practice where possible.

요약하면, 평가 구조를 설계할 때 교수자의 순진한 경험주의는 프로그램적 평가의 구현을 방해하고 학습자의 주체성과 학습을 위한 평가를 제약할 수 있습니다. 학생과 교수진 모두에게 효과적인 평가 시스템을 만들려면 평가 시스템을 의도적으로 설계해야 합니다. 필요한 역량을 다루기 위한 다양한 평가 과제의 필요성은 프로그램에 적합한 증거 기반 평가 과제를 의도적으로 선택하여 그 타당성을 확립함으로써 관리할 수 있습니다(Schuwirth & Van der Vleuten, 2012).
In summary, naïve empiricism from faculty in designing the structures of assessment can impede the implementation of programmatic assessment and constrain learner agency and assessment for learning. To make an assessment system work for both students and faculty, it must be purposefully designed. The need for an eclectic set of assessment tasks to address the required competencies is managed by establishing the validity of purposely chosen evidence-based assessment tasks appropriate to the program (Schuwirth & Van der Vleuten, 2012).

공유된 내러티브 개발
Developing a shared narrative

이 주제는 평가와 학습 간의 상호작용에 대해 서로 다른 생각을 가진 교육자 간의 격차를 해소하기 위한 공유 내러티브에 대한 합의가 부족하다는 점에 초점을 맞췄습니다. (웹스터-라이트, 2009). 
This theme focussed on the lack of a consensus on a shared narrative to bridge the gap between educators holding differing ideas about the interplay between assessment and learning. (Webster-Wright, 2009).

우리는 [학습을 위한 평가]에서 중요한 두 가지 문제, 즉 비례성(평가의 지분)과 학생 학습에 대한 데이터의 삼각 측정을 논의함으로써 평가 관행을 개혁하는 데 있어 공유된 내러티브의 필요성을 설명합니다. 프로그램식 평가에는 [학습을 위한 평가]의 구조를 제공하고, 여기에는 기존의 총괄 평가와는 크게 다른 여러 가지 핵심 타당도 주장이 있습니다. 여기에는 다음이 포함됩니다(Dijkstra 외., 2012; Van der Vleuten 외., 2015). 

  • 비례성의 원칙,
  • 피드백의 가치,
  • 학습에 대한 평가와 학습을 위한 평가 간의 균형


We illustrate the need for a shared narrative in reforming assessment practice by discussing two issues that are important in assessment for learning: proportionality (the stakes of an assessment) and triangulation of data on student learning. In programmatic assessment, there are a number of core validity claims which provide a structure for assessment for learning, and which differ substantially from traditional summative assessment. These include

  • the principle of proportionality,
  • the value of feedback, and
  • the balance between assessment of and assessment for learning (Dijkstra et al., 2012; Van der Vleuten et al., 2015).

[비례성]은 학습자의 진도에 대한 결정에 대한 이해관계가 정보의 신뢰성 또는 풍부함에 비례해야 한다는 것을 의미합니다. 프로그램 평가에서 형성적-총괄적 평가의 전통적인 이분법은 의사 결정의 이해관계에 따른 연속적인 평가로 대체됩니다(Schuwirth et al., 2017). 비판적 현실주의 관점에서 볼 때, 무엇이 누구에게 왜 효과가 있는지에 대한 실제 설명은 다소 다릅니다. 프로그램 설계자는 주로 평가의 구조, 즉 학습자 선택에 미치는 영향을 검토하지 않고 타당도 규칙을 다루는 것처럼 보입니다. 현재 프로그래밍 평가의 언어에서는 학습자 선택권을 최적화하기 위해 개별 평가에 "낮은 위험도" 또는 "중간 위험도"라는 라벨을 붙입니다. Schut 등(2018)과 Heeneman 등(2015)은 '낮은 위험도 평가'를 뒷받침하는 프로그램 설계자의 의도와 이러한 평가를 높은 위험도로 인식하는 학습자의 인식 사이에 불일치가 있음을 보여주었습니다.
Proportionality implies that the stakes of decisions about a learner’s progress should be proportional to the credibility or richness of information. The traditional dichotomy of formative-summative assessments in programmatic assessment is replaced with a continuum of assessments depending upon stakes of decisions (Schuwirth et al., 2017). From a critical realist perspective, the real explanation of what is working for whom and why is somewhat different. It would seem program designers mainly address the structure of assessment, i.e., the validity rules without examining impacts on learner agency. In the current language of programmatic assessment, individual assessments are labelled as “low stakes” or “medium stakes” to optimise learner agency. Schut et al. (2018) and Heeneman et al. (2015) have shown that there is a mismatch between program designers’ intentions underpinning ‘low stakes assessments,” and learners’ perceptions of these assessments as high stakes.

이로 인해 학습자가 평가 과제를 저부담, 중부담 또는 고부담으로 간주하는지 또는 위험도가 평가의 조합을 기반으로 한 판단 또는 진도 결정을 의미하는지에 대해 혼란을 야기할 수 있습니다. 프로그래밍 방식의 평가를 구현할 때 개별 평가 과제에 할당된 '부담'이라는 단어는 세 가지 문제를 야기합니다. 

  • 첫째, 평가에 '스테이크'가 부여되는 순간 학습자는 평가가 '합산'된 것으로 인식하여 의도한 교육 효과를 방해할 수 있습니다(Heeneman 외., 2015; Schut 외., 2018; Watling & Ginsburg, 2019). 
  • 둘째, 프로그램식 평가의 적용 가능성을 주로 의사 결정 또는 판단 목적으로 제한하고 평가의 학습 효과를 가려버립니다
  • 개별 과제에 '부담'을 할당할 때 의도하지 않은 세 번째 효과는 학습을 위한 평가의 [이중 목적]에 미치는 영향입니다. 학생은 저부담 과제를 작업량 측면에서 수행할 가치가 없는 것으로 인식하고, 고부담 과제로부터 주의를 분산시키는 과제일 뿐이어서, 학습 목적도 낮은 가치로 간주할 수 있습니다. 

This leads to confusion as to whether students considered assessment tasks as low, medium, or high stakes, or whether the stakes refer to the judgment or progression decision based on the collation of the assessments. When implementing programmatic assessment, the ‘stakes’ assigned to individual assessment tasks leads to three issues.

  • First, the moment ‘stakes’ are assigned to assessments, they are perceived as ‘summative’ by learners which may interfere with the intended educational effect (Heeneman et al., 2015; Schut et al., 2018; Watling & Ginsburg, 2019).
  • Second, it limits the applicability of programmatic assessment mostly to the decision-making or judgemental purpose of assessments and eclipses the learning impact of assessment.
  • A third and unintended effect of assigning ‘stakes’ to individual tasks is its impact on the dual purpose of the assessment for learning. Students may perceive a task with low stakes as not worth doing in terms of the workload and distracting them from tasks designated high stakes and inferring that the learning purpose of the task is also deemed as low value.

따라서 '부담'이라는 용어의 사용하는 것은 그 자체로 학생의 자율성에 부정적인 영향을 미칠 수 있으며, '위험도가 낮은' 과제에 대해 풍부한 피드백을 제공하는 등 교사의 자율성을 증진하는 데 거의 도움이 되지 않습니다.
Thus, counterintuitively the use of the term ‘stakes’ can have a negative effect on student agency, and do little to promote teacher agency in for example giving rich feedback for ‘low stakes’ tasks.

'스테이크'라는 용어는 학습 과제에 대한 개별 평가와 관련하여 사용하지 않는 것이 가장 좋으며, [학습의 평가]를 목적으로 제출된 작품 모음에 대한 의사 결정 및 진급 결정을 위해 유보되어야 한다.
The term ‘stakes’ is best avoided in relation to individual assessment for learning tasks, and reserved for the decision making and progression decisions on the collection of works submitted for the purpose of in assessment of learning.

평가 관행 개혁의 두 번째 사례는 [삼각 측량 개념]입니다.

  • [전통적인 평가]에서 삼각 측량이란 프로그램에서 각 평가 요소의 최종 결과를 조합하는 것을 말합니다. 이러한 정보 결합 방식은 학생의 전반적인 역량에 대한 논증과 추론의 근거가 취약할 수 있습니다. 이 공식은 [결합 및 보상]과 같은 개념을 기반으로 합니다(Schuwirth & Van der Vleuten, 2012). 역량 기반 의학교육에서 결합과 보상의 개념은 논쟁의 여지가 있습니다. 예를 들어, 지식이 부족한 학생에게 의사소통을 잘하는 것이 충분한 보상이 될 수 있다고 주장할 수는 없습니다.
  • [프로그램식 평가]에서는 평가 도구 간의 삼각 측정을 통해 학습자의 강점, 약점 및 진도에 대한 의미 있는 정보를 제공합니다. 기존 평가에서는 임상 검사에서 복부 검사 스테이션의 결과와 무릎 검사의 결과와 같이 동일한 형식이기 때문에 서로 다른 부분이 결합됩니다. 프로그래밍 방식 평가는 의미 있는 콘텐츠를 기반으로 삼각 측정을 추구한다는 점에서 근본적으로 다릅니다. 즉, 서로 다른 평가 도구의 일부가 의미 있게 결합되어 정보가 풍부한 내러티브로 이어질 수 있기 때문에 서로 삼각 측량됩니다. 

Our second example of a shared narrative issue in assessment practice reform is the concept of triangulation.

  • In traditional assessment, triangulation refers to a combination of the final results of each assessment element in the program. This way of combining information may provide a weak basis for argumentation and making inferences about the students’ overall competence. This formulation is based on notions such as conjunction and compensation (Schuwirth & Van der Vleuten, 2012). In competency-based medical education, the notions of conjunction and compensation are contentious; one cannot argue that for example, being a good communicator would be sufficient compensation for a student having insufficient knowledge.
  • In programmatic assessment, triangulation across assessment instruments provides meaningful information about a learner’s strengths, weaknesses and progress. In traditional assessment different parts are combined because they are of the same format, e.g., the results on an abdominal examination station in a clinical examination with those on a knee examination. Programmatic assessment is fundamentally different in that it seeks triangulation based on meaningful content. This means that parts of different assessment instruments are triangulated with each other because they can be meaningfully combined leading to information-rich narratives.

또한 삼각측정은 단일 평가 형식, 단독 평가자, 단일 관점의 이론이나 모델의 한계를 극복합니다. 삼각측량과 삼각측량 수행 방법에 대한 명확한 규칙은 평가 프레임워크의 타당성 논증의 요소를 강화합니다. 이 평가 시스템에서 나오는 풍부한 정보 결과와 내러티브에 중점을 두면 [학습을 위한 평가]에 더 잘 적용될 수 있습니다. 동시에 삼각측량 과정의 종단적 특성과 풍부한 정보는 환자에게 임상의의 역량을 확신시켜야 하는 면허 취득 상황과 같은 고위험 학습 평가에 더욱 적합하게 만듭니다. (Norcini 등, 2018)

  • 전통적인 시험에서는 각 역량 영역이 개별적으로 측정 가능한 구성 요소로 간주되는 반면,
  • 프로그램적 평가에서는 [역량]이라는 [다면적인 현상에 대한 다양한 렌즈]로 간주되며, 각 영역은 서로 보완적인 역할을 합니다(Schuwirth & Van der Vleuten, 2012).

Moreover, triangulation overcomes the limitation of singular assessment formats, lone assessors, and single-perspective theories or models. Clear rules on triangulation and how it was performed, will enhance elements of the validity argument of the assessment framework. The emphasis on information-rich results and narratives that emerge from this system of assessment make it more applicable for an assessment for learning. At the same time, the longitudinal nature and the information richness of the triangulation process make it more defensible for high-stakes assessment of learning, for example the licensing context, where patients need to be assured of their clinician’s’ competence. (Norcini et al., 2018)

  • Where in traditional testing, each competency domain is seen as a separately measurable construct,
  • in programmatic assessment, they are seen as a variety of lenses on the multifaceted phenomenon of competence, and each is complementary to the other (Schuwirth & Van der Vleuten, 2012).

종단적 평가 정보가 쉽게 접근 가능하고 상호 참조할 수 있는 형식으로 제공되고, 소위 '저부담' 평가를 포함한 모든 평가에서 삼각측정이 학생의 전반적인 역량에 기여하는 것을 볼 수 있는 경우, 교사와 학생은 비례성삼각측량의 개념을 더 잘 이해할 수 있습니다. 이는 의사 결정을 지원하기 위해 학생들의 정보 모음을 시각화하는 데 기술을 사용하는 것과 관련이 있습니다.
The concepts of proportionality and triangulation are more likely to be appreciated by teachers and students where longitudinal assessment information is in an easily accessible and cross-referential format and the contribution of triangulating across all assessment including so-called ‘low stakes’ assessments can be seen contributing to the overall capabilities of the students. This relates to the use of technology in visualising collections of astudents’ information to support decision-making.

요약하면, '스테이크' 및 '평가의 삼각 측량'과 같은 [주요 용어의 정의]에 대한 아이디어를 공유하면 교육자가 평가와 학습의 상호 작용을 더 잘 이해하고 다양한 평가 관행의 언어에 대한 유창성을 개발하는 데 도움이 될 수 있습니다. 이러한 방식으로 교수진과 학생 모두 몇 가지 지침과 이론적 틀을 통해 평가 관행을 변화시킬 수 있는 역량을 강화할 수 있습니다. 교수자와 학습자는 아이디어를 공유해야만 학습자의 선택권을 부여하고 평가 관행의 구조를 어느 정도 통제할 수 있는 선택권과 조치를 취할 수 있습니다(Archer, 2003).
In summary, sharing of ideas about the definitions of key terms such as ‘stakes’ and ‘triangulation of assessment’ could help educators better appreciate the interplay of assessment and learning and develop fluency in the language of differing assessment practices. In this way both faculty members and students might be better empowered to change assessment practice with some guidelines and theoretical framing. Only through sharing their ideas can faculty and learners have the power to make choices and take action to privilege learner agency and develop a degree of control over the structure of assessment practice (Archer, 2003).

토론
Discussion

우리의 비판적 현실주의 질문은 새로운 평가 관행을 구현하고 평가하는 맥락에서 [평가 관행의 구조]와 [학습자 주체성] 간의 상호 작용을 탐구했습니다. 우리는 개혁 관리의 문제에 대한 더 깊은 설명은 평가 관행을 뒷받침하는 구조와 학습자 선택의 어포던스 사이의 복잡한 역학 관계에 대한 설명에서 찾을 수 있다고 주장했습니다. 우리는 구조, 에이전시 및 이들 간의 상호 작용에 대한 정보를 제공함으로써 기존의 전통적인 평가 시스템이 있는 맥락에서 프로그램 평가와 같은 새로운 프로그램이 미치는 영향에 대한 초기 이해를 제공하고자 했습니다(De Souza, 2013; Pawson 외., 1997).
Our framing critical realist question explored the interplay between the structure of assessment practice and learner agency in the context of implementing and evaluating new assessment practices. We have argued that a deeper explanation of the problem in managing reform can be found in accounts of the complex dynamics between the structures underpinning assessment practice and the affordances of learner agency. We have sought to provide some initial understanding of the impact of a new program such as programmatic assessment in a context where there is a pre-existng traditional assessment system by providing information about the workings of structure, agency and the interplay between them (De Souza, 2013; Pawson et al., 1997).

네 가지 주요 관점에 주의를 기울이면 기존 평가 관행의 한계를 이해하고 새로운 평가 관행을 개발하는 데 도움이 될 것입니다. 

  • 학습자의 주체성 활성화, 
  • 새로운 타당성 논거의 재구성 및 확립, 
  • 목적에 맞는 평가 과제 설계, 
  • 이해관계자가 평가에 대한 새로운 접근 방식을 이해하고 수용할 수 있도록 교수자 개발을 촉진하는 것입니다. 

이러한 관점을 통해 교육자는 프로그램 평가와 같은 개입이 여러 맥락에서 누구를 위해, 어떤 상황에서, 왜 작동하는지에 대한 복잡성을 풀 수 있습니다.
Paying attention to four key perspectives will aid in both understanding the limitations of traditional assessment practice and in developing new assessment practice. These are:

  • enabling learner agency,
  • restructuring and establishing new validity arguments,
  • designing purposeful assessment tasks, and
  • promoting faculty development to enhance the fluency of stakeholders to understand and accept new approaches to assessment.

These perspectives allow educators to unpack the complexity of how interventions such as programmatic assessment work across multiple contexts, for whom, under what circumstances, and why?

비판적 현실주의 렌즈를 통해 평가 개혁을 실행하고 평가할 때 이론, 정책 및 실무에 대한 여러 가지 시사점이 있습니다. 구조와 행위자성 간의 상호 작용에 대한 이러한 접근 방식은 교육자가 자신의 교육기관에서 평가 관행에 대한 주요 변화의 복잡성을 이해하는 데 유용한 통찰력을 제공할 수 있습니다. 이는 정책, 커리큘럼 및 측정 의무를 충족하고 연구 질문, 데이터 수집 및 분석 방법 안내를 포함한 평가 연구 프로토콜을 개발할 때 실제 평가 고려 사항을 기반으로 할 수 있는 이론적 틀을 제공할 수 있습니다. 
There are a number of implications for theory, policy, and practice when implementing and evaluating assessment reforms when viewed through a critical realist lens. This approach around the interplay between structure and agency can give helpful insights for educators in making sense of the complexity of major changes to assessment practice in their own institutions. It can provide a theoretical framing on which to base practical assessment considerations in meeting policy, curriculum, and measurement imperatives, and in developing evaluative research protocols, including guiding research questions, data collection and analysis methods.

표 2에서는 프로그램 평가를 실행하거나 지속하는 데 관심이 있는 사람들을 위해 각 주제에서 다양한 실제 적용 사례를 제공하며, 비네트(표 1)에 요약된 문제 유형을 다룹니다.
In Table 2, we provide a number of practical applications from each theme for those interested in implementing or sustaining programmatic assessment and addressed the kinds of issues outlined in our vignette (Table 1).

[학습자 주도성을 촉진하기 위한 전략]에는 학습자와 함께 평가가 이루어질 수 있도록 평가 구조를 변경하는 것이 포함됩니다. 학생은 학습 기회를 극대화하고 성취도에서 표준을 입증할 수 있도록 연중 내내 자신의 책임에 대해 방향을 설정하고 지속적으로 참여해야 합니다.
Strategies to promote learner agency involve changing assessment structures so that assessment is done with learners. Students need to be orientated and continually engaged throughout the year around their responsibilities to maximize their learning opportunities and demonstrate standards in their achievements.

[Kane과 같은 논증적 프레임워크를 재구성]하면 교수자, 학생 및 인증 기관과 같은 외부 이해관계자 모두에게 평가의 타당성과 표준에 대한 명확성을 제공할 수 있습니다(Kane, 2013; Schuwirth & Van der Vleuten, 2012). 그러면 학생은 자신의 학습을 결정할 때 진정으로 얼마나 많은 권한을 가질 수 있는지, 학습 결과의 성취를 뒷받침하는 데 사용할 수 있는 증거의 종류에 대해 더 명확하게 알 수 있습니다. 학생 점수/측정/정보에 대한 의사 결정의 타당성을 명확히 하면 교수진, 학생, 인증 기관, 의료 서비스 제공자 및 커뮤니티에 확신을 줄 수 있습니다.
Restructuring an argumentative framework such as Kane’s gives clarity around validity and standards of assessment for both faculty, students and external stakeholders such as accreditation bodies (Kane, 2013; Schuwirth & Van der Vleuten, 2012). Students can then have more clarity around how much agency they can genuinely assume in determining their learning and the kinds of evidence they can use to underpin their achievements of learning outcomes. Clarity around the validity of decision making on student scores/measures/information in their collections of evidence gives assurity for faculty, students, accreditation bodies, healthcare providers, and the community.

[교수진 개발 전략]은 학생과 교사를 새로운 평가 관행의 변경된 언어와 규칙에 익숙하게 만드는 데 매우 중요합니다. 이러한 방식으로 교수진과 학생 모두 평가 관행의 변화를 지속할 수 있는 역량을 강화할 수 있습니다(Leibowitz et al., 2012).

  • 교수자 개발은 새로운 평가 관행의 변화된 언어와 규칙에 익숙해지도록 할 때 교사의 상황에 따라 영향을 받은 평가 경험뿐만 아니라 제도적 맥락을 고려해야 합니다.
  • 마찬가지로, 프로그램 평가에 대한 학습자 개발도 학습 활동에 포함되어야 합니다. 워크숍은 개인(교사와 학생)이 변화하는 평가 관행에 대한 우려를 해결하고 우선순위를 정하는 반성적 숙고, 즉 내부 대화를 촉진할 수 있습니다(Archer, 2003).

이러한 방식으로 교수진은 평가 관행에 대한 신뢰를 갖고 변화를 지속할 수 있으며(Schut et al., 2020), 교사-학습자 관계가 개선될 수 있습니다. (Schut 외, 2021) 
Strategies for faculty development are critical in socializing student and teachers into the changed language and rules of the new assessment practices. In this way both faculty members and students might be better empowered to sustain changes in assessment practice (Leibowitz et al., 2012).

  • Faculty development needs to take the institutional context as well as teachers’ contextually influenced experiences of assessment into account when socializing them into the changed language and rules of the new assessment practices.
  • Similarly, learner development around programmatic assessment needs to be embedded in the learning activities. Workshops can promote an internal conversation; the reflexive deliberations through which the individual (teachers and students) address and prioritise their concerns about changing assessment practice (Archer, 2003).

In these ways, both faculty members might be better empowered to have trust in and sustain changes in assessment practice (Schut et al., 2020) and have an improved teacher-learner relationship. (Schut et al., 2021)

비판적 현실주의 접근법은 의과대학 또는 대학 내의 다양한 주체(예: 프로그램 및 평가 위원회 구조)가 어떻게 인과적 힘을 행사하고 의도된(예: 통합 평가 시스템) 및 의도하지 않은 관찰 가능한 결과(예: 과도한 평가를 통한 학습자 이탈)로 이어지는 일련의 사건 패턴을 생성할 수 있는 메커니즘을 생성할 수 있는지를 설명하는 데 도움이 될 수 있습니다. 권력 문제가 평가 이외의 '현실' 구조의 겹겹이 쌓인 층과 상호작용하여 개혁에 영향을 미치는 방식에서 우리의 시나리오에 대한 추가적인 방법론적 탐구의 기회가 있습니다. 예를 들어, 교육 거버넌스, 그리고 선임 교수진(학장 및 부학장)과 같은 인과적 권한을 가진 기타 행위자들(Bhaskar, 1978)이 이에 해당합니다. 향후 연구에서는 이 방법론적 접근 방식을 평가 관행의 주요 변화에 관여하는 프로그램의 경험적 데이터에 적용하여 주요 기본 메커니즘을 밝힐 수 있습니다.

A critical realist approach might help explain how various entities within the medical school or university, for example, the program and assessment committee structures, are capable of exercising causal powers and generating mechanisms that can create a pattern of events leading to intended (e.g., an integrated assessment system) and unintended observable outcomes (e.g., learner disengagement through over assessment). There is an opportunity for further methodological exploration of our scenario in the ways in which issues of power interact with the laminated layers of ‘reality’ of structures other than assessment, to impact reform. For example, educational governance, and other agencies having causal powers, for example senior faculty (Deans and Associate Deans) (Bhaskar, 1978). Future research might apply this methodological approach to empirical data from programs engaged in major change in assessment practice to illuminate the key underlying mechanisms.

불확실성 영역
Areas of uncertainty

이 논문은 변화하는 평가 관행에 대한 비판적 현실주의적 탐구를 제공하는 몇 안 되는 문헌 중 하나입니다. 이 연구는 평가 관행의 구조 개혁에 대한 현재의 맥락적 영향, 평가와 학습 간의 관계에 대한 다양한 생각, 학습자의 선택권을 강화하기 위해 평가 구조와 프레임워크를 조정하는 데 필요한 여러 단계와 관련하여 설명된 구조와 선택권 간의 상호 작용이라는 개념을 지지합니다. 저희는 저희의 분석과 이론화가 데이터에 기반한 것이 아니라 반사적이라는 점을 인정하지만, 평가 관행 변화의 맥락에서 경험적 연구에 대한 중요한 노력을 고려하는 사람들에게 견고한 이론적 근거를 제공했다고 믿습니다. 저희는 근본적인 메커니즘을 제공하지 못했음을 인정합니다. 그보다는 맥락이 풍부한 연구 환경에서 경험적 데이터를 검토할 때 해결해야 할 가능성이 있는 논쟁의 종류를 설명했습니다.
This paper is one of few in the literature that provide a critical realist exploration of changing assessment practice. The study supports the notion of an interplay between structure and agency, which has been illustrated in relation to current contextual influences on reforming the structure of assessment practices, differing ideas about the relationship between assessment and learning, and the several steps required to adapt assessment structures and frameworks to enhance learner agency. We acknowledge that our analysis and theorising was reflexive and not data driven, but we believe we have provided a solid theoretical grounding for those contemplating significant efforts on empirical research in contexts of assessment practice change. We recognise that we have not provided underlying mechanisms as such. Rather, we have described the kinds of arguments that are likely to need addressing when reviewing empirical data in context rich research settings.

결론
Conclusion

비판적 현실주의 접근법은 교육자가 평가 실무에서 개혁을 반영하고, 실행하고, 평가하는 데 도움이 될 수 있습니다. 비판적 현실주의 접근법은 프로그램 평가와 같은 혁신이 여러 맥락에서 어떻게 작동할 수 있는지, 그리고 어떤 것이 누구에게 효과가 있는지 설명하는 데 도움이 될 수 있습니다. 저희는 교육자가 각자의 맥락에서 평가 관행의 구조를 개혁하는 데 도움이 되는 근본적인 설명을 찾는 데 도움이 될 수 있는 네 가지 관점을 제안합니다. 여기에는 다음이 포함됩니다. 

  • 학습자의 참여와 선택권 보장,
  • 논증에 기반한 타당도를 강조하는 타당도 프레임워크 수립,
  • 증거에 기반한 의도적이고 다양한 평가 과제 설계,
  • 평가 구조와 학습자의 선택권에 대한 대화의 결과를 통합할 수 있는 공유 내러티브 개발

이러한 관점과 관련된 전략을 적용하는 방법에 대한 몇 가지 지침을 제공합니다.
A critical realist approach may help educators in reflecting on, implementing, and evaluating reforms in assessment practice. It can help identify how innovations such as programmatic assessment might work across multiple contexts and explain how it works, and what works for whom. We propose four perspectives which can aid educators to find underlying explanations that are helpful in reforming the structures of assessment practice in their contexts. These include

  • enabling learner engagement and agency,
  • establishing validity frameworks that emphasizes argument-based validity,
  • designing purposeful and eclectic assessment tasks that are evidence-based, and
  • developing a shared narrative that can integrate outcomes of conversations about the structure of assessment and learner agency.

We provide some guidance on how to apply strategies related to these perspectives.


Adv Health Sci Educ Theory Pract. 2021 Dec;26(5):1641-1657. doi: 10.1007/s10459-021-10065-8. Epub 2021 Aug 24.

Exploring complexities in the reform of assessment practice: a critical realist perspective

Affiliations collapse

Affiliations

1Faculty of Medicine and Health, Education Office, Sydney Medical School, The University of Sydney, Sydney, NSW, Australia. christopher.roberts@sydney.edu.au.

2Faculty of Medicine and Health, Education Office, Sydney Medical School, The University of Sydney, Sydney, NSW, Australia.

3Centre for Research on Learning and Innovation (CRLI), The University of Sydney, Sydney, NSW, Australia.

4Prideaux Discipline of Clinical Education, College of Medicine and Public Health, Flinders University, Adelaide, South Australia, Australia.

PMID: 34431028

DOI: 10.1007/s10459-021-10065-8

Abstract

Although the principles behind assessment for and as learning are well-established, there can be a struggle when reforming traditional assessment of learning to a program which encompasses assessment for and as learning. When introducing and reporting reforms, tensions in faculty may arise because of differing beliefs about the relationship between assessment and learning and the rules for the validity of assessments. Traditional systems of assessment of learning privilege objective, structured quantification of learners' performances, and are done to the students. Newer systems of assessment promote assessment for learning, emphasise subjectivity, collate data from multiple sources, emphasise narrative-rich feedback to promote learner agency, and are done with the students. This contrast has implications for implementation and evaluative research. Research of assessment which is done to students typically asks, "what works", whereas assessment that is done with the students focuses on more complex questions such as "what works, for whom, in which context, and why?" We applied such a critical realist perspective drawing on the interplay between structure and agency, and a systems approach to explore what theory says about introducing programmatic assessment in the context of pre-existing traditional approaches. Using a reflective technique, the internal conversation, we developed four factors that can assist educators considering major change to assessment practice in their own contexts. These include enabling positive learner agency and engagement; establishing argument-based validity frameworks; designing purposeful and eclectic evidence-based assessment tasks; and developing a shared narrative that promotes reflexivity in appreciating the complex relationships between assessment and learning.

Keywords: Assessment reform; Clinical competence; Critical realism; Programmatic assessment; Structure and agency.

멀고 숨겨진 존재: 환자를 EPA의 개발, 내용, 도입의 전면에 두기(Acad Med, 2021)
Distant and Hidden Figures: Foregrounding Patients in the Development, Content, and Implementation of Entrustable Professional Activities 
Stefanie S. Sebok-Syer, PhD, Andrea Gingerich, ND, MMed, PhD, Eric S. Holmboe, MD, Lorelei Lingard, PhD, David A. Turner, MD, and Daniel J. Schumacher, MD, PhD, MEd 

 

역량 기반 의료 교육(CBME)개별 환자, 가족 및 지역사회의 의료 요구를 충족하기 위해 수련의가 습득해야 하는 능력을 강조함으로써 결과에 초점을 맞춘 학습자 중심 프레임워크입니다. 1 결과를 목표로 하는 CBME는 학습자에 대한 더 나은 교육과 평가를 통해 환자 치료를 개선하고자 합니다. 2 CBME에 대한 접근 방식으로서, 위임 가능한 전문 활동(EPA)은 자격을 갖춘 전문가가 환자에게 안전하고 효과적인 진료를 제공하기 위해 수행할 수 있어야 하는 활동을 설명합니다. 따라서 EPA는 환자 치료의 결과를 직장에서 전문가의 임상적 행동 및 결정과 연결할 수 있습니다. 3-5 이 논문에서는 EPA가 환자 중심주의에 대한 약속을 어느 정도 이행하는지 살펴봅니다. 
Competency-based medical education (CBME) is a learner-centered framework that focuses on outcomes by emphasizing the abilities a trainee must acquire to meet the health care needs of individual patients, families, and their communities. 1 By targeting outcomes, CBME seeks to enhance patient care through better teaching and assessment of learners. 2 As an approach to CBME, entrustable professional activities (EPAs) describe activities that qualified professionals must be able to perform to deliver safe and effective care to patients. Thus, EPAs can link outcomes of patient care with clinical actions and decisions of professionals in the workplace. 3–5 In this paper, we examine the extent to which EPAs deliver on their promise of patient-centeredness.

최근의 문헌은 환자를 평가 방정식에 포함시킬 수 있는 EPA가 제공하는 기회에 대해 이야기합니다. 6,7 EPA는 [학습자 결과]에만 초점을 맞추지 않고 [환자 결과]를 고려하는 최초의 평가 프레임워크입니다. EPA의 전문적 활동 측면은 커리큘럼을 개발하는 데 사용될 수 있으며, 이를 사용하여 내린 위임 결정은 확실히 평가에 사용될 수 있지만, 환자 중심의 교육 접근 방식을 달성하려면 환자의 요구에서 시작하여 이러한 요구를 충족하는 커리큘럼과 평가를 설계하는 것이 필요합니다. 8 이 작업은 말처럼 간단하지 않을 수 있습니다.

  • 한 가지 접근 방식은 EPA 개발에 환자를 포함하는 것입니다. 9
  • 또 다른 접근 방식은 EPA 실행에 환자를 치료의 공동 생산자로 포함시키는 것을 목표로 의도적으로 EPA를 개발하는 것입니다. 10

Recent literature speaks to the opportunities afforded by EPAs that allow us to bring the patient into the assessment equation. 6,7 EPAs are the first assessment framework to consider patient outcomes rather than focus solely on learner outcomes. The professional activity aspect of EPAs can be used to develop curricula, and entrustment decisions made using them can certainly be used for assessment, but achieving a patient-centered approach to education necessitates starting with the needs of the patient and designing curricula and assessments to meet those needs. 8 This undertaking may not be as simple as it sounds.

  • One approach is to include patients in the development of EPAs. 9 
  • Another approach is to intentionally develop EPAs with the goal of including patients as co-producers of their care when the EPA is executed. 10 

환자에게 보다 명시적으로 초점을 맞추기 위한 이러한 변화로 인해 우리는 다음과 같은 질문을 탐구하게 되었습니다: 환자를 수련의 및 감독자의 초점과 함께 전면에 내세우면 EPA 개발, 내용 및 실행에 어떤 변화가 있을까요? 
This shift toward focusing more explicitly on the patient led us to explore the question: What happens to EPA development, content, and implementation when we foreground the patient alongside trainee and supervisor foci?

먼저 우리 주장의 중요한 전제를 확립하는 것으로 시작하겠습니다: EPA는 (종종 그렇게 사용되기는 하지만) 단순한 평가 프레임워크가 아닙니다. 오히려 환자에게 필요한 것이 무엇인지 파악하는 것에서 시작하여, 그 정보를 사용하여 이러한 요구를 충족할 수 있는 커리큘럼과 평가 접근법을 개발하는 CBME를 설계하고 구현하는 접근 방식입니다. EPA를 평가 프레임워크로만 바라보는 것은 거의 전적으로 교육생에 초점을 맞추는 것입니다. 이렇게 교육생에게만 초점을 맞추면 환자는 그늘에 가려질 수 있습니다. 교육과 환자 결과를 성공적으로 연결하려면 평가 프레임워크의 관점을 넘어 EPA에 대한 관점을 넓혀서 교육 과정과 평가를 모두 대표하는 것으로 EPA를 고려해야 합니다. 그렇게 함으로써 환자에게 주의를 환기하고 수련의 중심 접근법과 환자 중심 접근법을 모두 지원하는 방식으로 수련의의 요구를 해결할 수 있습니다. 11 
Let us begin by establishing an important premise of our argument: EPAs are not simply an assessment framework, despite the fact that they are often used as such. Rather, they are an approach to designing and implementing CBME, which begins with determining what patients need and then uses that information to develop curricula and approaches to assessment that ensure those needs can be met. Viewing EPAs solely as an assessment framework foregrounds, almost exclusively, the trainee. Casting this spotlight on the trainee may leave the patient in the shadows. If we want to successfully link educational and patient outcomes, we need to broaden our perspective of EPAs beyond that of an assessment framework and consider EPAs as representative of both curricula and assessment. In doing so, we can return our attention to the patient and also address the needs of our trainees in a way that supports both trainee-focused and patient-centered approaches. 11

EPA에 환자를 포함해야 한다는 데 이의를 제기하는 사람은 거의 없지만, 그 여부와 방법을 조사한 경험적 연구는 거의 없습니다. EPA를 수련의에게 맡겨야 할 일련의 활동 이상으로 생각하면 환자의 역할에 대해 더 풍부한 질문을 할 수 있습니다. 최근의 한 논문이 이러한 이점을 잘 보여줍니다. El-Haddad와 동료 9는 질적 방법론을 사용하여 EPA 개발에서 환자의 역할을 조사했습니다. 이들의 연구는 EPA에서 환자의 역할을 스케치하기 시작하여 환자가 의사를 보완하는 고유한 관점을 제공하며, 진정한 환자 중심의 방식으로 기대치와 임상 실습을 조정하는 데 도움이 될 수 있다는 결론을 내렸습니다. 9 연구 결과에 따르면 의사와 환자는 수련의가 임상 진료를 제공할 때 무엇을 알고 어떻게 해야 하는지, 그리고 그러한 지식과 기술이 왜 중요한지 파악하는 데 중요한 역할을 합니다. 이 연구는 의사와 환자의 상호의존성을 어떻게 바라보는지에 대한 정보를 제공하고 EPA의 개발과 실행에 있어 환자에 대한 이해를 안내하는 데 도움이 될 수 있습니다.  
While few would disagree that we should include patients in EPAs, little empirical research has explored whether and how this occurs. If we think about EPAs as more than a set of activities to be entrusted to trainees, we can ask richer questions about the role of patients. A recent paper illustrates this benefit. El-Haddad and colleagues 9 used qualitative methodology to examine the role of patients in EPA development. Their work begins to sketch the role of patients in EPAs and concludes that patients offer a unique perspective, which is complementary to physicians, and can help align both expectations and clinical practice in a way that is truly patient centered. 9 Their findings demonstrate that physicians and patients are vital in identifying both what is important for a trainee to know and do in the context of providing clinical care and why that knowledge and those skills matter. This work can help inform how we view physician–patient interdependence and guide our understanding of patients in the development and implementation of EPAs.

그러나 의사-환자 상호의존성을 고려할 때, EPA가 수행되는 임상 현장에서 의사 수련의는 한 개인에게 성과를 쉽게 돌릴 수 없는 협업 팀의 일원이라는 점도 기억해야 합니다. 12

  • 수련의수퍼바이저와 불가분의 관계에 있으며,13 이러한 수퍼바이저와 수련의의 상호의존성 때문에 어떤 환자 결과가 특정 수련의, 수퍼바이저 또는 조직적 맥락과 같은 다른 측면에 기인하는지를 식별하는 것이 특히 어렵습니다. 14,15
  • 또한, 수퍼비전은 일부 수련자에 대한 책임 부여를 공식화한 위임 결정에 의해 EPA 프레임워크에 포함되었습니다. 16,17
  • 이러한 척도는 또한 수퍼바이저가 생각하는 방식에 맞춰 구성되도록 설계되어 있어,18 수퍼바이저-수련생-환자의 상호의존성을 삼원적으로 만듭니다.

따라서 우리는 EPA에 존재하는 상호 의존성환자, 수련의, 감독자라는 세 가지 독립된 주체의 삼중적 관계로 특징짓습니다. 
As we consider physician–patient interdependence, however, we need to also remember that in the clinical workplace where EPAs are carried out, physician trainees are part of a collaborative team where performance often cannot be easily attributed to a single individual. 12 

  • Trainees are inseparably tied to their supervisors, 13 and this supervisor–trainee interdependence makes discerning which patient outcomes can be attributed to specific trainees, supervisors, or other aspects such as organizational context particularly challenging. 14,15 
  • Furthermore, supervision is baked into the EPA framework by virtue of entrustment decisions that some have formalized with statements of awarded responsibility for the trainee. 16,17 
  • These scales are also designed to be construct aligned with how supervisors think, 18 which makes the supervisor–trainee–patient interdependence triadic.

Thus, we characterize the interdependence that exists in EPAs as a triad of 3 distinct entities: patient, trainee, and supervisor.

EPA에서 '환자' 검색하기
Searching for “the Patient” in EPAs

이러한 프레임을 염두에 두고, 저희는 3개 전문 분야(소아과, 응급의학과, 정신과)와 4개 국가(미국, 캐나다, 호주, 뉴질랜드)의 EPA를 검토하여 현재 환자가 이 세 가지 영역의 일부로 표현되는 다양한 방식을 종합적으로 살펴보기 시작했습니다. 우리가 식별하고 선정한 특정 EPA는 개발 및 시행에 관련된 프로세스 및 개인에 대한 우리 팀의 지식과 친숙함을 기반으로 합니다. 선택된 EPA는 편의상 소규모 표본을 반영한 것으로, 의학의 모든 전문 분야 또는 EPA를 사용하는 국가를 대표하기 위한 것이 아니라 현재 사용되는 일부 EPA에서 환자를 어떻게 설명할 수 있는지를 엿볼 수 있도록 하기 위한 것입니다. 표본은 작지만, 관찰과 해석을 통해 주목과 논의가 필요한 패턴이 드러나며, 특히 이러한 패턴은 이 백서의 표본 외에서도 관찰되었기 때문에 더욱 그러합니다. 
With this framing in mind, we collectively set out to explore the various ways that patients are currently represented as part of this triad by reviewing EPAs from 3 different specialties (i.e., pediatrics, emergency medicine, and psychiatry) and 4 different countries (i.e., the United States, Canada, Australia, and New Zealand). The particular EPAs we identified and selected were based upon our team’s knowledge and familiarity with the processes and individuals involved in their development and implementation. The selected EPAs reflect a small convenience sample, not intended to be representative of all specialties within medicine or countries that employ EPAs, but rather to offer a glimpse into how patients can be described in some currently used EPAs. Although the sample is small, we expect that our observations and interpretations reveal patterns that warrant attention and discussion, especially since these patterns have been observed outside our sampling for this paper.

소아과(미국), 응급의학과(캐나다) 19, 20, 정신과(호주/뉴질랜드) 21의 EPA를 하나의 문서로 통합하여 검토할 수 있도록 했습니다. 각 전문 분야의 EPA는 한 쌍의 저자(소아과: D.A.T. 및 D.J.S.; 응급의학: S.S.-S. 및 E.S.H.; 정신과: A.G. 및 L.L.)를 통해 환자의 상태를 분석했습니다. 이러한 EPA의 분석과 해석에 참여한 팀은 환자와 직접적으로 상호작용할 뿐만 아니라 수련의의 환자와의 상호작용을 감독하는 의사(E.S.H., D.A.T., D.J.S.)와 감독 및 평가 전문가인 연구 과학자(S.S.-S., A.G., L.L., D.J.S.)로 구성되어 다양한 시각과 경험을 대변합니다. 또한 CBME 인가 및 인증 영역에서 일하는 개인(E.S.H., D.A.T.)도 있었습니다. 
EPAs from pediatrics (USA), 19 emergency medicine (CAN), 20 and psychiatry (AUS/NZ) 21 were consolidated into one document for our team to review. Each specialty’s EPAs were assigned to a pair of authors (pediatrics: D.A.T. and D.J.S.; emergency medicine: S.S.S.-S. and E.S.H.; psychiatry: A.G. and L.L.) for analysis of how the patient was represented. The team involved in the analysis and interpretation of these EPAs represent diverse perspectives and experiences, consisting of physicians who not only directly interact with patients but also supervise trainees’ interactions with patients (E.S.H., D.A.T., D.J.S.) as well as research scientists who are experts in supervision and assessment (S.S.S.-S., A.G., L.L., D.J.S.). We also had individuals who work within the CBME accreditation and certification realm (E.S.H., D.A.T.).

연구팀의 정기 회의는 3단계로 진행되었습니다: EPA와 관련 교육 문서 및 활동에서 환자가 묘사되는 방식에 대한 일반적인 논의, 할당된 EPA의 짝 분석, 짝 분석에서 분석 범주를 논의하고 구체화하는 전체 그룹 회의가 있었습니다. 또한 S.S.-S.와 E.S.H.는 문헌에 존재하는 일반적인 EPA 평가 척도(예: EQual, 22 O-SCORE, 23 및 Chen 감독 척도24)를 검토하여 '환자'를 찾고 '환자' 대신 '임상 치료'라는 문구가 더 자주 사용되었으며, 환자가 언급된 경우에도 주체가 아닌 객체로 위치하는 경우가 더 많았다는 점에 주목했습니다. 예를 들어, 정신과 EPA의 "정신분열증/정신병 환자에게 항정신병 약물 사용"에서는 교육생이 동사의 주어(즉, 행위의 주체)로 암시되어 있어 환자는 간접 목적어(약물은 직접 목적어이므로)의 역할에 그치고 있습니다. 
Regular meetings of the research team involved 3 steps: general discussion of the ways patients are portrayed in EPAs and related educational documents and activities, paired analysis of assigned EPAs, and full-group meeting to discuss and refine analytical categories from the paired analysis. S.S.S.-S. and E.S.H. also reviewed common EPA assessment scales present in the literature (i.e., EQual, 22 O-SCORE, 23 and the Chen Supervisory Scale 24) looking for “the patient” and noted the phrase “clinical care” was used more often in place of “the patient,” and even in instances where the patient was mentioned, they were positioned more as an object rather than the subject. For instance, the psychiatry EPA, “Use of an antipsychotic medication in a patient with schizophrenia/psychosis,” has the trainee implied as the subject of the verb (i.e., agent of the action), which renders the patient into the role of an indirect object (because medication is the direct object).

EPA에서 환자 특성화
Characterizing the Patient in EPAs

분석 결과, EPA에서 환자는 다음 4가지 중 한 가지 방식으로 특성화될 수 있는 것으로 확인되었습니다:

  • (1) 공동 생산자,
  • (2) 직접 객체,
  • (3) 간접 수혜자,
  • (4) 숨겨진 인물.

이러한 범주를 EPA 내 환자 참여 스펙트럼의 한 지점으로 개념화하고, 이러한 범주가 환자 초점이 다양한 정도를 어떻게 강조하는지 고려합니다. 또한 성찰의 기회로 프로세스, 구조 및 언어의 일부 조정이 어떻게 환자를 더 잘 부각하고 EPA 실행에서 환자, 수련의 및 감독자를 위한 공유 공간을 촉진할 수 있는지 보여줍니다. 
Our analysis identified that patients could be characterized in 1 of 4 ways in EPAs:

  • (1) as a co-producer,
  • (2) as a direct object,
  • (3) as an indirect beneficiary, or
  • (4) as a hidden figure.

We conceptualize these categories as points on the spectrum of patient involvement within EPAs and consider how they highlight the extent to which patient foci vary. As an opportunity for reflection, we also showcase how some adjustments in the process, structure, and language could better foreground the patient and promote a shared space for patients, trainees, and supervisors in EPA implementation.

  • 첫 번째 범주는 공동 생산자로서의 환자로, 의사가 환자와 협력하여 활동을 수행하는 경우입니다. 25-27 이 범주에는 환자-수련의-감독자 삼자 관계의 행위가 전면에 부각되는 EPA가 포함됩니다. 이 응급의학 EPA에서 관찰된 바와 같이 세 주체는 모두 활동의 공동 생산자입니다:
    • "응급실에 있는 환자, 간병인, 의료진 간의 정보 소통을 촉진하여 환자 치료 및 처치를 조직하는 것".
  • 두 번째 범주는 직접적인 대상인 환자로, 환자의 명시적인 참여 없이 환자에게 직접적인 혜택을 제공하는 방식으로 활동이 수행되는 경우를 설명합니다. 예를 들어 소아과에서 시행하는 EPA를 들 수 있습니다:
    • "환자를 소생시키고 안정화시킨 다음 질병의 중증도에 따라 치료를 조정하기 위해 분류합니다."
  • 세 번째 범주는 간접 수혜자인 환자로, 일반적으로 환자 치료를 개선하기 위해 환자를 위한 활동이 수행되는 경우입니다. 정신과 EPA의
    • "학교 출석에 위험이 있는 청소년을 위한 관리 계획 개발"이 이 범주의 예입니다.
  • 마지막 범주는 환자가 상태나 증상을 통해 암시되는 숨겨진 인물로서의 환자입니다. 응급의학 EPA에 설명된 것처럼:
    • "집중 병력 및 신체 검사 문서화."
  • 또한 일부 EPA는 환자 치료와 직접 관련이 없는 전문적 활동을 나타내며, 이는 종종 수련의 중심이고 환자는 전혀 고려되지 않는 직업의 중요한 구성 요소를 반영합니다.
    • (예: 의료 시스템 내 연구 또는 품질 개선에 집중)

 

  • The first category is patients as co-producers, where activities are performed by a physician in collaboration with the patient. 25–27 This category captures EPAs where the actions of the entire patient–trainee–supervisor triad are foregrounded. The 3 entities are all co-producers of an activity as observed in this emergency medicine EPA:
    • “Facilitating communication of information between a patient in the emergency department, caregivers, and members of the health care team to organize care and disposition the patient.”
  • The second category is patients as a direct object and describes instances where activities are performed on a patient without their explicit engagement, but in a way that offers a direct benefit. An example of this is an EPA from pediatrics:
    • “Resuscitate, initiate stabilization of the patient, and then triage to align care with severity of illness.”
  • The third category is patients as an indirect beneficiary, where an activity is done for the patient, usually to enhance patient care. The psychiatry EPA,
    • “Develop a management plan for an adolescent where school attendance is at risk,” is an example from this category.
  • Finally, the last category is patients as a hidden figure where the patient is implied through a condition or presentation, as described in this emergency medicine EPA:
    • “Documenting focused histories and physical exams.”
  • We also noted that some EPAs represent professional activities (e.g., focus on research or quality improvement within a health system) that are not directly related to patient care; these were often trainee focused and reflect important components of the profession where the patient is a distant consideration, if at all.

 

이러한 EPA에 환자의 참여를 고려하면서 환자의 위치가 달라지는 것과 관련된 사항을 고려할 수 있었습니다. 그림 1은 상호 의존성의 세 가지 구성이 EPA에서 각각 어떻게 나타나는지 시각적 일러스트레이션을 통해 강조합니다. EPA 자체를 삼각형으로 생각하고 삼각형의 모서리에 환자, 수련생, 감독자를 배치하면 환자의 위치가 달라질 수 있는 다양한 방법을 확인할 수 있습니다. 임상 활동을 설명, 실행 및 평가하는 방법의 변화에 따라 환자가 전경에 더 많이 배치될 수도 있고, 배경에 더 많이 배치될 수도 있습니다. 임상 활동에서 환자, 실습생, 감독자의 위치는 임상 활동 자체에 따라 그리고 주어진 임상 활동의 다양한 시점에 따라 바뀔 수 있습니다. 환자, 수련생 및 감독자의 거리가 같지 않을 수 있다는 개념화는 스펙트럼이 존재할 수 있다는 것을 의미하며, 이는 트라이어드의 개체 간의 이동도 가능하다는 것을 시사합니다.
Contemplating the involvement of patients in these EPAs allowed us to consider what is associated with the patient being positioned differently. Figure 1 highlights, using visual illustrations, how each configuration of the triad of interdependence might look in EPAs. If we think about EPAs themselves as triangles and position patients, trainees, and supervisors at the corners of those triangles, you can see various ways the patient can be positioned. The patient can shift from being more in the foreground to more in the background based on changes in how we describe, enact, and assess a clinical activity. Rotation of patients, trainees, and supervisors around a clinical activity also occurs depending on the encounter itself and at various points of time throughout a given encounter. The conceptualization that patients, trainees, and supervisors can have unequal distances allows for a spectrum to exist, which suggests that movement between entities of the triad is also possible.




이러한 순환과 이동을 보여주는 사례의 예로는 패혈성 쇼크 환자가 의식이 있고 합리적으로 일관성이 있을 때 응급실에서 소생술에 중점을 둔 EPA를 들 수 있습니다.

  • 이 경우 처음에는 환자와 실습생이 함께 이 EPA를 실행하는 모습을 전면에 내세우지만,
  • 실습생이 해당 상태의 환자를 관리한 경험이 충분하지 않아 환자가 더 이상 치료에 참여할 수 없는 심각한 저혈압 및 무반응 상태가 되면 감독자가 전면에 내세워질 수 있습니다.
  • 이러한 상황은 가족이 병상에 입회하고 참여하기 시작하면 환자의 가족 및 감독자와의 공동 작업으로 전환될 수도 있습니다.

An example case depicting such rotation and movement surrounds an EPA that focuses on resuscitation in the emergency department when a patient with septic shock is conscious and reasonably coherent.

  • In this instance, the patient and trainee may initially be foregrounded through the co-production of executing this EPA,
  • but the supervisor may then be foregrounded when the patient becomes profoundly hypotensive and unresponsive because the trainee may not have enough experience managing a patient in that state and the patient can no longer participate in his/her care.
  • This situation could even shift to co-production with the patient’s family and the supervisor when family presence and participation at the bedside is initiated.

이 예에서 삼각형은 몇 분 안에 여러 가지 다른 위치로 회전할 수 있습니다. 스펙트럼을 따라 가장 환자 중심적인 위치에서 가장 환자 중심적이지 않은 위치로 이동함에 따라 [EPA]와 [제공되는 임상 치료] 사이의 거리가 멀어지고 환자에게 도움이 되기 위해 더 많은 개입 단계가 필요합니다. 의도적인 경우에는 거리가 멀다고 해서 반드시 문제가 되는 것은 아니지만, 의도하지 않은 경우에는 거리가 멀면 아무도 눈치 채지 못한 채 환자가 완전히 제거될 수 있습니다. 
In this example, the triangle may rotate through several distinct positions within a matter of minutes. As one moves along the spectrum from most to least patient centered, the distance between an EPA and the clinical care provided is greater and requires more intervening steps to benefit the patient. We believe that a larger distance is not necessarily problematic when intentional; however, when unintentional, a larger distance can allow the patient to be completely removed without anyone even noticing.

EPA에서 환자를 전면으로 이동하기
Shifting the Patient to the Foreground in EPAs

마지막 예에서 알 수 있듯이, 우리는 EPA에서 환자, 실습생 및 감독자의 위치에 어느 정도 영향을 미칠 수 있습니다. 따라서 환자 중심주의의 목표는 가능한 한 환자를 전면에 배치하는 것을 고려할 것입니다. EPA의 설계, 언어, 운영에서 환자가 얼마나 쉽게 뒷전으로 밀려나고 축소될 수 있는지를 보여주는 분석적 통찰력에도 불구하고, 이러한 결과가 불가피한 것은 아닙니다. 따라서 환자가 여전히 중요한 부분을 차지할 수 있도록 돕는 방법에 대한 제안을 제공합니다. EPA의 세 가지 상호의존성에서 환자에게 초점을 맞추려면 환자가 어떤 방식으로 결국 배경으로 사라지는지를 고려하는 것부터 시작해야 합니다. 또한 이러한 분석 및 성찰 과정의 일환으로, 저희 팀은 EPA에 환자를 더 잘 통합하기 위해 적용할 수 있는 몇 가지 조정 사항에 대해서도 논의했습니다.
As the last example shows, we have some influence over how we position the patient, trainee, and supervisor in EPAs. Therefore, a goal of patient-centeredness would consider foregrounding the patient whenever possible. Despite our analytical insights revealing how easily the patient can be backgrounded and diminished in the design, language, and operationalization of EPAs, this outcome is not inevitable. We offer suggestions for how to help the patient remain an important part of our focus. To shift the focus of the patient in an EPA’s triad of interdependence, we need to start by considering some of the ways the patient ends up fading into the background. And as part of this analytical and reflective process, our team also discussed a few adjustments that could be made to better incorporate the patient in EPAs.

개발 관점에서 볼 때, EPA는 커리큘럼과 프로그램의 필요에 따라 설계되는 경우가 많고, 환자가 [설계 위원회의 구성원]이 되는 경우는 드물기 때문에, [EPA의 설계 및 작성]에 [환자와 커뮤니티 구성원을 포함하는 것을 표준 관행]으로 삼을 수 있습니다. 환자에게 주어진 EPA의 맥락에서 안전하고 효과적인 진료 제공을 가능하게 하는 필수 과제를 설명하도록 요청하면 환자에게 주도권을 부여할 뿐만 아니라 커리큘럼과 평가를 인구 건강 요구사항에 맞추는 데 도움이 됩니다. 개발 시 고려해야 할 또 다른 중요한 측면은 치료의 특정 측면(예: 만성 질환 및 당뇨병 프로그램)을 둘러싼 협의체에 환자를 포함하는 등 이해관계자 사이에서 환자를 대표하는 것입니다. EPA 개발 시 환자 대표성을 강화하는 것은 의도적인 것이어야 하며 억지로 만들어서는 안 됩니다.
From a development perspective, EPAs are often designed based upon curricular and program needs, and patients are rarely members of design committees, we could start making it standard practice to include patients and members of the community in the design and creation of EPAs. Asking patients to describe the essential tasks that enable safe and effective care delivery within the context of a given EPA not only gives agency to the patient but also helps to align our curricula and assessments with population health needs. Another important aspect to consider in development is representation of patients among stakeholders, such as including patients on councils surrounding specific aspects of care (e.g., chronic disease and diabetes programs). Our augmentation of the patient in development of EPAs needs to be purposeful, not forged.

내용적 관점에서 보면, EPA의 표현은 해당 활동을 수행하는 전문가인 교육생에 초점을 맞추는 경향이 있습니다. 그러나 전문가 활동을 설명하는 데 사용되는 언어는 좀 더 사람 우선적이고 환자에 민감하도록 개선될 수 있습니다. 예를 들어,

  • 간접적 활동으로 분류한 정신과 EPA("학교 출석에 위험이 있는 청소년을 위한 관리 계획 개발")
  • "학교 출석에 위험이 있는 청소년과 협력하여 관리 계획 개발"로 수정할 수 있습니다.

전자의 표현은 활동 자체를 강조하고 환자의 결과를 암시하므로 연수생은 모범 사례를 기반으로 관리 계획을 개발할 수 있으며, 이는 특정 청소년이 학교에 더 효과적으로 출석하거나 참여하는 데 도움이 되지 않을 수 있음을 의미합니다. 
From a content perspective, the wording of EPAs tends to focus on trainees because they are the professionals performing the activity. But, the language used to describe professional activities could be improved to be more person-first and patient-sensitive. For instance,

  • the psychiatry EPA we categorized as indirect—“Develop a management plan for an adolescent where school attendance is at risk”—
  • could be rephrased to “Collaborate with an adolescent to develop a management plan where school attendance is at risk.”

The former language foregrounds the activity itself and implies the patient outcome, which means a trainee can develop a management plan based on best practices which may not help the specific adolescent attend or engage more effectively in school.

앞서 언급한 바와 같이, 모든 EPA가 환자의 전경화(또는 공유 전경화)에 적합한 것은 아니지만(예: "소생, 환자 안정화 시작 후 질병의 중증도에 따라 치료를 조정하기 위해 분류"), 질병의 중증도를 고려할 때 환자와 함께 일하는 것이 불가능해 보이는 경우에도 환자를 전면에 배치할 기회는 여전히 존재합니다. 예를 들어, 심정지 상태의 어린이를 소생시킬 때 가족(즉, 환자의 연장선상)을 포함시킬 수 있는 옵션이 있으며, 대부분의 상황에서 가족이 침상 옆에 있고 인식하는 것이 더 바람직한 것으로 나타났습니다. 28 따라서 기억해야 할 중요한 점은 대부분의 EPA 설명에 환자의 역할에 대한 명시적인 언급이 있어야 한다는 것입니다.

As previously stated, not all EPA lend themselves to foregrounding (or shared foregrounding) with the patient (e.g., “resuscitate, initiate stabilization of the patient and then triage to align care with severity of illness”); however, despite instances where working with patients seem impossible given the severity of their illness, opportunities still exist to shift patients closer to the forefront. For example, when resuscitating a child in full arrest, there is the option of incorporating the family (i.e., an extension of the patient), and family presence and awareness at the bedside has been shown to be preferable in most circumstances. 28 Important to remember, therefore, is that the majority of EPA descriptions should make explicit reference to patients’ roles.

마지막으로, 실행 관점에서 볼 때, 무엇보다도 평가 접근법으로서 EPA에 초점을 맞추는 경향이 있는데, 이는 수련의와 감독자를 우선시하고 환자에게는 주의를 분산시킵니다. 여기서 첫 번째 단계는 환자와 환자 결과를 평가 관행에 의미 있게 통합하는 것입니다. 이 접근 방식은 단순한 환자 만족도 측정 그 이상입니다. 임상 수행의 어떤 측면을 평가하기에 가장 적합한지 결정하여 수련의 평가에 환자를 포함시킬 수 있습니다. 또한 환자에게 가장 중요한 측면을 반영하는 결과 측정을 식별하고 개발하는 과정에 환자를 포함시킬 수 있습니다. 이를 통해 반복적인 프로세스에 참여하여 커리큘럼이 환자의 요구를 충족하는지 확인할 수 있습니다. 또한 설계, 콘텐츠 및 결과에 대한 정보를 정기적으로 수집하고 검토하는 평가 프로세스가 내장되어 있으면 EPA에서 환자의 역할과 묘사를 지속적으로 모니터링할 수 있습니다.

  • 슈마허와 동료 29의 연구에 따르면 23개 레지던트 프로그램에서 소아과 레지던트 졸업생 중 53%만이 행동 및 정신 건강에 중점을 둔 EPA의 무감독 실습을 할 준비가 된 것으로 간주된다는 사실이 이 점을 잘 보여줍니다.
  • 일반 소아과 의사가 이 활동을 일상적으로 수행해야 한다는 점을 고려할 때, 전체 졸업생 중 절반만이 무감독 EPA를 수행할 수 있다는 것은 행동에 대한 촉구입니다.

Finally, from an implementation perspective, the focus tends to be on EPAs as an assessment approach above all else, which prioritizes trainees and supervisors and distracts from the patient. The first step here is to meaningfully incorporate patients and their outcomes into our assessment practices. This approach goes beyond mere measures of patient satisfaction. We can include patients in the assessments of our trainees by determining what aspects of clinical performance they are ideally suited to assess. We can also incorporate patients in the process of identifying and developing outcome measures that reflect aspects most important to them. In doing so, we can engage in an iterative process to ensure that our curricula are meeting the needs of our patients. Furthermore, built-in assessment processes that regularly collect and review information about design, content, and outcomes would allow for continual monitoring of the role and portrayal of patients in our EPAs.

  • The research of Schumacher and colleagues 29 illustrates this point, finding that only 53% of graduating pediatric residents across 23 residency programs were deemed ready for unsupervised practice for an EPA focused on behavioral and mental health.
  • Given that a general pediatrician is expected to perform this activity routinely, having only half of all graduates capable of executing this EPA unsupervised is a call to action.

이 연구에서 알 수 있듯이, EPA를 시행하면 평가하기 어렵거나 현재 평가되고 있지 않은 활동에 대한 교육 격차를 발견할 수 있습니다. 이러한 경우, 수련생이 수련생-환자 경험을 통해 스스로를 발전시킬 수 있도록 커리큘럼을 수정해야 합니다(즉, 환자와 직접 협력하여 전문직에 필요한 활동을 수행하는 데 필요한 지식과 기술을 학습하는 것). 
Implementation of EPAs can uncover training gaps for activities that are difficult to assess or are not currently being assessed, as this study illustrates. In these instances, curricular modifications are needed to ensure trainees develop themselves through the trainee–patient experience (i.e., working directly with patients to learn the knowledge and skills necessary to perform these necessary activities of the profession).

임상 활동에 환자와 함께 참여하면 최적의 환자 치료가 가능할 뿐만 아니라,30 환자에게 주체성을 부여하여 환자를 전면에 내세울 수 있습니다. 공동 제작이 잘 이루어지면 수련의와 환자 모두 전면에 나설 수 있는 여건이 조성됩니다. 일부 활동은 환자에게 덜 주체성을 부여할 수 있지만, 사실 의사가 수행하는 모든 활동의 수혜자는 환자입니다. 환자 중심의 의료 행위 접근 방식으로서 EPA의 진화하는 잠재력에 대해 더 깊이 생각하면서, 우리는 위임 결정이 결국 환자가 의사를 신뢰하는 것으로 이어져야 한다고 제안합니다. 7 이 메시지는 현재 EPA에 적용될 수도 있고 적용되지 않을 수도 있지만, 신중한 개발, 목적에 맞는 언어, 사려 깊은 실행을 통해 환자의 위치를 전면에 내세우고 환자 중심 진료의 약속을 이행하는 방식으로 EPA 커리큘럼과 평가에 환자를 통합할 수 있습니다.
Engaging with patients in clinical activities not only facilitates optimal patient care, 30 but brings the patient into the forefront by offering them a sense of agency. When done well, co-production grants the conditions for both the trainee and patient to be foregrounded. Although some activities may offer patients less agency, patients are in fact the recipients of all the activities that physicians perform. As we think further about the evolving potential of EPAs as a patient-centered approach to health care practice, we also suggest that entrustment decisions should eventually translate to patients trusting doctors. 7 This message may or may not translate in our current EPAs, but through careful development, purposeful language, and thoughtful implementation, we can shift the position of patients to the foreground and incorporate patients in our EPA curricula and assessments in a way that fulfills the promise of patient-centered care.

결론
Concluding Thoughts

우리가 CBME에 대한 접근 방식의 일부로 EPA를 계속 발전시키고 있는 가운데, EPA에서 환자의 입장에 대한 평가는 우리가 교육과 환자 결과를 연결하는 데 어려움을 겪는 이유에 대한 중요한 논의를 뒷받침합니다. 이 분석은 교육적 결과와 환자 결과를 연결하는 것의 중요성을 강조하고, EPA의 설계와 실행을 통해 연결을 강화할 수 있는 기회를 확인합니다. 이 사례는 역사적으로 더디게 진행되어 온 EPA의 개발, 콘텐츠 및 구현에서 환자를 의미 있게 전경화할 수 있는 기회를 보여줍니다. 이러한 전경화의 속도를 높이려면 [환자, 수련생, 감독자]가 EPA에서 가장 중요한 위치를 공유하는 [상호 의존성]의 삼위일체를 인식하는 데 달려 있습니다. 수련의, 수퍼바이저, 환자를 포그라운딩하는 것이 적절한 시점을 결정하는 것은 간단한 일이 아니지만, 삼중적 개념화는 어느 한 쪽을 우선시하는 것에서 벗어나는 데 도움이 됩니다. 또한, 스펙트럼 전반에서 삼자의 역동적인 관계를 파악하면 EPA가 감독자나 심지어 수련의를 제쳐두는 위험을 감수하지 않고도 환자를 더 두드러지게 전경에 배치할 수 있습니다. EPA에서 항상 환자를 전면에 내세울 필요는 없지만, 그렇다고 해서 그림에서 완전히 배제되어서도 안 됩니다. 상호 의존성의 세 가지 요소를 인식하고, 가능하면 환자를 포 그라운드하고, 적절한 경우 공유 포 그라운드를 고려한다면 교육 결과와 환자 결과를 연계하는 바람직한 목표에 한 걸음 더 다가갈 수 있습니다. 

As we continue to advance EPAs as part of our approach to CBME, an evaluation of the position of patients in our EPAs supports a critical discussion of why we struggle to connect educational and patient outcomes. Our analysis highlights the importance of bridging educational and patient outcomes and identifies an opportunity to bolster the connection through both our design and our implementation of EPAs. Our examples illustrate opportunities to meaningfully foreground the patient in the development, content, and implementation of EPAs, which has historically progressed slowly. Speeding up this foregrounding depends upon recognition of a triad of interdependence in which patients, trainees, and supervisors share the foremost position in our EPAs. Determining when it is appropriate to foreground the trainee, the supervisor, and the patient is no simple task, but a triadic conceptualization helps us shift away from having to prioritize one over another. Further, viewing the dynamic relationship of the triad across a spectrum allows for EPAs to more prominently foreground the patient without running the risk of pushing aside the supervisor, or perhaps even the trainee. Patients need not always be foregrounded in EPAs, but they should never be entirely absent from the picture either. If we remain cognizant of the triad of interdependence, foreground the patient when possible, and consider shared foregrounding when appropriate, we can move one step closer to the desired goal of linking educational outcomes with patient outcomes.

 


Acad Med. 2021 Jul 1;96(7S):S76-S80. doi: 10.1097/ACM.0000000000004094.

Distant and Hidden Figures: Foregrounding Patients in the Development, Content, and Implementation of Entrustable Professional Activities

Affiliations collapse

Affiliations

1S.S. Sebok-Syer is instructor, Department of Emergency Medicine, Stanford University School of Medicine, Palo Alto, California; ORCID: http://orcid.org/0000-0002-3572-5971.

2A. Gingerich is assistant professor, Division of Medical Sciences, University of Northern British Columbia, Prince George, British Columbia, Canada; ORCID: http://orcid.org/0000-0001-5765-3975.

3E.S. Holmboe is chief research, milestones development, and evaluation officer, Accreditation Council for Graduate Medical Education, Chicago, Illinois; ORCID: http://orcid.org/0000-0003-0108-6021.

4L. Lingard is professor, Department of Medicine and Faculty of Education, and senior scientist, Centre for Education, Research, and Innovation, Schulich School of Medicine and Dentistry, Western University, London, Ontario, Canada; ORCID: http://orcid.org/0000-0002-1524-0723.

5D.A. Turner is vice president for competency-based medical education, American Board of Pediatrics, Chapel Hill, North Carolina.

6D.J. Schumacher is associate professor, Department of Pediatrics, Cincinnati Children's Hospital Medical Center and the University of Cincinnati College of Medicine, Cincinnati, Ohio; ORCID: http://orcid.org/0000-0001-5507-8452.

PMID: 34183606

DOI: 10.1097/ACM.0000000000004094

Abstract

Entrustable professional activities (EPAs) describe activities that qualified professionals must be able to perform to deliver safe and effective care to patients. The entrustable aspect of EPAs can be used to assess learners through documentation of entrustment decisions, while the professional activity aspect can be used to map curricula. When used as an assessment framework, the entrustment decisions reflect supervisory judgments that combine trainees' relational autonomy and patient safety considerations. Thus, the design of EPAs incorporates the supervisor, trainee, and patient in a way that uniquely offers a link between educational outcomes and patient outcomes. However, achieving a patient-centered approach to education amidst both curricular and assessment obligations, educational and patient outcomes, and a supervisor-trainee-patient triad is not simple nor guaranteed. As medical educators continue to advance EPAs as part of their approach to competency-based medical education, the authors share a critical discussion of how patients are currently positioned in EPAs. In this article, the authors examine EPAs and discuss how their development, content, and implementation can result in emphasizing the trainee and/or supervisor while unintentionally distancing or hiding the patient. They consider creative possibilities for how EPAs might better integrate the patient as finding ways to better foreground the patient in EPAs holds promise for aligning educational outcomes and patient outcomes.

학부의학교육에서 적극적 환자 참여의 역할: 체계적 문헌고찰(BMJ Open. 2020)
Role of active patient involvement in undergraduate medical education: a systematic review
Stijntje Willemijn Dijk ,1 Edwin Johan Duijzer,2 Matthias Wienold3 

 

소개 및 근거
Introduction and rationale

최근 수십 년 동안 의학교육에 환자의 참여가 점점 더 많이 주장되어 왔으며, 의학교육 개혁가들이 채택하는 일반적인 관행이 되었습니다.1 환자와 환자의 이야기는 더 이상 임상 교육에서 '학습 자료'의 대상으로만 사용되는 것이 아닙니다. Towle 등은 교육, 평가 및 커리큘럼 개발에서 환자 교사로서 지속적인 참여와 더불어 의학 커리큘럼의 코드 설계자로서 기관 차원의 참여까지 다양한 수준의 참여를 확인했습니다.1-4 이제 의학교육자들은 의대생과 환자 및 그 가족, 지역사회를 연결하여 종적 관계의 중요성을 인식하고, 학생들의 사회적 상호작용 기술을 향상시키고, 실제 세계에서 질병에 대처하는 학습을 촉진하는 것의 가치를 깨닫고 있습니다.5 6 
In recent decades, the involvement of patients in medical education has been advocated for increasingly and has become common practice adopted by reformers of medical education.1 Patients and their narratives are no longer just used as subjects for ‘learning material’ in clinical training. Towle et al identified different levels of involvement, from paper-based involvement to involvement at the institutional level as codesigners of the medical curriculum in addition to sustained involvement as patient teachers in education, evaluation and curriculum development.1–4 Medical educators are now seeing the value of linking medical students with patients and their families and communities to foster awareness of the importance of longitudinal relationships, to improve students’ social interaction skills and to facilitate learning of coping with illness in the real world.5 6

의학교육에서 환자의 협력적 역할이 점점 더 커지고 있지만, 이를 어떻게 포함시킬지, 모든 수준의 의학교육에서 환자 참여를 계획하기 위한 체계적이고 기관 차원의 접근법을 개발하는 방법에 대해 배워야 할 것이 많습니다.3 7 임상 의사 결정과 환자 중심 치료에서 보다 동등한 파트너십을 향한 노력은, 국내 및 국제 지침에 힘입어, 의학교육에서 파트너십이 표준이 되는 문화를 개발하기 위한 노력의 확대를 촉진하고 있습니다.3 5-15 
Despite an increasingly collaborative role of patients in medical education, there is much to be learnt about how to embed it, and how to develop systematic, institution-wide approaches to planning patient involvement in all levels of medical education.3 7 The drive towards a more equal partnership in clinical decision-making and patient-centred care, fuelled by national and international guidelines, promotes the expansion of the efforts towards developing a culture where partnership in medical education becomes the norm.3 5–15

연구자들은 의학교육에서 환자의 적극적인 참여에 대한 다양한 정의를 사용함에 따라 문헌 검색에서 다양한 포함 기준을 사용했습니다. 이로 인해 포함된 논문이 중복되고 공통된 이론적 틀과 용어가 생성되는 데 한계가 있었습니다.3 16 이전 연구에서는 장단기 학습 결과, 윤리적 문제, 심리적 영향 및 모집, 선발, 준비와 같은 주요 절차적 기여자와 관련된 지식 기반에 큰 격차가 있는 것으로 확인되었습니다. 또한 적극적인 환자 참여의 비용 효율성에 대한 정보도 제한적입니다. 
As researchers have used varying definitions of active patient involvement in medical education, they have used varying inclusion criteria in their literature searches. This has resulted in overlap of included papers, and limited the generation of a common theoretical framework and terminology.3 16 Previous studies have identified major gaps in the knowledge base relating to short and long-term learning outcomes, ethical issues, psychological impact and key procedural contributors like recruitment, selection and preparation. There is also limited information of the cost-effectiveness of active patient involvement.

의학교육에서의 환자 참여에 대한 체계적 문헌고찰1 17-19 및 비체계적 문헌고찰2 8 20이 발표된 이후, 많은 새로운 연구가 발표되었습니다. 이전 리뷰에서는 환자 교사 역할만 다루거나,1 특정 술기(친밀한 검사)를 가르치고 평가하거나,17 모의 환자를 포함하거나,17 대학원 의학교육을 포함하거나18 모든 의료 전문직을 다루었습니다.2 17 20 최근의 체계적 고찰에서는 주로 학습자의 관점에서 환자 참여, 결과 및 참여 이유에 대한 포괄적인 개요를 제공했습니다.19 
Since the publication of the last systematic reviews1 17–19 and non-systematic reviews2 8 20 of patient involvement in medical education, many new studies have been published. Previous reviews addressed only the patient teacher role,1 teaching and assessing one specific skill (intimate examination),17 included simulated patients,17 included postgraduate medical education18 or addressed all healthcare professions.2 17 20 A recent systematic review provided a comprehensive overview of the involvement, outcome and reason behind involvement mainly from learner’s perspective.19

본 논문에서는 학부 의학교육의 모든 분야 또는 환경에서 단순히 우발적인 수동적 참여를 넘어서는 모든 형태의 참여를 포함하는 환자 참여에 대한 광범위한 정의를 사용하여 가장 최근의 문헌을 검토하고 요약합니다. 이 정의를 채택함으로써 우리는 연구 데이터의 범위와 양을 확장하여 적극적인 환자 참여에 대한 실질적인 지식 기반을 늘리고 개선된 이론적 틀과 공통 용어의 근거를 마련할 수 있습니다. 본 연구는 주로 환자의 참여에 대한 환자의 관점에 초점을 맞춘 새로운 접근 방식을 취합니다.  
Our paper reviews and summarises the most recent literature using a broad definition of patient involvement consisting of any form of involvement that is beyond merely incidental passive involvement, in any field or setting of undergraduate medical education. By adopting this definition we are able to extend the scope and amount of research data in order to increase the practical knowledge base on active patient involvement and in order to give ground to an improved theoretical framework and common terminology. Our study takes a novel approach by focusing primarily on the patients’ perspective on their involvement.

방법
Methods

설계
Design

문헌 검색은 [의대생 학부 교육의 개발, 전달 또는 평가에 실제 환자 및 지역사회 구성원이 직접 참여하는 것]으로 정의되는 [의학교육에 대한 적극적인 환자 참여]를 찾기 위해 체계적인 검토 방법을 사용했습니다. 
Our literature search employed a systematic review method looking for active patient involvement in medical education defined as the direct involvement of real patients and community members in the development, delivery or evaluation of undergraduate education of medical students.

검색 전략
Search strategy

저자들은 2018년 7월 12일에 PubMed를 통해 검색을 수행했습니다. 검색어에는 환자*, 커뮤니티*, 참여, 참여*, 협력*, 협력*, 대표*, 의학교육, 커리큘럼, 의대생* 등이 단독으로 또는 조합하여 사용되었습니다. 검색 기준은 병원 정보 전문가가 검토했습니다. 전체 검색 전략은 온라인 부록 파일 1에서 확인할 수 있습니다. 
The authors performed a search through PubMed on 12 July 2018. The search terms used on their own and in combination included: patient*, communit*, involvement, engag*, cooperat*, collaborat*, represent*, medical education, curriculum, medical student*. Search criteria were reviewed by a hospital information specialist. The full search strategy can be found in online supplementary file 1.

2003년 1월 1일부터 2018년 7월 12일 사이에 영어로 출판되고 학부 의학교육에 환자의 적극적인 참여를 다룬 1차 경험적 연구를 보고한 모든 논문을 포함할 수 있었습니다. 모의 환자 또는 배우, 검사만 받은 환자 또는 병동에서 관찰만 받은 환자를 대상으로 한 연구는 검토 대상에서 제외되었습니다. 
All articles published in English between 1 January 2003 and 12 July 2018 and reporting primary empirical research that addressed the active participation of patients in undergraduate medical education were eligible for inclusion. Studies with simulated patients or actors, patients solely undergoing examinations or patients who were only being observed in wards were excluded from the review.

1차에서는 제목과 초록을 기준으로, 2차에서는 전문을 기준으로 논문을 평가했습니다. 포함된 모든 논문의 참고문헌을 분석하여 원래 포함 기준에 부합하는 추가 연구가 있는지 확인했습니다. 포함 기준을 준수한 모든 리뷰는 관련 참고 문헌에 대해 추가적으로 평가되었습니다. 최종 분석에는 원저 연구 논문만 포함되었습니다. 
We assessed articles based on title and abstract in the first round, and in a second round based on full text. References of all included articles were analysed for additional studies that matched the original inclusion criteria. All reviews that complied with the inclusion criteria were additionally assessed for relevant references. Only original research articles were included in the final analysis.

포함된 연구의 질 평가
Quality assessment of included studies

질적, 양적 및 혼합 방법 설계의 논문이 포함되었으므로 두 명의 저자(SWD 및 ED)가 혼합 방법 평가 도구를 적용하여 연구의 방법론적 품질을 평가했습니다(온라인 보충 파일 2).21 평가된 방법론적 품질에 따라 연구는 제외되지 않았습니다. 
As our review included papers of qualitative, quantitative and mixed methods designs, two authors (SWD and ED) applied the Mixed Methods Appraisal Tool to assess the methodological quality of studies (online supplementary file 2).21 Studies were not excluded based on assessed methodological quality.

데이터 추출 및 합성
Data extraction and synthesis

모든 저자(SWD, ED, MW)는 포함된 연구에서 데이터를 추출하기 위해 미리 준비된 표준화된 양식을 사용했습니다. 두 번째 저자는 추출된 데이터가 정확한지 확인했으며, 불일치하는 부분은 토론을 통해 해결했습니다. 저자, 출판 연도, 저널, 개입 국가, 연구 유형, 초록, 연구 환경, 보고된 재정적 측면, 개입 환자 수, 개입 학생 수, 환자 특성, 환자 참여 동기, 모집 관행, 교육 및 준비 관행, 환자 단체의 역할, 환자 참여 유형, 결과 측정, 지속 가능성에 대한 조직의 언급 및 추구한 학습 결과 등의 데이터가 기록되었습니다. 이러한 학습 결과의 범위를 탐색하기 위해 의과대학에서 일반적으로 적용되는 역량 프레임워크인 CanMEDS 프레임워크에 따라 의도된 학습 결과를 분류했습니다.22 추출된 데이터를 관련 테마로 구성하여 데이터 요소 간의 연결과 불일치를 탐색했습니다. 환자가 맡을 수 있는 잠재적 역할을 그룹화할 때 기존의 프레임워크를 사용하지 않기로 결정했습니다. 토론 섹션에서는 역할의 다양성에 대한 우리의 연구 결과를 Towle 등의 기존 분류체계와 비교합니다.2 
All authors (SWD, ED, MW) used a prepiloted standardised form to extract data from included studies. A second author checked if the extracted data were accurate, and discrepancies were resolved through discussion. The following data were recorded: authors, year of publication, journal, country intervention, study type, abstract, study setting, financial aspects reported, number of patients in intervention, number of students in intervention, patient characteristics, patient motivations to join, recruitment practices, training and preparation practices, role of patient organisations, type of patient involvement, outcome measurement, organisational remarks on sustainability and pursued learning outcomes. To explore this range of learning outcomes, we categorised intended learning outcomes according to the CanMEDS framework as a commonly applied competency framework within medical schools.22 We organised extracted data in related themes to explore connections and discrepancies between data elements. We opted not to use any of the existing frameworks for grouping potential roles patients take on. In the Discussion section, we compare our findings of the diversity of roles with the existing taxonomy by Towle et al.2

환자 및 대중 참여 성명서
Patient and public involvement statement

이 연구 이니셔티브의 초기 동기는 국제환자단체연합(IAPO)과 국제의대생협회연맹(IFMSA)의 회원인 저자들 간의 협력에 따른 것입니다. 환자 대표이자 공동 저자인 MW는 프로젝트 시작, 연구 설계, 데이터 분석, 토론 및 논문 작성 등 연구 프로젝트의 모든 단계에 동등한 파트너로서 참여했습니다. 이 논문의 초안은 2019 유럽 환자 포럼에서 300명의 환자 대표들이 참석한 가운데 본회의에서 발표되고 논의되었으며, 그 피드백이 최종 논문에 반영되었습니다. 
The initial impulse for this study initiative followed a collaboration between the authors as members of the International Alliance of Patients’ Organizations (IAPO) and the International Federation of Medical Students' Associations (IFMSA). MW, patient representative and co-author, was involved as an equal partner in all stages of the research project including project initiation, study design, data analysis, discussion and writing of the paper. The initial draft of this paper was presented and discussed at the European Patient Forum 2019 in a plenary session with 300 patient representatives present, feedback from which has been incorporated into the final paper.

 

결과
Results

연구 선택
Study selection

초기 검색 결과 769개의 논문이 검색되었으며, 이 중 제목 및/또는 초록을 기준으로 95개의 논문이 추가 검토 대상으로 선정되었습니다. 이 95개 논문은 두 명의 저자(SWD 및 ED)가 독립적으로 읽었으며 지정된 기준에 따라 포함되었습니다. 검토자 간의 합의율은 91.2%였습니다. 나머지 논문은 짧은 토론 후 합의에 따라 포함되었습니다. 주요 제외 사유는 논문 유형이 잘못된 경우(학회 초록 또는 논평), 환자의 적극적인 참여와 관련이 없는 연구였습니다. 참고문헌 검토 결과 11편의 논문이 추가로 포함되었습니다. 포함 기준을 충족한 49개 연구의 특징은 그림 1과 온라인 부록 파일 3에 제시되어 있습니다. 

The initial search resulted in 769 articles, of which 95 were selected for further review based on the title and/or abstract. These 95 articles were independently read by two authors (SWD and ED) and included based on the specified criteria. Consensus between reviewers was 91.2%. The remaining articles were included based on consensus after a short discussion. Main reasons for exclusion where wrong article type (conference abstracts or commentaries) and studies that did not concern active participation of patients. The review of references resulted in 11 additional articles for inclusion. The characteristics of the 49 studies that met inclusion criteria are presented in figure 1 and online supplementary file 3.

포함된 연구에서 추출한 데이터를 사용하여 다음 하위 섹션의 증거를 종합했습니다:
We used the extracted data from included studies to synthesise the evidence in the following subsections:

  • 환자를 의학교육에 참여시키는 근거.
  • 환자 모집 및 선정.
  • 의학 교육에 참여하기 위한 환자의 준비 과정.
  • 의학 교육에서 환자가 수행하는 역할.
  • 환자 참여를 통해 추구하는 학습 목표.
  • 환자 참여에 대한 우려 사항.
  • 환자 참여의 영향에 대한 환자들의 견해.
  • 환자 참여의 재정적 영향.
  • 환자 단체의 역할.
  • 환자 참여의 지속 가능성을 보장하기 위한 조치.
  • Rationale for involving patients in medical education.
  • Patient recruitment and selection.
  • Patients’ preparation to participate in medical education.
  • Roles patients take on in medical education.
  • Learning objectives pursued through patient involvement.
  • Concerns about the involvement of patients.
  • Patients’ views on the impact of their involvement.
  • Financial implications of patient involvement.
  • Roles of patient organisations.
  • Measures to ensure the sustainability of patient involvement.

의학 교육에 환자를 참여시키는 근거
Rationale for involving patients in medical education

몇몇 저자는 의학교육에 환자 참여를 의무화하는 정부 정책, 즉 영국 보건부 및 영국 일반의학회,16 23 24 호주 의학회25 및 WHO5 를 환자 참여 이니셔티브의 근거로 언급했습니다. 이러한 정치적 동인 외에도

  • 학생들에게 환자 중심 및 전문가 간 치료를 가르치기 위해24 26-28,
  • 학생들에게 만성 질환 치료를 소개하기 위해29-31,
  • 다문화 학습 환경을 조성하기 위해,
  • 사회적 책임과 포용 의제를 실천하기 위해32-37,
  • 교육을 더욱 매력적이고 강력하며 혁신적으로 만들기 위해38,
  • 환자에게 권한을 부여하기 위해39

환자들은 미래의 의료 인력을 양성하고24 32 의료 시스템을 개선한다는 점에서 더 광범위한 커뮤니티에 대한 책임감을 느낀다고 언급했습니다.24 33 40 
Several authors referred to government policy mandating patient participation in medical education, namely the UK Department of Health and the UK General Medical Council,16 23 24 the Australian Medical Council25 and the WHO5 as a rationale for their patient involvement initiatives. Besides these political drivers, cited rationales were:

  • to teach students patient-centred and interprofessional care24 26–28;
  • to introduce students to chronic illness care29–31;
  • to create a multicultural learning environment;
  • to practise social accountability and an inclusion agenda32–37;
  • to make education more engaging, powerful and transformative38; and
  • to empower patients.39 

Patients mentioned that they felt a sense of responsibility to the broader community in shaping the future health workforce24 32 and improving the healthcare system.24 33 40

환자 모집 및 선택
Patient recruitment and selection

가장 자주 보고된 환자 모집 방법은

  • 기존 대학 파트너십 및 기존 프로그램,28 33 41-46
  • 언론 또는 소셜 미디어를 통한 광고 또는 보건 시설에 게시,26 34-36 47 48
  • 지역사회 및 환자 단체,
  • 개인적 인맥 및 이전 참가자,32 38 49-51 및
  • 보건 전문가를 통하여 52-55

학생들이 만성 질환 환자를 섀도잉하는 한 프로젝트에서는 학생들이 직접 환자를 모집하도록 요청받았습니다.30
The most frequently reported methods to recruit patients were through

  • existing university partnerships and existing programmes,28 33 41–46 
  • advertisements through press or social media or posted in health facilities,26 34–36 47 48 
  • through community and patient organisations,
  • through personal connections and previous participants,32 38 49–51 and
  • through health professionals.52–55 

In one project, where students shadowed a patient with a chronic condition, students were asked to recruit patients themselves.30

선발 기준은 일반적으로 광범위하게 설정되어 모든 환자 또는 지역사회 구성원을 초대했습니다. 환자 교육자의 일반적인 기준에는 다음 등이 포함되었습니다.

  • 좋은 의사소통 능력,
  • 교육에 대한 선호,
  • 추가 학습에 대한 적성,
  • 열정,
  • 연구에 전념할 수 있는 시간,
  • 완전히 움직일 수 있고 반복적인 신체 검사에 대처할 수 있는 능력 48 54 56

어떤 경우에는 환자에게 질병의 대표적인 신체적 징후가 요구되기도 했습니다.53 54 56 57 
Selection criteria were generally set up broadly, inviting any patient or community member. General criteria for patient educators included

  • good communication skills,
  • affinity for teaching,
  • aptitude for further learning,
  • enthusiasm,
  • time to commit to the study, as well as
  • being fully mobile and being able to cope with repeated physical examinations.48 54 56 

In some cases patients were required to have representative physical signs of their disease.53 54 56 57

의료 교육에 참여하기 위한 환자의 준비 과정
Patients’ preparations to participate in medical education

20개의 논문에서 환자 교사의 준비에 대해 언급하고 있습니다. 준비 기간은 총 100시간의 상당한 교육 세션42 에서부터 서면 정보만 제공하거나 1~1.5 시간의 단일 오리엔테이션 세션이 대부분이었습니다.24 30 32 39 43 49 52 57 근골격계(MSK) 술기를 교육하는 환자 대상 교육 프로그램이 가장 광범위했으며 의료 또는 교육 전문가가 제공했습니다.42 47 54 56 58 59 다른 준비 세션은 덜 형식적이며 교수진 교육 전문가, 학생 또는 동료에 의해 진행되었습니다. 
Twenty articles mention preparation of patient teachers. The duration of the preparation ranged from substantial training sessions of 100 hours in total42 to the majority of programmes providing solely written information or a single orientation session of 1–1.5 hours.24 30 32 39 43 49 52 57 Training programmes for patients in teaching musculoskeletal (MSK) skills were the most extensive and were delivered by medical or educational experts.42 47 54 56 58 59 Other preparatory sessions were less formal and were facilitated by faculty educational experts, students or peers.

준비 프로그램의 주요 목표는 다양했습니다. 한 연구에서는 학생의 학습 결과를 보호하기 위해 환자 교육자가 코스 목표를 인식하는 것이 중요하다고 강조했습니다.53 다른 연구에서는 환자들의 자신감 향상, 기술 훈련 제공57, 교육 과정과 관련된 지식 제공을 목표로 언급했습니다.23 39 48 60 이러한 세션에서는 문제 기반 학습, 프레젠테이션 전달 방법, 코퍼실리테이션 방법, 효과적인 피드백 제공 방법과 같은 접근법을 다루었습니다. 또한 준비 과정은 코칭, 감독 및 디브리핑 방법을 사용하여 갈등, 감정, 충족되지 않은 기대와 같은 [이점과 과제를 예측할 수 있는 기회]를 제공했습니다.40 61
The primary aims of the preparatory programmes varied. One study underscored the importance of patient educators being aware of the course goals in order to safeguard student learning outcomes.53 Others mentioned aiming to serve the needs of patients in building their confidence, providing skills training57 and providing knowledge related to the educational process.23 39 48 60 These sessions addressed approaches such as problem-based learning, how to deliver a presentation, cofacilitation methods and how to provide effective feedback. Preparation also provided opportunities for anticipating benefits and challenges such as conflicts, emotions, unmet expectations, using methods of coaching, supervision and debriefing.40 61

한 논문에서는 평가 도구로 짧은 퀴즈를 통해 환자 교육자의 준비를 마무리하고 교육을 시작하기 전에 졸업 세션을 갖는 것을 언급했습니다.28 또 다른 논문에서는 환자 교육자가 교육할 준비가 되었는지 확인하기 위해 교육 후 만족도 설문지를 사용하는 것을 설명합니다.47 대부분의 논문에서는 환자 교육자가 역할을 수행하기 전에 환자를 평가하는 것에 대해 다루지 않았습니다.
One article mentions finishing the preparation of patient educators with a short quiz as an assessment tool and having a graduation session before starting to teach.28 Another article describes the use of a post-training satisfaction questionnaire to help ensure that patient educators felt ready to teach.47 The majority of papers did not address assessing patients prior to them taking on their roles.

Towle 등은 [환자 교육자의 준비성]과 [교육의 형식과 내용의 실제성authenticity] 사이의 긴장에 대해 논의했습니다.49 이들은 [학계]와 [지역사회]라는 두 문화 사이에서 중개자 역할을 할 수 있는 지역사회 조직 대표의 중요한 역할을 강조합니다. 한 논문에서는 환자 교육자가 명시적으로 교육을 받지 않았기 때문에 학생과 환자의 만남이 가능한 한 진정성 있게 이루어질 수 있다고 설명합니다.53 또 다른 논문에서는 환자와 교육자 간의 협업을 통해 진정한 환자 관점을 잃지 않고 상호 학습이 가능하다고 설명합니다.39
Towle et al discussed the tension between preparedness of patient educators and authenticity of education in both form and content.49 They highlight the critical role of the community organisation representatives who can be brokers between the two cultures of academia and community. One article describes an intervention in which the patient educators explicitly have not received training, so the student–patient encounters would be as authentic as possible.53 Another article described that the collaboration between patients and educators allows for mutual learning without an authentic patient perspective being lost.39

의학 교육에서 환자가 맡는 역할
Roles patients take on in medical education

우리가 확인한 주요 역할 범주는 교사, 평가자, 커리큘럼 개발자 및 학생 선발자로서의 환자 영역으로 나뉩니다(표 1).
The main categories of roles that we identified are divided in the areas of a patient as a teacher, an assessor, a curriculum developer and a student selector (table 1).

 

교사로서의 환자
Patient as teacher

교사의 역할이 가장 많이 언급되었습니다. 환자 교사들은 다발성 경화증과 같은 자신의 질환에 대한 병력 청취 및 신체 검사 세션에 대한 임상 기술 실습을 진행했습니다.42 47 54 56 58 59 62 환자 교사들은 학생들에게 기술을 가르치고 즉각적인 피드백을 제공하여 추가 학습을 촉진하도록 훈련받았습니다.56 
The role of a teacher was cited most frequently. Patient teachers gave clinical skills practicals on history taking and physical examination sessions on their own condition such as MSK disorders.42 47 54 56 58 59 62 They were trained to teach students skills and deliver immediate feedback, which stimulated further learning.56

장애인 환자 교사로 활동하는 여러 장애인 환자 그룹이 의사소통 기술과 병력 청취에 대한 실습을 진행했습니다.48 53 57 HIV 감염인이 교사로 참여하여 학생들이 상담을 제공하는 모의 임상 상황에 참여했습니다.60 
Several groups of patients with disabilities acting as patient teachers with disabilities gave practicals on communication skills and history taking.48 53 57 People living with HIV participated as teachers during a simulated clinical encounter in which students provided counselling.60

환자들은 임상 기술 외에도 전반적인 치료 관리 경험과 삶의 개인적인 측면에 대해 학생들에게 가르쳤습니다. 여기에는 실제적인 신체적, 가정적 적응부터 자신과 가족에게 영향을 미치는 심리적, 사회적, 행동적 문제까지 다양한 내용이 포함되었습니다. 교육은 패널 토론과 소그룹 세션24 43 63뿐만 아니라 지역사회와 환자 가정 방문을 통해 이루어졌습니다.31 52 만성 질환을 가진 환자 교사가 멘토 역할을 하고 학생들과 정기적으로 만났습니다.27 29 환자들은 학생들에게 환자 중심주의와 전문직 간 전문성,23 24 31 38 49 지역사회 중심주의문화적 역량 및 윤리를 가르쳤습니다.33 64 환자들은 자신의 이야기를 들려주고 성찰을 자극하는 등 자신의 교육 방법을 선택할 수 있었습니다.38 지적 및 발달 장애가 있는 환자 그룹도 자신의 관점과 이야기를 공유하는 비디오를 통해 학습 자료 제작에 참여했습니다.65
In addition to clinical skills, patients taught students about their experiences of overall management of care, and the personal aspects of their lives. These ranged from practical physical and home adjustments, to psychological, social and behavioural issues impacting them and their family. Teaching was done through panel discussions and small group sessions24 43 63 as well as visits to the community and patients’ homes.31 52 Patient teachers with chronic conditions acted as mentors, and met regularly with students.27 29 Patients taught students patient-centredness and interprofessionalism,23 24 31 38 49 community-centredness, cultural competence and ethics.33 64 Patients could choose their own teaching method, such as telling their stories and stimulating reflection.38 A group of patients living with intellectual and developmental disabilities also participated in the creation of learning materials, through videos sharing their perspectives and stories.65

Jackson 등의 연구에 참여한 대부분의 환자는 자신을 가르치는 것이 아니라 파트너십을 맺고 질병의 특정 측면을 설명하고 공유하는 역할을 한다고 생각했습니다.52 
Most patients in the study by Jackson et al considered themselves not as teaching, but having a role of partnership, explanation and sharing certain aspects of their illness.52

학생의 역량 평가자로서의 환자
Patient as assessor of students’ competence

교육 세션 중 피드백과 같은 형성적 학생 평가 외에도 환자는 학년 말 객관적 구조화 임상시험(OSCE)과 같은 고부담 총괄 평가에 참여했습니다.39 환자는 공식 평가에 사용되는 학생 에세이에 대한 서면 피드백도 제공했습니다.39 환자는 학생 수행의 대부분 비인지 영역을 평가했습니다.35 46 Jha 등이 인터뷰한 의학교육자들은 학생들이 편안하게 느끼는지, 학생들이 올바른 질문을 했는지 평가하는 데 환자의 역할이 있다고 생각했습니다.66 Raj 등의 연구에 참여한 환자와 의학교육자들은 환자의 평가를 칭찬했지만, 학생들은 환자가 임상 기술을 신뢰할 수 있는지 또는 너무 관대할 수 있는지에 대해 우려를 표명했습니다.54
In addition to formative student assessment, such as feedback during teaching sessions, patients participated in high stakes summative assessments, such as the final year objective structured clinical examination (OSCE).39 Patients also provided written feedback to student essays, which were used for formal assessment.39 Patients assessed mostly non-cognitive domains of student performance.35 46 Medical educators interviewed by Jha et al believed there was a role for patients in assessing whether students made them feel at ease and whether students asked the right questions.66 While patients and medical educators in the study by Raj et al54 praised patients’ assessments, students expressed their concerns whether patients could reliably assess clinical skills or whether they were likely to be too lenient.54

커리큘럼 개발에서 환자의 역할
Patient role in curriculum development

커뮤니티 구성원들은 커리큘럼 개발에 참여하고자 하는 동기가 있었습니다.35 의료 전문가로 간주되지는 않았지만, 자신과 가족을 위한 최적의 의료 서비스를 보장하는 데 관심이 있었습니다.26 
Community members were motivated to participate in curriculum development.35 While they were not seen as medical experts, they did have an interest in ensuring optimal healthcare for themselves and their families.26

지역사회 구성원은 교육 프로그램의 계획, 실행 및 평가에 적극적으로 참여했습니다.46 49 환자 교사는 계획 단계부터 교육 전달 단계까지 자율성을 가졌습니다.43 67 여러 환자가 정신과 교육과정28 및 전문직 간 교육 교육과정의 운영위원회 위원으로 활동했습니다.49
Community members actively participated in the planning, implementation and evaluation of the educational programme.46 49 Patient teachers had autonomy from the stage of planning to the stage of delivery of teaching.43 67 Several patients were members of the steering committee for the psychiatry curriculum28 and the interprofessional education curriculum.49

환자들은 자신의 질병 또는 사회적 상태와 관련된 교육과정 개발에 참여했습니다.35 원주민 대표들은 원주민 건강 교육과정 개발 및 지역사회 배치에 대한 권장 사항을 제공했습니다.33 44 하와이 원주민과의 포커스 그룹 회의는 문화적 역량 및 건강 격차 교육과정을 정의하기 위해 열렸습니다.37 트랜스젠더 건강 교육과정에 대한 월드 카페 토론도 마찬가지로 열렸습니다.36 소수자 커뮤니티 회원들은 교육과정 설계, 특히 암 격차 교육과정의 내용에 대한 의견을 제공했습니다.35 
Patients were involved in the development of courses related to their illness or social conditions.35 Aboriginal delegates provided recommendations for the development of an Aboriginal health curriculum and community placement.33 44 Focus group meetings with Native Hawaiians were held to define a cultural competencies and health disparities curriculum.37 A world café discussion was similarly held to inform the curriculum on transgender health.36 Minority community members provided input on curricular design, especially on the content of the cancer disparities curriculum.35

질병 또는 역량별 교육과정 외에도 이해관계자와의 자문 회의에 참여하여 바람직한 특성, 졸업생의 역량, 지역사회 기반 학습 환경 개발을 파악했습니다.45 환자들도 교육과정의 바람직한 특성에 대해 자문을 구했습니다.68 한 의과대학에서는 포커스 그룹을 통해 신설 의과대학의 인구보건학과의 전략적 발전을 위한 의견을 구했습니다.69
Beyond the disease or competency-specific courses, patients were involved in consultative meetings with stakeholders to identify desirable attributes, competencies of graduates and development of a community-based learning environment.45 Patients were also consulted on the desired characteristics of the curriculum.68 One medical school sought input for the strategic development of the department of population health in a new medical school through focus groups.69

의과대학 학생 선발에서 환자의 역할
Patient role in selection of students to medical schools

커뮤니티 구성원은 임상의사 및 교수진과 함께 의학전문대학원 입학 프로그램에 지원하는 학생을 선발하는 패널에 초대되었습니다.32 커뮤니티 구성원은 특히 지원자의 의사소통 능력과 사회적 맥락 및 사회적 요구에 대한 민감성, 연민, 공감 능력을 평가하는 학생 선발 과정 및 팀의 일원으로 초대되었습니다.45
Community members were invited to join a panel together with clinicians and academic staff members to select students applying for the Graduate Entry Medical Program.32 Members of the community were invited to be a part of the student selection process and team, especially in assessing candidates’ communication skills as well as sensitivity, compassion and empathy towards social contexts and societal needs.45

교수진과 환자 간의 협력
Collaboration between faculty and patients

환자 교사와의 협업에서 교수진의 역할은 다양했습니다. 일부 환자 교사는 임상 프리셉터의 감독 하에 일했습니다.60 다른 세션에서는 환자가 실무자와 공동 촉진자 역할을 했습니다.23 워크샵은 환자 교사가 주도하고 교수진이 진행했지만 통제하지는 않았습니다. 교수진의 역할은 학생과 멘토 간의 직접적인 학습을 지원하는 것이었습니다.49 교수진은 회의에 항상 참석하지는 않았지만 토론을 위한 광범위한 주제 설정과 같은 배경 지원을 제공할 수 있었습니다.46 환자 교사들은 프로그램 지원이 참여에 필수적이며, 개별 메시지를 가르치는 것에서 보편적인 수업을 가르치는 것으로 전환할 수 있게 해준다고 말했습니다.43
The role of faculty members in the collaboration with patient teachers varied. Some patient teachers worked under the supervision of a clinical preceptor.60 In other sessions, patients were cofacilitators with practitioners.23 Workshops were led by patient teachers and facilitated, but not controlled, by faculty. The faculty member’s role was to support the direct learning between students and mentors.49 Faculty was not always present in meetings but could provide background support, such as setting broad topics for discussions.46 Patient teachers stated that programme support was essential for participation, allowing them to transform from teaching individual messages to teaching universal lessons.43

한 연구에 참여한 일반인 참가자들은 일반인 관점의 중요성을 인정하기 위해 커리큘럼 소유권 공유가 필요하다고 생각한 반면, 교수진 참가자들은 커리큘럼 개발의 소유권을 추정했습니다.26 Jha 등의 연구에 참여한 [교수진]은 평가 또는 코스 개발에 환자를 더 많이 참여시키는 방법에 대해 명확하지 않았고, 그렇게 하는 것이 적절하다고 확신하지도 못했습니다. 일부 교수진은 환자 평가자 및 코스 개발자와 함께 일한 경험을 토큰주의적이라고 표현했습니다.66
Lay participants of one study regarded sharing of curriculum ownership as necessary to acknowledge the importance of lay perspectives, whereas faculty participants presumed ownership of curriculum development.26 Faculty in the study by Jha et al were not clear on how to involve patients more fully in assessments or course development, nor were they convinced of the appropriateness of doing so. Some faculty members expressed their experiences of working with patient assessors and course developers as tokenistic.66

환자의 적극적인 참여를 통한 학습 목표 달성
Learning objectives pursued through active patient involvement

환자 참여의 학습 결과는 4건의 무작위 대조 실험42 47 54 62 과 2건의 추가 연구에서 MSK 검사 기술을 주제로 정량적으로 평가되었습니다.56 59 훈련된 환자 교육자가 제공한 세션과 류마티스내과 직원이 수동적으로 검사를 받는 환자와 함께 제공한 세션54 62 및 비 MSK 전문 의사와의 세션을 비교했을 때 구조화된 임상 검사(OSCE) 점수 증가에 차이가 관찰되지 않았습니다.42 Humphrey-Murto 등의 실험에서 교수진이 가르친 학생은 환자 교육자가 가르친 학생(30명 중 5명)보다 낙제한 학생이 훨씬 적었습니다(32명 중 0명).62 학생들은 환자 교육자보다 교수진 교육자를 더 높게 평가했습니다(5점 리커트 척도에서 4.13점 대 3.58점).62 
Learning outcomes of patient participation were quantitatively assessed on the subject of MSK examination skills in four randomised controlled experiments42 47 54 62 and two further studies.56 59 No difference was observed in increased structured clinical examination (OSCE) scores when comparing sessions delivered by trained patient educators with sessions delivered by rheumatology staff together with a passive patient undergoing examination54 62 and sessions with a non-MSK specialist physician.42 In the experiment by Humphrey-Murto et al, significantly fewer faculty-taught students failed (0 out of 32) than patient educator-taught students (5 out of 30).62 Students rated faculty educators higher than patient educators (4.13 vs 3.58 on a 5-point Likert scale).62

교수진이 주도하는 정규 세션 외에 환자 교사의 교육을 받은 학생은 정규 커리큘럼에 참여한 학생에 비해 OSCE 점수가 더 많이 상승했습니다.47 de Boer 등59 의 중재는 학생들에게 전임상 MSK 장애 블록에서 두 번의 비의무적 실제 환자 학습 세션에 참여할 기회를 제공했습니다.59 참여한 학생들은 블록 종료 테스트에서 훨씬 더 높은 점수를 얻었습니다 
When students were taught by a patient teacher in addition to the regular faculty-led sessions, their OSCE scores increased more compared with students participating in the regular curriculum.47 An intervention by de Boer et al59 offered students the opportunity to participate in two non-obligatory real patient learning sessions in the preclinical MSK disorders block.59 Students who participated scored significantly higher at the end-of-block test.

오스왈드 등은 MSK 신체 검사 술기를 가르칠 때 환자 교육자와 의사 교육자 간에 교육 방식이 어떻게 다른지 조사했습니다.58 비디오 녹화에 따르면 교육을 받은 환자 교육자는 모든 주요 관절을 일관되게 다룸으로써 내용과 스타일이 더 일관된 것으로 나타났습니다. Bokken 등53 은 실제 환자와 시뮬레이션 환자의 교육 효과에 대한 학생의 관점을 평가했습니다.53 전반적으로 교육 효과가 높은 것으로 나타났습니다. 학생들은 실제 환자를 더 현실감 있고 신체 검사 실습에 더 유용하다고 생각했습니다. 
Oswald et al examined how teaching was different between patient educators and physician educators when teaching MSK physical examination skills.58 Video recordings show that trained patient educators were more consistent in content and style by consistently covering all major joints. Bokken et al53 assessed student’s perspectives on instructiveness of real patients versus simulated patients.53 Overall instructiveness was marked high. Students regarded real patients as more authentic and the encounters more useful in practising physical examination.

Jaworksy 등의 중재 연구에서는 의대생이 환자 강사에게 HIV 검사 상담을 제공했습니다.60 검증된 의료진 HIV/AIDS 낙인 척도70의 중재 전 및 중재 후 점수는 유의미한 감소(68.74점 대 61.81점)를 보여주었습니다. 학생들은 HIV 관련 치료를 제공할 때 편안함이 증가했다고 보고했습니다(10.24점 대 18.06점). 마찬가지로 신체적 또는 정신적 장애를 가진 환자 교사를 대상으로 한 중재 연구에 참여한 학생들은 태도 개선,28 의사소통의 편안함 증가,57 자기 효능감과 자신감 증가,63 모든 인터뷰 스테이션에서 평균 수행 점수가 대조군에 비해 높은 것으로 나타났습니다.65 
In the intervention study by Jaworksy et al, medical students provided HIV test counselling to patient instructors.60 Preintervention and postintervention scores of the validated Health Care Provider HIV/AIDS Stigma Scale70 demonstrated a significant decrease (68.74 vs 61.81). Students reported increased comfort in providing HIV-related care (10.24 vs 18.06). Similarly, students in intervention studies with patient teachers living with physical or mental disabilities demonstrated an improved attitude,28 increased levels of comfort in communication,57 increased levels of self-efficacy and confidence,63 65 and higher mean performance scores across all interview stations when compared with a control group.65

Rees 등의 연구에 참여한 학생들은 교과서 학습에 비해 환자와의 만남이 더 동기 부여가 되었다고 설명했습니다.71
Students in the study by Rees et al described the encounters with patients as more motivating compared with textbook learning.71

이 검토에서 확인된 질적 연구에서는 환자 참여 교육의 광범위한 학습 성과가 언급되었습니다. 이러한 다양한 결과를 탐색하기 위해 캐나다 왕립 의사 및 외과의사 대학에서 개발한 CanMEDS 프레임워크에 따라 분류를 사용했습니다22(표 2).
Wide ranges of learning outcomes of education with patient participation were mentioned in the qualitative studies identified by this review. To explore this range of outcomes a categorisation is used according to the CanMEDS framework, developed by the Royal College of Physicians and Surgeons of Canada22 (table 2).

 

커뮤니케이터
Communicator

몇몇 저자는 환자 참여 교육의 주요 성과로 환자 중심 진료를 언급했습니다.23 27 29 49 67 환자 중심에는 다음이 포함되었습니다.

  • 환자 멘토를 개인으로 보는 능력,27
  • 치료에서 환자 자율성과 전문성의 중요성,64
  • 동정적이지 않고 판단하지 않는 태도,55
  • 환자의 필요를 인식하고41
  • 환자를 팀의 유능한 일부로 보는 것41

Jha 등은 적극적인 환자 참여 자체가 동등한 파트너십을 보여준다고 지적했으며66 Rees 등은 이 접근법이 학생들이 의료에 대한 총체적인 시각을 개발하는 데 도움이 된다고 결론을 내렸습니다.71 
Several authors mentioned patient-centred care as the main outcome of education involving patients.23 27 29 49 67 Patient-centredness included

  • the ability to see patient mentors as individuals,27 
  • the importance of patient autonomy and expertise in care,64 
  • adopting a non-patronising and non-judgemental attitude,55 
  • recognising patients’ needs41 and
  • seeing the patient as a capable part of the team.41 

Jha et al pointed out that active patient involvement by itself demonstrates an equal partnership66 and Rees et al concluded that this approach helps students to develop a holistic perspective of healthcare.71

보다 일반적으로 환자 참여는 의사소통의 중요성에 대한 이해 증가,27 29 의사소통 기술 구축 및 향상,55 71 공감, 경청 기술 및 존중과 관련이 있습니다.71
More generally, patient participation was associated with increased understanding of the importance of communication,27 29 building and improving communication skills,55 71 empathy, listening skills and respect.71

협력자
Collaborator

McKinlay 등은 학생들이 만성 질환을 앓고 있는 환자의 가정 방문을 수행하는 교육 프로그램을 설명했는데,31 학생들은 반성적 평가에서 만성 질환 관리에서 다학제간 팀에 대한 이해가 높아진 것으로 나타났습니다. 4명의 저자가 환자가 참여하는 전문가 간 교육 프로그램을 설명했습니다.23 27 44 49 
McKinlay et al described an education programme in which students undertake a home visit to a patient with a chronic condition,31 where students demonstrated increased understanding of interdisciplinary teams in management of chronic conditions in their reflective assays. Four authors described interprofessional education programmes in which patients are involved.23 27 44 49

리더
Leader

의학, 물리치료, 작업치료, 간호학, 약학 학생이 환자 멘토와 팀을 이룬 종단적 멘토 프로그램에서 학생들은 의료 시스템에 대해 더 깊이 이해했다고 보고했습니다.27 1년 동안 진행된 학생 멘토 프로그램은 학생들에게 치료의 연속성에 대한 경험과 인식을 제공했습니다.55
In a longitudinal mentor programme with medical, physical therapy, occupational therapy, nursing and pharmacy students teaming up with a patient mentor students reported a deeper understanding of the healthcare system.27 A yearlong student mentor programme gave students an experience in and appreciation of continuity of care.55

전문가
Professional

다양한 질적 연구에 따르면 환자 참여는 임상 실무의 윤리적 복잡성과 임상의 오류에 대한 환자의 관점을 다루고64 반성적 기술을 개발하는 데 기여할 수 있다고 합니다.29 48 55 역할 모델에 대한 고찰 일부 저자는 의료인의 역할에 대한 이해의 폭을 넓히고,27 개인적 이상향에 관하여 알려주는 뛰어난 임상의의 자질,64 미래의 전문직 모델55 및 전문직 정체성을 만드는 것을 언급했습니다.71 실제 환자 교육자와의 만남 경험도 불확실성, 감정 및 스트레스에 대처하는 데 도움이 될 수 있습니다.71 
Various qualitative studies suggested that patient involvement can attribute to dealing with ethical complexity in clinical practice and patients’ perspectives on clinician error64 and developing reflective skills.29 48 55 Reflecting on role models some authors referred to broadening understanding of the role of the healthcare provider,27 qualities of remarkable clinicians that inform personal ideals,64 creating a future professional model55 and professional identity.71 Experiences with real patient educator encounters could also help in coping with uncertainties, emotions and stress.71

특정 환자 또는 소수자 그룹의 환자 교육자에 대한 노출은 학생들이 만성 질환과 노인,27 31 정신 건강 문제29 또는 장애를 가진 환자에 대한 긍정적인 태도를 높이는 데 도움이 되었습니다.48 
Exposure to patient educators from within specific patient or minority groups helped students increase positive attitude towards chronic conditions and elderly,27 31 patients with mental health problems29 or disabilities.48

건강 옹호자
Health advocate

환자 멘토와의 윤리 및 전문성 모듈 경험에 대한 연구에서 학생들은 일상적인 실무에서 환자 옹호의 중요성에 대해 성찰했습니다.64 보다 구체적으로, 학생들은 취약한 상황에 처한 환자를 옹호할 수 있는 역량을 갖추게 되었습니다. 사켓쿠 등이 설명한 교육 프로그램의 목표 중 하나는 특히 장애인의 맥락에서 의사 옹호의 영향에 대한 인식을 개발하는 것이었습니다.63 사전 테스트와 사후 테스트에 따르면 이 프로그램을 통해 이러한 인식이 크게 증가한 것으로 나타났습니다. 
Students reflected on the importance of patient advocacy in day-to-day practice in a study on experiences within an ethics and professionalism module with patient mentors.64 More specifically, students were empowered to advocate for patients when they are in vulnerable situations. One of the aims of the education programme described by Saketkoo et al was to develop an awareness of the impact of physician advocacy, specifically in the context of people with disabilities.63 A pretest and post-test showed that this awareness increased significantly with the programme.

학자Scholar

학자 역량 개발의 목표를 명시적으로 설명한 프로그램은 없습니다.
No programmes have explicitly described the aim of developing the competency of scholar.

의료 전문가
Medical expert

의료 전문가의 역할은 환자 중심의 고품질의 안전한 진료를 제공하기 위해 의학 지식, 임상 기술 및 전문적 가치를 적용함으로써 다른 모든 역할을 통합하는 것입니다. 두 논문에서는 환자 참여가 의학 실습의 복잡성을 인식함으로써 학생의 학습을 지원한다고 언급했습니다.41 48 Jha 등은 환자가 실습에서 이론의 예시를 제공함으로써 학생의 이해와 기억을 향상시킨다고 설명했습니다.66 
The role of medical expert integrates all other roles by applying medical knowledge, clinical skills and professional values in the provision of high-quality and safe patient-centred care. Two articles mentioned that patient participation supports students’ learning by recognising the complexity of practising medicine.41 48 Jha et al described the patient as providing an illustration of the theory in practice, thus enhancing students’ understanding and recall.66

 

환자 참여에 대한 우려
Concerns about the involvement of patients

여러 저자들은 의학교육에 환자를 교사로 참여시킬 때 우려를 제기하기도 했습니다. 일부 교수진 교육자들은 환자의 이야기가 너무 충격적이어서 학생들이 그로 인한 감정을 다루기 위해 지원이나 디브리핑이 필요할 수 있다고 우려했습니다.66 71 질적 연구에서 학생들은 서비스 사용자가 임상의에게 물어보지 않고 정보와 조언을 요청하거나 서비스 사용자가 임상의에게 말하지 않은 정보를 학생들에게 누설할 때 '압박'을 느꼈습니다.71 학생들은 환자에게 잘못된 정보를 제공할까 봐 걱정했습니다. 
Various authors have also raised concerns when involving patients as teachers in medical education. Some faculty educators were concerned that patient stories might be so traumatic that students would require support or debriefings to deal with the resulting emotions.66 71 In a qualitative study, students felt ‘pressured’ by service users asking them for information and advice, rather than asking their clinicians, or when service users divulged information to students that they had not told their clinicians.71 Students worried about giving incorrect information to patients.

학생들은 한 사람의 관점만 보게 되어 편향된 시각을 갖게 될 수 있다는 우려를 표명했습니다.23 학생들은 또한 환자들이 좋은 퍼포먼스와 안 좋은 퍼포먼스를 구별하기 어렵고, 피드백이나 평가에 너무 관대할 수 있다는 점을 우려했습니다.54 Henriksen과 Ringsted67 의 연구에 참여한 학생들은 환자의 지식에 대해 회의적인 태도를 보였으며67 환자가 교육 환경을 계획하고 전달하는 데 자율성을 갖는 맥락에서 구조화되지 않은 체험 학습에 대해 우려를 표명했습니다.67 다른 연구에서 직원들은 동일한 환자가 동일한 프로그램에 너무 자주 참여하면 환자 경험의 영향이 감소될 수 있다는 우려를 표명했습니다.66 
Students expressed reservations that they were only getting the view of one person, which could lead to a biased perspective.23 Students were also concerned that patients might have difficulty discriminating between poor and good performance, and are likely to be too lenient in feedback or assessment.54 Students in the study by Henriksen and Ringsted67 expressed scepticism about patients’ knowledge67 and expressed concerns about unstructured experiential learning in a context where patients had autonomy in both planning and delivering the teaching encounter.67 In a different study, staff members expressed the concern that the impact of the patient experience might be reduced if the same patient was involved in the same programme too often.66

환자 참여의 영향에 대한 환자들의 견해
Patients’ views of the impact of their involvement

환자들은 환자 중심주의를 가르치고,24 72 자신의 몸과 진정성을 제공하고, 학생들의 자신감을 강화하며,72 더 넓은 지역사회에 대한 책임을 다하고,24 32 의료 시스템을 개선하기 때문에 의미 있는 기여와 개인적인 성취감을 느낀다고 설명했습니다.24 33 40
Patients described a strong sense of having a meaningful contribution and personal fulfilment, because they were teaching patient-centredness,24 72 offering their body and authenticity, bolstering students’ confidence,72 fulfilling their responsibility to the broader community24 32 and improving the healthcare system.24 33 40

환자들은 개인적 차원에서 물질적, 직업적, 개인적, 정서적 혜택을 설명했습니다. 물질적 혜택에는 선물과 같은 유형의 보상43 을 받고 전체 건강 검진을 받은 것이 포함되었습니다.40 의대생 교육에 참여한 요통 환자 교육자들은 참여로 인해 자신의 요통 관리가 개선되었고, 의사와의 상담에서 자신의 요구를 표현하는 데 자신감이 향상되었다고 말했습니다.47 일부 환자들은 교육할 때 의료 전문가로부터 더 많은 시간과 관심을 받았다고 느꼈습니다.71
On an individual level patients described material, professional, personal and emotional benefits. Material benefits included receiving tangible rewards such as gifts43 and receiving a full medical check-up.40 Patient educators with back pain involved in teaching medical students stated their participation improved the management of their own back pain, and improved confidence in voicing their needs in consultations with physicians.47 Some patients felt that they received more time and attention from their healthcare professionals when they were teaching.71

환자들은 학생 선발 과정에 참여함으로써 직업적 성장과 개인적 성취감을 느꼈다고 설명했습니다.32 하템 등은 의료 서비스 제공자를 더 찾는 것을 더 잘 하게 되었다거나, 자신의 건강 상태에 대한 지식이 증가하는 등 환자에게 실질적인 이점이 있다고 보고했습니다.43
Patients described professional growth and personal fulfilment from being involved in the selection process of students.32 Hatem et al reported practical benefits for patients including getting better at finding healthcare providers and increased knowledge of their medical condition.43

환자가 의료 교육에 참여하는 것과 관련된 단점과 위험에는 낙인찍는 가정, 자기 공개에 대한 취약성자발적인 질의응답 교환에 직면하는 것이 포함됩니다. 예를 들어, HIV를 주제로 강의하는 한 환자 교육자는 한 남성이 질병에 대한 지식 부족과 무지로 인해 매우 좌절했던 경험을 설명했습니다. 환자들은 또한 예상치 못한 질병의 진행이 교육 능력에 영향을 미친다는 사실에 주목했습니다. 어떤 경우에는 이로 인해 교사로서의 헌신에서 손을 떼는 경우도 있었는데, 이는 환자 교육자들 사이에서 불가피한 손실이었습니다.43 환자들은 또한 학생들의 부정적이고 인정하지 않는 반응에 대한 취약성을 설명했습니다.40 47 멘토들은 초기에 일반적으로 자신이 공유하는 내용이 학생들에게 도움이 되는지에 대해 불안하고 확신하지 못했습니다.29 사회경제적으로 열악한 지역에서 지역사회 기반 개입에 참여한 환자의 절반은 인터뷰 전에 불안, 걱정 또는 긴장감을 표현했지만, 모든 경우에서 환자는 이것이 정상이라고 느꼈습니다.52 환자들은 '취약하다'는 단어 외에도 임상 교육 환경 내에서 서비스 이용자의 감정을 강조하기 위해 '노출', '두려움', '피곤함', '스트레스' 및 '괴로움'과 같은 용어를 사용했습니다.71 일부는 정신건강 서비스 이용자가 자신의 이야기를 반복적으로 말하는 것이 트라우마틱하다고 설명하기도 했습니다.71 
The drawbacks and risks associated with patient involvement in medical education included being confronted with stigmatising assumptions, vulnerability of self-disclosure and spontaneous question-answer exchanges. A patient educator teaching on the subject of HIV, for example, described the experience of being very frustrated with one man’s lack of knowledge and ignorance about the disease. Patients also drew attention to the fact that unanticipated disease progression had an impact on their ability to teach. In some cases, this resulted in them pulling out of their commitment as teachers, an inevitable loss among patient educators.43 Patients also described a sense of vulnerability to negative and non-appreciative reactions from students.40 47 Initially, mentors were commonly anxious and unsure about whether what they shared was of benefit to students.29 Half of the patients involved in the community-based intervention in a socioeconomic-deprived area expressed feelings of anxiety, apprehension or nervousness prior to the interview, although in all cases patients felt that this was normal.52 In addition to the word ‘vulnerable’, patients employed terms like ‘exposed’, ‘frightened’, ‘tired’, ‘stressed’ and ‘harrowing’ to emphasise service users’ feelings within the clinical education environment.71 Some even described it as traumatic for mental health service users to repeatedly tell their stories.71

의학 교육에 대한 환자 참여의 재정적 영향
Financial implications of patient involvement in medical education

19편의 논문에서 환자 비용 상환, 사례금 지급, 조직 비용 또는 비용에 대한 인식 등 개입의 재정적 측면에 대해 언급했습니다. 그러나 경제적 및 재정적 자원은 체계적으로 비교할 수 있는 방식으로 조사되지 않았습니다. 보고된 재정적 비용에는 장애 기술 워크숍에 800달러,63 신체 검사 교육에 800파운드54, 전체 환자 파트너 프로그램에 2640파운드가 포함되었습니다.47 10개의 연구에서 참여 환자에게 시간당 8유로에서 하루 350파운드43 53 54 또는 불특정 금액의 보수를 제공했습니다.28 35 39 48 50 57 4개의 연구에서 여행, 전화 또는 우편 비용과 같은 환자 비용을 환급했습니다.24 34 47 65 일부 직원은 환자가 다과나 대중교통에 지출해야 하는 현금이 참여에 장벽이 될 것을 두려워했습니다.73 
Nineteen articles commented on any financial aspect of the interventions, ranging from reimbursement of patients’ expenses, payments of honoraria, organisational costs or perceptions of cost. Economic and financial resources, however, have not been explored in a way that they can be systematically compared. Reported financial costs included $800 for a disability skills workshop,63 £800 for a physical examination training54 and £2640 for the overall Patient Partners programme.47 Ten studies offered participating patients remuneration through honoraria between €8 per hour and £350 per day43 53 54 or an unspecified amount.28 35 39 48 50 57 Four studies offered reimbursement of patient expenses such as travel, phone or mail costs.24 34 47 65 Some staff feared that cash patients needed to spend on refreshments or public transport would be a barrier for their participation.73

의학교육자들은 환자 참여가 재정적 투자와 직원 시간 모두에서 비용이 많이 드는 노력이라고 제안했습니다.71 비용 효과성에 대해 언급한 논문은 한 편뿐으로, 환자 주도 교육이 의사 주도 교육에 비해 비용 효율적인 방법이라고 언급했지만 경제적 평가는 제공하지 않았습니다.54 비용 효과성 또는 비용 편익 분석을 제공한 논문은 없었습니다. 
Medical educators suggested that patient involvement was a costly endeavour, both in financial investment as well as staff time.71 Only one article commented on cost-effectiveness, noting that patient-led teaching is a cost-effective method compared with physician-led teaching, but did not provide an economic evaluation.54 No paper provided a cost-effectiveness or cost-benefit analysis.

환자 단체의 역할
Roles of patient organisations

환자 단체와 의료진 간의 가장 많이 언급된 상호작용은 개별 환자를 지역사회 구성원 또는 질환별 지원 그룹의 구성원으로 모집하기 위해 환자 단체와 네트워크를 활용하는 것이었습니다.24 28 29 36 49 55 56 60 69 Baral 등의 논문에서는 의과대학 운영위원회가 보건과학 아카데미 교육과정 개발을 위해 농촌 지역사회 및 소비자 단체의 대표들에게 자문을 구했습니다.45 지역사회에 기반을 둔 환자 옹호 및 지원 단체의 대표들이 Towle과 Godolphin의 연구에서 중재 자문 그룹에 참여했습니다.49 이러한 환자 단체의 대표들이 모두 환자 본인이었던 것은 아닙니다. 이들은 학계와 지역사회라는 두 문화 사이의 중개자 역할을 수행했습니다. 
The most cited interaction between patient organisations and medical faculties was the use of patient organisations and their networks for the recruitment of individual patients as community members or members of condition-specific support groups.24 28 29 36 49 55 56 60 69 In the paper by Baral et al, representatives from rural communities and consumer groups were consulted by the medical school steering committee for the development of the Academy of Health Sciences curriculum.45 Representatives of community-based patient advocacy and support organisations took part in the advisory group of the intervention in the study by Towle and Godolphin.49 Not all of these representatives of patient organisations were patients themselves. They were described as brokers between two cultures of academia and community.

리즈 대학교는 자신의 경험을 바탕으로 교사와 연구자로서의 역할을 알리는 일반인들로 구성된 '환자 목소리 그룹'이라는 내부 환자 전담 그룹과 협력했습니다. 이 그룹은 형성 평가와 종합 평가에 참여했습니다. 또한 학교 내에 지속적인 지원을 제공하는 환자 및 대중 참여 관리자가 배정되었습니다.39
The University of Leeds worked with a dedicated internal patient group named ‘The Patient Voice Group’, consisting of lay people who use their experiences to inform their roles as teachers and researchers. This group was involved in formative and summative assessment. Additionally, a patient and public involvement manager who provided ongoing support was assigned within the school.39

일부 의학교육자들은 정치적인 단체와 협력하는 것에 대한 두려움 때문에 환자 단체와 협력하지 않기로 명시적으로 선택했습니다.71 그들은 토큰주의에 대한 비판을 방지하기 위해 다양한 목소리를 들을 수 있도록 여러 환자 그룹을 포함시키기로 결정했습니다. 환자들은 그룹에 참여함으로써 동료들로부터 지지와 우정을 얻을 수 있었다고 말했습니다.71
Some medical educators made the explicit choice not to collaborate with patient organisations, due to a fear of working with politicised groups.71 They did make a decision to include groups of patients to allow multiple voices to be heard in order to prevent criticisms of tokenism. Patients stated that participating in groups gave them support and companionship from their peers.71

의료 교육에 참여하고자 하는 환자 단체는 상자 1 에 설명된 몇 가지 실용적인 사항을 고려할 수 있습니다.
Patient organisations wishing to engage in medical education may wish to consider some of the practical points as described in box 1 .


박스 1. 의료 교육에 참여하고자 하는 환자 단체를 위한 실용적인 사항
Box 1. Practical points for patient organisations wishing to engage in medical education

환자 단체를 위한 실용적인 포인트
Practical points for patient organisations

  • 회원들에게 의학교육 참여의 잠재적 이점을 강조하고 공통된 동기에 부응합니다.
  • 의학교육에 참여하는 환자들 간에 동료 지원 네트워크를 활성화합니다.
  • 해당 지역의 교육 기관에 환자 및 대중의 참여를 위한 기존 프레임워크가 있는지, 환자 및 대중의 참여를 명시적으로 약속했는지, 환자 및 대중의 참여를 전담하는 임원을 임명하여 협력의 시작점을 제공할 수 있는지 조사합니다.
  • 의료진 내에 환자 위원회 또는 자문 그룹을 구성하여 교육 과정에 구조적으로 환자의 의견을 반영하고 보호할 수 있도록 요청합니다.
  • 교육 과정의 어느 단계가 목표에 가장 적합한지 결정합니다. 학교의 사명과 비전, 학습 목표, 교육 전략, 교수법, 학습자 평가 또는 코스 평가에 노력을 집중할 수 있습니다.
  • 모든 의미 있는 개입이 전체 교수진 수준에서 이루어져야 하는 것은 아니며, 하나의 학습 목표 또는 특정 질병 관련 코스와 같은 특정 영역에 집중할 수도 있습니다.
  • 대학과의 성공적인 협업과 모범 사례를 더 많은 환자 및 학계 커뮤니티에 소개하세요.
  • Highlight potential benefits of involvement in medical education to members and respond to common motivations.
  • Facilitate peer support networks among patients involved in medical education.
  • Investigate whether teaching facilities in your area have existing frameworks for patient and public involvement, have made explicit commitments to patient and public involvement and/or have appointed officers specifically assigned to patient and public involvement that could provide you with a starting point for collaboration.
  • Call for the creation of a patient committee or advisory group within a medical faculty to enable and safeguard structural input into the educational process.
  • Determine which step of the educational process is best suited for your goals. You might choose to focus your efforts on the mission and vision of the school, learning objectives, educational strategies, teaching, assessment of learners or the evaluation of the course.
  • Not every meaningful intervention has to be at an overall faculty level, they may also be on specific areas such as one learning objective or a disease-specific course.
  • Showcase successful collaborations with universities and best practices to the wider patient and academic communities.

환자 참여의 지속 가능성을 보장하기 위한 조치
Measures to ensure the sustainability of patient involvement

환자 참여를 유지하는 데 있어 핵심 요소로 확인된 것은

  • 적절한 자원 지원 제공,
  • 일반인 기여의 가치에 대한 공식적인 인정,
  • 일반인 의견에 따른 변화에 대한 교수진의 명확한 약속이었습니다.26

The key factors identified in sustaining patient involvement were

  • the provision of adequate resource support,
  • formal acknowledgement of the value of lay contributions and
  • a clear faculty commitment to change following lay input.26

환자와 지역사회 참여를 보장하기 위해 대학의 사명과 비전 선언문 또는 전략 계획에 사회적 책임 또는 환자 중심 교육과 의료를 기관 전체에 통합하는 것이 여러 논문에서 인용되었습니다. 저자들은 환자 참여 프로그램의 자원 집약도가 높기 때문에 대학이 지속 가능하려면 [환자 중심의 토대]를 중시해야 한다고 강조했습니다.26 44 45 61 커리큘럼에 지속적인 모듈로 이니셔티브를 통합하면 산발적인 환자 참여가 아닌 지속 가능한 환자 참여를 달성할 수 있습니다.53 63
Institution-wide incorporation of social accountability or patient-centred education and medicine in the university’s mission and vision statement or strategic plan was cited in several papers to ensure patient and community involvement. The authors emphasised that the resource intensity of a patient involvement programme requires the university to value its patient-centred underpinnings in order to be sustainable.26 44 45 61 The incorporation of initiatives as ongoing modules in the curriculum achieved sustainable patient involvement rather than sporadic involvement.53 63

리즈대학에서는 기관에 [상설 환자 의견 그룹]이 설립되었습니다.39 이 기관은 환자 및 대중 참여 관리자를 임명하여 지속적인 지원을 제공했습니다. 일부 이니셔티브는 기존 기관과 협력하거나, 학교 전체 수준에서 시행하거나45 관절염과 같은 한 가지 질환에 초점을 맞추었습니다.58 Gaver 등55 은 환자 참여의 지속 가능성을 위한 핵심 과제로 자원봉사 조직과 가족 간의 약속을 수립하는 과정을 확인했습니다.55 
In the University of Leeds, a permanent patient voice group was incorporated in the institution.39 The institution appointed a patient and public involvement manager to provide ongoing support. Some initiatives chose to work in partnership with existing institutions, implemented at a school-wide level45 or focused on one condition such as arthritis.58 Gaver et al55 identified the process of establishing commitment among volunteering organisations and families as a key challenge to the sustainability of patient involvement.55

의학교육자들은 환자 교육자가 급여를 받고 의과대학의 직원으로 간주된다면, 그들은 더 진지하게 역할을 수행하고 더 신뢰할 수 있을 뿐만 아니라 교육 팀의 존경받는 일부로 여겨질 수 있다고 언급했다.66
Medical educators commented that if patient educators were paid and seen as an employee of the medical school, they might take on the role more seriously and become more reliable, as well as being seen as a respected part of the educational team.66

토론
Discussion

이 검토는 49개의 주요 경험적 연구를 체계적으로 평가했으며, 의대 학부 교육에서 환자의 적극적인 참여의 역할과 영향에 대한 업데이트된 통합 증거를 제공하는 것을 목표로 했습니다. 새로운 경험적 증거는 학부 의학교육에서 환자가 적극적으로 참여하는 학습 목표와 교육 환경의 범위가 점점 더 넓어지고 있음을 보여줍니다. 
This review systematically evaluated 49 primary empirical studies and was aimed at providing updated integrated evidence on the role and impact of the active involvement of patients in medical undergraduate education. The new body of empirical evidence shows the increasing range of learning objectives and educational settings in which patients play an active part in undergraduate medical education.

우리 연구에 따르면 환자들은 의학교육에 참여함으로써 물질적, 전문적, 개인적, 정서적 이점을 얻었다고 설명했습니다. 기대되는 혜택 외에도 여러 저자는 환자 참여 프로그램을 시작해야 하는 근거로 정책적 의무를 언급했습니다. 그러나 몇몇 연구에서는 낙인찍힘에 대한 두려움, 토큰주의 또는 교육 세션의 구조 부족과 같은 잠재적인 피해와 부정적인 경험에 대해 보고했습니다. 학생, 교수진, 환자 스스로가 제기하는 환자 참여와 관련된 우려는 체계적인 방식으로 면밀히 모니터링하고 적절하게 해결해야 합니다. 
Our study found that patients described material, professional, personal and emotional benefits of participating in medical education. In addition to expected benefits, several authors mentioned policy mandates as rationale for initialising patient involvement programmes. Several studies however reported on the potential harms and negative experiences, such as fear of stigmatisation, tokenism or lacking structure of teaching session. Concerns related to patient involvement coming from students, faculty and patients themselves should remain closely monitored in a systematic manner and addressed appropriately.

포함된 논문은 다양한 유형의 환자 역할을 설명했지만, 이전 검토와 유사하게 대부분의 논문에서 환자 교사의 역할을 언급했습니다.1 17-19 최근 논문에서는 환자가 커리큘럼 개발에 점점 더 많이 관여하고 있음을 시사합니다. 이러한 이니셔티브의 대부분은 부수적인 것이었으며 제도적 통합과 장기적인 참여가 부족했습니다. 
Included papers described various types of roles for patients, but the vast majority of papers cited the role of a patient teacher, similarly to previous reviews.1 17–19 More recent papers suggest that patients are increasingly involved in curriculum development. Most of these initiatives were incidental and were lacking institutional incorporation and longitudinal involvement.

이 검토에서 확인된 환자의 역할은 주로 Towle 외2(상자 2)의 검토에서 제안한 스펙트럼의 3~6단계와 일치합니다. 또한 이 스펙트럼에서 특정 레벨로 분류할 수 없는 새로운 역할도 추가로 확인했습니다. 한 역할에서 환자는 커리큘럼 개발에서 동등한 역할을 수행했지만, 전체 커리큘럼이 아닌 특정 코스에 대해서만 역할을 수행했으며, 이는 Towle의 레벨 4와 5 사이에 속했습니다. 또 다른 역할에서는 환자가 특정 과정을 넘어 교육기관 차원의 주제와 커리큘럼 개발에 대해 자문을 받았지만, 동등한 파트너라기보다는 교수진이 주도하는 이니셔티브에서 자문을 받아 Towle의 레벨 4, 5, 6.2의 부분적인 요소를 나타냈습니다. 
The patients’ roles identified in this review are largely in accordance with levels 3–6 of the spectrum proposed in the review by Towle et al2 (box 2). We additionally identified new roles that could not be ascribed to one specific level on this spectrum. In one role, patients did take on roles as equal in curriculum development, but only to specific courses rather than the curriculum as a whole, falling between Towle’s levels 4 and 5. In another role, patients were consulted in institution-level topics and curriculum development beyond specific courses, but rather than being equal partners, they were consulted in a faculty-driven initiative, displaying partial elements of Towle’s levels 4, 5 and 6.2


박스 2. Towle 등이 정의한 의학교육에 대한 환자 참여의 스펙트럼2
Box 2. Spectrum of patient involvement in medical education by Towle et al2

Towle 등이 정의한 의학교육에서의 환자 참여 수준2
Levels of patient involvement in medical education as defined by Towle et al2

  1. 종이 기반 또는 전자 사례의 초점으로서의 환자.
  2. 임상 환경에서 표준화되거나 자원한 환자.
  3. 교수진이 주도하는 커리큘럼에서 학생들과 자신의 경험을 공유하는 환자.
  4. 학생을 가르치거나 평가하는 데 관여하는 환자 교사.
  5. 학생 교육, 평가 및 커리큘럼 개발에서 특정 과정을 넘어 커리큘럼 전체에 대한 동등한 파트너로서의 환자 교사.
  6. 교육, 평가 및 커리큘럼 개발에 환자 교사로서 지속적으로 참여하는 것 외에도 기관 차원에서 참여하는 환자.

 

  1. Patients as focus of a paper-based or electronic case.
  2. Standardised or volunteer patients in clinical settings.
  3. Patients sharing their experiences with students in a faculty-directed curriculum.
  4. Patient teachers involved in teaching or evaluating students.
  5. Patient teachers as equal partners in student education, evaluation and curriculum development beyond specific courses, to the curriculum as a whole.
  6. Patients involved at the institutional level in addition to sustained involvement as patient teachers in education, evaluation and curriculum development.

이 검토에서 확인된 학습 목표는 미래의 의사를 위한 CanMEDS의 역할 중 한 가지를 제외하고 모두 포함되었습니다. 이는 환자의 참여가 의료 커리큘럼의 다양한 측면에 지속적으로 더 큰 영향을 미치고 있음을 보여줍니다.
The learning objectives identified in this review encompassed all but one of the CanMEDS roles for future physicians. This demonstrates that patient’s involvement is continuing to gain a larger influence on a diverse range of aspects of the medical curriculum.

[지속 가능한 환자 참여를 지원하기 위한 조치]에는 다음 등이 포함되었습니다.

  • 장기적인 기관 통합,
  • 환자 모집 및 교육,
  • 자원 지원,
  • 교수진의 명확한 약속 

학생, 교수진, 환자 모두 환자 참여의 중요성과 이점을 강조했지만, 기관은 환자와 학생에게 발생할 수 있는 잠재적 피해를 완화하기 위해 지속적으로 고려하고 모니터링하며 조치를 취해야 합니다. 환자 참여와 관련된 재정적 측면에 대해 보고한 논문은 소수에 불과했으며, 이는 타당성을 뒷받침하기 위해 추가 조사가 필요합니다.
Measures to support sustainable patient involvement included

  • longitudinal institutional incorporation,
  • patient recruitment and/or training,
  • resource support and
  • clear commitment by faculty.

The importance and advantages of patient involvement were highlighted by students, faculty and patients themselves; however, organisations must continue to consider, monitor and take steps to mitigate any potential harms to patients and students. Only few papers reported on the financial aspects related to patient involvement, which should be further investigated to help support feasibility.

중요한 한계는 기존 문헌에 공통된 용어가 없어 관련 논문이 누락될 위험이 높다는 점이며, 이는 이전에 다른 검토 저자들이 한계로 보고한 바 있습니다.2 
An important limitation was the lack of common terminology in the existing literature, potentiating the risk of missing relevant articles, which has been previously reported as a limitation by other review authors.2

저희의 체계적 문헌고찰에는 동료 심사를 거친 저널의 원저 문헌만 포함되었습니다. 환자가 의학교육 참여에 대해 쓴 많은 논문은 블로그 게시물, 학회 성명서, 환자 단체 뉴스레터 등 회색문헌에서만 찾을 수 있으므로, 이 리뷰는 이 주제에 대한 환자 관점의 중요한 측면을 놓칠 수 있는 위험이 있습니다. 또한 영어로 작성된 연구만 포함되었기 때문에 선정된 논문이 편향되었을 수 있습니다. 포함된 논문의 대부분은 북미(n=23), 유럽(n=17), 호주 및 뉴질랜드(n=7)에서 수행되었습니다. 
Our systematic review included only original literature from peer-reviewed journals. As many articles written by patients on their involvement in medical education may only be found in grey literature, including blog posts, conference statements and patient organisation newsletters, this review runs a risk of having missed important aspects of patient views on this topic. Additionally, only studies written in English were included, which may have led to bias in selected papers. The majority of included papers were from North America (n=23), Europe (n=17) and Australia and New Zealand (n=7).

포함된 연구의 대부분은 질적 연구(n=38)였으며, 그 외에는 혼합 방법(n=5) 또는 질적 연구(n=6)였습니다. 포함된 논문 중 대조군을 사용한 논문은 소수에 불과했습니다(n=7). 대부분의 경우 학생과 환자가 자발적으로 중재에 참여했기 때문에 연구 결과를 더 많은 인구에 일반화하기에는 한계가 있을 수 있습니다. 편견을 유발했을 수 있는 또 다른 중요한 요인은 대부분의 연구가 익명화되지 않았거나(익명화가 불가능) 학생 평가의 일부였기 때문에 응답에 사회적 선호 편향이 적용될 위험이 있다는 점입니다. 
The majority of included studies were qualitative (n=38), others were mixed methods (n=5) or qualitative (n=6). Only few of the included papers used control groups (n=7). In most cases, students and patients participated in interventions on a voluntary basis, which may limit the generalisability of findings to the wider population. Another important factor that may have introduced bias is that most studies were not (possible to be) anonymised, or were part of student assessments, which introduces a risk of responses being subject to social desirability bias.

향후 연구는 환자, 학생, 의료 시스템에 대한 장기적인 영향, 특히 환자 중심주의와 공동 의사 결정에 대한 주제에 초점을 맞춰야 합니다. 이러한 연구 격차로 인해 현재 문헌에 기반한 권장 사항은 제한적입니다. 또한 교육 정책의 의사 결정자에게 중요한 요소일 수 있는 환자 참여에 대한 경제적 평가를 수행한 논문은 없습니다. 마지막으로, 환자 참여에 대한 기존 프레임워크를 의학교육에서 새롭게 확인된 환자의 역할과 필요에 맞게 업데이트하려면 더 많은 연구가 필요합니다. 
Future research should focus on the long-term effects for patients, students and the healthcare system, especially on the subjects of patient-centredness and shared decision-making. This gap in research limits recommendations that can be made based on current literature. Additionally, no paper performed an economic evaluation of patient involvement, which may be a critical factor for decision makers in educational policy. Finally, more research is needed to update existing frameworks for patient involvement to the newly identified roles and needs patients have in medical education.

결론Conclusion

의학교육에서의 환자 참여에 관한 첫 번째 논문이 발표된 지 40년이 넘었습니다. 오늘날 의학교육계와 환자 커뮤니티는 모두 환자 중심주의를 촉진하기 위한 움직임에 함께 동참하고 있습니다. 이 체계적 문헌고찰은 교육기관에 적극적인 환자 참여를 지속적으로 통합하고자 하는 커리큘럼 개발자와 의학교육에 참여하고자 하는 환자 단체에 도움이 될 수 있는 지식과 실질적인 고려사항을 제공합니다. 
It has been over 40 years since the first article on patient involvement in medical education was published. Today, both the medical education community and the patient community have joined together in the movement to promote patient-centredness. This systematic review provides knowledge and practical considerations that can aid curriculum developers who wish to sustainably incorporate active patient involvement in their institutions, and patient organisations wishing to engage in medical education.

 


BMJ Open. 2020 Jul 27;10(7):e037217. doi: 10.1136/bmjopen-2020-037217.

Role of active patient involvement in undergraduate medical education: a systematic review

Affiliations collapse

Affiliations

1Erasmus University Medical Center, Rotterdam, Zuid-Holland, Netherlands 340974sd@student.eur.nl.

2University Medical Centre Groningen, Groningen, Netherlands.

3Wissenschaftliche Dienste und Projektberatung, Berlin, Germany.

PMID: 32718925

PMCID: PMC7389514

DOI: 10.1136/bmjopen-2020-037217

Free PMC article

Abstract

Objectives: To identify the scope of active patient involvement in medical education, addressing the current knowledge gaps relating to rationale and motivation for involvement, recruitment and preparation, roles, learning outcomes and key procedural contributors.

Methods: The authors performed a systematic search of the PubMed database of publications between 2003 and 2018. Original studies in which patients take on active roles in the development, delivery or evaluation of undergraduate medical education and written in English were eligible for inclusion. Included studies' references were searched for additional articles. Quality of papers was assessed using the Mixed Methods Appraisal Tool.

Results: 49 articles were included in the review. Drivers for patient involvement included policy requirements and patients' own motivations to contribute to society and learning. Patients were engaged in a variety of educational settings in and outside of the hospital. The vast majority of studies describe patients taking on the role of a patient teacher and formative assessor. More recent studies suggest that patients are increasingly involved in course and curriculum development, student selection and summative assessment. The new body of empirical evidence shows the wide range of learning objectives was pursued through patient participation, including competencies as professional, communicator, collaborator, leader and health advocate, but not scholar. Measures to support sustainable patient involvement included longitudinal institutional incorporation, patient recruitment and/or training, resource support and clear commitment by faculty. The importance and advantages of patient involvement were highlighted by students, faculty and patients themselves; however, organisations must continue to consider, monitor and take steps to mitigate any potential harms to patients and students.

Discussion: This systematic review provides new knowledge and practical insights to physicians and faculty on how to incorporate active patient involvement in their institutions and daily practice, and provides suggested action points to patient organisations wishing to engage in medical education.

Keywords: education & training (see medical education & training); health services administration & management; medical education & training.

졸업후 윤리 교육 프로그램: 체계적 스코핑 리뷰(BMC Med Educ, 2021)
Postgraduate ethics training programs: a systematic scoping review
Daniel Zhihao Hong1,2, Jia Ling Goh1,2, Zhi Yang Ong1,2, Jacquelin Jia Qi Ting1,2, Mun Kit Wong1,2, Jiaxuan Wu1,2, Xiu Hui Tan1,2, Rachelle Qi En Toh1,2, Christine Li Ling Chiang1,2, Caleb Wei Hao Ng1,2, Jared Chuan Kai Ng1,2, Yun Ting Ong1,2, Clarissa Wei Shuen Cheong1,2, Kuang Teck Tay1,2, Laura Hui Shuen Tan1,2, Gillian Li Gek Phua2,3, Warren Fong1,3,4, Limin Wijaya3,5, Shirlyn Hui Shan Neo2, Alexia Sze Inn Lee6, Min Chiam6, Annelissa Mien Chew Chin7 and Lalit Kumar Radha Krishna1,2,6,8,9,10* 

 

소개
Introduction

"의료 행위를 규율하는 도덕적 의무"[1]를 지키기 위한 수단으로 간주되는 의사의 윤리 교육은 의료 행위가 직면한 윤리적 문제에 대처하기 위해 발전해 왔습니다. 기본적인 수준의 윤리 지식과 기술은 캐나다 왕립 의사 및 외과의 대학, 일반 의사회, 미국 가정의학회(AAFP), 의학전문대학원 교육 인증위원회(ACGME) 등의 인증 기관에서 규정하고 있지만, 많은 윤리 프로그램은 임상 실무의 요구에 민감하면서도 변화에 발맞추는 데 어려움을 겪어 왔습니다. 최근 미국의 레지던트 프로그램에서 가정의학과 의사에 관한 리뷰에서 의사들 사이에서 윤리 교육의 내용과 기간에 불가피한 차이가 있다는 사실이 드러났습니다[2]. 
Seen as a means of ensuring that “obligations of moral nature which govern the practice of medicine” [1] are maintained, ethics training amongst physicians have evolved to contend with ethical issues facing medical practice. Whilst basic levels of ethics knowledge and skills have been stipulated by accreditation bodies such as The Royal College of Physicians and Surgeons of Canada, The General Medical Council, the American Academy of Family Physicians (AAFP) and the Accreditation Council for Graduate Medical Education (ACGME), many ethics programs have struggled to keep pace with change whilst remaining sensitive to the demands of clinical practice. Inevitable variations in the content and duration of ethics education amongst physicians have been laid bare in a recent review pertaining to family physicians in residency programs in the United States [2].

법적 책임을 두려워하고 대중의 신뢰와 사회적 기대에 부응하기 위해 고군분투하는 의료계에서 의사에게 윤리 지식, 기술 및 전문직업적 행동을 효과적으로 교육하기 위한 리트머스 시험지[3-7]는 아마도 코로나19 팬데믹이었을 것입니다. 그러나 코로나19 팬데믹 기간 동안 의사의 의심스러운 행동과 임상적 결정에 대한 보고가 드러나면서 기존의 교육 프로그램을 점검하고, 윤리 교육 프로그램의 내용과 구조의 격차를 검토하며, 근거에 기반하고 임상적으로 관련성이 높은 학습자 중심 교육 이니셔티브를 업데이트하고 도입할 수 있는 기회가 되었습니다. 
The litmus test for effectively educating physicians in ethics knowledge, skills and professional conduct in a medical field trepidatious of legal recourse and struggling to meet public trust and societal expectations [37] has perhaps been the COVID-19 pandemic. Yet, the surfacing of reports of questionable physician conduct and clinical decisions during the COVID-19 pandemic also offers an opportunity to take stock of prevailing education programs, review gaps in content and structure of ethics education programs as well as update and instil more evidence based, clinically relevant, learner centred education initiatives.

이 검토의 필요성
The need for this review

의사를 위한 윤리 교육 프로그램을 재구성하는 과정을 안내하기 위해, 동료 검토 문헌에서 현재 [졸업후교육 의학 윤리 교육 및 평가 프로그램]을 분석하는 체계적인 범위 검토를 제안합니다. 
To guide this process of retooling ethics education programs for physicians, a systematic scoping review is proposed to analyse current postgraduate medical ethics training and assessment programs in peer-reviewed literature.

방법론
Methodology

이 체계적 문헌고찰(이하 SEBA의 SSR)[8-14]을 안내하기 위해 크리슈나의 체계적 근거 기반 접근법(SEBA)을 채택하고 광범위한 문헌을 면밀히 조사합니다[15-17]. 구성주의적 관점상대주의적 렌즈를 통해 SEBA의 SSR은 실무에 영향을 미치는 복잡하고 다양한 역사적, 사회문화적, 이념적, 맥락적 요인을 매핑하여 의과대학 졸업생들을 위한 의료윤리 교육 프로그램에 대한 전체적인 그림을 제공합니다[17-24]. 
We adopt Krishna’s systematic evidence-based approach (SEBA) to guide this systematic scoping review (henceforth SSRs in SEBA) [814] and scrutinise a broad range of literature [1517]. With its constructivist perspective and relativist lens, SSRs in SEBA map the complex and diverse historical, socio-cultural, ideological and contextual factors that impact practice to provide a holistic picture of medical ethics training programs for graduates beyond medical school [1724].

연구팀은 결과의 신뢰성을 더욱 높이기 위해 싱가포르국립대학교(NUS) 및 싱가포르국립암센터(NCCS)의 용루린 의과대학(YLLSoM)의 의학 사서, NCCS, 리버풀 완화치료연구소, YLLSoM 및 듀크-NUS 의과대학의 현지 교육 전문가 및 임상의(이하 전문가 팀)의 자문을 구했습니다. SEBA의 체계적 접근, 분할 접근, 직소 관점, , 퍼널링 프로세스, 토론 단계(그림 1. SEBA 프로세스)가 전체 연구 프로세스를 안내하는 데 사용되었습니다. 
To further improve the reliability of the results, the research team consulted medical librarians from the Yong Loo Lin School of Medicine (YLLSoM) at the National University of Singapore (NUS) and the National Cancer Centre Singapore (NCCS), and local educational experts and clinicians at NCCS, Palliative Care Institute Liverpool, YLLSoM and Duke-NUS Medical School (henceforth the expert team). The Systematic Approach, Split Approach, Jigsaw Perspective, , Funnelling Process, and Discussion stages of SEBA (Fig. 1. The SEBA Process) were used to guide the entire research process.

 

1단계: 체계적 접근
Stage 1: Systematic approach

검토 제목 및 배경 결정
Determining the title and background of the review

연구팀은 전문가 팀과 현지 의료윤리 교육 프로그램의 이해관계자들과 협의하여 SEBA에서 SSR의 중요한 목표와 평가 대상 인구, 상황 및 의료윤리 교육 프로그램을 결정했습니다. 
The research team consulted the expert team and stakeholders from a local medical ethics training program to determine the overarching goals of the SSR in SEBA as well as the population, context and medical ethics training programs to be evaluated.

연구 질문 파악
Identifying the research question

포함 기준[25]의 인구, 중재, 비교 및 결과(PICOS) 요소에 따라 1차 연구 질문은 "졸업후 의학 교육 프로그램은 윤리적 기술을 어떻게 가르치는가?"입니다. 두 번째 질문은 "핵심 주제는 무엇인가?""졸업후 의학교육에서 프로그램을 구성하는 데 사용되는 방법은 무엇인가?"입니다. 
Guided by the Population, Intervention, Comparison and Outcome (PICOS) elements of the inclusion criteria [25], the primary research question is “How do postgraduate medical training programs teach ethical skills?” The secondary questions are “What are the core topics included?” and “What are the methods used to structure the program in postgraduate training?”

SEBA 방법론의 반복적 과정의 일환으로, 이 검토의 초기 결과를 논의할 때 전문가 팀은 윤리 교육 평가에 대한 데이터 부족을 해결하기 위해 현재의 윤리 평가 방법에 대한 연구를 수행할 것을 권고했습니다. 이에 따라 SEBA에서 두 번째 SSR이 수행되었습니다. PICOS의 지침에 따라 1차 연구 질문은 "졸업후교육에서 윤리 지식, 기술 및 역량은 어떻게 평가되는가?"입니다. 두 번째 질문은 "어떤 영역이 평가되는가?"입니다. 
As part of the SEBA methodology’s iterative process, when the initial results of this review were discussed, the expert team advised that a study of current methods of assessing ethics be conducted to address the lack of data on assessments of ethics education. Thus, a second SSR in SEBA was carried out. Similarly guided by PICOS, the primary research question is “How is ethics knowledge, skills, and competencies assessed in postgraduate training?” The secondary question is “What domains are assessed?”

포함 기준
Inclusion criteria

연구팀은 전문가 팀의 안내에 따라 의료 윤리를 가르치고 평가하기 위한 SEBA의 SSR에 대한 포함 기준을 다음과 같이 만들었습니다(표 1).
Guided by the expert team, the research team created the inclusion criteria for the SSRs in SEBA for teaching and assessing medical ethics, as outlined (Table 1).

 

검색
Searching

전반적으로 두 검색에는 16명의 연구팀이 참여했으며, 이들은 검토를 위해 PubMed, Embase, PsycINFO 및 ERIC 데이터베이스에서 독립적으로 검색을 수행했습니다. 연구팀은 실행 가능하고 지속 가능한 연구 프로세스를 보장하기 위한 Pham, Rajic[26]의 접근 방식에 따라 일반적인 인력 및 시간 제약을 고려하여 1990년 1월 1일부터 2019년 12월 31일 사이에 출판된 논문으로 검색을 제한했습니다. 영어로 출판되었거나 영어 번역본이 있는 논문의 모든 연구 방법론이 포함되었습니다. 독립적인 검색은 2020년 2월 14일부터 2020년 4월 9일 사이에 수행되었습니다. 전체 PubMed 검색 전략은 추가 파일 1에서 확인할 수 있습니다. 
Overall, both searches involved 16 members of the research team who carried out independent searches of PubMed, Embase, PsycINFO, and ERIC databases for the review. In keeping with Pham, Rajic [26]’s approach to ensuring a viable and sustainable research process, the research team confined the searches to articles published between 1 January 1990 and 31 December 2019 to account for prevailing manpower and time constraints. All research methodologies in articles published in English or had English translations were included. The independent searches were carried out between 14 February 2020 and 9 April 2020. The full PubMed search strategy may be found in Additional File 1.

그런 다음 연구팀은 최종 목록의 모든 제목을 독립적으로 검토하고, 검토에 포함될 개별 논문 목록을 비교하고, '합의된 합의 검증'을 사용하여 윤리 교육(그림 2.) 및 윤리 평가(그림 3)에 대해 분석할 최종 논문 목록에 대한 합의를 도출했습니다.The research team then independently reviewed all the titles on the final list, compared their individual lists of articles to be included in the review and employed ‘negotiated consensual validation’ to achieve consensus on the final list of articles to be analysed on the teaching of ethics (Fig. 2.) and assessing of ethics (Fig. 3).

 
 
 
 
 

2단계: 분할 접근법
Stage 2: Split approach

SEBA의 각 SSR에 대해 5명의 연구원으로 구성된 두 팀이 분석의 신뢰성을 높이는 데 초점을 맞춘 크리슈나의 분할 접근 방식에 따라 동시에 독립적으로 전체 논문을 검토했습니다[27, 28]. 첫 번째 팀은 브라운과 클라크[29]의 주제 분석 접근법을 사용하여 포함된 기사를 면밀히 조사했고, 두 번째 팀은 시에와 섀넌[30]의 지시적 콘텐츠 분석 접근법을 사용했습니다. 분할 접근법의 결과 간의 비교는 방법 삼각 측량을 제공하는 반면, 각 검토자가 동일한 데이터를 독립적으로 분석하도록 하는 것은 연구자 삼각 측량을 제공합니다[27, 28]. 삼각 측량은 외부 타당도를 높이고 이 접근법을 보다 객관적으로 만들 수 있습니다. 
For each SSR in SEBA, two teams of five researchers concurrently and independently reviewed the full-text articles in keeping with Krishna’s Split Approach that is focused on enhancing the reliability of the analyses [27, 28]. The first team scrutinised the included articles using Braun and Clarke [29]’s approach to thematic analysis whilst the second team employed Hsieh and Shannon [30]’s approach to directed content analysis. Comparisons between the results of the Split Approach provides method triangulation whilst having each reviewer independently analyse the same data provides investigator triangulation [27, 28]. Triangulation augments external validity and allows this approach to be more objective.

브라운과 클라크(2006)의 주제별 분석 접근법
Braun and Clarke (2006)’s approach to thematic analysis

의사를 대상으로 의료 윤리를 가르치거나 평가하기 위한 선험적 프레임워크가 없는 상황에서, 우리는 의학에서 의료 윤리의 상황별 특성을 탐색하면서 다양한 목표와 다양한 학년, 경험, 전공을 가진 의사 집단에서 공통된 주제를 찾아내기 위해 브라운과 클라크의 주제 분석 접근법을 사용했습니다[29, 31-37]. 또한 통계적 풀링 및 분석의 사용을 방지하고[29, 38-42] 의료 윤리와 같이 사회문화적으로 영향을 받는 교육 과정에 대한 적절한 분석을 용이하게 하기 위해 포함된 논문들 사이에 존재하는 광범위한 연구 방법론을 수용합니다. 
Without an a priori framework for either teaching or assessing medical ethics amongst physicians, we employed Braun and Clarke’s approach to thematic analysis to single out common themes across varying goals and populations of physicians of different grades, experiences and specialties whilst circumnavigating the context-specific nature of medical ethics in Medicine [29, 3137]. It also accommodates for a wide range of research methodologies present amongst the included articles which prevents the use of statistical pooling and analysis [29, 3842] and facilitates appropriate analysis of socio-culturally influenced educational processes such as medical ethics.

'코드'는 반복적인 단계별 주제 분석을 통해 텍스트의 '표면적' 의미로부터 구성되었습니다. 이러한 코드는 데이터를 가장 잘 표현할 수 있는 테마로 재구성되었습니다. 개별적으로 검토한 다음 그룹으로 검토했습니다. 그 후, 이 소팀원들은 각자의 개별 결과를 온라인으로 심의하고 최종 주제에 대한 합의를 도출하기 위해 '합의된 합의 검증'을 활용했습니다. 
‘Codes’ were constructed from the ‘surface’ meaning of the text through a reiterative step-by-step thematic analysis. These were re-organised into themes that were best able to represent the data. They were reviewed individually and then as a group. Subsequently, the members of this sub-team deliberated their separate findings online and utilised ‘negotiated consensual validation’ to achieve consensus on the final themes.

Hsieh와 섀넌(2005)의 지시적 콘텐츠 분석 접근 방식
Hsieh and Shannon (2005)’s approach to directed content analysis

Hsieh와 섀넌의 지시적 내용 분석 접근법은 주제의 타당성을 높이고 브라운과 클라크가 상대적으로 모순된 데이터를 활용하지 못한 문제를 해결하기 위해 사용되었습니다. 
Hsieh and Shannon’s approach to directed content analysis was employed to increase the validity of the themes and to address Braun and Clarke’s relative failure to engage contradictory data.

윤리 교육과 관련하여 두 번째 하위 팀은 Sutton [43]의 '학부 의학에서의 윤리 및 법 교육 및 학습'이라는 제목의 논문과 McKneally와 Singer [44]의 '임상의사를 위한 생명윤리 25'에서 코드와 범주를 도출했습니다. 임상 환경에서의 생명윤리 교육'. 
With regards to the teaching of ethics, the second sub-team drew codes and categories from Sutton [43]’s article entitled ‘Ethics and law teaching and learning in undergraduate medicine’ and McKneally and Singer [44]’s ‘Bioethics for clinicians 25. Teaching bioethics in the clinical setting’.

윤리, 강령 및 범주 평가와 관련하여 Norcini, Anderson [35]의 '2018 좋은 평가를 위한 합의 프레임워크 초안', Veloski, Boex [45]의 '평가, 피드백 및 의사의 임상 수행에 관한 문헌의 체계적 검토'를 참조하세요: BEME 가이드 7'과 Watling과 Ginsburg[46]의 '평가, 피드백 및 학습의 연금술'이 사용되었습니다. 
With regards to the assessing of ethics, codes and categories from Norcini, Anderson [35]’s ‘Draft 2018 Consensus Framework for Good Assessment’, Veloski, Boex [45]‘s ‘Systematic review of the literature on assessment, feedback and physicians’ clinical performance: BEME Guide No. 7′ and Watling and Ginsburg [46]’s ‘Assessment, feedback and the alchemy of learning’ were used.

이 코드는 포함된 논문을 검토하기 위한 프레임워크로 채택되었습니다. 기존 코드에 포함되지 않은 관련 데이터는 연역적 범주 적용을 통해 새로운 코드를 할당했습니다. 독립적인 연구 결과는 온라인에서 논의되었고, 최종 '코드북'에 대한 합의를 도출하기 위해 '협상된 합의 검증'이 다시 사용되었습니다. 
These codes were adopted as a framework for reviewing the included articles. Any relevant data not captured by existing codes were assigned a new code through deductive category application. The independent findings were discussed online and ‘negotiated consensual validation’ was again used to achieve consensus on the final ‘code book’.

3단계: 직소 관점
Stage 3: The jigsaw perspective

분할 접근법의 결과와 반복적인 프로세스를 통해 도출된 결과를 한데 모아 데이터에 대한 균형 잡힌 관점을 확보했습니다. 여기서는 각 SSR 내의 공통 주제와 카테고리를 비교했습니다. 카테고리와 테마가 겹치는 부분을 결합하여 직소 퍼즐의 보완적인 조각을 모으는 것처럼 데이터에 대한 더 넓은 관점을 만들었습니다. 이 프로세스를 직소 관점이라고 하며, 일관성을 보장하기 위해 전문가 팀이 감독합니다. 
The findings of the Split Approach and its reiterative process were then pooled together to ensure a well-rounded perspective of the data. Here, common themes and categories within each SSR were compared. Overlaps between the categories and themes were combined to create a wider perspective of the data, much like bringing together complementary pieces of a jigsaw. This process is called the Jigsaw Perspective and is overseen by the expert team to ensure consistency.

결과
Results

윤리 교육과 관련된 첫 번째 검색에서는 7669개의 초록이 검색되었으며, 573개의 전체 텍스트 논문이 검토되었고 66개의 논문이 포함되었습니다. 분할 접근법의 일부로 식별된 범주와 주제를 비교한 결과, 직소 관점을 사용하여 주제/카테고리로 결합된 유사한 범주와 주제가 발견되었습니다. 이러한 주제/범주에는 교육 목표, 내용, 사용된 교수법, 윤리를 가르치는 데 있어 조력자 및 장벽이 포함됩니다.  
The first search involving the teaching of ethics retrieved 7669 abstracts, with 573 full-text articles reviewed and 66 articles included. Comparison of the categories and themes identified as part of the Split Approach revealed similar categories and themes which were combined into themes/categories using the Jigsaw Perspective. These themes/categories include the goals, content, teaching methods employed, and enablers and barriers to teaching ethics.

윤리 평가를 위해 검색을 통해 9919개의 초록이 확인되었고, 333개의 전문 논문이 검토되었으며, 29개의 논문이 포함되었습니다. 평가 방법에 대한 SEBA의 SSR의 분할 접근 방식은 평가되는 유형과 영역, 다양한 평가 방법의 장단점을 포함하는 세 가지 주제/범주를 드러냈습니다. 
For the assessment of ethics, the search saw 9919 abstracts identified, 333 full-text articles reviewed and 29 articles included. The Split Approach from the SSR in SEBA of assessment methods revealed three themes/categories which included the types and domains assessed and the pros and cons of various assessment methods.

4단계: 퍼널링 프로세스
Stage 4: The funnelling process

또한, 세 번째 소팀은 포함된 논문의 중요한 논의 개념과 모순되는 견해가 포함되었는지 확인하기 위해 포함된 논문의 전문을 요약하고 표로 작성했습니다. 또한 표로 정리된 요약은 확인된 결과가 기존 데이터의 정확한 표현인지 확인하는 역할도 합니다. 윤리 교육 및 평가에 대한 표로 정리된 요약본은 각각 추가 파일 2와 3에서 확인할 수 있습니다. 연구팀은 전문가 팀의 감독 하에 퍼널링 프로세스에 따라 유사성과 중복되는 영역을 기준으로 SEBA의 두 SSR에서 주제/범주를 결합했습니다.
In addition, a third sub-team summarised and tabulated the included full-text articles to ensure that important concepts of discussion and contradictory views within the included articles were retained. The tabulated summaries also serve to verify that the results ascertained are an accurate representation of the existing data. The tabulated summaries for the teaching and assessing of ethics may be found in Additional File 2 and 3 respectively. Under the oversight of the expert team, the research team combined themes/categories from the two SSRs in SEBA based upon their similarities and their areas of overlap in keeping with the Funnelling Process.

두 검색에서 도출된 5가지 주제/카테고리는 목표와 목적, 콘텐츠, 교수법, 활성화 및 제한 요소, 평가 도구입니다.
The five funnelled themes/categories from the two searches are the goals and objectives, the content, pedagogy, enabling and limiting factors, and assessment tools.

목표 및 목적
Goals and objectives

의사를 위한 윤리 교육 프로그램의 목표와 목적은 아래 표 2에 강조 표시되어 있습니다.
The goals and objectives of ethics training programs for doctors are highlighted in Table 2 below.

 

전반적으로 대부분의 윤리 프로그램의 목표는 

  • 의과대학에서 다루는 주요 윤리적 원칙을 되새기고[51], 
  • 의사가 윤리적 딜레마에 대처할 수 있도록 준비시키며, 
  • 그렇게 할 수 있는 자신감을 향상시키는 것이었습니다[59, 71, 77]. 

일부 프로그램은 다음 섹션에서 다루는 내용에서 강조한 바와 같이 상황 및 전문 분야별 윤리적 딜레마를 소개하기도 했습니다[48, 53, 56, 70, 78-80].
Overall, the goal of most ethics programs was to

  • refresh key ethical principles covered in medical schools [51],
  • prepare physicians to tackle ethical dilemmas, and
  • improve their confidence in doing so [597177].

Some programs also introduced context and specialty-specific ethical dilemmas as highlighted in the next section on content covered [48, 53, 56, 70, 7880].

다루는 콘텐츠
Content covered

다루는 콘텐츠는 표 3에 요약되어 있습니다.
Content covered is outlined in Table 3.

 

대부분의 교육 프로그램은 다양한 주제를 다루고 있습니다.
Most training programs covered a varying number of topics.

Carrese, Malek [96]은 의사를 위한 윤리 교육과 의대생을 위한 윤리 교육에서 다루는 주제의 범위가 겹친다고 지적했지만, 저자들은 "레지던트에게 제공되는 교육 자료는 일반적으로 의대생을 위한 것보다 더 복잡하고 상황에 따라 달라질 수 있으며 윤리적 문제는 더 미묘하고 심도 있게 논의될 수 있다"고 설명합니다.
Whilst Carrese, Malek [96] noted an overlap in the range of topics covered in ethics training for doctors and those for medical students, the authors explain that “educational materials offered to residents can typically be more complex and contextual than those intended for medical students, and ethical issues can be more nuanced and discussed in greater depth”.

교수법
Pedagogy

아래 표 4에는 다양한 교육법이 강조되어 있습니다.
The diverse pedagogies are highlighted in Table 4 below.

 

이러한 교육 세션의 시기와 기간에는 큰 차이가 있습니다. 주최 기관이나 기관에서 운영하는 공식 교육은 몇 년[62, 82] 또는 하루[67]에 걸친 의무 교육 프로그램[80, 81]의 형태로 제공되는 경향이 있었습니다. 일부 프로그램은 매년 몇 시간 동안 진행되거나[58, 94], 더 광범위한 레지던트 교육 프로그램의 일부로 매달 또는 몇 달에 한 번씩 진행되기도 합니다[49, 59, 83]. 
There is great variation in the timing and duration of such training sessions. Formal teaching run by the host organisation or institution tended to come in the form of mandatory training programmes [80, 81] that span the course of a few years [62, 82] or  a single day [67]. Some programs are held over a few hours each year [58, 94], or each month or every few months as part of a wider residency training program [49, 59, 83].


비공식 프로그램은 다과가 제공되고 위계가 최소화되는 보다 비공식적인 환경에서 이루어지는 경향이 있었습니다[49, 59]. 
Informal programs tended to be situated in more informal settings where refreshments are served and hierarchies are minimised [49, 59].

다양한 교육 프로그램은 목표를 달성하기 위해 여러 가지 접근 방식을 조합하여 활용했습니다[82]. 토론토 대학교의 하워드 맥닐리[70]는 공식적인 생명윤리 교육과 "윤리적 행동의 역할 모델링 및 윤리적 문제에 대한 병상 교육"의 통합에 대해 설명합니다. 이러한 조합의 영향은 소아과 프로그램 디렉터를 대상으로 윤리를 가르치는 방법에 대한 Lang, Smith[97]의 설문조사에서 잘 드러납니다. 의료 윤리 교육에 대한 Carrese, Malek [96]의 문헌 검토에서도 마찬가지로 공식, 비공식 및 숨겨진 커리큘럼의 시너지 효과를 강조했습니다[77]. 
Different training programs utilised a combination of approaches to meet their objectives [82]. At the University of Toronto, Howard, McKneally [70] describes integrating formal bioethics teaching with “role modelling of ethical behaviour and bedside teaching around ethical issues”. The impact of this combination is echoed by Lang, Smith [97]’s survey of paediatric programme directors on how ethics is taught. Carrese, Malek [96]’s literature review of medical ethics training similarly highlighted the synergistic nature of the formal, informal and hidden curricula [77].

다른 저자들은 의료 환경에서 팀 기반 업무의 복잡성을 설명하기 위해 다학제적 접근 방식을 사용할 것을 제안했습니다[59, 69, 73, 101, 102]. 
Other authors have proffered the use of a multidisciplinary approach to illustrate the intricacies of team based working in the healthcare setting [59, 69, 73, 101, 102].

활성화 요인 및 장벽
Enabling factors and barriers

윤리 교육 프로그램의 성공적인 실행을 위한 지원 요인과 장벽은 다음과 같이 나타날 수 있습니다(표 5): 
Enabling factors and barriers to the successful execution of ethics training programs may present themselves as follows (Table 5):

새로운 학습자들은 "임상 의학에 대한 노출이 제한적이거나 관련 가치에 대한 직업적 정체성이 아직 완전히 형성되지 않았기 때문에 윤리적 갈등의 실제적인 측면을 인식하지 못하는 경우가 많다"고 생각한 Grace와 Kirkpatrick [68]은 윤리적 사고를 실무에 적용하기 위해 윤리적 비네트 및 윤리적 추론 기법을 시범적으로 사용했습니다. 윤리 교육에 대한 외과 레지던트의 태도에 대한 Howard, McKneally [84]의 연구에 따르면 일반적으로 사례 토론과 윤리적 문제의 실제 적용에 대한 준비가 부족하고 상대적으로 경험이 부족하다는 인식이 드러났습니다. 
Believing that new learners often “do not appreciate the practical side of ethical conflicts as they have had limited exposure to clinical medicine or have not yet fully formed a professional identity with its associated values,” Grace and Kirkpatrick [68] piloted ethical vignettes and ethical reasoning technique to acculturate ethical thinking into practice. Howard, McKneally [84]’s study of surgical resident’s attitudes towards ethics teaching revealed a general sense of being poorly prepared and relatively inexperienced for case discussions and practical application of ethical issues.

Carrese, McDonald [60]와 Chandra, Ragesh [69]도 윤리적 문제가 발생하더라도 제대로 모델링되지 않았고 교육 순간으로 거의 사용되지 않았다고 지적합니다.
Carrese, McDonald [60] and Chandra, Ragesh [69] also note that even in the event that ethical issues did arise, they were poorly modelled and rarely used as teaching moments.

평가 도구
Assessment tools

평가 도구는 사용되는 평가 방법의 유형, 평가되는 해당 영역, 장단점으로 구성됩니다(표 6). 이러한 평가 방법은 밀러의 임상 역량 피라미드에 매핑될 수 있습니다[131].
Assessment tools comprise the type of assessment method employed, corresponding domains assessed and their pros and cons (Table 6). These assessment methods may be mapped onto the Miller’s pyramid of clinical competency [131].

5단계: 5단계: SEBA의 SSR에 대한 토론 및 종합
Stage 5: Discussion and synthesis of SSRs in SEBA

의료 윤리 교육 및 평가에 경험이 있는 현지 교육자, 임상의, 연구자들과 함께 결과를 검토하고 자문을 구하여 이 검토의 완성도를 다시 한 번 확인했습니다. 작성된 내러티브는 STORIES(보건의료 교육에서의 근거 종합 보고에 대한 구조화된 접근법) 성명서[38] 및 근거 중심 의학 교육(BEME) 협력 가이드[39]에 따라 작성되었으며, 의사들의 윤리 교육 및 평가에 대한 이 새로운 검토는 여러 가지 통찰력을 보여줍니다. 여기에서는 쉽게 참조할 수 있도록 주요 결과 중 일부를 나열하고 특히 관심 있는 세 가지 영역을 자세히 살펴봅니다. 
A review of the results and consultation with local educationalists, clinicians and researchers experienced in medical ethics teaching and assessment reiterated the completeness of this review. The narrative produced was guided by the STORIES (Structured approach to the Reporting In healthcare education of Evidence Synthesis) statement [38] and Best Evidence Medical Education (BEME) Collaboration guide [39].This novel review of teaching and assessment of ethics amongst physicians reveals a number of insights. Here we list some of the key findings for ease of reference and will delve into three areas of particular interest.

  • 대부분의 윤리 프로그램의 공통적인 목표는 [윤리적 원칙에 대한 인식]과 [윤리적 딜레마를 재치 있고 전문적으로 해결하는 기술을 향상]시키는 것입니다. 그러나 최근의 논문들은 관행을 바꾸고, 태도를 형성하며, 사회적 및 직업적 의무를 충족하는 데 초점을 맞추고 있습니다.
  • 윤리를 가르치고 평가하는 최근의 사례는 맥락 및 전공 관련 영향의 영향을 드러냅니다.
  • 대부분의 프로그램의 핵심 요소는 다음에 관한 것이었습니다.
    • 자율성, 유익성, 비악용성, 정의의 네 가지 원칙, 
    • 의사-환자 관계, 
    • 의사소통, 
    • 임종 간호
  • 전공 분야 또는 맥락 특이적 정보 내용에는 다음이 포함됩니다.
    • 연구 윤리,
    • 전공 분야 관련 주제,
    • 수련의 관련 고려 사항,
    • 사회적 및 제도적 상호 작용
  • 윤리를 가르치는 데는 여러 가지 접근법이 사용되었지만, 모두 학습자에게 그룹 토론에 선택적으로 참여하는 것부터 사례 토론 및 성찰을 유도하는 것까지 다양한 방식으로 지식을 적용할 수 있는 기회를 제공하는 데 중점을 두었습니다. 
  • 윤리 교육과 평가를 촉진하는 요소로는 체계적인 프로그램, 양육 문화, 우려 사항과 문의 사항을 논의할 수 있는 안전한 환경이 꼽혔습니다.
  • 윤리 교육에서 중요한 것은 역할 모델링, 사례 기반 토론, 윤리적 감수성 및 윤리적 문제 해결에 대한 교육입니다.
  • 일반적으로 평가 방법이 부족합니다.
  • The common objective across most ethics programs is to improve awareness of ethical principles and skills in resolving ethical dilemmas tactfully and professionally. More recent articles however focused on changing practice, shaping attitudes and meeting social and professional obligations.
  • Recent accounts of teaching and assessing ethics reveal the impact of context and speciality related influences.
  • The core elements of most programs concerned
    • the four principles of autonomy, beneficence, non-maleficence and justice;
    • the doctor- patient relationship;
    • communication; and
    • end of life care.
  • Speciality or context specific information contents include
    • research ethics;
    • speciality related topics;
    • trainee related considerations and
    • social and or institutional interactions.
  • There were a number of approaches employed to teach ethics yet all were focused upon providing learners with an opportunity to apply their knowledge in a variety of ways, ranging from optional participation in group discussions to guided case discussions and reflections.
  • Factors facilitating ethics education and assessments were a structured program, a nurturing culture and a safe environment to discuss concerns and enquiries.
  • Important in ethics training are role modelling, case-based discussions and instruction on ethical sensitivity and resolving ethical issues.
  • There is a general lack of assessment methods.

각 교육 프로그램에는 고유한 차이가 있지만, 지식 구축에서 실천으로, 궁극적으로는 [학습자의 직업적 정체성]을 키우는 데까지 학습자를 안내하는 연속선상에 있다고 볼 수 있습니다. 실제로 많은 프로그램에서 학습자가 사회적 책임[49, 60, 70-76, 85]과 '실무 커뮤니티'[69, 70]에 대한 멤버십을 갖출 수 있도록 준비시키려고 노력합니다. 이는 크루스, 크루스[131]가 수정한 밀러 피라미드의 정점에 있는 "Is" 단계와 일치합니다. 이러한 가설에 대한 증거는 윤리 교육이 가르치는 내용과 방식에서 확인할 수 있습니다.
While there are inherent differences to each of the training programs, they may be seen to lie on a continuum of guiding the learner from knowledge building to practice and ultimately to nurturing the learner’s professional identity. Indeed, many programs seek to prepare learners for their societal responsibilities [49, 60, 7076, 85] and their membership to their ‘community of practice’ [69, 70]. This would be consistent with Cruess, Cruess [131]’s “Is” level at the apex of their amended Miller’s pyramid. With this in mind, evidence for this posit is visible from the contents and manner that ethics education is taught.

교육 프로그램의 종단적 성격, 재교육 세션 및 자율성, 유익성, 비악의성, 정의, 임종 간호, 의사 환자 관계 및 치료 의무와 같은 '핵심' 주제를 포함하는 세션의 존재에 대한 면밀한 연구는 [일반적인 지식의 강화]를 시사합니다 [48, 50, 51, 56, 57, 59, 67, 73, 81, 82, 86, 97, 98]. 보다 전문화된 전공과목 지식, 임상 및 연구 콘텐츠의 도입은 [새로운 지식과 경험적 학습의 계층화]를 시사합니다. 이러한 [기존 지식에 대한 구축 과정]은 [의과대학에서 받은 교육과 임상 환경에서 윤리적 문제에 대한 인식을 심화하려는 노력]에 기반한 것으로 보이는 [교육의 종단적 특성]을 입증합니다. 이는 학습자를 교육하는 데 사용되는 방법에서도 입증됩니다. 교훈적인 강의, 온라인 비디오, 병상 윤리 토론은 사례 토론과 프레젠테이션으로 이어지며, 학습자는 지식과 자신감을 쌓고 윤리적 문제를 해결하는 데 지식과 기술을 적용할 수 있습니다[58, 69, 74, 78, 86, 99]. 
Careful study of the longitudinal nature of training programs, the presence of refresher sessions and/or sessions involving ‘core’ topics such as autonomy, beneficence, non-maleficence, justice, end of life care,the doctor patient relationship and the duty of care suggests a reinforcement of prevailing knowledge [48, 50, 51, 56, 57, 59, 67, 73, 81, 82, 86, 97, 98]. The introduction of more specialised speciality, clinical and research content suggests a layering of new knowledge and experiential learning. This process of building on prevailing knowledge evidences the longitudinal nature of training that would seem to build on training received in medical school and efforts to deepen appreciation of ethical issues in the clinical setting. This is also evidenced by the methods used to train the learners. Here didactic lectures, online videos and bedside ethics discussions give way to case discussions and presentations, allowing the learner to build their knowledge and confidence and apply their knowledge and skills in addressing the ethical issues [58, 69, 74, 78, 86, 99].

이러한 고려사항은 또한 대부분의 프로그램에서 사용하는 [나선형 커리큘럼의 수직적 측면]을 강조하고 지식 및 술기 평가의 중요성을 높입니다. 임종 간호와 특히 관련이 있거나 중환자실 배치와 같은 연명 치료 중단 및 보류에 대한 논의가 이루어지는 등 특정 진료 단계에 윤리 교육이 도입되었다는 증거는 윤리 교육 프로그램이 [수평적으로 통합]되어 있음을 시사합니다.
These considerations also highlight the vertical aspect of the spiral curriculum employed by most programs and raise the importance of knowledge and skills assessments. Evidence that ethical training is introduced at specific stages of practice such as during postings where end of life care is especially relevant, or where discussions of withdrawing and withholding life sustaining treatment, such as intensive care placements, suggest horizontal integration of the ethics training programs.

일반적인 지식을 바탕으로 상황에 맞는 학습을 통합하고자 하는 나선형 커리큘럼의 존재는 두 가지 고려 사항을 강조합니다. 첫 번째는 교육의 다음 단계로의 진도를 결정하기 위해 적절한 평가를 사용하는 것이고, 두 번째는 주최 기관의 프로그램 지원입니다.
The presence of a spiral curriculum that seeks to build on prevailing knowledge and integrate context specific learning highlights two considerations. The first is the use of pertinent assessments to determine progress to the next stage of the training and the second is the support of the program by the host organizsation.

교육 후에는 Norcini [132]가 제안한 것처럼 지식이 효과적으로 동화되고 적절하게 적용되었는지 확인하고 마이크로 크레딧을 촉진하기 위한 평가가 뒤따라야 합니다. 이와 함께 윤리 교육에서 명확한 위탁 가능한 전문 활동(EPA)을 수립할 필요성도 있는데, 현재로서는 실무의 다양성, 전문성, 사회 문화적 고려 사항 및 학습자의 일반적인 지식, 기술, 태도 및 경험 측면에서 학습자의 다양성을 고려할 때 추가 연구와 고려가 필요합니다[133]. 교육 포트폴리오의 일부로서 종단적 평가 프로세스의 필요성과 그것이 직업 정체성 형성(PIF) 개발에 미치는 영향에 대해서도 면밀한 조사가 필요합니다[131, 134]. 
Training should be followed by assessments to ensure that knowledge has been effectively assimilated and applied appropriately, and to facilitate micro-credentialling, as suggested by Norcini [132]. In tandem with this, there is also the need to establish clear Entrustable Professional Activities (EPA) s in ethics education which, at present, will require further research and consideration given the diversity of practice, specialities, socio-cultural considerations and learner variability in terms of their prevailing knowledge, skills, attitudes and experience [133]. The need for a longitudinal assessment process as a part of an education portfolio and their impact on the development of professional identity formation (PIF) also demands closer scrutiny [131, 134].

여기서 학습 포트폴리오는 학부 교육과 대학원 교육에서 윤리 교육을 원활하게 통합할 수 있으며[51, 83, 97, 98], 윤리 교육이 종단적 교육 경험의 일부라는 개념[4, 135]에 부합하여 PIF[131, 134]를 육성하는 데 도움이 될 수 있습니다. 포트폴리오는 윤리적 역량을 종단적으로 평가하는 데 유용한 평가 방식일 뿐만 아니라 지식 결핍을 인식하고 올바른 행동을 강화함으로써 지속적인 자기 학습을 촉진합니다[63, 136-143]. 
Here, learning portfolios will allow seamless integration between ethics training in undergraduate and postgraduate training [51, 83, 97, 98] and would be in keeping with the notion of ethics training being part of a longitudinal training experience [4, 135] that nurtures PIF [131, 134]. Portfolios not only serve as a valuable assessment modality for longitudinal evaluation of ethical competency but also promotes continuous self-learning through the recognition of knowledge deficits while reinforcing good behaviour [63, 136143].

그러나 효과적인 윤리 교육 프로그램을 위해서는 레지던트 프로그램, 의료 기관 및 교육 기관이 전담 자원, 인력 및 교수진 교육을 할당하는 등 지원이 필요합니다[64, 70, 79]. 주최 기관은 이러한 교육을 조율하고 프로그램의 궤적을 세심하게 감독해야 합니다. 마찬가지로 중요한 것은 임상의가 '진료 커뮤니티'에서 자신의 역할과 책임을 인정하고 받아들이도록 하는 노력이 필요하다는 것입니다[144]. 선택된 주제는 제한된 시간 내에 실용적이고 실현 가능하면서도 임상 실무와 관련이 있어야 합니다[52, 55, 59, 60, 73, 79, 83, 96-98]. 
Yet an effective ethics training program requires support fromresidency programs, healthcare institutes and educational institutes through the allocation ofallocating dedicated resources, manpower and faculty training [64, 70, 79]. The host organisation must orchestrate this training and provide careful oversight of the program's trajectory. Perhaps just as important is that there are efforts to ensure that clinicians acknowledge and adopt their roles and responsibilities in their ‘communities of practice’ [144]. The topics chosen should be practical and feasibly covered within the limited time allotted yet be relevant to clinical practice [52, 55, 59, 60, 73, 79, 83, 9698].

또한 프로그램과 주최 기관은 핵심 가치를 전파하고 "의료 기관의 일상 생활에 이러한 가치를 적극적으로 통합"하는 인프라를 도입하여 윤리적 분위기를 조성해야 합니다[145]. 윤리적 분위기는 직업적 정체성 형성에 도움이 될 것입니다[131, 134, 146]. 
The programs and host organisations must also instil a nurturing ethical climate through the dissemination of core values and introduction of infrastructure that “proactively incorporates these values in the daily life of the healthcare organi[z]ations” [145]. An ethical climate would aid in professional identity formation [131, 134, 146].

한계점
Limitations

대학원 의학 교육에서 윤리 교육에 관한 다양한 문헌을 살펴보고자 했지만, 각 논문을 더 깊이 있게 연구할 수 있음은 분명합니다. 이러한 한계는 주로 현재의 교육 접근 방식과 커리큘럼, 프로그램 수행 및 평가 방식에 대한 보고가 불완전하기 때문입니다. 
Whilst it was our intention to appreciate the range of available literature on ethics education in postgraduate medical education, it is evident that each paper could be studied in greater depth. This limitation is mainly due to incomplete reporting of the current training approaches and their curriculum, as well as the way in which the programs are is carried out and evaluated.

또한, 선택된 논문의 범위가 주로 북미와 유럽에서 작성된 논문에서 비롯되었습니다. 따라서 다양한 지리적 경계를 가로지르는 다양한 문화가 고려되지 않았기 때문에 이러한 연구 결과의 적용 가능성에 한계가 있습니다. 
Furthermore, the range of selected articles chosen originates from papers that were largely written in North America and Europe. This limits the applicability of these findings, as the different cultures across the different geographical boundaries are not accounted for.

그러나 이러한 한계에도 불구하고 이 범위 검토는 Arksey와 O'Malley [21], Pham, Rajic [26], Levac, Colquhoun [147]이 주장한 엄격함과 투명성을 바탕으로 수행되었습니다. 서지 관리 프로그램인 Endnote를 사용하여 여러 데이터베이스의 모든 인용을 적절히 설명할 수 있었습니다. 
However, despite these limitations, this scoping review was carried out with the necessary rigour and transparency advocated by Arksey and O’Malley [21], Pham, Rajic [26], and Levac, Colquhoun [147]. The use of Endnote, a bibliographic manager, ensured that all the citations from the different databases were properly accounted for.

결론
Conclusion

우리는 이 범위 검토의 결과 분석이 전 세계 대학원 의료 환경의 교육자 및 프로그램 설계자와 관련이 있을 것이라고 믿습니다. 그러나 접근 방식, 내용 및 질 평가에 대한 합의가 부족하고 관점의 차이가 있으며 윤리, 의사소통 및 전문성 간의 내재적 연관성을 탐구할 필요가 있다는 점[63, 148, 149]은 의학에서 의사소통 기술과 전문성을 향상시키는 데 초점을 맞춘 프로그램을 포함시키는 것을 정당화합니다. 또한 학습자의 다양한 특성, 환경, 경험 수준, 특정 의료 시스템 및 문화 속에서 윤리에 대한 EPA를 수립하기 위한 연구를 더 많이 수행해야 합니다. 또한 대학원 환경에서 윤리 실습에 대한 포트폴리오 설계, 실행 및 평가와 PIF 및 마이크로 크레딧에 대한 연구도 진행되어야 합니다. 
We believe the analysis of our findings in this scoping review will be relevant to educators and program designers in postgraduate medical settings around the world. However, the lack of consensus and difference in perspectives regarding the approach, content and quality assessments as well as the need to explore the inherent link amongst ethics, communication and professionalism [63, 148, 149] justifies inclusion of programs focused on enhancing communication skills and professionalism in medicine. In addition, more needs to be done to research on establishing EPAs in ethics amidst the diverse characteristics of learners, their settings and their levels of experience as well as the particular healthcare system and culture that they practice in. Research should also look into portfolio design, implementation and assessment of PIF and micro-credentialling in ethics practice in the postgraduate setting.

 


BMC Med Educ. 2021 Jun 9;21(1):338. doi: 10.1186/s12909-021-02644-5.

Postgraduate ethics training programs: a systematic scoping review

Affiliations collapse

1Yong Loo Lin School of Medicine, National University of Singapore, NUHS Tower Block, 1E Kent Ridge Road, Level 11, Singapore, 119228, Singapore.

2Division of Supportive and Palliative Care, National Cancer Centre Singapore, 11 Hospital Cr, Singapore, 169610, Singapore.

3Duke-NUS Medical School, National University of Singapore, 8 College Rd, Singapore, 169857, Singapore.

4Department of Rheumatology and Immunology, Singapore General Hospital, 16 College Road, Block 6 Level 9, Singapore General Hospital, Singapore, 169854, Singapore.

5Department of Infectious Diseases, Singapore General Hospital, Outram Road, Singapore, 169608, Singapore.

6Division of Cancer Education, National Cancer Centre Singapore, 11 Hospital Cr, Singapore, 169610, Singapore.

7Medical Library, National University of Singapore Libraries, Blk MD6, Centre, 14 Medical Dr, #05-01 for Translational Medicine, Singapore, 117599, Singapore.

8Yong Loo Lin School of Medicine, National University of Singapore, NUHS Tower Block, 1E Kent Ridge Road, Level 11, Singapore, 119228, Singapore. lalit.radha-krishna@liverpool.ac.uk.

9Division of Supportive and Palliative Care, National Cancer Centre Singapore, 11 Hospital Cr, Singapore, 169610, Singapore. lalit.radha-krishna@liverpool.ac.uk.

10Division of Cancer Education, National Cancer Centre Singapore, 11 Hospital Cr, Singapore, 169610, Singapore. lalit.radha-krishna@liverpool.ac.uk.

11Palliative Care Institute Liverpool, Cancer Research Centre, University of Liverpool, 200 London Rd, L3 9TA, Liverpool, UK. lalit.radha-krishna@liverpool.ac.uk.

12Centre of Biomedical Ethics, National University of Singapore, 21 Lower Kent Ridge Rd, Singapore, 119077, Singapore. lalit.radha-krishna@liverpool.ac.uk.

13PalC, The Palliative Care Centre for Excellence in Research and Education, PalC c/o Dover Park Hospice, 10 Jalan Tan Tock Seng, Singapore, 308436, Singapore. lalit.radha-krishna@liverpool.ac.uk.

PMID: 34107935

PMCID: PMC8188952

DOI: 10.1186/s12909-021-02644-5

Abstract

Background: Molding competent clinicians capable of applying ethics principles in their practice is a challenging task, compounded by wide variations in the teaching and assessment of ethics in the postgraduate setting. Despite these differences, ethics training programs should recognise that the transition from medical students to healthcare professionals entails a longitudinal process where ethics knowledge, skills and identity continue to build and deepen over time with clinical exposure. A systematic scoping review is proposed to analyse current postgraduate medical ethics training and assessment programs in peer-reviewed literature to guide the development of a local physician training curriculum.

Methods: With a constructivist perspective and relativist lens, this systematic scoping review on postgraduate medical ethics training and assessment will adopt the Systematic Evidence Based Approach (SEBA) to create a transparent and reproducible review.

Results: The first search involving the teaching of ethics yielded 7669 abstracts with 573 full text articles evaluated and 66 articles included. The second search involving the assessment of ethics identified 9919 abstracts with 333 full text articles reviewed and 29 articles included. The themes identified from the two searches were the goals and objectives, content, pedagogy, enabling and limiting factors of teaching ethics and assessment modalities used. Despite inherent disparities in ethics training programs, they provide a platform for learners to apply knowledge, translating it to skill and eventually becoming part of the identity of the learner. Illustrating the longitudinal nature of ethics training, the spiral curriculum seamlessly integrates and fortifies prevailing ethical knowledge acquired in medical school with the layering of new specialty, clinical and research specific content in professional practice. Various assessment methods are employed with special mention of portfolios as a longitudinal assessment modality that showcase the impact of ethics training on the development of professional identity formation (PIF).

Conclusions: Our systematic scoping review has elicited key learning points in the teaching and assessment of ethics in the postgraduate setting. However, more research needs to be done on establishing Entrustable Professional Activities (EPA)s in ethics, with further exploration of the use of portfolios and key factors influencing its design, implementation and assessment of PIF and micro-credentialling in ethics practice.

Keywords: Ethics curriculum; Ethics education; Ethics training program; Medical ethics; Physicians; Postgraduate medical education; SEBA; Scoping review; Systematic scoping review.

프로페셔널리즘 달성에 의료윤리교육의 필수적 역할: 로마넬 보고서(Acad Med, 2015)
The Essential Role of Medical Ethics Education in Achieving Professionalism: The Romanell Report
Joseph A. Carrese, MD, MPH, Janet Malek, PhD, Katie Watson, JD, Lisa Soleymani Lehmann, MD, PhD, Michael J. Green, MD, MS, Laurence B. McCullough, PhD, Gail Geller, ScD, MHS, Clarence H. Braddock III, MD, MPH, and David J. Doukas, MD 

 

1985년, 디캠프 보고서로 알려진 "의료 윤리의 기본 커리큘럼 목표"라는 획기적인 논문에서 [의료 윤리에 대한 기본 교육이 모든 미국 의과대학에서 필수 과목이 되어야 한다]고 주장했습니다.1 같은 해 의학교육연락위원회(LCME)는 미국 의과대학에서 "의학과 관련된 윤리적, 행동적, 사회경제적 과목이 커리큘럼에 포함되어야 하며, 의료윤리와 인간의 가치에 관한 자료가 제시되어야 한다."2 최근에는 의학교육자 및 인증기관이 윤리 교육 지침의 범위를 확대했으며, 이는 모든 수준의 학습자가 환자 치료, 교육 및 연구 분야에서 평생 전문직업성을 위해 헌신할 수 있도록 준비할 수 있도록 전문가 양성에 관한 교육을 받을 것을 요구함으로써 부분적으로 나타났습니다.3 의사의 도덕적 규범과 가치에 따라 행동하려는 의사의 능력과 의지는 전문직 행동의 핵심 요소 중 하나이며, 그 결과 윤리와 관련된 교육 목표는 이제 전문직업성 교육의 더 넓은 목표에 통합되는 경우가 많습니다.
In 1985, the landmark article “Basic Curricular Goals in Medical Ethics,” known as the DeCamp Report, argued that basic instruction in medical ethics should be a requirement in all U.S. medical schools.1 That same year, the Liaison Committee on Medical Education (LCME) introduced standards stipulating that in U.S. medical schools “ethical, behavioral, and socioeconomic subjects pertinent to medicine must be included in the curriculum and that material on medical ethics and human values should be presented.”2 More recently, medical educators and accrediting organizations have expanded the scope of ethics education guidelines, manifested in part by requirements that learners at all levels receive instruction addressing professional formation to prepare them for a lifelong commitment to professionalism in patient care, education, and research.3 A physician’s ability and willingness to act in accordance with accepted moral norms and values is one key component of professional behavior; as a result, educational objectives relating to ethics are now often incorporated into broader goals for professionalism education.


의료 윤리 및 전문직업성 교육의 중요성에 대한 광범위한 합의에도 불구하고 미래의 의사를 위한 다음에 대한 합의가 이루어지지 않고 있습니다.4 -8 

  • 의료 윤리 교육의 구체적인 목표,
  • 학습자가 습득해야 하는 필수 지식 및 기술,
  • 최적의 교육 방법론 및 프로세스,
  • 최적의 평가 전략

또한, 특히 졸업후교육 교육(GME) 수준에서 교육의 질과 범위는 교육기관과 레지던트 수련 프로그램 내에서 그리고 프로그램 간에 다양합니다.9-11 이러한 다양성은 교육적 맥락의 차이를 고려할 때 적절할 수 있지만, 의료윤리 교육 노력은 [궁극적으로 인증기관이 명시한 전반적인 기대치]를 충족해야 합니다. 다양한 접근 방식은 모든 접근 방식이 기본적인 교육 목표를 달성하는 데 성공하는지에 대한 우려를 불러일으키며, 이는 "의료 윤리 교육에 대한 어떤 접근 방식이 가장 효과적인가?"라는 질문으로 이어집니다.
Despite broad consensus on the importance of teaching medical ethics and professionalism, there is no consensus about the specific goals of medical ethics education for future physicians, the essential knowledge and skills learners should acquire, the best methodologies and processes for instruction, and the optimal strategies for assessment.4–8 Moreover, the quality and extent of instruction, particularly at the graduate medical education (GME) level, varies within and across institutions and residency training programs.9–11 Although such variation may be appropriate in light of differences in educational contexts, medical ethics education efforts must ultimately address the overarching expectations articulated by accrediting organizations. Variation raises concerns about whether all approaches succeed in meeting basic educational objectives, which leads to the question, “Which approaches to medical ethics education are most effective?”

이 글, 로마넬 보고서는 생명윤리 교육에 대한 패트릭 및 에드나 로마넬 기금의 지원을 받아 의학교육 재균형 및 통합 프로젝트(PRIME)의 결과물입니다. PRIME은 의과대학 및 레지던트 수련의 전문직업성 교육과 관련된 의료 윤리 및 인문학 교육에 초점을 맞춘 전국적인 실무 그룹이었습니다.12,13 PRIME은 전문직업성 교육 전담 조직인 의료 전문직업성 아카데미의 설립을 이끌어 냈습니다.14
This article, the Romanell Report, is a product of the Project to Rebalance and Integrate Medical Education (PRIME), funded by the Patrick and Edna Romanell Fund for Bioethics Pedagogy. PRIME was a national working group that focused on medical ethics and humanities education as they relate to professionalism education in medical school and residency training.12,13 PRIME led to the founding of the Academy for Professionalism in Health Care as an organization devoted to professionalism education.14

의료 윤리 교육에 특별한 관심을 가지고 있는 PRIME의 회원으로서, 우리는 이 보고서에서 의료 학습자들의 전문직업성 함양에 있어 이러한 교육의 필수적인 역할을 다루고 있습니다. 우리는 앞서 의료 전문직업성을

  • (1) 과학적, 임상적으로 유능해지는 것,
  • (2) 임상 지식과 기술을 주로 환자의 건강 관련 이익을 보호하고 증진하기 위해 사용하고 사익은 체계적으로 부차적인 것으로 유지하는 것,
  • (3) 의료를 구성원들의 경제적, 정치적, 사회적 권력을 보호하는 데 주로 관심을 갖는 길드가 아니라 공공의 신뢰로서 유지하는 것으로 설명한 바 있습니다.13

As members of PRIME with a particular interest in medical ethics education, we address in this report the essential role of such education in cultivating professionalism among medical learners. We previously described medical professionalism as

  • (1) becoming scientifically and clinically competent;
  • (2) using clinical knowledge and skills primarily for the protection and promotion of the patient’s health-related interests, keeping self-interest systematically secondary; and
  • (3) sustaining medicine as a public trust, rather than as a guild primarily concerned with protecting the economic, political, and social power of its members.13

저희는 저명한 임상윤리 교과서에서 '의료 윤리'에 대한 실무적 정의를 따릅니다: 

  • "임상 윤리는 모든 임상 상황에 존재하는 [윤리적 특징]과 그러한 상황에서 때때로 발생하는 [윤리적 문제] 모두에 관한 것이다."15 

또한, 우리는 의료 윤리에 문제나 가치 갈등이 있을 때 어떻게 해야 하는지를 결정하는 것, 즉 가능한 선택지 중에서 올바른 행동 방침이나 도덕적으로 [수용 가능한 선택을 결정]하는 것에 대한 관심도 포함한다고 생각합니다.
We take our working definition of “medical ethics” from a prominent textbook on clinical ethics:

  • “Clinical ethics concerns both the ethical features that are present in every clinical encounter and the ethical problems that occasionally occur in those encounters.”15 

In addition, we consider medical ethics to include attention to determining what ought to be done when problems or values conflicts are present: that is, determining the right course of action or a morally acceptable choice from among the available options.

[의료 윤리]와 [전문직업성] 사이에는 밀접한 관계가 있으며, 의료 윤리에 관한 광범위한 학문이 전문직업성에 대한 우리의 생각을 알려준다는 것은 자명한 사실이라고 생각합니다. 그러나 이러한 관계에 대한 철저한 분석은 이 글의 범위를 벗어납니다. 이 보고서에서는 전문가 양성을 추구하는 데 있어 인문학 교육의 중요한 역할에 대해서는 다루지 않았으며, 향후 연구에서 이에 대해 집중적으로 다룰 계획입니다. 또한 이 글에서는 의과대학 및 레지던트 수련 과정의 의료 윤리 교육에 초점을 맞추고 있지만, 교육의 연속성이 양쪽 모두에 걸쳐 있음을 인정합니다. 우리는 의료 윤리와 전문직업성은 의과대학 전 교육 과정에서도 우선순위가 되어야 하며, 레지던트 이후에도 계속 의학 교육(CME)을 통해 강화되어야 한다고 믿습니다.
We consider it self-evident that there is a close relationship between medical ethics and professionalism and that the extensive body of scholarship on medical ethics informs how we think about professionalism. However, a thorough analysis of this relationship is beyond the scope of this article. We do not address the important role of humanities education in the pursuit of professional formation in this report; we plan to focus on that in future work. Additionally, although our focus in this article is on medical ethics education during medical school and residency training, we acknowledge that the educational continuum extends on either side of this focus. We believe that medical ethics and professionalism should also be made a priority during premedical studies and reinforced post residency through continuing medical education (CME).

이 보고서에서는 의료윤리 교육자들이 인증기관의 명확한 기대치를 충족하는 데 도움을 주기 위해 의과대학 및 레지던트 프로그램에서 [의료윤리 교육의 목표와 목적, 교수법, 평가 전략, 추가적인 도전과 기회]에 대해 다룹니다. 다음 단계와 향후 연구 분야를 권장하는 것으로 마무리합니다.
In this report, with the aim of aiding medical ethics educators in meeting the articulated expectations of accrediting organizations, we address the following aspects of medical ethics education in medical schools and residency programs: goals and objectives, teaching methods, assessment strategies, and additional challenges and opportunities. We conclude by recommending next steps and areas for future study.

목표와 목적
Goals and Objectives

대부분의 교육자들은 [의료 윤리 교육의 핵심 목표][환자 치료의 우수성을 증진하는 것]이라는 데 동의하지만, 이 목표를 달성하는 최선의 방법에 대해서는 다양한 견해가 있습니다.4

  • 일부 교육자들은 미래 의사의 인성 개발의 중요성을 강조하는 반면, 다른 교육자들은 행동 형성에 더 초점을 맞추는 것이 더 적절하다고 주장합니다.
  • 또 다른 사람들은 윤리와 전문직업성은 가르칠 수 없으며, 의대 입학 과정을 통해 도덕적인 인재를 선발해야 한다고 생각합니다.

이러한 학파를 지지하는 사람들 사이의 논쟁은 가까운 시일 내에 해결될 것 같지 않습니다.
Although most educators agree that the central goal of medical ethics education is to promote excellence in patient care, there are diverse views about how best to achieve this aim.4 Whereas some educators emphasize the importance of developing future physicians’ character, others hold that shaping their behavior is a more appropriate focus. Still others believe that ethics and professionalism cannot be taught; rather, virtuous individuals must be selected through the medical school admission process. The debate among proponents of these schools of thought is unlikely to be resolved in the near future.

의과대학은 '올바른' 인성과 태도를 갖춘 학생을 선발하기 위해 노력해야 하지만, 이러한 자질을 정확하게 평가하기는 어렵습니다. 또한 의료 윤리 및 전문직업성 교육에 사용할 수 있는 제한된 시간 내에 인성 변화에 영향을 미치는 것은 어려운 일입니다. 인성 특성을 형성하고 평가하는 데 따르는 현실적인 어려움은 [윤리적 및 전문직업적 덕목을 모범적으로 보여주는 행동]을 배양하는 대안으로 이어집니다. 이 접근법의 기본은 [임상 의학에서 직면하는 윤리적 딜레마를 보고, 예방하고, 분석하고, 해결하기 위한 개념적 도구를 수련생에게 제공하는 것]입니다. 이러한 실용적인 접근 방식이 이상적이지 않다는 주장이 있을 수 있지만, 기존의 제약 조건을 고려할 때 최선의 선택이 될 수 있는 실행 가능한 절충안입니다.
Although medical schools should seek to select students with the “right” character and attitudes, those qualities are difficult to assess accurately. Further, effecting character change in the limited time available for medical ethics and professionalism education seems challenging at best. The practical challenges of shaping and evaluating character traits logically lead to the alternative: cultivating behavior that exemplifies ethical and professional virtues. The foundation of this approach is to provide trainees with conceptual tools for seeing, preventing, analyzing, and resolving the ethical dilemmas encountered in clinical medicine. Although an argument can be made that this pragmatic approach is not ideal, it is a workable compromise that may be the best available option given existing constraints.

[행동 목표]에 대한 이러한 초점은 학습자가 교육 중에 달성해야 하는 행동 기반 표준 및 역량을 확립한 미국 의과대학 및 레지던트 프로그램의 주요 인증 기관에 의해 뒷받침됩니다. 예를 들어, LCME 표준 ED-23에는 다음과 같이 명시되어 있습니다: "의학교육 프로그램에는 의료 윤리 및 인간 가치에 대한 교육이 포함되어야 하며, 의대생이 환자를 돌보고 환자의 가족 및 환자 치료에 관여하는 다른 사람들과의 관계에서 세심한 윤리적 원칙을 보여줄 것을 요구해야 합니다."16 LCME는 [학생들의 행동]이 허용되는 윤리 지침에 부합하는지 확인하기 위해 관찰하고 평가해야 한다고 명시하고 있습니다.
This focus on behavioral goals is supported by the major accrediting bodies for U.S. medical schools and residency programs, which have established behavior-based standards and competencies that learners must achieve during training. For example, LCME standard ED-23 states: “A medical education program must include instruction in medical ethics and human values and require its medical students to exhibit scrupulous ethical principles in caring for patients and in relating to patients’ families and to others involved in patient care.”16 The LCME specifies that students’ behavior must be observed and assessed to ensure that it is in line with accepted ethical guidelines.

마찬가지로, 미국의학전문대학원교육인증위원회(ACGME)는 6가지 핵심 역량을 정의하고17 각 역량에 대해 미국 레지던트 프로그램을 이수하는 의사가 [입증해야 하는 행동에 대한 기준을 설정하는 이정표]를 개발할 것을 촉구했습니다. 6가지 핵심 역량 중 하나는 특히 전문직업성에 초점을 맞추고 있는데, "레지던트는 전문적 책임을 수행하고 윤리적 원칙을 준수하겠다는 의지를 보여야 합니다."라고 명시되어 있습니다. 레지던트는

  • 타인에 대한 연민과 존중을 보여주고,
  • 환자의 필요를 자신의 필요보다 우선시하며,
  • 환자의 자율성을 존중하고,
  • 책임감 있게 행동하고,
  • 다양한 배경을 가진 환자에 대한 민감성을 보여야 합니다.

ACGME는 이 핵심 역량을 구성하는 마일스톤을 정의하는 것을 개별 전문 분야에 맡겼습니다. 예를 들어, 미국 내과학회에서 규정한 전문직업성 마일스톤은 표 1에 제시되어 있습니다. ACGME 핵심 역량 6가지 모두 명시적 또는 암묵적으로 전문직업성의 다양한 측면을 포함한다는 점에 유의해야 합니다.
Similarly, the Accreditation Council for Graduate Medical Education (ACGME) has defined six core competencies17 and has called for the development of milestones that establish benchmarks for the behaviors that physicians completing U.S. residency programs must demonstrate for each competency. One of the six core competencies specifically focuses on professionalism, stating, “Residents must demonstrate a commitment to carrying out professional responsibilities and an adherence to ethical principles.” Residents are expected to

  • show compassion and respect for others,
  • put patients’ needs above their own,
  • respect patients’ autonomy,
  • act accountably, and
  • demonstrate sensitivity to patients from diverse backgrounds.

The ACGME has left it to individual specialties to define the milestones that compose this core competency. As an example, the professionalism milestones identified by the American Board of Internal Medicine are presented in Table 1. It should be noted that all six of the ACGME core competencies involve various aspects of professionalism, explicitly or implicitly.

 

의학 학습의 연속성과 관련하여 역량과 이정표에 대한 초점을 GME 이상으로 확장하는 데 관심이 있습니다. 일부 교육자들은 이를 학부 의학교육(UME)에 통합하고 CME 및 인증 유지의 일부로 다룰 것을 제안합니다.18
With respect to the continuum of medical learning, there is interest in extending the focus on competencies and milestones beyond GME. Some educators suggest integrating them into undergraduate medical education (UME) as well as addressing them as part of CME and maintenance of certification.18

또한, 위임가능 전문활동(EPA)을 정의하고 이를 학습자 성과 평가의 근거로 사용함으로써 마일스톤을 실제 임상 실습 사례와 연결하는 데 관심을 기울이고 있습니다.19 이러한 핵심 임상 활동 중 하나를 성공적이고 독립적으로 수행하려면 학습자는 [필수 지식, 태도 및 기술을 입증]해야 할 뿐만 아니라 [역량, 하위 역량 및 마일스톤을 원활하게 통합]해야 합니다. 일부 교육자18는 학습자의 발달 수준에 맞게 EPA를 조정하여 학습 연속체를 더욱 통합하는 데 도움이 될 수 있다고 주장했습니다.
In addition, attention has been directed at linking milestones to instances of actual clinical practice by defining entrustable professional activities (EPAs) and using them as a basis for assessing learner performance.19 To successfully and independently perform one of these core clinical activities, learners must not only demonstrate the requisite knowledge, attitudes, and skills but also seamlessly integrate competencies, subcompetencies, and milestones. Some educators18 have argued for tailoring EPAs to the learner’s developmental level, which could serve to further integrate the learning continuum.

EPA, 이정표 및 역량은 학습자가 [교육이 끝날 때까지 도달할 것으로 예상되는 위치]를 정의하지만, 교육자가 학습자를 [그곳으로 이끌기 위해 사용해야 하는 세부 목표]를 명시하지는 않습니다. 윤리 교육자들 사이에서 의료윤리 교육을 위한 구체적인 목표에 대한 합의는 없지만, [핵심 역량과 주제]에 대한 몇 가지 목록이 제시되었습니다.20-22 [의료윤리 교육을 위한 최소한의 목표]에 대한 현재의 생각을 종합하려는 시도는 목록 1에 제시되어 있습니다. 이러한 목표는 의대생과 레지던트 모두에게 적용되며, 더 높은 수준의 수련생에게는 더 높은 숙련도가 기대됩니다. 이 목록은 경험이 풍부한 교육자 그룹이 공동으로 개발했으며 관련 경험적 연구 및 기타 출판된 문헌을 바탕으로 합니다.4,6,9,20,21,23 이 목록은 우리가 의료윤리 교육의 기본 요건으로 간주하는 것을 나타낸다는 점을 강조하는 것이 중요합니다. 우리는 학습자의 전문직업성을 증진하기 위한 다른 목표(즉, 다른 특정 술기 및 주제를 포함하는 목표)가 이 목록에 추가될 수 있음을 인정합니다.
EPAs, milestones, and competencies define where learners are expected to be by the end of their training, but they do not specify the detailed objectives that educators should use to lead them there. Among ethics educators, there is no consensus on a set of specific objectives for medical ethics education, although several lists of key skills and topics have been put forward.20–22 Our attempt to synthesize current thought on a minimum set of objectives for medical ethics education is presented in List 1. These objectives apply to both medical students and residents, with greater proficiency expected of higher-level trainees. This list was developed collaboratively by our group of experienced educators and draws on relevant empirical studies and other published literature.4,6,9,20,21,23 It is important to emphasize that this list represents what we consider to be the basic requirements for medical ethics education. We acknowledge that other objectives to promote professionalism in learners (i.e., objectives incorporating other specific skills and topics) could be added to this list.


목록 1 의료 윤리 교육에 대한 제안된 목표
List 1 Proposed Objectives for Medical Ethics Education

의과대학 또는 레지던트 수련 프로그램을 마친 학습자는 적절한 수준의 숙련도를 갖추게 됩니다:
Upon completion of medical school or a residency training program, learners will, with an appropriate level of proficiency:

  • 수탁자로서의 의사의 개념과 전문직으로서의 의학의 역사적 발전에 대한 이해를 입증합니다.
  • 환자 치료 과정에서 발생할 수 있는 윤리적 문제를 인식합니다.
  • 전문 협회의 관련 윤리 선언문을 활용하여 임상 윤리적 판단 및 의사 결정을 가이드합니다.
  • 생명윤리 원칙 및 기타 윤리적 분석 도구를 사용하여 윤리적 문제를 비판적이고 체계적으로 사고합니다.
  • 윤리적 문제에 대한 전문적 책임 관리에 대한 합리적인 설명을 제공하고 이러한 판단에 따라 행동합니다.
  • 윤리적 추론을 일관성 있고 정중하게 다른 사람에게 명확하게 설명합니다.
  • Demonstrate an understanding of the concept of the physician as fiduciary and the historical development of medicine as a profession
  • Recognize ethical issues that may arise in the course of patient care
  • Utilize relevant ethics statements from professional associations to guide clinical ethical judgment and decision making
  • Think critically and systematically through ethical problems using bioethical principles and other tools of ethical analysis
  • Provide a reasoned account of professionally responsible management of ethical problems and act in accordance with those judgments
  • Articulate ethical reasoning to others coherently and respectfully

의과대학 또는 레지던트 수련 프로그램을 수료하면 학습자는 적절한 수준의 숙련도를 갖추고 [다음 영역]에서 윤리적 문제를 전문적으로 관리할 수 있습니다:
Upon completion of medical school or a residency training program, learners will, with an appropriate level of proficiency, manage ethical challenges in a professional manner in the following areas:

  • 환자의 프라이버시 및 기밀 보호
  • 의료 오류 및 나쁜 소식 전달을 포함한 환자에 대한 정보 공개
  • 환자의 의사 결정 능력 평가 및 대리 의사 결정과 관련된 문제
  • 환자의 사전 동의 및 사전 정보에 입각한 의료 개입 거부를 포함한 공동 의사 결정
  • 환자의 사전 의료 지시서, 연명 치료 중단 및 철회, 임종 간호, 사망 결정을 포함한 임종 시 간호
  • 생식 기술 및 임신 중절을 포함한 산모-태아 의학
  • 소아 및 신생아 의학
  • 의료 서비스 격차, 의료 시스템, 희소 자원 배분을 포함한 의료 서비스 접근성
  • Protection of patient privacy and confidentiality
  • Disclosure of information to patients, including medical errors and the delivery of bad news
  • Assessment of patient decision-making capacity and issues related to surrogate decision making
  • Shared decision making, including informed consent and informed refusal of medical interventions by patients
  • Care at the end of life, including patient advance directives, withholding and withdrawing life-sustaining interventions, care for the dying, and determination of death
  • Maternal–fetal medicine, including reproductive technologies and termination of pregnancy
  • Pediatric and neonatal medicine
  • Access to health care, including health care disparities, the health care system, and the allocation of scarce resources
  • 문화적 역량과 겸손을 포함한 이문화 간 의사소통
  • 양심권의 범위와 한계를 포함하여 임상 진료에서 의료 전문가의 개인적 가치관의 역할
  • 교육, 임상 진료 및 연구에서의 이해관계 및 의무의 충돌
  • 기관 심의 위원회를 포함한 인간 피험자를 대상으로 한 연구
  • 전문가 간 상호 작용을 포함한 의료팀 내 업무
  • 장애, 무능력, 실수 등 동료에 대한 우려 사항
  • 학생 신분 공개, 교육과 환자를 위한 최선의 치료 사이의 긴장, 숨겨진 커리큘럼, 도덕적 고통 등 의료 수련생 문제
  • 직업적 정체성 및 자기 관리를 포함한 자기 인식
  • 임상의가 어려움에 기여할 수 있는 요인에 대한 인식을 포함하여 까다로운 환자/가족 관리
  • 소셜 미디어
  • 종교 및 영성
  • 감사하는 환자 자선 활동을 포함한 환자의 선물 수락
  • Cross-cultural communication, including cultural competency and humility
  • Role of the health care professional’s personal values in the clinical encounter, including the extent and limits of the right of conscience
  • Conflicts of interest and of obligation in education, clinical practice, and research
  • Research with human subjects, including institutional review boards
  • Work within the medical team, including interprofessional interactions
  • Concerns about colleagues, including impairment, incompetence, and mistakes
  • Medical trainee issues, including disclosure of student status, the tension between education and best care for patients, the hidden curriculum, and moral distress
  • Self-awareness, including professional identity and self-care
  • Management of challenging patients/family members, including recognition of what the clinician may be contributing to the difficulty
  • Social media
  • Religion and spirituality
  • Acceptance of gifts from patients, including grateful patient philanthropy

비교를 위해 1985년 디캠프 보고서1에 제시된 의료 윤리 교육의 목표를 목록 2에 요약했습니다. 이 보고서(목록 1)에서 제안한 목표는 몇 가지 점에서 이전 목표와 다릅니다. 

  • 첫째, 우리의 목표는 [보다 포괄적]이며, 이는 의료 교육에서 윤리와 전문직업성에 대한 강조가 증가하여 이러한 주제에 더 많은 커리큘럼 시간이 할애될 것이라는 기대를 반영할 수 있습니다. 또한 아직 발전 중인 생명윤리 분야의 범위가 넓어지고 있음을 반영하는 것일 수도 있습니다.
  • 디캠프 보고서와 로마넬 보고서의 두 번째 차이점은 의료가 행해지는 맥락, 특히 [의료 접근성 및 문화적 역량 문제]를 고려한 항목을 포함시켰다는 점입니다. 이러한 항목의 포함은 사회경제적 불평등에 대한 인식 확대, 건강의 사회적 결정 요인 강조, 다양성 존중 및 가치 중시 등 최근의 사회적 추세를 반영한 것입니다.
  • 셋째, 윤리적 고려 사항을 환자-의사 간 관계를 넘어 [전문가 간 상호 작용과 자기 관리]로 확장한 점도 주목할 만합니다. 의료 오류를 예방하고 환자에게 우수한 치료를 제공하는 데 있어 효과적인 팀의 중요한 역할에 대한 이해가 높아지면서 의료팀 내 업무에 대한 항목을 추가한 것으로 설명할 수 있습니다. 자기 관리에 대한 관심은 임상 실무에서 의미 상실과 부적절한 일과 삶의 균형을 경험하는 것이 헌신의 약화, 불만, 소진으로 이어질 수 있으며,24 이는 결국 전문직업성 저하와 연관될 수 있다는 인식의 발전을 반영합니다.25,26
  • 넷째, 디캠프 보고서 목표는 특정 내용 영역에서 습득해야 할 도덕적 추론과 지식을 강조하지만 개발해야 할 특정 기술에 대해서는 덜 관심을 기울이고 있습니다. 로마넬 보고서 목표에 [더 많은 술기 기반 항목을 포함]시킨 것은 인증 기관이 임상 상황에서 학습자의 실제 수행 능력과 해당 이정표 달성을 평가하는 방향으로 나아가고 있음을 반영한 것입니다.

For comparison, we have summarized the objectives for medical ethics education presented in the 1985 DeCamp Report1 in List 2. The objectives proposed in this report (List 1) differ from the earlier objectives in several ways.

  • First, our objectives are more comprehensive, which may reflect an increased emphasis on ethics and professionalism in medical training and therefore an expectation that more curricular time will be devoted to these topics. It may also reflect the broadening scope of the still-developing field of bioethics.
  • A second difference between the objectives offered in the DeCamp and Romanell Reports is our inclusion of items that take into account the context in which medicine is practiced, particularly issues of access to health care and cultural competence. The inclusion of these items mirrors recent social trends—expanding awareness of socioeconomic inequalities, emphasizing the social determinants of health, and increasingly respecting and valuing diversity.
  • Third, our expansion of ethical considerations beyond the patient–physician dyad to interprofessional interaction and self-care should be noted. An improved understanding of the important role of effective teams in preventing medical errors and in offering patients excellent care can explain our addition of an item on working within the medical team. The attention to self-care reflects a developing awareness that experiencing a loss of meaning in clinical practice and inadequate work–life balance can lead to waning commitment, dissatisfaction, and burnout,24 and these in turn can be associated with lapses in professionalism.25,26 
  • Fourth, the DeCamp Report objectives emphasize moral reasoning and knowledge to be acquired in specific content areas, but devote less attention to specific skills to be developed. Our inclusion of more skills-based items in the Romanell Report objectives reflects accrediting agencies’ move toward evaluation of learners’ actual performance in clinical encounters and their achievement of corresponding milestones.

목록 2 디캠프 보고서의 제안된 의료윤리 교육 목표
List 2 The DeCamp Report’s Proposed Objectives of Medical Ethics Education

  • 의료 행위의 도덕적 측면을 식별하는 능력
  • 유효한 동의 또는 유효한 치료 거부를 얻을 수 있는 능력
  • 환자가 부분적으로만 동의하거나 치료를 거부할 수 있는 능력이 있거나 무능력한 경우 어떻게 진행해야 하는지에 대한 지식
  • 환자가 치료를 거부하는 경우 진행 방법에 대한 지식
  • 환자로부터 정보를 보류하는 것이 도덕적으로 정당한지 판단할 수 있는 능력
  • 기밀을 위반하는 것이 도덕적으로 정당한 경우를 판단할 수 있는 능력
  • 말기 환자를 포함하여 예후가 좋지 않은 환자 치료의 도덕적 측면에 대한 지식
  • 포함을 고려하는 추가 영역:
  • The ability to identify the moral aspects of medical practice
  • The ability to obtain a valid consent or a valid refusal of treatment
  • Knowledge of how to proceed if a patient is only partially competent or incompetent to consent or to refuse treatment
  • Knowledge of how to proceed if a patient refuses treatment
  • The ability to decide when it is morally justified to withhold information from a patient
  • The ability to decide when it is morally justified to breach confidentiality
  • Knowledge of the moral aspects of the care of patients with a poor prognosis, including patients who are terminally ill
  • Additional areas considered for inclusion:

컬버 외, 1985.1에 명시된 목표
aObjectives articulated in Culver et al, 1985.1


이러한 학습 목표의 차이 외에도 로마넬 보고서에서는 교육 방법, 평가 전략, 추가적인 도전과 기회 등 디캠프 보고서에서 다루지 않은 몇 가지 영역에 주의를 기울이고 있습니다. 이제 이러한 문제에 주목해 보겠습니다.
In addition to these differences in learning objectives, the Romanell Report devotes attention to several areas not addressed by the DeCamp Report: methods of teaching, assessment strategies, and additional challenges and opportunities. We now turn our attention to these issues.

교육 방법
Teaching Methods

의료 윤리와 전문직업성을 가르치는 데 가장 좋은 단일 교육적 접근 방식은 없습니다. 학습 스타일과 교육 기관의 리소스는 다양하므로 이러한 다양성을 반영하여 교육 방법도 유연하고 다양해야 합니다. 예를 들어, ACGME 전문직업성 하위 역량인 "다양한 환자 집단에 대한 민감성과 반응성"17 을 다루기 위해 교육자는 전통적인 교훈적인 강의를 하거나 임상 사례를 제시하거나 토론과 논쟁을 유도하기 위한 "트리거 테이프"를 보여줄 수 있습니다.27 마찬가지로, 환자의 관점을 제시하여 다양성 문제를 조명하거나28,29 건강과 치유에 관한 다양한 "세계관"의 진화를 다루는 기사를 할당하고 토론할 수도 있습니다.30 또 다른 교육 기법은 학습자에게 윤리 문제를 제기한 사례에 대해 성찰적 내러티브를 쓰도록 하는 것입니다.31,32 가능하면 의료 윤리 및 전문직업성 교육에는 임상 실무에서 필요한 팀 접근법을 강화하기 위해 여러 분야의 교수진 간의 협력이 포함되어야 합니다. 최근 몇 년 동안 전문직업성 교육에 대한 다학제적 기여는 철학, 역사, 문학, 법학, 사회과학의 전통적인 분야를 넘어 즉흥 연극 연습,33 만화 그리기,34 창의적 글쓰기 연습,35 미술 연구와 같은 예술의 응용 방법을 포함하도록 확장되었습니다.36-38
There is no single, best pedagogical approach for teaching medical ethics and professionalism. Learning styles and institutional resources vary, so teaching methods need to be flexible and varied to reflect this diversity. For example, to address the ACGME professionalism subcompetency “sensitivity and responsiveness to a diverse patient population,”17 an educator could deliver a conventional didactic lecture, present clinical cases, or show a “trigger tape” intended to inspire discussion and debate.27 Similarly, articles that illuminate issues of diversity by presenting patient perspectives28,29 or that address the evolution of different “worldviews” on health and healing could be assigned and discussed.30 Another pedagogical technique is to invite learners to write reflective narratives about cases they have been involved in that have raised ethics issues.31,32 Whenever possible, medical ethics and professionalism instruction should involve collaboration among faculty from different disciplines to reinforce the team approach required in clinical practice. In recent years, multidisciplinary contributions to professionalism teaching have expanded beyond the traditional fields of philosophy, history, literature, law, and social sciences to include applied methods from the arts such as improvisational theater exercises,33 comics drawing,34 creative writing practices,35 and fine art study.36–38

교육 이론에 따르면 콘텐츠의 간격두기와 반복은 학습을 향상시킨다고 합니다.39 따라서 의료 윤리 및 전문직업성 커리큘럼은 초기 교육 개입이 후속 노출을 통해 강화되거나 발전하는 등 [종단적]일 때 추론과 행동에 지속적인 변화를 가져올 가능성이 가장 높습니다. 예를 들어, 의과대학 1학년 때 도입한 윤리 사례 분석 방법을 임상 실습에서 강화하여 학생들이 임상 환경에서 직면하는 윤리적 문제를 분석하는 데 이 방법을 적용하도록 요청할 수 있습니다. 또한 학습자 중심의 교육 전략도 고려해야 합니다. 예를 들어, 학습자는 토론할 윤리 문제가 있는 임상 사례를 파악하고 사례 토론을 촉진하는 데 적극적인 역할을 할 수 있습니다.
Educational theory suggests that spacing and repetition of content improve learning.39 A medical ethics and professionalism curriculum is therefore most likely to result in sustained changes in reasoning and behavior when it is longitudinal, such that early educational interventions are reinforced or advanced by subsequent exposures. For example, a method for ethics case analysis introduced in the first year of medical school could be reinforced in clinical clerkships by asking students to apply that method to analyze ethical issues they are encountering in clinical settings. In addition, learner-driven teaching strategies should be considered. For example, learners could identify clinical cases with ethics issues for discussion and take an active role in facilitating case discussions.

윤리 및 전문직업성 교육은 학습자가 지식 습득과 술기 개발에서 우수한 환자 치료가 목표인 [행동 변화]로 나아갈 수 있도록 노력해야 합니다(ACGME 핵심 역량 달성을 통한 방법). 이는 어려운 일이지만, 연극의 언어를 빌리자면 [리허설 없이는 대본이 공연이 되지 않는다]는 의미이다. 학생들이 강의실에서 의학 지식을 습득한 후, 교육자들은 일반적으로 역할극 시나리오(종종 모의 환자 또는 "윤리 OSCE"[객관적 구조화 임상시험]40)를 사용하여 학생들이 실제 환자의 복잡한 상황에 직면하기 전에 의학 지식을 술기로 전환하는 연습(그리고 그 지식을 시연하는 수단으로)을 할 수 있도록 지원합니다.41-43 이 접근법은 윤리 및 전문직업성을 교육하는 데 매우 효과적입니다.44,45
Ethics and professionalism education must strive to move learners from knowledge acquisition and skills development to behavior change in which excellent patient care is the goal (by way of achieving the ACGME core competencies). This is challenging, but—to borrow from the language of theater—script does not become performance without rehearsal. After students gain medical knowledge in the classroom, educators commonly employ role-play scenarios (often with simulated patients or in an “ethics OSCE” [objective structured clinical examination]40) to help students practice translating their medical knowledge into skills (and as a means for demonstrating that knowledge) before they encounter the complexity of actual patients.41–43 This approach is highly effective for teaching ethics and professionalism.44,45

기술의 발전으로 윤리 및 전문직업성을 교육할 수 있는 다양한 옵션이 늘어났습니다. 이제 녹화된 강의46 또는 공식 윤리 과정과 같은 일부 자료를 온라인으로 이용할 수 있습니다.47 교육자들은 또한 자신의 수업을 위한 온라인 콘텐츠를 제작하고 있으며, "플립드 클래스 룸" 접근 방식(학생들이 온라인에서 혼자 강의를 시청하여 수업 시간을 절약하고 자료를 토론 및 적용하는 방식)은 전통적인 대면 강의 방식을 보완하거나 대체할 수 있습니다.48 교육자들은 이러한 혁신에 개방적이며 새로운 기술을 통해 어떤 콘텐츠가 가장 효과적으로 전달되는지 신중하게 평가해야 합니다. 강의를 온라인으로 전환하면 그룹 토론 시간을 늘리고 수업 중 학생들의 비판적 사고와 행동 능력에 집중할 수 있다는 장점이 있습니다. 그러나 혁신적인 교육 기술의 사용은 학습자가 지속적으로 외부 준비에 참여하지 않는 상황(예: 보호 학습 시간이 제한되어 있는 바쁜 레지던트 프로그램)에는 적합하지 않을 수 있습니다. 사용 가능한 교육 방법이 다양하기 때문에 교육자는 자신이 맡은 업무에 가장 적합한 교육 도구를 선택할 수 있지만, 이러한 다양성은 어떤 방법이 가장 효과적인지에 대한 의문을 제기하기도 합니다(향후 연구의 중요한 영역).
Technological advances have increased the variety of options for teaching ethics and professionalism. Some materials are now available online, such as recorded lectures46 or formal ethics courses.47 Educators are also creating online content for their own classes, and the “flipped classroom” approach (where students watch lectures online, on their own, saving class time for discussion and application of the material) may complement (or replace) the traditional approach of in-person lectures.48 Educators should be open to these innovations and carefully evaluate which content is best delivered by new technologies. Advantages of moving lectures online include increasing both time for group discussions and the focus on students’ critical thinking and behavioral skills during class. However, the use of innovative educational technologies may not be suited to situations in which learners do not consistently engage in outside preparation (e.g., busy residency programs with limited protected learning time). The wide range of available teaching methods gives educators opportunities to choose the pedagogical tools that are best suited to the jobs they are asked to do, but this variety also raises questions about which methods are most effective (an important area for future research).

이 보고서에서 전공의와 레지던트 수준에 따라 의료윤리 교육 노력이 어떻게 달라져야 하는지에 대해 자세히 설명하는 것은 불가능하지만, 몇 가지 주요 차이점에 주목할 가치가 있습니다.

  • 첫째, 레지던트에게 제공되는 교육 자료는 일반적으로 의대생을 대상으로 하는 것보다 더 복잡하고 상황에 따라 달라질 수 있으며, 윤리적 이슈는 더 미묘하고 심도 있게 논의될 수 있습니다. 일반적으로 교육자는 모든 교육 세션에 다양한 수준의 학습자가 참여할 수 있다는 점을 인식해야 하며, 따라서 교육자는 함께 일하는 학습자의 범위에 적합한 자료를 제공하기 위해 사례와 교육 요점을 조정해야 합니다.
  • 둘째, 일정과 책임의 차이로 인해 교육자는 의대생과 레지던트에게 윤리와 전문직업성을 가르치기 위해 다양한 접근 방식을 채택해야 합니다. 일반적으로 의과대학 커리큘럼에는 종단형 과정을 포함한 다양한 형식이 포함될 수 있지만, 대면 교육 세션이 산발적이고 불규칙적으로 참석하는 1시간짜리 컨퍼런스 형태로 이루어지는 레지던트 교육 프로그램에서는 공식적인 윤리 및 전문직업성 교육의 기회를 찾는 것이 더 어려울 수 있습니다. 이러한 컨퍼런스 형식은 사례 기반 토론에 도움이 될 수 있지만, 교육자는 이러한 기회를 [일관된 커리큘럼으로 전환]하기 위해 창의력을 발휘해야 합니다.

Although it is not feasible in this report to offer a full account of how medical ethics education efforts should vary between GME and UME levels, it is worth noting some key differences. First, educational materials offered to residents can typically be more complex and contextual than those intended for medical students, and ethical issues can be more nuanced and discussed in greater depth. As a general point, educators must recognize that any teaching session may involve learners at different levels of sophistication; accordingly, educators should tailor cases and teaching points to offer material appropriate to the range of learners with whom they are working. Second, differences in schedules and responsibilities require educators to adopt different approaches for teaching ethics and professionalism to medical students and residents. Whereas a variety of formats, including longitudinal courses, can generally be included in a medical school curriculum, finding opportunities for formal ethics and professionalism instruction can be more challenging in residency training programs where face-to-face educational sessions tend to take the form of sporadic, irregularly attended one-hour conferences. Although this conference format can be conducive to case-based discussions, educators need to be creative in turning these opportunities into a coherent curriculum.

평가 전략
Assessment Strategies

교수자가 윤리와 전문직업성을 가르친다고 해서 자신의 교육 기법이 의도한 목표를 달성한다고 가정해서는 안 됩니다. 오히려 의학교육의 광범위한 추세에 따라 교수들은 자신이 하고 있는 일이 효과가 있음을 입증해야 합니다. 점점 더 의료 윤리 및 전문직업성 교육에 [할당된 커리큘럼 시간의 양]과 이러한 [노력에 대한 재정적 지원]정당화해야 합니다.
Faculty teaching ethics and professionalism cannot just assume that their pedagogical techniques achieve the intended goals. Rather, consistent with a broader trend in medical education, they are expected to demonstrate that what they are doing is working. Increasingly, they must justify the amount of curricular time allotted for medical ethics and professionalism education as well as any financial support they receive for such efforts.

이러한 목적을 위해 의료 윤리 교육이 특정 결과를 개선한다는 증거가 있습니다. 특히 학습자의 인식,49 태도,50 지식,51 자신감,52 의사 결정,53 도덕적 추론이 개선된다는 연구 결과가 있습니다.54 그러나 의료 윤리 교육과 의사의 성과, 그리고 이상적으로는 환자의 결과 사이의 관계를 조사하기 위해서는 보다 강력한 증거 기반이 필요합니다. 인증 기관, 의과대학 학장, 레지던트 프로그램 책임자는 [교육 프로그램이 준비된 임상의를 배출하는 데 효과적인지 평가할 수 있는 평가 도구]를 찾고 있습니다. 또한 의사가 [윤리와 전문직업성에 대해 적절한 교육을 받았다는 (정당한) 확신을 갖는 것]이 환자의 이익에 부합합니다. 
Toward these ends, there is evidence that medical ethics education improves certain outcomes. Specifically, studies have shown an improvement in learner awareness,49 attitudes,50 knowledge,51 confidence,52 decision making,53 and moral reasoning.54 However, a more robust evidence base is required to examine the relationships between medical ethics education, physician performance, and—ideally—patient outcomes. Accrediting bodies, medical school deans, and residency program directors seek assessment tools to evaluate whether educational programs are effective in producing prepared clinicians. Further, it is in patients’ interests to have (justified) confidence that their physicians have been trained adequately in ethics and professionalism.

평가의 출발점은 평가가 가능하고 합리적일 때 [평가를 학습 목표와 연결]하는 것입니다. 이를 위해서는 개별 목표의 성격, 개별 목표를 평가할 수 있는지 여부, 가르치는 자료의 복잡성 등을 신중하게 고려해야 합니다. 평가가 가능하다고 판단되는 경우, 학습 목표를 평가와 연계하는 한 가지 모델로는 구체적이고 측정 가능하며 행동 지향적이고 합리적이며 시간 제한이 있는 목표를 설정하는 SMART 접근법55이 있습니다. 예를 들어, "이 세션이 끝나면 참가자들은 문화 간 커뮤니케이션을 위한 R.E.S.P.E.C.T. 모델의 5가지 구성 요소를 설명할 것입니다."56 이러한 유형의 목표는 앞서 설명한 것처럼 인성 발달에 중점을 두기보다는 행동 기반 교육 목표에 초점을 맞추는 것을 반영합니다.
A starting point for assessment is linking evaluation to learning objectives when doing so is possible and sensible. This requires careful consideration of the nature of individual objectives, whether individual objectives can be evaluated, and the complexity of the material being taught. If assessment is viewed as feasible, one model for linking learning objectives to assessment is the SMART approach55—creating objectives that are specific, measurable, action oriented, reasonable, and time bound. For example, “At the end of this session participants will describe the 5 components of the R.E.S.P.E.C.T. model for cross-cultural communication.”56 Objectives of this type reflect a focus on behavior-based educational goals, as discussed earlier, rather than an emphasis on character development.

윤리 및 전문직업성 학습 목표가 달성되었는지 여부를 판단하기 위해 다양한 평가 전략이 필요할 수 있습니다. 가능한 전략에는 학습자 자기 평가, 학습자 성찰, 학습자 공감, 냉소주의, 태도 변화 평가, 수행 포트폴리오, 전통적인 지식 기반 시험, 임상 평가 연습 사용, 모의 환자를 이용한 OSCE 및 기타 연습 사용, 소그룹 토론 모듈 후 교수진의 서면 피드백, 환자 치료 맥락에서 동료, 교수진, 간호사, 직원, 환자 및 가족으로부터 360도 피드백 등이 포함되지만 이에 국한되지는 않습니다.57 -62 위에서 언급한 바와 같이, 새로운 평가 전략은 실제 임상 활동의 맥락에서 학습자의 성과를 평가하기 위해 정의된 EPA를 사용하는 것입니다. 최근 기사에서는 의료윤리 교육과 관련하여 목표, 방법론, 커리큘럼, 평가를 통합하기 위한 한 기관의 노력을 소개합니다.23 
Varied assessment strategies may be needed to determine whether ethics and professionalism learning objectives have been met. Possible strategies include, but are not limited to,

  • learner self-assessment;
  • learner reflection;
  • evaluation of changes in learner empathy, cynicism, and attitudes;
  • performance portfolios;
  • traditional, knowledge-based exams;
  • use of clinical evaluation exercises;
  • use of OSCEs and other exercises with simulated patients;
  • written feedback from faculty after small-discussion-group modules; and
  • 360-degree feedback from peers, faculty, nurses, staff, patients, and families in the patient care context.57–62 

As noted above, an emerging assessment strategy is using defined EPAs to evaluate learner performance in the context of actual clinical activities. A recent article presents one institution’s efforts related to medical ethics education to integrate goals, methodology, curriculum, and assessment.23

성과 중심 평가에 대한 기대는 의학 커리큘럼의 많은 영역에서 어렵지만, 특히 윤리와 전문직업성에서는 더욱 어렵습니다: 윤리와 전문직업성의 일부 측면은 성과와 관련이 없으며, "행동적"인 측면조차도 측정하기 어려울 수 있습니다.63 예를 들어, 일부 저자는 의료 전문가에게 바람직한 특정 인성 자질(예: 겸손, 연민, 성실, 이타주의)은 기존의 정량적 의미로는 측정할 수 없다고 지적했습니다.64,65 
Although an expectation of performance-oriented assessment is challenging for many areas of the medical curriculum, it is especially challenging for ethics and professionalism: Some aspects of ethics and professionalism are not performance related, and even those aspects that are “behavorial” may be difficult to measure.63 For example, some authors have pointed out that certain qualities of character desirable in any health care professional (e.g., humility, compassion, integrity, altruism) are not measureable in any conventional, quantitative sense.64,65

또한 교육 프로그램 평가자들은 숨겨진 커리큘럼보다는 공식적인 교육과정에 초점을 맞추고,66 부족한 부분보다는 개선점을 찾는 경향이 있습니다. 그러나 의료 사회화 과정에서 전문적 행동의 발현이 감소한다는 상당한 증거가 있습니다.64,67 따라서 평가에는 교육 기관의 [학습 환경 평가]68 및 학습자의 [전문직업적 행동 감소를 예방하기 위한 개입]의 능력을 측정하는 것도 포함되어야 합니다.69 
Further, evaluators of educational programs tend to focus on formal course work rather than the hidden curriculum,66 and to look for improvement rather than lack of erosion. Yet, there is substantial evidence that manifestations of professional behaviors decrease throughout the medical socialization process.64,67 Arguably, evaluation should also include assessing the learning environment of educational institutions68 and measuring the ability of interventions to inoculate learners against diminishment of professional behaviors.69

또한 의료 윤리 및 전문직업성 교육의 주요 목표가 환자 치료 개선이라면, 교육 개입을 환자 치료 결과와 연결하는 방법을 개발해야 합니다. 최근 한 연구에서는 의사의 공감과 혈당 조절 개선 사이의 관계를 문서화하여 이에 대한 예를 제시했습니다.70 
Additionally, if the primary goal of medical ethics and professionalism education is improved patient care, we need to develop methods of connecting educational interventions to patient outcomes. One recent study provides an example of this by documenting a relationship between physician empathy and improved glucose control.70

평가가 [공식적으로 가르친 내용]과 [정량적으로 평가할 수 있는 내용]으로 제한되거나, [긍정적인 변화가 요구되는 경우], 전문직업성의 가장 중요한 자질 중 일부를 뉘앙스를 포착하지 못한 방식으로 평가할 위험이 있습니다. 이러한 점을 감안하여 일부 저자는 이러한 자질의 존재 여부와 학습자에게 이러한 자질을 함양하려는 교육자의 노력의 성공 여부를 평가하기 위한 대안 전략을 주장하기도 했습니다.71,72 분명히 평가 대상과 평가에 사용되는 전략이 잘 맞아야 합니다. 정량적 평가가 전문직업성의 우수성을 평가하는 유일한 수단이 되어서는 안 되며, 질적 평가로 보완되어야 합니다. 이러한 결합된 접근 방식은 보다 풍부하고 맥락에 맞는 평가를 가능하게 하지만, 이러한 평가를 수행할 수 있는 관찰력, 지각력, 분석력을 갖춘 평가자를 식별해야 하는 과제를 안고 있습니다. 
If assessment is limited to what is formally taught and to what can be quantitatively assessed, or there is a requirement of positive change, we risk evaluating some of the most important qualities of professionalism in ways that fail to capture their nuances. Given this, some authors have argued for alternative strategies to assess the presence of such qualities and the corresponding success of educators’ efforts to cultivate them in learners.71,72 Clearly, there needs to be a good fit between what is being assessed and the strategies used to assess it. Quantitative ratings should not be the sole means to evaluate excellence in professionalism; rather, they should be complemented by qualitative assessments. This combined approach will enable richer, contextualized evaluations, but it also presents the challenge of identifying evaluators with the observational, perceptual, and analytical capabilities to conduct these assessments.

윤리 및 전문직업성 교육 평가 시에도 [지연 현상]을 고려해야 합니다: 관심 있는 결과가 수년 동안 나타나지 않을 수 있습니다. 의료 윤리 교육의 목표 중 하나는 학습자가 어려운 윤리적 문제가 발생했을 때 이를 해결할 수 있도록 준비하는 것이지만, 학습자는 강의실에서 특정 윤리 문제에 대해 배운 후 몇 년이 지나야 이를 접할 수 있습니다. 그러나 학습자의 나중에 수행 능력은 멀리 떨어진 독서나 수업 중 토론을 회상하는 것에 의해 크게 영향을 받을 수 있습니다. 이 시나리오는 평가에 어려움을 초래합니다. 따라서 전문직업성 및 윤리 교육자는 보다 즉각적인 결과에 대한 평가를 보완하기 위해 장기적인 평가 및/또는 연구 전략을 개발해야 합니다. 목록 3에서는 의료윤리 교육 평가와 관련하여 평가할 항목과 "해야 할 일" 목록(즉, 수행해야 할 작업)을 제안합니다.
The phenomenon of latency also must be considered in the assessment of ethics and professionalism instruction: Outcomes of interest may not manifest themselves for years. One goal of medical ethics education is to prepare learners to address difficult ethical issues when they arise, yet learners may not encounter a particular ethics problem until years after they were taught about it in the classroom. However, their later performance may be profoundly affected by recollecting a distant reading or in-class discussion. This scenario creates challenges for evaluation. Accordingly, professionalism and ethics educators should develop long-term evaluation and/or research strategies to supplement the assessment of more immediate outcomes. In List 3, we propose items to assess and a “to-do” list (i.e., work to be done) with respect to assessment in medical ethics education.


목록 3 의료 윤리 교육에서의 평가: 의료 학습자를 평가할 항목 및 '해야 할 일' 목록
List 3 Assessment in Medical Ethics Education: Items to Assess in Medical Learners and a “To-Do” List

평가할 항목
Items to assess

  • 기본적인 의료 윤리 콘텐츠의 숙달 여부
  • 윤리적 분석 및 추론/주장을 위한 지적 기술의 숙달 정도
  • 핵심 생명윤리 행동 기술 수행: 의미 있는 사전 동의 또는 사전 동의 거부 얻기, 의사결정 능력 평가하기, 나쁜 소식 전하기, 윤리 문제가 있는 사례 분석하기, 환자와 공동의 의사결정 접근 방식 사용하기
  • Mastery of a basic body of medical ethics content
  • Mastery of the intellectual skills for ethical analysis and reasoning/argument
  • Performance in core bioethics behavioral skills: obtaining meaningful informed consent or informed refusal, assessing decision-making capacity, breaking bad news, analyzing a case with ethics issues, and using a shared decision-making approach with patients

평가 '할 일' 목록
Assessment “to-do” list

  • 임상 동료와 협력하여 패스포트 및 기타 학습자 자가 평가 도구의 의료 윤리 구성 요소와 교수진이 임상 로테이션에서 의대생 및 레지던트의 학습을 평가하는 데 사용할 수 있는 도구를 개발합니다.
  • 임상 동료와 협력하여 의대생 및 레지던트 종합 평가 도구의 의료 윤리 구성 요소에 대해 작업한다.
  • 의학교육 전문가인 동료들과 협력하여 의료윤리 커리큘럼 설계 및 평가가 성인 학습자의 다양한 학습 스타일을 고려하도록 보장합니다.
  • 의료윤리 교육과 의사의 성과 및 환자 결과 사이의 관계를 다루는 평가 전략을 개발합니다.
  • 평가 대상과 평가에 사용되는 전략 간의 "적합성"을 보장하기 위해 양적 및 질적 평가 전략을 다양하게 활용한다.
  • 보다 즉각적인 결과에 대한 평가를 보완하기 위한 장기 평가/연구 전략 개발(지연 현상 해결)
  • Work with clinical colleagues to develop medical ethics components of passports and other learner self-assessment tools, as well as tools for faculty to use in assessing medical students’ and residents’ learning on clinical rotations
  • Work with clinical colleagues on medical ethics components of tools for summative assessment of medical students and residents
  • Work with colleagues who are specialists in medical education to ensure that medical ethics curricular design and assessment take into account variation in learning styles of adult learners
  • Develop assessment strategies that address the relationships between medical ethics education and physician performance and patient outcomes
  • Utilize a range of assessment strategies, both quantitative and qualitative, to ensure a “goodness of fit” between what is being assessed and the strategies used to assess it
  • Develop long-term evaluation/research strategies to supplement assessment of more immediate outcomes (to address the phenomenon of latency)

추가적인 과제와 기회
Additional Challenges and Opportunities

목표와 목적, 교수법, 평가 전략과 관련하여 이미 언급한 과제 외에도 의료 윤리 교육에는 기회와 함께 추가적인 도전과제가 존재합니다. 첫째, 윤리 및 전문직업성 교육은 의료 시스템과 의료 행위라는 더 큰 맥락에서 이루어집니다. 의사와 환자의 만남에 영향을 미치는 수많은 외부 요인은 학습 환경에 부정적인 영향을 미치며, 따라서 의학교육의 기반을 약화시킬 가능성이 있습니다. 학습자가 강의실에서 배운 내용이 임상 환경에서 지켜지지 않는다면 어려운 선택을 하게 됩니다.73,74 [교육기관의 학습 환경]은 학습자의 도덕적 침식, 소진 및 손상을 악화시킬 수도 있고, 학습자의 웰빙을 우선시하는 문화를 조성하여 학습자를 지원할 수도 있습니다.3 이러한 도전에 대응하여 의료 윤리 및 전문직업성 교육자는

  • (1) 학습자가 수신할 수 있는 혼합된 메시지를 조정하는 데 도움이 되는 도구를 제공하고,
  • (2) 학습 환경을 측정, 모니터링 및 개선해야 합니다.68

Beyond the challenges we have already noted related to goals and objectives, teaching methods, and assessment strategies, additional challenges—as well as opportunities—exist in medical ethics education. First, training in ethics and professionalism exists within the larger context of the health care system and medical practice. Numerous external factors affecting doctor–patient encounters have negative influences on the learning environment and, thus, have the potential to undermine the foundation of medical education. When learners do not see what is taught in the classroom being honored in the clinical setting, they have difficult choices to make.73,74 An institution’s learning environment can either exacerbate moral erosion, burnout, and impairment among learners, or it can support learners by creating a culture that prioritizes learner well-being.3 In response to this challenge, medical ethics and professionalism educators need to

  • (1) provide learners with tools that can help them reconcile the mixed messages they may be receiving, and
  • (2) measure, monitor, and improve their learning environments.68

둘째, 의과대학 또는 레지던트 수련 프로그램의 전체 커리큘럼에서 의료 윤리 및 전문직업성 교육을 어디에 어떻게 배치할 것인가는 중요하고 논쟁의 여지가 있는 문제입니다. 이 교육의 시기와 구조, 그리고 교육을 제공하는 데 필요한 전문직업성 수준을 신중하게 고려해야 합니다. 
Second, where and how to locate medical ethics and professionalism education in the overall curriculum of a medical school or residency training program is an important—and contested—issue. Careful consideration should be given to the timing and structure of this instruction and the level of expertise needed to deliver it.

최근 일부 의과대학에서는 윤리 교육에 대한 접근 방식의 철학적 변화를 반영하는 커리큘럼 개정에 착수했습니다.75 의료 윤리와 전문직업성을 개별 과목으로 제공하는 대신, 전체 커리큘럼에 걸쳐 발달에 적합한 방식으로 교육을 구성하고 있습니다. 이러한 [통합적 접근 방식]의 정당성은 윤리가 병상에서의 임상적 의사 결정과 임상 조사부터 의료 전달 시스템 차원의 정책적 고려에 이르기까지 모든 의료와 밀접한 관련이 있으며, 언제 어디서나 관련성이 있는 경우 커리큘럼에 통합되어야 한다는 점입니다.  
Some medical schools have recently undertaken curricular revisions that reflect a philosophical change in approach to ethics education.75 Instead of offering medical ethics and professionalism as a discrete course, instruction is woven throughout the entire curriculum in a developmentally appropriate way. The justification for this integrated approach is that ethics is germane to all of medicine—from clinical decision making at the bedside and clinical investigations to policy considerations at the health care delivery system level—and should be incorporated into the curriculum wherever and whenever it is relevant.

통합 접근법의 장단점에 대해서는 상당한 논쟁이 있습니다. 학생들이 의료 윤리라는 분야를 식별하고 관련 문헌에 익숙해져야 하므로 윤리 및 전문직업성 교육이 커리큘럼에 보이지 않을 정도로 통합되지 않는 것이 중요합니다. 우리가 보기에 가장 좋은 방법은 윤리와 전문직업성 교육을 강조하는 것과 이를 임상 교육에 원활하게 통합하는 것 사이에서 건강한 균형을 찾는 것입니다. 그러나 이러한 내용을 적절히 통합하려면 해당 내용을 포함하지 않을 수 있는 다른 코스 디렉터와의 조율이 필요합니다. 다른 과목의 시험에 윤리 관련 내용이 포함되도록 보장하고, 매년 학생들의 윤리 및 전문직업성 숙달 정도를 종합적으로 판단할 수 있는 메커니즘을 마련해야 합니다. 또한 윤리 및 전문직업성 교육이 비윤리 교수진이 담당하는 교과목 및 학사 과정에 포함될 경우, 누가 이 자료를 가르칠 책임이 있는지, 어떤 수준의 전문직업성이 필요한지, 바쁜 일정 속에서 이 교육을 위해 얼마나 많은 시간을 할애해야 하는지 등에 대한 의문이 제기됩니다.
There is considerable debate about the benefits and disadvantages of integrated approaches. It is important that ethics and professionalism education not be integrated into the curriculum to the point of being invisible, because students need to be able to identify the discipline of medical ethics and be familiar with its literature. In our view, the best practice may be to seek a healthy balance between emphasizing ethics and professionalism instruction and seamlessly integrating it into clinical education. However, appropriate incorporation of this content requires coordination with other course directors who may not be committed to its inclusion. Mechanisms must be put in place to ensure the inclusion of ethics material on other courses’ exams and to enable formative and summative determinations of students’ mastery of ethics and professionalism on an annual basis. Further, when ethics and professionalism teaching is woven into courses and clerkships directed by non-ethics faculty, there are questions about who will be responsible for teaching this material, what level of expertise is needed, and how much time should be set aside for this teaching (in the context of busy schedules).

셋째, 의료 학습자의 교육 및 평가에는 교수진의 고려 사항이 크게 작용합니다. 성공적인 의료 윤리 및 전문직업성 교육을 위해서는 적절한 교육을 받은 충분한 수의 교수진이 역할 모델 역할을 하고, 자신의 경험을 공유하며, 학습자를 가르치고, 관찰하고, 피드백을 제공하고, 궁극적으로 평가할 수 있도록 학습자와 의미 있고 지속적인 관계를 구축하는 데 전념할 수 있는 충분한 수의 교수진이 필요합니다. 성공을 달성하려면 교수진 교육자에게 재정적 지원, 인정, 보상이 필요합니다. 비의사 교수진 교육자(즉, 박사 및 법학박사 학위를 소지한 교육자)는 임상 수익을 창출하지 않는 반면, 임상의사 교육자는 교육이 아닌 환자 진료를 통해 수익을 창출하는 경향이 있기 때문에 재정적 제약이 있는 시대에는 특히 더 어려운 과제입니다. 일부 의과대학 환경에서는 교육이 자발적이고 보수가 없는 활동이라는 이유로 의학교육 참여가 암묵적으로 평가절하되는데, 이는 가장 헌신적인 교육자를 제외한 모든 교육자에게 실망스러운 메시지입니다. 교육자에게 보수를 지급하고 그들의 노력에 대해 학문적으로 보상하는 문제가 해결되지 않는 한, 의료 윤리 및 전문직업성 교육의 질은 떨어질 가능성이 높습니다. 
Third, faculty considerations factor significantly into the teaching and evaluation of medical learners. Successful medical ethics and professionalism education efforts require a sufficient number of faculty with appropriate training who are committed to establishing meaningful, ongoing relationships with learners to act as role models, share their own experiences, and teach, observe, give feedback to, and ultimately evaluate learners. Achieving success requires financial support, recognition, and reward for faculty educators. This is particularly challenging in an era of fiscal constraint because nonphysician faculty educators (i.e., those with PhDs and JDs) do not generate clinical revenue, whereas clinician educators tend to generate revenue by seeing patients, not by teaching. In some medical school settings, participation in medical education is implicitly devalued by the fact that teaching is a voluntary, nonremunerated activity—a discouraging message for all but the most committed educators. Until the issue of how to pay educators and reward them academically for their efforts is resolved, the quality of medical ethics and professionalism education efforts is likely to suffer.

마지막으로, 교수진은 평가에 대한 기대치를 다룰 때 고려해야 할 사항이 있습니다. 교육기관이 학습자 행동에 대한 방어 가능한 정량적 평가를 위해 노력하는 경우, 신뢰성을 확보하기 위해 충분한 수의 관찰을 수행할 수 있는 자격을 갖춘 교수진 관찰자가 충분한지 확인해야 합니다.60 마찬가지로, 원하는 결과 및 자질에 대한 평가에 질적 접근이 필요한 경우 교수진 평가자는 학습자를 진정으로 이해하고 "보기" 위해 듣고 관찰하고 "읽어내기"에 숙련되어 있어야 합니다.76 
Finally, faculty considerations are relevant when addressing expectations for assessment. If institutions strive for defensible quantitative evaluations of learner behavior, they need to ensure that there are enough qualified faculty observers to make a sufficient number of observations to achieve reliability.60 Similarly, if assessment of some desired outcomes and qualities requires a qualitative approach, then faculty evaluators must be skilled at listening, observing, and “reading” learners to truly understand and “see” them.76

앞으로 나아가기: 다음 주요 단계 및 고려 사항
Moving Forward: Key Next Steps and Considerations

우리는 의료윤리 교육 현황에 대한 이 보고서가 낙관적인 전망을 제시한다고 믿습니다. 디캠프 보고서가 발표된 이후 30년 동안1 의료 윤리는 의과대학 커리큘럼의 핵심 요소로 자리 잡았습니다. 또한 ACGME의 핵심 역량, 특히 전문직업성 역량에서 윤리를 강조하는 것은 의료 윤리 교육이 레지던트 교육에서도 중요한 구성 요소임을 나타냅니다. 
We believe that this report on the state of medical ethics education offers cause for optimism. In the three decades since publication of the DeCamp Report,1 medical ethics has become a core component of the medical school curriculum. Further, the emphasis on ethics in the ACGME’s core competencies—especially the professionalism competency—indicates that medical ethics education is a valued component of residency training as well.

그러나 이 보고서는 또한 의료 윤리 교육자들이 직면한 많은 과제를 확인했습니다.

  • 첫째, 의료 윤리 및 전문직업성에 대한 구체적인 교육 목표에 대한 합의가 이루어지지 않고 있습니다.
  • 둘째, 몇 가지 교육적 방법이 학습자에게 어느 정도 도움이 되는 것으로 나타났지만, 이를 뒷받침하는 데이터는 거의 없으며 교육 접근 방식은 프로그램과 기관에 따라 크게 다릅니다.
  • 셋째, 이러한 교육적 노력이 항상 단기간에 정량적으로 측정 가능한 개선을 가져오는 것은 아니기 때문에 효과를 입증해야 한다는 압박이 증가함에 따라 의료 윤리 및 전문직업성을 가르치는 교수진에게 특히 어려움이 가중되고 있습니다.
  • 마지막으로, "숨겨진 커리큘럼"66은 학습자의 전문직업성 개발을 저해할 수 있으며, 학습 환경에 대한 관심과 상당한 자원과 전문직업성을 필요로 하는 광범위한 교수진 개발의 필요성을 야기할 수 있습니다.

However, our report also identifies many challenges facing medical ethics educators.

  • First, there is no consensus about specific educational objectives for medical ethics and professionalism.
  • Second, several pedagogical methods have been shown to offer some benefit to learners, but the supporting data are rarely robust, and educational approaches vary greatly between programs and institutions.
  • Third, increasing pressure to demonstrate effectiveness raises particular challenges for faculty teaching medical ethics and professionalism because these educational efforts do not always produce short-term, quantitatively measurable improvements.
  • Finally, the “hidden curriculum”66 can undermine learners’ professional development, creating a need for attention to the learning environment and for widespread faculty development that would require significant resources and expertise.

이러한 문제를 해결하려면 엄격하고 체계적이며 학제 간 접근 방식이 필요합니다. 이는 어려운 과제이지만, 우리는 경험적(질적 및 양적 방법 포함) 및 개념적 학문을 위한 포괄적인 의제를 향한 첫 단계로 다음과 같은 연구 질문을 제안합니다:
Addressing these challenges requires a rigorous, systematic, and interdisciplinary approach. Although this is a daunting task, we propose the following research questions as first steps toward a comprehensive agenda for scholarship, both empirical (including qualitative and quantitative methods) and conceptual:

  • 의료 윤리 교육이 전문가 양성을 지원하는 데 구체적으로 어떤 역할을 하나요? 이 질문에 답하는 연구는 전문직업성과 관련된 의료 윤리 교육 노력에 초점을 맞추고 잠재적으로 재정적 지원의 근거를 제공하는 데 도움이 될 것입니다.
  • What specific role does medical ethics education play in supporting professional formation? Research that answers this question would help focus medical ethics education efforts as they relate to professionalism and potentially provide a rationale for financial support.
  • 의료 윤리 교육의 구체적인 교육 목표에 대한 합의된 목록은 어떻게 구성되나요? 이러한 목록을 수립하고 보급하는 연구는 모든 학습자가 합의된 기본 수준의 의료윤리 교육을 받을 수 있도록 하는 데 도움이 될 것입니다.
  • What constitutes a consensus list of specific educational objectives for medical ethics education? Research that establishes and leads to the dissemination of such a list would help ensure that all learners receive an agreed-upon basic level of medical ethics education.
  • 의료 윤리 교육에 사용되는 교육적 접근법의 강점과 약점은 무엇이며, 어떤 접근법이 학습자의 더 나은 성과와 관련이 있나요? 이 문제를 다루는 연구는 교육자가 여러 가지 가능한 교육 전략 중에서 정보에 입각한 선택을 하는 데 도움이 될 것입니다.
  • What are the strengths and weaknesses of pedagogical approaches used in medical ethics education, and which are associated with better learner outcomes? Research that addresses this issue would help educators make informed choices from a long list of possible teaching strategies.
  • 의료 윤리 및 전문직업성 교육은 학습자의 성과 및 환자 치료 결과와 어떤 관련이 있나요? 이 질문에 답하는 연구는 교육과 결과를 연결하는 절실히 필요한 증거 기반을 구축하는 데 도움이 될 것입니다. 이러한 근거 기반은 이러한 노력에 대한 재정적 지원을 위한 추가적인 근거를 제공할 수 있습니다.
  • How are medical ethics and professionalism education associated with learner performance and patient outcomes? Research that answers this question would help establish a much-needed evidence base linking education to outcomes. Such an evidence base could, in turn, provide additional rationale for financial support of these efforts.
  • 의대생, 레지던트, 의사 교수진, 개업의를 위한 효과적인 의료 윤리 교육 개입의 증거 기반 포트폴리오를 구성하는 요소는 무엇일까요? 이 문제에 대한 연구를 통해 포트폴리오를 직접 개발할 시간이 없는 교육자에게 유용한 리소스를 만들 수 있습니다.
  • What constitutes an evidence-based portfolio of effective medical ethics educational interventions for medical students, residents, physician faculty, and practicing physicians? Work on this issue could lead to the creation of a helpful resource for educators who do not have time to develop a portfolio themselves.
  • 의료 윤리 교육에서 관심 있는 결과를 측정하는 데 가장 효과적인 평가 도구는 무엇인가요? 어떤 평가 전략이 어떤 학습자 및 환자 결과와 짝을 이루어야 할까요? 이러한 질문에 답하는 연구는 교육자가 관심 있는 결과에 적합하고 효과가 입증된 평가 전략을 선택하는 데 도움이 될 것입니다. 이 분야의 연구는 위에서 언급한 지연 문제를 해결하고 윤리 및 전문직업성의 특정 측면과 관련하여 정량적 측정의 한계를 인식해야 합니다.
  • Which assessment tools are most effective at measuring outcomes of interest in medical ethics education? Which assessment strategies should be paired with which learner and patient outcomes? Research that responds to these questions would help educators select assessment strategies that are appropriate for the outcome of interest and proven to be effective. Work in this area should address the latency challenge noted above and recognize the limitations of quantitative measurement with respect to certain aspects of ethics and professionalism.

또 다른 과제는 의학교육자들이 이러한 문제를 탐구하고 이러한 질문에 대한 답을 찾을 수 있는 기관 간 기회가 거의 없다는 것입니다. 의료 전문직업성 아카데미의 목표 중 하나는 의료 윤리, 인문학, 전문직업성 교육자를 포함한 모든 이해관계자가 함께 모여 이러한 도전적인 문제를 해결할 수 있는 포럼을 제공하는 것입니다.14
Another challenge is that few interinstitutional opportunities exist for medical educators to explore these problems and seek answers to these questions. One goal of the Academy for Professionalism in Health Care is to provide a forum for all stakeholders—including medical ethics, humanities, and professionalism educators—to come together to work on these challenging issues.14

결론적으로, 우리는 의료윤리 커리큘럼이 환자 치료, 교육 및 연구 분야에서 전문직업성을 위해 평생 헌신하기 위한 준비로서 전문가 양성에 초점을 맞춤으로써 개선될 수 있다고 믿습니다. 의학이 환자의 요구를 최우선으로 생각하는 돌봄 전문직으로서의 지위를 유지하기 위해서는 많은 사람들의 노력이 필요할 것입니다.In conclusion, we believe that the medical ethics curriculum can be improved by focusing it on professional formation as preparation for a lifelong commitment to professionalism in patient care, education, and research. It will require the hard work of many to ensure that medicine preserves its status as a caring profession that situates the needs of patients as its top priority.


Acad Med. 2015 Jun;90(6):744-52. doi: 10.1097/ACM.0000000000000715.

The essential role of medical ethics education in achieving professionalism: the Romanell Report

Affiliations collapse

Affiliation

1J.A. Carrese is professor, Division of General Internal Medicine, Department of Medicine, Johns Hopkins University School of Medicine, and core faculty, Johns Hopkins Berman Institute of Bioethics, Johns Hopkins University, Baltimore, Maryland. J. Malek is associate professor, Department of Bioethics and Interdisciplinary Studies, Brody School of Medicine, East Carolina University, Greenville, North Carolina. K. Watson is assistant professor, Medical Humanities and Bioethics Program, Feinberg School of Medicine, Northwestern University, Chicago, Illinois. L.S. Lehmann is associate professor, Center for Bioethics, Brigham and Women's Hospital, and Division of Medical Ethics, Harvard Medical School, Boston, Massachusetts. M.J. Green is professor, Department of Humanities and Department of Medicine, Penn State College of Medicine, Hershey, Pennsylvania. L.B. McCullough is professor and Dalton Tomlin Chair in Medical Ethics and Health Policy, Center for Medical Ethics and Health Policy, Baylor College of Medicine, Houston, Texas. G. Geller is professor, Division of General Internal Medicine, Department of Medicine, Johns Hopkins University School of Medicine, and core faculty, Johns Hopkins Berman Institute of Bioethics, Johns Hopkins University, Baltimore, Maryland. C.H. Braddock III is professor and vice dean for education, David Geffen School of Medicine at UCLA, Los Angeles, California. D.J. Doukas is William Ray Moore Endowed Chair of Family Medicine and Medical Humanism and director, Division of Medical Humanism and Ethics, Department of Family and Geriatric Medicine, University of Louisville School of Medicine, Louisville, Kentucky.

PMID: 25881647

DOI: 10.1097/ACM.0000000000000715

Abstract

This article-the Romanell Report-offers an analysis of the current state of medical ethics education in the United States, focusing in particular on its essential role in cultivating professionalism among medical learners. Education in ethics has become an integral part of medical education and training over the past three decades and has received particular attention in recent years because of the increasing emphasis placed on professional formation by accrediting bodies such as the Liaison Committee on Medical Education and the Accreditation Council for Graduate Medical Education. Yet, despite the development of standards, milestones, and competencies related to professionalism, there is no consensus about the specific goals of medical ethics education, the essential knowledge and skills expected of learners, the best pedagogical methods and processes for implementation, and optimal strategies for assessment. Moreover, the quality, extent, and focus of medical ethics instruction vary, particularly at the graduate medical education level. Although variation in methods of instruction and assessment may be appropriate, ultimately medical ethics education must address the overarching articulated expectations of the major accrediting organizations. With the aim of aiding medical ethics educators in meeting these expectations, the Romanell Report describes current practices in ethics education and offers guidance in several areas: educational goals and objectives, teaching methods, assessment strategies, and other challenges and opportunities (including course structure and faculty development). The report concludes by proposing an agenda for future research.

환자로부터 환자중심성에 대해서 배우기: 실재주의자 문헌고찰 (BEME Guide No. 60) (Med Teach, 2020)
Learning from patients about patient-centredness: A realist review: BEME Guide No. 60
Esther de Groota , Johanna Sch€onrock-Ademab , Dorien Zwarta, Roger Damoiseauxa, Kristin Van den Bogerdc, Agnes Diemersb , Christel Grau Canet-Wittkampfa, Debbie Jaarsmab, Saskia Mola and Katrien Bombekec

소개
Introduction

현대 의학에서 의사는 환자의 상황을 고려하고 환자의 가치, 선호도 및 요구에 귀를 기울여야 합니다. 이러한 환자 중심 접근 방식은 환자 중심이라는 용어가 의미하는 바는 다양하지만 의사의 직업 윤리에서 필수 불가결한 것으로 인식되어 왔습니다(Mead and Bower 2000; Michie 외. 2003; Robinson 외. 2008; Smith 외. 2011). 기존의 개념 정의를 분석한 결과, Scholl 등은 [환자 중심주의의 15가지 차원]을 확인했으며, 이 차원은 원칙, 활동, 조력자의 세 가지 클러스터로 묶을 수 있습니다(2014).

  • [원칙 클러스터]에 속하는 차원의 예로는 임상의의 본질적 특성, 고유한 인격체로서의 환자, 생물심리사회적 관점이 있으며,
  • [활동 클러스터]에 속하는 차원의 예로는 환자의 치료 참여와 정서적 지원이 있으며,
  • [조력자 클러스터]에 속하는 차원의 예로는 임상의-환자 의사소통, 팀워크 및 팀빌딩이 있습니다(Scholl 외. 2014).

최근의 리뷰 업데이트에서는 상담 중 환자 중심주의에 대한 의사소통 기술 개입의 긍정적인 효과를 보여주었습니다(Dwamena 외. 2012). 환자 수준에서의 결과(예: 치료 계획 준수)는 부분적으로 긍정적이었지만 명확하지는 않았습니다. 포함된 연구의 이질성이 컸음에도 불구하고 Dwamena 등(2012)이 내린 결론은 교육적 개입(이제부터는 '개입'이라는 용어를 사용)이 의료진이 새로운 기술을 전수하여 환자 중심 치료를 개선하는 데 도움이 된다는 것이었습니다. 
In contemporary medicine, a doctor should consider the patient’s situation and listen to his or her values, preferences and needs. Such a patient-centred approach has been recognised as indispensable in the work ethics of physicians, even though what is meant with the term patient-centredness varies (Mead and Bower 2000; Michie et al. 2003; Robinson et al. 2008; Smith et al. 2011). After an analysis of existing conceptual definitions, Scholl et al. identified 15 dimensions of patient-centredness, which could be packed together into three clusters: principlesactivities and enablers (2014).

  • Examples of dimensions within the principles cluster are the essential characteristics of the clinician, the patient as a unique person and the bio-psychosocial perspective,
  • examples of dimensions within the activities cluster are patient involvement in care and emotional support, and
  • examples of dimensions within the enablers cluster are clinician-patient communication as well as teamwork and teambuilding (Scholl et al. 2014).

A recent review update showed positive effects of communication skills interventions on patient-centredness during consultations (Dwamena et al. 2012). Outcomes on the patient level (e.g., adherence to treatment plans) were partly positive but less evident. Even though there was a large heterogeneity of the studies included, the conclusion based on Dwamena et al. (2012) was that educational interventions (from now on we will use the term ‘intervention’) help providers to improve patient-centred care through transferring new skills.

[환자 중심주의]의 중요성이 일반적으로 인정되고 있고 환자 중심 접근법을 교육하는 데 초점을 맞춘 연구들이 긍정적인 효과를 보인다는 사실에도 불구하고(Dwamena 외. 2012; Maatouk-Bürmann 외. 2016), 환자 중심주의 학습을 최적화하는 좋은 디자인을 만드는 방법에 관한 지식은 아직 충분하지 않습니다. 예를 들어, 어떤 상황에서 어떤 개입이 환자 중심주의에 기여하는지는 아직 명확하지 않습니다. 이를 위해서는 [학습자가 환자 중심적이 되는 메커니즘]에 대한 이해를 높여 특정 디자인이 특정 상황에서 효과적인 이유를 이해해야 합니다. 이러한 질문에는 무엇이 누구에게, 어떤 상황에서, 왜 효과가 있는지에 초점을 맞춘 현실주의 패러다임에 기반한 개입에 대한 연구가 필요합니다(Pawson and Tilley 2004).
Despite the facts that the importance of patient-centredness is generally acknowledged and that studies focusing on teaching a patient-centred approach show positive effects (Dwamena et al. 2012; Maatouk-Bürmann et al. 2016), there is still insufficient knowledge regarding how to create good designs to optimise learning patient-centredness. For instance, it is not yet clear how interventions contribute to patient-centredness and under which circumstances. To this end, we need to improve our understanding of the mechanisms through which learners become patient-centred to understand why a certain design is effective in a specific context. Such questions call for research into interventions that is grounded in a realist paradigm, focused on what works for whom, in which situation, and why (Pawson and Tilley 2004).

[환자]를 [교육, 연구 및 치료의 파트너]로 보는 최근의 경향을 고려하여(Bleakley and Bligh 2008; Bell 외. 2009), 이번 리뷰에서는 [환자가 적극적인 역할을 하는 개입]에 초점을 맞췄습니다. 많은 연구에서 실제 환자가 의학교육에서 교육에 보다 적극적인 역할을 맡는 것이 가치 있다고 설명하는데, 이는 [환자가 자신의 질병과 함께 살아가는 전문가]로 간주되기 때문입니다(Wykurz and Kelly 2002; Jha 외. 2009a, 2009b; Towle 외. 2010; Henriksen and Ringsted 2014). 교육자로서의 실제 환자의 적극적인 역할은 그들의 기여가 학습을 더욱 현실적으로 만들었기 때문에 중요하게 평가되었습니다(Bell et al. 2009). Towle 등(2010)은 환자 중심 학습은 교육에 실제 환자의 참여를 요구한다고 말했습니다. 이 현실주의 리뷰에서는 환자가 학습 경험의 필수적인 부분인 다양한 설계의 중재를 평가한 논문을 포함했습니다.
Given recent developments – implying that patients are seen more as a partner in education, research and care (Bleakley and Bligh 2008; Bell et al. 2009) – we focused this review on interventions in which patients play an active role. Many studies describe the value of giving real patients a more active role in teaching within medical education because patients are seen as experts in living with their medical conditions (Wykurz and Kelly 2002; Jha et al. 2009a, 2009b; Towle et al. 2010; Henriksen and Ringsted 2014). An active role of real patients as an educator was valued because their contributions made learning more real (Bell et al. 2009). Towle et al. (2010) stated that learning patient-centredness asks for the involvement of real patients in education. In this realist review, we included articles that evaluated interventions, diverse in design, in which patients were an essential part of the learning experience.

이 검토의 목적은 참가자가 환자 중심이 되도록 돕는 것을 목표로 하는 다양한 개입을 설명하고 이러한 개입에 참여한 참가자가 다양한 맥락에서 어떻게, 즉 어떤 메커니즘을 통해 반응하는지 알아보는 것입니다(Wong 2012; Wong 외. 2016). 이러한 설명, 맥락, 메커니즘 및 (학습) 이론에 대한 광범위한 문헌의 비교를 통해 특정 맥락에서 참가자가 환자 중심성을 개발하는 방법에 대한 프로그램 이론을 생성합니다. 
The purpose of our review is to describe different interventions that aimed to help participants to become patient-centred and to find out how, i.e., through which mechanisms, participants in these interventions respond in different contexts (Wong 2012; Wong et al. 2016). Through a comparison of these descriptions, the contexts, the mechanisms and the broader literature about (learning) theories, a program theory will be generated on how participants develop patient-centredness within specific contexts.

방법
Methods

우리의 검토는 다양한 맥락이 왜 결과로 이어지는 메커니즘을 촉발하는지에 대한 설명을 생성하는 것을 목표로 하는 현실주의적 검토 접근법을 사용하여 수행되었습니다(Pawson and Tilley 2004). 개입이 어떻게 효과로 이어지거나 이어지지 않는지 이해하려면 인과관계를 명확히 하는 것이 필요합니다(Wong 2012). 현실주의적 검토는 연구에서 개입의 존재 여부와 중심성에 따라 CIMO, CMO, CICMO와 같은 다양한 약어로 설명되는 구성을 공식화하는 것을 목표로 합니다(Ellwood 외. 2017; Emmel 외. 2018, 88쪽). 이 약어에서 C는 맥락, I는 개입, M은 메커니즘, O는 결과를 나타냅니다. 우리는 맥락-개입-맥락-기전-결과(CICMO)를 사용하는데, 이 구성은 특정 맥락에서 개입이 이 맥락(C-I-C)에 영향을 미치고, 그 결과 결과(O)를 가져오는 메커니즘(M)을 촉발하는 경로를 식별하는 것을 의미하기 때문입니다(우리의 경우 환자 중심이 됨). CICMO 구성에서 우리는 [부분적인 프로그램 이론]을 개발했습니다. 프로그램 이론은 (교육) 개입의 구성 요소, 맥락 및 결과에 책임이 있는 것으로 추정되는 기본 메커니즘 간의 연관성을 설명합니다(Lacouture 외. 2015). 이러한 개념에 대한 정의는 용어집(보충 부록 1)을 참조하세요. 
Our review was performed using a realist review approach which aims to produce explanations of why different contexts trigger mechanisms that lead to outcomes (Pawson and Tilley 2004). To understand how interventions lead or do not lead to an effect, clarification of causation is necessary (Wong 2012). Realist reviews aim for formulating configurations described with different abbreviations such as CIMO, CMO, CICMO, depending on the presence and the centrality of an intervention in the study (Ellwood et al. 2017; Emmel et al. 2018, p. 88). In these abbreviations, C stands for Context, I = Intervention, M = Mechanism and O = Outcome. We use Context-Intervention-Context-Mechanism-Outcome (CICMO) because this configuration implies the identification of paths through which interventions in a particular context affect this context (C-I-C), which in turn trigger mechanisms (M) that bring about the outcomes (O), in our case becoming patient-centred. From the CICMO configurations, we developed partial program theories. A program theory explains the connections between components in (educational) interventions, contexts, and the underlying mechanisms that are presumed to be responsible for outcomes (Lacouture et al. 2015). For a definition of these concepts, we refer to the glossary (Supplementary Appendix 1).

검색 출처 및 전략
Search sources and strategy

우리는 환자 중심성에 대한 정의와 참여자의 환자 중심성 개발을 목표로 하는 다양한 중재 유형을 파악하는 데 도움이 되는 초기 문헌 검색을 수행했습니다. 의대생의 환자 중심성에 관한 박사 학위 논문을 쓴 저자(KB), 전임상 환자 접촉을 통한 학습에 관한 저자(AD), 네덜란드 의학 커리큘럼 내 환자 패널 중재에 참여한 두 저자(SM, DZ)와 함께 팀 내에서 이러한 문헌 범위 설정의 결과에 대해 논의했습니다(Mol 외. 2019). 이러한 논의 과정에서 주요 논의 사항은 환자 패널로만 검색을 제한할지(수율이 미미할 것으로 예상되는 경우) 아니면 사전에 제한 없이 광범위한 개입을 할지에 대한 것이었습니다. 처음에는 후자로 결정했습니다. 또한 연구팀은 프로그램 이론을 수립하는 데 유용할 것으로 보이는 몇 가지 이론을 확인했습니다. 저자들이 이미 알고 있는 이론에 대해 논의하고 이러한 이론에 대한 문헌을 검색했습니다. 이 검색은 문헌을 찾아보는 비공식적인 방식이었습니다. 
We conducted an initial scope of the literature which helped us to identify which definitions for patient-centredness were used and the different intervention-types described that aim for developing patient-centredness in participants. We discussed the outcomes of this scoping of the literature within the team, in which one author has written her doctoral thesis on patient-centredness in medical students (KB), another author on learning from preclinical patient contacts (AD), and two others (SM, DZ) were involved in a patient panel intervention within a Dutch medical curriculum (Mol et al. 2019). During these discussions, a major discussion point was whether we should limit our search to patient panels only (where we expected the yield to be minimal) or a broad range of interventions without limitation up-front. We initially decided upon the latter. Also, the team identified several theories that seemed useful for formulating a program theory. We discussed theories already known to the authors and searched for literature about these theories. This search was informal, browsing the literature.

처음에는 상황적 학습사회적 학습 이론(Bandura 1977; Lave와 Wenger 1991)과 같이 환자 중심을 목표로 하는 중재에 관련성이 있을 것으로 예상되는 학습에 관한 기존 이론에 대한 지식을 바탕으로 프로그램 이론을 수립했습니다. 그런 다음 환자 중심 학습에 관한 문헌에서 보다 구체적인 메커니즘을 확인했습니다.

  • 첫 번째는 중재를 통해 환자의 의견을 들을 수 있을 때 학습자는 개선이 필요한 부분을 더 잘 인식하게 된다는 것이었습니다(Wykurz와 Kelly 2002).
  • 두 번째는 환자와의 접촉을 경험함으로써 학습자가 환자를 숫자나 질병이 아닌 인간으로 더 많이 보게 된다는 것입니다(Henry-Tillman 외. 2002).

이 두 가지 메커니즘, 즉 더 큰 인식질병의 맥락화를 대략적인 프로그램 이론으로 고려했습니다. 맥락의 중요성은 아직 고려되지 않았기 때문에 대략적인 수준입니다. 방향성 검색 후, 프로그램 이론을 구체화하고 설명할 논문을 찾기 위해 검색 문자열을 구성했습니다(보충 부록 2 참조). 검색에 대한 세부 사항과 그 과정에서 내린 결정은 가독성을 위해 보충 부록 3에 포함되어 있습니다. 
Initially, our program theory was informed by our knowledge of existing theories on learning that we expected to be relevant in interventions aiming for patient-centredness, such as situated learning and social learning theory (Bandura 1977; Lave and Wenger 1991). Then we identified more specific mechanisms from the literature on learning patient-centredness.

  • The first was that when the intervention allows for hearing patients’ opinions, learners will become more aware of what they need to improve (Wykurz and Kelly 2002).
  • The second one was that by experiencing the patient-contact, learners saw patients more as human beings than as numbers or diseases (Henry-Tillman et al. 2002).

These two mechanisms, greater awareness and contextualising disease, were considered for a rough program theory. Rough, since the importance of the context was not considered yet. After our orienting search, we constructed a search string to find papers to refine and explicate our program theory (see Supplementary Appendix 2). The details on our search and the decisions we took along the way are, for the sake of readability, included in Supplementary Appendix 3.

전체 텍스트 심사
Full-text screening

590개의 논문 중 554개의 전체 텍스트를 확보하는 데 성공했습니다. 여러 연구팀원(KB, KvdB, EdG, CW, SM, JSA, AD)이 개별적으로 전체 텍스트를 검토하여 실제 환자에 대한 중재와 관련된 논문인지 여부를 확인했습니다. 이 과정에서 204편의 논문이 포함 및 제외 기준에 부합하지 않아 제외했습니다. 나머지 350편의 전체 논문 중 122편에서 환자가 개입의 일부가 아닌 것으로 밝혀져 실제 환자 개입이 포함된 논문은 228편이었습니다. 
Out of 590 articles, we succeeded to obtain 554 full-text articles. Several research team members (KB, KvdB, EdG, CW, SM, JSA, AD) individually judged the full texts to find out whether the articles concerned interventions with real patients. During this process, we excluded 204 articles because the articles did not meet the in- and exclusion criteria. In 122 of the remaining 350 full-text articles, patients turned out not to be part of the intervention, resulting in 228 articles with real patient interventions.

메커니즘에 대한 전체 텍스트 심사
Full-text screening for the mechanisms

연구팀은 메커니즘의 존재 여부에 대한 논문을 판단하는 방법론적 논문과 전체 텍스트가 메커니즘을 충분히 상세히 기술했는지에 대한 방법론적 논문을 바탕으로 무엇을 메커니즘으로 간주해야 하는지에 대한 상호 견해에 도달했습니다(Astbury and Leeuw 2010). 이 토론이 끝난 후, 모든 팀원들은 개별적으로 동일한 5개의 논문을 읽고 판단한 다음, 설명된 메커니즘의 존재 여부에 대해 동의할 수 있는지 토론했습니다. 
As a research team, we came to a mutual view on what to consider a mechanism, based on methodological papers on how to judge an article on the presence of a mechanism, and whether the full text described a mechanism in sufficient detail (Astbury and Leeuw 2010). After this discussion, all team members individually read and judged the same set of five articles and then discussed these to see if they could agree upon the presence of a described mechanism.

이 연습이 끝나면 나머지 논문을 세트로 나누었습니다. 세 커플은 각각 한 세트의 논문을 개별적으로 읽고 메커니즘의 존재 여부, 즉 저자가 논문 어딘가에서 개입이 어떻게 작용할 것으로 예상했는지 또는 그 결과가 메커니즘으로 볼 수 있는 결과를 수반하는지 여부를 선별했습니다. 개별적으로 논문을 검토한 후, 커플 내 구성원들이 서로의 결과를 비교하고 이견이 있는 부분에 대해 토론한 후, 합의에 이르지 못한 경우 팀에 논문을 전달하여 더 큰 팀에서 독립적으로 검토하고 토론하도록 했습니다. 질적 논문의 경우, 추가 분석을 위해 설명이 충분히 풍부한지도 평가했습니다(부스 2016). 이러한 단계를 거친 후 32개의 논문이 남았습니다.
After this exercise, we divided the remaining articles into sets. Each of three couples individually read a set of articles to screen on the presence of a mechanism; in other words, to find out whether the authors mentioned, somewhere in the article, how they expected their intervention to work or whether the results entailed findings that might be seen as mechanisms. After screening the articles individually, the members within a couple compared their outcomes and discussed disagreements and, if the couples did not reach consensus, passed the articles on to the team to have them independently screened and discussed in the larger team. For the qualitative papers, we also evaluated whether the descriptions were rich enough for further analysis (Booth 2016). After these steps, a set of 32 papers remained.

데이터 추출
Data extraction

Excel을 사용하여 개입의 특성에 대한 개요를 작성했습니다. NVivo 11을 사용하여 메커니즘, 맥락적 요소(이제부터는 '맥락' 또는 단수인 경우 '컨텍스트'라고 함), 개입 구성 요소 및 결과에 대한 최종 분석을 수행했습니다. 최종 세트는 두 명의 연구자(EdG, KB)가 자세히 읽고 분석했습니다. 추가 분석을 위해 충분히 풍부한 것으로 간주되는 연구의 품질은 조안나 브릭스 연구소의 연구 설계 방법에 따라 연구팀의 다른 구성원들이 평가하여 포함된 논문에 대한 정보를 추가했습니다. 
We used Excel to make an overview of the characteristics of the interventions. With NVivo 11, we performed the final analysis of mechanisms, contextual elements (from now on called ‘contexts’ or, when singular, ‘context’), intervention components and outcomes. The final set was read in detail and analysed by two researchers (EdG, KB). The quality of the studies considered as rich enough for further analysis was appraised by the other members of the research team, based on the methods of the Joanna Briggs Institute for the distinct research designs (http://joannabriggs.org/) to add information about the included papers.

추출된 증거의 종합
Synthesis of extracted evidence

아래에서는 맥락, 메커니즘, 개입 구성 요소 및 결과를 식별한 방법과 이러한 요소들이 조정된 프로그램 이론에 어떻게 기여했는지 설명합니다. 
Below, we describe the way in which we identified the contexts, mechanisms, intervention components and outcomes, plus how these contributed to an adjusted program theory.

맥락
Contexts

[맥락]은 개입 외부의 [개인적, 제도적, 인프라적 특성]으로 간주되었습니다. 맥락은 '개입이 실행되는 환경'과 동일하지 않으며, 개입이 효과가 있을 수 있는 특정 견해를 가진 사람들의 그룹이기도 합니다. 우리는 논문에서 저자가 직장 문화와 같은 맥락을 설명한 부분 중 메커니즘과 관련이 있다고 생각되는 부분을 코딩했습니다. 대부분의 경우, 보유된 논문이 맥락에 대한 충분한 정보를 포함하지 않거나 개입의 맥락이 동일하지 않아 동일한 개입을 맥락과 관련하여 비교할 수 없었습니다. 종종 맥락은 논문의 도입부에서 개입을 시작하는 근거로 설명되었지만 개입이 실행된 환경의 일부로 명시적으로 언급되지 않았습니다. 개입이 이러한 맥락에 어떤 영향을 미쳤는지에 대한 증거를 항상 찾을 수는 없었고, 대부분의 논문에서 맥락을 설명하는 방식이 제한적이어서 어려움을 겪었습니다. 결과 또는 토론 섹션에 개입이 시행된 원래의 맥락이 아니라 개입이 시행된 후의 맥락이 어떻게 변했는지에 대해 설명한 논문은 분석에서 더 많은 가중치를 부여했습니다.
Contexts were considered to be individual, institutional and infrastructural characteristics external to the intervention. Context is not the same as ‘the environment in which an intervention is implemented’ but also, a group of people with specific views for whom an intervention might work. We coded fragments in the papers in which authors described contexts, for example, the workplace culture, that we considered as relevant for the mechanisms. In most instances, we could not compare the same interventions with regard to context because the papers retained did not contain enough information about the contexts or because the contexts of the interventions were not equal. Often, the contexts were described in the introduction of a paper as a rationale for starting with the intervention but not mentioned explicitly as part of the environment in which the intervention was implemented. We could not always find evidence for the way in which interventions affected these contexts, and we were hampered by the limited manner in which the contexts were described in most papers. Papers in which the results or discussion sections also described what the contexts looked like after the intervention had taken place, as opposed to the original contexts in which the intervention was implemented, were given additional weight in our analysis.

개입
Intervention

저희는 기준에 부합하는 모든 개입을 검토에 포함할 수 있도록 허용했습니다(부록 3의 표). 따라서 최종적으로 포함된 중재의 세트는 이질적이었습니다. 우리는 광범위한 문헌을 바탕으로 학생, 환자 및 교사의 역할(예: 결과 섹션 참조)과 같이 맥락에 영향을 미칠 것으로 예상되는 중재 내 구성 요소를 확인했습니다(봄베케 외. 2010; 토울 외. 2010). 개입 구성 요소는 참가자가 (행동 또는 다른 태도로) 반응하는 리소스를 제공합니다. 이러한 반응은 학습을 방해하는 맥락적 문제를 해결하거나 학습을 자극하는 맥락을 개선하는 데 도움이 됩니다.
We allowed for any intervention suiting our criteria to be included in our review (Table in Supplementary Appendix 3). Therefore, the final set of interventions included was heterogeneous. We identified components within these interventions that we expected to affect the context, such as the role of the student, the patient and the teacher of the intervention (for an example, see the results section), based on the broader literature (Bombeke et al. 2010; Towle et al. 2010). The intervention component provides a resource to which participants respond (with behaviour or different attitudes). This response, in turn, helps to remediate contextual problems that hinder learning or to enhance contexts that stimulate learning.

메커니즘
Mechanism

우리는 사람들이 이러한 개입에 참여할 때 활성화되고, 그럼으로써 환자 중심적이 되는 성공적인 결과를 가져올 가능성을 높일 것으로 예상되는 메커니즘을 확인했습니다. 이러한 메커니즘은 개인적 차원에서 존재하며, 변화된 맥락에서 비롯된 개별 참가자의 추론과 반응으로 설명할 수 있습니다(Astbury와 Leeuw 2010). 연구자들 간의 논의를 통해 확인된 메커니즘은 [개입의 구성 요소로서 '설계' 또는 '실행'된 것]이 아니라 [결과를 초래하는 과정과 관련된 것]임을 확인했습니다. 이러한 고려사항의 복잡성을 보여주는 예는 '반영'이라고 표시된 메커니즘에 대한 설명에서 볼 수 있으며, 연구자들이 데이터에서 추론한 '과제'로서의 반영이 아니라는 점을 분명히 해야 했습니다(보충 부록 4 참조).
We identified mechanisms expected to be activated when people participate in these interventions and hence increase the chance for a successful outcome, being patient-centredness. These mechanisms existed on a personal level and can be described as the reasoning and responses of individual participants that result from the changed context (Astbury and Leeuw 2010). Through discussions among the researchers, we made sure that mechanisms identified were not intervention components and as such would have been ‘designed’ or ‘implemented’, but rather concerned the processes that cause the outcomes. An illustration of the complexity of these considerations can be seen in the description of the mechanism labelled ‘reflection’; we had to be very clear that reflection as an ‘assignment’ was not what the researchers inferred from the data (see Supplementary Appendix 4).

의도한 결과
Intended outcomes

Scholl의 환자 중심성 차원은 논문의 결과 및 토론 섹션에 코딩되었습니다(Scholl 외. 2014). 포함된 논문은 주로 질적 논문이었기 때문에 많은 경우 결과가 명확하게 정의되고 측정되지 않았습니다. 
The patient-centredness dimensions of Scholl were coded in the results and discussion sections of the papers (Scholl et al. 2014). Because the included papers were primarily qualitative papers, outcomes were, in many instances, not clearly defined and measured.

프로그램 이론 개발 및 조정
Developing and adjusting the program theory

현실주의적 검토의 결과를 바탕으로 다음 단계를 통해 대략적인 프로그램 이론을 개발하고 조정했습니다. 귀납적 코딩을 통해 논문에서 개입 구성 요소, 맥락, 메커니즘 및 결과를 식별했습니다. 다음으로, 32편의 논문에서 이러한 맥락이 충분히 발생했는지 판단하여 맥락의 관련성을 결정했습니다. 이 판단은 특정 컨텍스트가 언급된 논문의 수, 코딩 밀도, 조각의 풍부함을 기준으로 했습니다. 그 후, 어떤 개입 구성 요소가 컨텍스트에 영향을 미쳤는지 여부와 그 방법을 분석하여 구성configurations을 개발했습니다. 이러한 영향은 논문에 설명되어 있거나(연구에서 예상되거나 연구 결과), 연구자로서 NVivo 내 쿼리를 사용하여 연관성을 뒷받침하여 추론할 수 있습니다. 중재 구성요소의 영향을 설명하기 위해, 우리는 중재 구성요소가 맥락이 환자 중심성 학습을 제약하는 상황에 대한 해결책을 제공했는지, 아니면 맥락이 환자 중심성 학습에 도움이 되는 경우에 개선 효과를 제공했는지 해석했습니다. 예를 들어, 일부 개입은 권력이 학습을 제약하는 맥락에서 권력 균형을 바꾸는 것을 명시적으로 목표로 했습니다.
Based on the findings of our realist review, we developed and adjusted our rough program theory through the following steps. Through inductive coding, we identified intervention components, contexts, mechanisms, and outcomes in the papers. Next, we determined the relevance of the contexts through a judgement whether these contexts occurred sufficiently in our 32 papers. This judgement was based on the number of papers in which a specific context was mentioned, the coding density and the richness of the fragments. After this, we developed configurations by analysing whether and which intervention components influenced the context and how. This influence could be described in the paper (anticipated or resulting from the study) or could be inferred by us as researchers, using queries within NVivo to support the linkages. To elucidate the influence of an intervention component, we interpreted whether the intervention component provided a solution for those situations where the context constrained the learning of patient-centredness, or an enhancement when the context was beneficial for learning patient-centredness. For example, some interventions explicitly aimed to change the power balance in a context where power constrained learning.

데이터에서 개입 구성 요소맥락 사이의 연관성을 발견한 경우, 해당 논문에서 어떤 메커니즘이 설명되었는지 고려했습니다. 여기에서도 NVivo 쿼리를 통해 나타난 코딩 밀도와 논문 내 코딩의 동시 발생이 분석을 뒷받침했습니다. 연구자(KB, EdG) 간의 토론과 논문의 일부분을 데이터로 삼아 이러한 메커니즘 중 어떤 메커니즘이 이러한 맥락에서 개입 구성 요소에 대한 반응으로 예상되는지 설정했습니다. 예를 들어, 환자가 교사 역할을 하는 환자의 전체 질병 궤적에 대한 통찰력으로 분류된 실제 학습 상황의 맥락은 학습자에게 정서적 헌신의 반응을 불러일으켰습니다. 마지막으로, NVivo 기능의 지원과 논문을 면밀히 읽고 연구자(KB, JS, EdG)들 간의 토론을 통해 포함된 논문의 결과를 메커니즘과 연결했습니다. 이러한 방식으로 [CICMO 구성configurations]을 수립한 후 전체 팀과 논의하고, 이 구성을 바탕으로 부분 프로그램 이론을 개발하고 시각화하여 무엇이 누구에게, 왜, 어떻게, 어떤 상황에서 효과가 있는지를 알려주는 보다 발전된 전체 프로그램 이론에 기여하고자 했습니다. 이 단계에서는 첫 번째 제목-초록 심사 단계에서 관련성이 있다고 판단했지만 경험적 논문이 아닌 이론적 논문이기 때문에 포함 기준에 부합하지 않는 배경 논문으로 돌아갔습니다. 이러한 논문과 스노우볼 샘플링을 통해 추가로 발견된 논문을 검토하여 부분적인 프로그램 이론을 구체화하고 결과적으로 전체 프로그램 이론을 개선하는 데 도움이 될 수 있는지 확인했습니다(토론 섹션 참조). 
When our data indicated a link between the intervention component and the context, we considered which mechanisms were described in those papers. Here also, the coding density and co-occurrence of coding within papers, as shown through NVivo queries, supported our analysis. In discussions between researchers (KB, EdG) and grounded in fragments from the papers as the data, we established which of these mechanisms were expected to be a response to the intervention component within this context. For example, the context of an authentic learning situation labelled as insight into the whole illness trajectory of the patient, in which patients act as a teacher, evokes a response in learners of emotional commitment. Finally, also with the support of the NVivo functionality, close-reading of the papers and discussion among the researchers (KB, JS, EdG), we linked outcomes in the included papers to mechanisms. In this manner, CICMO configurations were established, which we subsequently discussed with the whole team, and from the configurations, we developed and visualised partial-program theories, intended to contribute to a more advanced whole program theory that informed us about what works for whom, why and how, and under which circumstances. At this stage, we returned to background papers that we identified in the first title-abstract screening phase as relevant but that did not meet our inclusion criteria as they were theoretical rather than empirical papers. We explored those papers (and additional ones found through snowball sampling) to check whether they could be helpful to refine our partial program theories and consequently our whole program theory (see discussion section).

부분적인 프로그램 이론을 보다 간결하게 만들기 위해 메커니즘을 클러스터로 배치하는 것에 대해 논의했습니다. 이 과정에서 귀납적으로 도출된 메커니즘을 중간 범위의 이론과 연결하기 위해 이론에 대한 모든 연구자의 집단적 지식을 사용했습니다. 이러한 이론에 대한 추가 문헌을 비공식적으로 검색하면서 어떤 이론이 가장 적합한지 확인했습니다. 
We discussed the placement of mechanisms into clusters, to make the partial program theories more succinct. For this process, the collective knowledge of all researchers about (learning) theories was used to connect the inductively derived mechanisms to middle-range theories. In searching informally for additional literature on these theories, we established which theories were most appropriate.

결과
Results

검색 전략과 적격성 평가 결과 32편의 논문이 이 검토에 포함되었습니다. 가독성을 위해 PRISMA 체계는 부록 3에 그림 1로 포함되어 있습니다. 부록 5에는 포함된 연구에 대한 개요가 제시되어 있습니다. 대부분의 연구는 북미의 병원 환경에서 수행되었습니다. 대부분의 논문은 레지던트나 전문가가 아닌 학생에 관한 것이었으며, 32개 논문 중 25개 논문이 질적 설계를 사용했습니다. 

The search strategy and the assessment for eligibility resulted in a set of 32 articles that were included in this review. For reasons of readability, the PRISMA scheme is included as Figure 1 in Supplementary Appendix 3. In Supplementary Appendix 5, an overview is presented of the studies included. Most studies were performed in hospital settings in North America. The majority of papers were about students, not residents or professionals, and 25 out of the 32 papers had a qualitative design.

개입
Interventions

개입의 다양한 특성이 확인되었습니다(보충 부록 6 참조). 아래에서는 [부분 프로그램 이론을 위해 살펴본 개입 구성 요소]에 대해 설명합니다. 이러한 개입 구성 요소는 전체 교육 프로그램 개발자가 설계할 수 있는 문제입니다. 예를 들어 학습자에게 기록자의 역할을 부여하거나 교사에게 보호자의 역할을 수행하도록 요청하고 훈련할 수 있습니다. 
Different characteristics of the interventions were identified (see Supplementary Appendix 6). Below, we describe the intervention components that we have explored for our partial program theories. These intervention components are issues that a developer of a whole educational program might design for: learners can be given the role of a chronicler for example, or teachers can be asked and trained to take up the role of a carer.

학습자 역할
Learner role

학습자는 개입에서 다양한 역할을 수행할 수 있습니다. 의료적 역할, 기록자 역할, 교육생 역할, 환자 옹호자 역할 또는 경청자 역할을 할 수 있습니다. 처음 두 가지 역할(의료 및 기록자)은 여러 논문에서 언급되었으므로 이 두 가지 역할을 모두 설명하며 부분적인 프로그램 이론에 포함시켰습니다. 중재 내에서 학습자는 환자의 인생 이야기를 묻고 제시해야 할 때 [기록자 역할]을 맡을 수 있습니다.
Learners may have different roles in the intervention. They can have a medical role, the role of a chronicler, a trainee, a patient advocate, or a listener. The first two roles (medical and chronicler) were present in several papers, and therefore both of these roles will be described and were included in our partial program theories. Within the intervention, a learner may have a chronicler role when he or she is required to ask about and present a patient’s life story.

기록자 역할에 대한 8편의 논문과 달리, 학습자가 [수련생 및 청중으로서의 역할]을 맡은 논문에서는 환자의 질병에 대한 정보를 수집하는 데 중점을 두었습니다. 학습자가 개입 내에서 환자 치료에 대한 실질적인 책임을 부여받고 의료팀의 일원이 될 때 학습자는 [의료 역할](13개 논문에서 볼 수 있음)을 수행합니다.
In contrast to the eight papers about the chronicler role, in the papers with learners’ roles as trainees and listeners, the focus was on gathering information about a patient’s illness(es). The learner has a medical role (seen in 13 papers) when, within the intervention, the learner is given real responsibilities for patient care and becomes part of the healthcare team.

환자 역할
Patient role

많은 논문에서 환자는 명시적인 역할이 없고 학습자가 환자를 돌보는 존재로 등장합니다. 그럼에도 불구하고 환자의 역할은 크게 [정보 제공자]와 [교사]라는 두 가지 범주로 구분할 수 있습니다. 9편의 논문에서 환자는 자율적인 역할을 수행하며, 교사는 학습자와 환자 사이의 중개자 역할을 수행하지 않습니다. 환자는 일상 생활에서 질병에 대한 [정보 제공자]로서의 역할을 수행하며, 학습자는 환자의 삶과 질병이 환자에게 의미하는 바에 대해 질문합니다. 환자는 종종 자신의 가정 환경에서 학습자와 일상 생활에서 질병에 대한 정보 제공자로서 상호 작용합니다.
In many papers, the patient does not have an explicit role, he or she is present, and the learner cares for the patient. Nevertheless, two broad categories of patient roles were identified, namely as an informant and as a teacher. In nine papers, the patient has an autonomous role, without the teacher playing a role as a go-between the learner and the patient. The patient has a role as an informant about the disease in daily life, where the learner asks questions about the patient’s life and what the disease means for the patient. The patient interacts, often in the own home environment, as an informant about the disease in daily life with the learner.

다른 논문(6)에서는 환자가 [교사]로서 더 적극적인 역할을 수행합니다. 교사 역할을 맡은 환자는 때때로 개입을 설계하고 학습 목표를 결정하는 데도 참여했습니다.
In other papers (6), the patient has a more active role as a teacher. Patients with a teacher role were sometimes also involved in designing the intervention and deciding on the learning goals.

교사의 역할
Teacher role

우리는 교사 또는 감독자가 가질 수 있는 다양한 역할, 즉 보호자 또는 평가자의 역할을 확인했습니다. 

  • 7개의 논문에서 확인된 역할 중 하나는 교사가 '모든 답을 줄 전문가'가 아니라 학습자를 돌보고 중증 환자와의 첫 상호작용이 어려울 수 있음을 인식하는 사람이라는 것을 의미합니다. 따라서 학습자 중심의 역할을 수행하며, 이를 우리는 [보호자의 역할]이라고 설명합니다.
  • 7편의 논문에서 교사는 [평가자로서의 역할]을 수행했습니다. 이 논문에서 교사는 학생이 충분한 진전을 이루었는지, '충분히 잘하고 있는지'를 평가하는 것이 주된 책임이라고 생각하는 것으로 나타났는데, 이는 학습자를 지원하는 것이 주된 책임이라고 생각하는 [보호자의 역할]과는 대조적인 모습입니다.

We identified different roles that a teacher or supervisor might have, namely the roles of carer or assessor.

  • One of the roles, identified in seven papers, implies that the teacher is not primarily the ‘expert who will give all the answers’ but rather someone who cares for the learners and recognizes that interactions with severely ill people for the first time might be challenging. As such, he or she takes on a learner-centred role, which we describe as the role of the carer.
  • In seven papers, the teacher had a role as an assessor. The teachers in these papers did seem to feel that their primary responsibility was to assess whether the students made sufficient progress and were ‘good enough’, which is in contrast with the role of a carer where teachers feel their primary responsibility is to support the learners.

기타 개입 구성 요소
Other intervention components

우리는 개입이 자발적 개입이 아닌 의무적 개입인지, 환자와의 연결이 더 관례적인 환경에서 의도적으로 수행되는지 또는 커리큘럼에서 인정된 요소인지와 같은 다양한 개입 구성 요소를 확인했습니다. 이러한 측면은 시간 부족과 관련된 맥락의 관련성으로 인해 관련성이 있는 것으로 간주되었습니다(아래 맥락 아래 참조). 이러한 개입 구성 요소에 대한 명확한 CICMO 구성을 확인할 수 없었으므로 이러한 구성 요소는 더 이상 포함되지 않습니다.
We identified different intervention components such as whether the intervention is mandatory (as opposed to voluntary), performed purposely in a setting where connecting with patients is more customary or a recognised element in the curriculum. These aspects were considered pertinent due to the relevance of a context related to lack of time (see below, under contexts). We could not identify clear CICMO configurations around these intervention components and therefore these components will not be included further.

맥락
Contexts

우리는 환자 중심 학습과 관련된 메커니즘을 이끌어내는 데 중요해 보이는 [다섯 가지 맥락]을 확인했습니다. 첫 번째 맥락은 가치 있는 지식을 제공하는 것으로 간주되는 [합법적인 사람(교사 또는 환자)의 존재] 또는 배울 가치가 있는 것으로 간주되는 [합법적인 지식]으로서 [지식의 존재]였습니다
We identified five contexts that seem important for eliciting mechanisms that are relevant for learning patient-centredness. The first context was the presence of legitimate people (teachers or patients) who are considered as providing worthwhile knowledge or of legitimate knowledge, knowledge considered as being worthwhile to learn.

학생들은 간호에 대해 대체로 생의학적이고 하이테크적인 비전을 가지고 있으며, 이는 임상 분야에서 '높은 지위'로 여겨지는 것에 의해 강화됩니다. 징후, 증상 및 증후군에 대한 이해는 학생들의 목적의식, 성취감 및 중요성에 대한 주요 초점이 됩니다. 이러한 가치관이 강화되면 학생들이 자격을 취득하는 데 있어 지배적인 가치로 남게 됩니다. [E] 
Students hold a largely bio-medical, hi-tech vision of nursing, reinforced by what is seen as being ‘high status’ in clinical areas. Understanding signs, symptoms and syndromes become the main focus for students’ sense of purpose, achievement and significance. If reinforced, this remains the dominant value that students hold as they qualify. [E]

[정당성]에 대한 단편에서 얽혀 있는 이러한 맥락의 뚜렷한 측면은 [학습자와 환자 간의 권력 관계 또는 위계]가 맥락에 존재하는 방식이었습니다. 
A distinct aspect of this context that was intertwined in the fragments about legitimacy was the way power relations or hierarchy between learners and patients are present in the context.

학생들이 [환자 지식의 신뢰성]이나 [환자 주도 교육의 정당성]에 의문을 제기하는 것은 무의식적으로 전통적인 권력 관계를 유지하거나 보호하려는 측면으로 해석할 수 있습니다. [Q] 
Students questioning the credibility of patients’ knowledge or the legitimacy of patient-led teaching can be interpreted in terms of their unconsciously trying to maintain or protect traditional power relations. [Q]

두 번째 맥락은 [비인간적인 영향의 발생]으로, 예를 들어 [환자를 인간이 아닌 물건으로 보는 것]의 부정적인 결과입니다. 많은 논문에서 이러한 영향은 [효율성과 행정적 부담]의 필요성을 언급하거나 숨겨진 커리큘럼을 언급하는 단편적인 내용에서 언급되었습니다. 
The second context was the occurrence of dehumanizing influences; for instance, the negative consequences of seeing patients as objects instead of human beings. In many papers, such influences were referred to in fragments mentioning a need for efficiency and administrative burden or referring to the hidden curriculum.

의학은 기술적이고 경험적인 것에 초점을 맞춰 [환자를 대상화]하고, 기술적인 능력을 가진 사람에게 더 많은 '명성'을 부여하는 것으로 묘사되었습니다: 환자를 [사람이 아닌 의료적인 것]으로 간주하고... '우리는 당신의 심부전을 고치고 싶다'는 메시지를 전달합니다. [AA]] 
Medicine was described as objectifying patients with its focus on the technical and empirical, rewarding those with technical prowess with more ‘‘prestige’’: They are seen as a medical thing and not a human, not a person…the message is: we want to fix your heart failure. [AA]

세 번째 맥락은 개입이 이루어지기 전에 좋은 역할 모델의 가용성과 관련이 있습니다. 논문의 결과 또는 토론 섹션에서 [롤모델을 보고 해석하는 것]이 결과에 도달하는 과정으로 보이는 경우, 우리는 롤모델에 대한 이러한 반응을 맥락이 아닌 메커니즘으로 코딩했습니다. 예를 들어, 다음 인용문에서 롤 모델과 개입의 성공에 대한 관련성을 확인할 수 있습니다: 
The third context pertained to the availability of good role models before the intervention took place. When seeing and interpreting role models was visible in the results or discussion sections of a paper as the process by which the outcome was reached, we coded this response to role models as a mechanism rather than as context. Role models and their relevance for the success of an intervention was, for example, visible in the following quotation:

또한 인터뷰 기법을 시연할 수 있는 능력에 자신감을 가진 강사가 부족했기 때문일 수도 있습니다. 의사들은 필요한 임상 경험을 가지고 있었지만 환자 중심 면담 기법에 익숙하지 않았습니다. [C] 
It may have also been due to a lack of instructors who felt confident in their ability to demonstrate interviewing techniques. Although physicians had the necessary clinical experience, they were rarely familiar with patient-centred interviewing techniques. [C]

네 번째 맥락은 학습 상황의 통합성과 일관성에 관한 것으로, 우리는 이를 '전체 질병 궤적에 대한 통찰력'이라고 명명했으며, 특히 학습자가 [아직 임상 교육을 받지 않은 커리큘럼 기간]과 관련이 있습니다. 이 개입은 학습자가 가상의 사례로 연습만 하는 것이 아니라, 동시에 임상 실습에 들어갈 때 경험하게 될 모든 것에 압도되지 않고 환자 중심주의를 잊어버릴 위험이 없는 [일종의 '중간' 공간]을 만들어냈습니다. 
The fourth context concerned the integrality and coherence of the learning situation, which we labelled as ‘insight in the whole illness trajectory’, especially relevant in those periods in a curriculum where learners did not have clinical education yet. The interventions generated a kind of ‘in-between’ space where learners were not just practising with hypothetical cases, but at the same time were not overwhelmed by all the things they would experience when entering clinical practice with the risk of forgetting about patient-centredness.

어려운 대화를 나누는 방법에 대해 배우는 것은 임상 업무와 동떨어진 것으로 여겨졌습니다: "임상 교육에는 실제로 포함되어 있지 않습니다... 튜토리얼과 가상 사례에서 이야기합니다... 처음 임종 환자를 만나면 말이 없고 무슨 말을 해야 할지 몰라서 대화를 피하게 됩니다. 그래서 그냥 물러서게 되죠." [AA] 
Learning about how to have difficult conversations was seen as disconnected from clinical work: “It’s not really built into clinical education…we talk about it in tutorials and hypothetical cases…when you first have a patient who is dying, you avoid talking about it because you don’t have the words and don’t know what to say. So you just withdraw”. [AA]

환자와 소통할 시간을 갖는 것은 우리가 관련성이 있다고 생각한 다섯 번째 맥락이었습니다. 분석 결과, 억제하는 맥락의 핵심은 환자와 소통할 시간이 없거나 제한적이라는 것을 확인했습니다. 논문에서 이러한 종류의 내러티브는 실제 시간보다는 지각된 시간과 관련이 있었습니다. '시간이 없다'는 내러티브의 한 측면은 바쁜 병동에서 감정과 환자와의 연결이 시간이 많이 걸리고 전문가로서의 자기 보존에 위험하다고 인식된다는 것입니다. 
Having time to connect with patients was the fifth context that we considered relevant. In our analysis, we identified that the core of an inhibiting context was that there was no (or limited) time to connect with patients. In the papers, this kind of narratives related to perceived rather than to actual time. An aspect of this ‘no time’ narrative is that emotions and connecting with patients are perceived as time-consuming and as a risk for your self-preservation as a professional in a busy ward.

... 그리고 환자의 이야기를 들어줄 시간이 없다는 메시지입니다. 또한, 그들은 '감정적'이거나 '친밀감'을 느끼는 것이 실제로 비전문적이거나 심지어 위험할 수 있으며, 특히 '자기 보존'을 위험에 빠뜨릴 수 있다는 것을 배우고 있습니다. [...] 이는 숙련된 의사는 감정을 표현해서는 안 되며, 표현할 시간도 없다는 메시지를 강력하게 전달합니다... [AA]
… and a message that there isn’t time to listen to a person’s story. Further, they are learning that getting ‘emotional’ or ‘close’ may actually be seen as unprofessional or even dangerous, especially when it puts ‘‘self-preservation’’ at risk. […] This strongly articulated message that an experienced physician shouldn’t or doesn’t have time to show emotion… [AA]

메커니즘 식별
Identification of mechanisms

부록 4는 확인된 메커니즘에 대한 개요와 각 메커니즘에 대한 설명을 제공합니다. 메커니즘의 개념을 더 자세히 설명하기 위해 아래 두 가지 예를 설명합니다. 첫 번째 예는 "다른 사람과 정서적으로 관여하고 정서적 동일시 경험"이라는 메커니즘으로, 다음과 같은 단편에서 볼 수 있습니다: 

Supplementary Appendix 4 provides an overview of the mechanisms identified and a description for each mechanism. We describe two examples below to explain the concept of mechanisms further. A first example is a mechanism called “Being emotionally involved with others and experiencing affective identification”, which is seen in fragments such as:

자원봉사자들의 이야기가 주는 [정서적 영향]은 [학생들에게 의사로서 변화하고 발전할 수 있는 동기를 부여]했습니다. 학생들은 자신이 들은 이야기에서 얻은 '교훈'을 환자에 대한 자신의 접근 방식에 어떻게 적용하고 싶은지 명시적이든 암묵적이든 일관되게 표현했습니다. [T]
The emotional impact of the volunteers’ stories gave students motivation to change and develop as physicians. Students consistently expressed—either explicitly or implicitly—how they wished to incorporate ‘‘lessons’’ from the stories they heard into their own approach to patients. [T]

이야기는 개입의 한 측면이었으며, 이러한 이야기의 영향은 특정 맥락에서 참가자의 무의식적 반응(메커니즘)을 이끌어낸다는 것입니다.
The stories were an aspect of the intervention while the impact of these stories is that they elicit an unconscious response from the participants (a mechanism) within certain contexts.

메커니즘의 두 번째 예는 '실제 사람들의 삶의 이야기로 질병을 맥락화하기'입니다:
The second example of a mechanism is ‘Contextualizing disease with the life stories of real people’:

모든 포커스 그룹에서 공통적으로 발견한 것은 인생에서 많은 트라우마와 좌절을 겪으면서도 긍정적인 태도를 유지한 환자들에 대한 존경과 동경의 감정이었습니다. 인생의 역경에 대한 경험이 상대적으로 부족한 어린 학생들에게 인생의 관점이 훨씬 더 긴 사람과의 만남은 강력한 경험이었습니다. '긍정적으로 생각하고 낙관적인 태도를 잊지 말라'고 직접적으로 말씀하시지는 않았지만, '저 분은 에너지가 넘치시네, 나도 저 분 나이가 되면 저렇게 되고 싶다'는 생각을 여러 번 했어요. [N]
A central finding in all focus groups was feelings of respect and admiration for patients, who had often endured many traumas and setbacks during life, yet still maintained a positive attitude. For young students relatively inexperienced in life’s adversities, the meeting with a person who had a much longer life perspective was a powerful experience. ‘Think positively and remember to be optimistic’, my patient did not say it directly, but I thought to myself several times ‘she has lots of energy – I hope I will be like that when I get to her age’. [N]

이러한 메커니즘은 인지적 수준일 수도 있고 더 심리적이고 무의식적인 수준일 수도 있습니다. 프로그램 이론을 개발하면서 우리는 [네 가지 메커니즘 클러스터]를 확인했습니다:
These mechanisms may be on a cognitive level or a more psychological, unconscious, level. In developing the program theory, we identified four clusters of mechanisms:

  1. 비교와 결합, 그리고 관점의 확장. 포함된 논문 중 일부는 주로 다른 관점을 관찰하는 데 중점을 둔 반면, 다른 논문에서는 이전에 알려진 관점을 새로운 관점과 대조하고 연결하는 데 중점을 두었습니다. 이 클러스터에는 변화된 맥락과 개입 요소에 대한 인지적 또는 합리적 반응을 가정하는 모든 메커니즘이 포함되었습니다.
  2. 내러티브 개발 및 환자 참여. 이 클러스터에서 메커니즘은 관계적이며 사회적 학습을 가정합니다. 일부 메커니즘은 의미 형성으로서의 내러티브 개발에 중점을 두는 반면, 다른 메커니즘은 관계를 발전시키거나 변화시키는 데 중점을 둡니다.
  3. 자아실현. 이 군집에는 개별 학습 과정을 통해 유용함이나 영감을 얻는 등 자아의 학습에 초점을 맞춘 메커니즘이 포함됩니다.
  4. 사회화. 이 군집은 사회적 환경에 대한 적응 또는 조정과 관련된 메커니즘으로 구성됩니다.

 

  1. Comparing and combining as well as broadening perspectives. Some of the included papers focused on observing other perspectives primarily, while in other papers the emphasis was on contrasting and connecting previously known perspectives with new ones. In this cluster, we included all mechanisms that presume a cognitive or rational response to the changed context and the intervention component.
  2. Developing narratives and engagement with patients. In this cluster, the mechanisms are relational and assume social learning. Some of the mechanisms have a focus on developing narratives as meaning making while others were about developing or changing relationships.
  3. Self-actualisation. This cluster includes mechanisms that have a focus on the learning of the self, through individual learning processes, such a feeling useful or inspired.
  4. Socialisation. This cluster consists of mechanisms that pertain to adaptation or adjustment to the social environment.

결과
Outcomes

부록 7(A)에 제시된 개요는 포함된 논문에서 환자 중심주의의 어떤 차원(Scholl 외. 2014)이 다루어졌는지 전반적으로 보여줍니다. 요약하면, 대부분의 중재에서 의도한 결과는 '고유한 인격체로서의 환자'와 같은 차원에 관한 것이었고, 소수의 중재에서 의도한 결과는 조정 및 연속성, 팀워크와 같은 조력자 클러스터의 차원에 관한 것이었습니다. 부록 7(B)에는 특정 결과가 기술된 논문에서 가장 자주 추론한 메커니즘에 대한 개요가 제시되어 있습니다. Scholl 등(2014)의 모든 차원이 우리 데이터에 나타난 것은 아닙니다.
The overview shown in Supplementary Appendix 7(A) gives a general impression which dimensions of patient-centredness (Scholl et al. 2014) were covered in the included papers. In sum, in the majority of the interventions, the intended outcomes concerned dimensions such as the patient as a unique person while in the minority of the interventions, the intended outcomes concerned dimensions from the enablers cluster, such as coordination and continuity and teamwork. In Supplementary Appendix 7(B) an overview is presented of the mechanisms that we inferred most frequently from those papers in which a specific outcome was described. Not all dimensions of Scholl et al. (2014) were seen in our data.

부분 프로그램 이론
Partial program theories

CICMO의 개별 요소를 식별한 후, 포함된 코딩된 논문으로 돌아가 CICMO configuration을 개발했습니다(보충 부록 8 참조). 포함된 논문 세트의 컨텍스트와 개입 구성 요소에 대해 [세 가지 부분 프로그램 이론]을 개발했습니다. 이를 다음과 같은 방식으로 시각화(아래 그림 참조)했습니다. 먼저 왼쪽부터 기존 컨텍스트가 표시되고, 데이터에서 발생한 개입 구성 요소와 개입 구성 요소로 인해 변경된 컨텍스트가 그 뒤를 따릅니다. 그런 다음 맥락의 변화를 통해 개입 구성요소에 의해 촉발된 메커니즘이 표시되어 환자 중심성을 학습하는 데 기여합니다. 선의 굵기는 해당 구성이 데이터에 존재하는 정도를 반영합니다.
After identifying the separate elements of the CICMO, we went back to the included coded papers and developed CICMO configurations (see Supplementary Appendix 8). For the contexts and the intervention components in our set of included papers, we developed three partial program theories. These are visualised (see the Figures below) in the following manner: first, starting at the left, the existing context is depicted, followed by the intervention component which occurred in the data and the changed context resulting from the intervention component. Then the mechanisms are shown that were triggered by the intervention component through the change in context, which contributes to learning patient-centredness. The boldness of the lines reflects the degree to which the configurations were present in the data.

부분 프로그램 이론 1: 학습자 역할
Partial program theory 1: learner role

데이터에서 우리는 중재에서 (1) 의료 전문가로서의 학습자와 (2) 기록자로서의 학습자라는 두 가지 학습자 역할을 확인했습니다.  의료 전문가의 역할(그림 2 참조)의 경우, 가장 많이 추론된 메커니즘은 [자신이 하는 일에 유능하다고 느끼는 것, 환영받고 안전하며 안정감을 느끼는 것('자아실현'), 롤모델을 보고 동화되는 것('사회화')] 등이었습니다. 종단적 개입에 관한 논문에서는 [의료진의 역할]이 명시적으로 언급되었지만, 종단적 개입이 아닌 단기간에 여러 번의 만남으로 이루어진 논문에서도 동일한 메커니즘과 결과가 나타나는 경우가 많았습니다. 그림에 표시되지 않은 [기록자의 역할]에서 맥락에 가장 중요한 영향을 미친 것은 [어떤 지식이 가치 있는 것으로 간주되는지, 환자를 객체로 볼 것인지 주체로 볼 것인지]의 변화였습니다. 이는 다음 인용문에서 잘 설명되어 있습니다:
In our data, we identified two learner roles in the interventions: the learner (1) as a medical professional and (2) as a chronicler. Although we also identified other roles in the papers, the accompanying CICMO’s were less clear. For the role of a medical professional (depicted in Figure 2), the mechanisms inferred the most are: feeling competent in doing what you do; feeling welcome, safe, and secure [both in ‘self-actualisation’]; and seeing and assimilating role models [socialisation]. Even though the medical role was mentioned explicitly in papers about longitudinal interventions, often the same mechanisms and outcomes were seen in papers in which the setting was not longitudinal but consisted of several meetings in a short time span. For the role of the chronicler (not depicted in the Figure), the most important influence on the context was the change in what knowledge was considered valuable and whether the patient was seen as an object or a subject. This is illustrated by the following quote:

우리 연구에서 레지던트들은 [환자의 질병에 대한 정보를 수집하는 질문자]에서 [한 사람의 인생 이야기를 듣는 청중]으로 모자를 바꿨습니다. [...] 전통적인 의료 면담은 의사가 '퍼즐'을 푸는 것을 전제로 하며, 본질적으로 환자에게 문제를 제기하고 의사의 목표에 초점을 맞추기 때문에 환자를 대상화할 수 있습니다. [AA]
In our study, the residents switched hats, from being an interrogative gatherer of information about a patient’s illness to being the audience of a person’s life story. […] The traditional medical interview is predicated on the physician’s solving of a ‘puzzle’; it inherently problematizes the patient, and has its focus on the physician’s goals, which can result in an objectification of the patient. [AA]

우리가 가장 많이 추론한 [기록자 역할의 메커니즘]

  • 의미 있는 내러티브에 기여하고 발전시키는 것, 정서적 동일시를 보여주는 것, 다른 사람과 정서적으로 관여하는 것[모두 '내러티브 개발/환자와의 참여'],
  • 다른 사람이 되는 상상[자아실현] 등입니다.

The mechanisms for the chronicler role that we inferred the most are:

  • contributing and developing meaningful narratives, showing affective identification, being emotionally involved with others [all in ‘developing narratives/ engagement with patients’], and imagining
  • to be someone else [self-actualisation].

부분 프로그램 이론 2: 환자의 역할
Partial program theory 2: patient role

우리는 중재에서 환자가 (1) 일상 생활에서 질병에 대한 정보 제공자로서의 역할(2) 교사로서의 역할이라는 두 가지 환자 역할을 확인했습니다. 첫 번째 역할(그림 3에 표시)의 경우, 가장 많이 유추된 메커니즘은 다른 사람과 정서적으로 관여하고 정서적 동일시, 다른 사람과의 상호작용에서 자신의 생각과 성찰을 표현하고 환자의 삶을 더 완벽하게 상상하는 것[모두 '내러티브 개발 및 참여' 클러스터에 속함]이었습니다. [환자를 주체로 인식하는 것]과 관련하여 변화된 맥락은 다음과 같습니다:
We identified two patient roles in the interventions: the patient as an (1) informant about the disease in daily life and (2) as a teacher. For the first role (depicted in Figure 3), the mechanisms inferred the most are: being emotionally involved with others and showing affective identification, articulating your thoughts and reflections in interaction with others, and imagining a patient’s live more completely [all in the cluster ‘developing narratives and engagement’]. The changed context concerning the perception of patients as subjects is illustrated below:

[...] 만성 질환 환자의 모든 요인(재정적 부담, 가족 역할, 환자/의사 관계 등)에 노출되었습니다. 둘째, 가족이 엄청난 스트레스, 슬픔, 두려움에 어떻게 대응하는지, 그리고 그러한 상황이 가져다주는 힘, 희망, 용기, 삶에 대한 놀라운 전망을 매우 친밀한 환경에서 보여주었습니다. 놀랍고 감동적이었으며 환자를 의료 차트가 아닌 한 인간으로서 고려하는 것이 얼마나 중요한지 잘 보여줍니다. [Z]
[…] it exposed me to all the factors of chronically ill patients … financial strain, family roles, patient/physician relationship, etc. Second, it showed me, in a very intimate setting, how a family responds to tremendous stress, grief, and fear, and the strength, hope, courage, and incredible outlook on life that are brought about by such a situation. It was remarkable and inspiring and speaks volumes to the importance of considering the patient as a human being as opposed to a medical chart. [Z]

[교사로서의 환자 역할](그림에 표시되지 않음)에서 가장 영향력 있는 맥락[어떤 지식이 가치 있는 것으로 간주되는지], [환자를 객체 또는 주체로 보는지] 여부의 변화였습니다.
For the patient role as a teacher (not depicted in the Figure), the most influential contexts were the change in what knowledge was considered valuable and whether the patient was seen as an object or a subject.

가장 많이 추론된 메커니즘은 실제 사람들의 삶의 이야기를 통해 질병을 맥락화하고[내러티브 개발], 다양한 관점을 통합하며[관점], 환경의 기대에 부응해야 한다는 압박감을 느끼지 않는 것[사회화]이었습니다.
The mechanisms inferred the most are contextualising disease with the life stories of real people [developing narratives], integrating different perspectives [perspectives], and feeling no pressure to live up to expectations in their environment [socialisation].

부분 프로그램 이론 3: 교사의 역할
Partial program theory 3: teacher role

우리는 중재에서 여러 가지 교사 역할을 확인했으며, 가장 두드러진 두 가지 역할은 (1) 보호자 역할과 (2) 평가자 역할입니다. 보호자의 역할(그림 4)에서 가장 많이 추론된 메커니즘은 환영받고, 안전하며, 안정감을 느끼는 것[자아실현], 스트레스가 많은 환경에서 편안함을 경험하는 것, 역할 모델을 보고 동화되는 것[모두 '사회화'] 등입니다. 다음 인용문은 이 인용문의 출처인 논문의 다른 곳에서 설명한 역할 모델로서의 배려하는 교사가 학습자의 관점에 어떤 영향을 미치는지 보여줍니다:
We identified several teacher roles in the interventions, with the most prominent two roles being that of (1) a carer and of (2) an assessor. For the role of the carer (Figure 4), the mechanisms inferred the most are: feeling welcome, safe and secure [self-actualisation]; experiencing a sense of comfort with a stressful environment; and seeing and assimilating role models [both in ‘socialisation’]. The following quote illustrates how the caring teacher as a role model, described elsewhere in the paper where this quote comes from, influences the perspectives of learners:

이러한 경험의 결과로 학생들은 '자비롭고 배려심 많은 의사가 될 수 있고 정서적 안정을 유지할 수 있다'는 것을 배웠습니다. [FF]
As a result of such experiences, students learned that they could be ‘compassionate and caring physicians and keep their emotional sanity’. [FF]

평가자로서의 교사의 역할에 대해서는 모든 맥락이 관련성이 있는 것으로 보였지만 그다지 강하지는 않았습니다. 가장 많이 추론된 메커니즘은 자신이 하는 일에 유능하다고 느끼는 것, 직업적 정체성을 개발하는 것['자아실현'], 롤모델을 보고 동화되는 것['사회화']이었습니다.
For the role of the teacher as an assessor, all contexts were seen as relevant but not very strong. The mechanisms inferred the most are: feeling competent in doing what you yourself do, developing your professional identity [both in ‘self-actualisation’], and seeing and assimilating role models [socialisation].

전체 프로그램 이론
Whole program theory

검토를 통해 상황과 개입 구성 요소가 모두 서로 다른 조합으로 발생한다는 것이 분명해졌기 때문에 데이터에 의해 충분히 뒷받침되는 하나의 포괄적인 프로그램 이론을 개발하는 것은 물론 시각화하는 것조차 불가능했습니다. 그럼에도 불구하고 실제로는 세 가지 이해관계자 그룹(환자, 학습자, 교사)의 대표자가 모두 개입에 참여하므로 부분적인 프로그램 이론을 함께 고려해야 합니다.
As it became clear from our review that the contexts and intervention components all occur in different combinations, it was not possible to develop, let alone visualise, one comprehensive program theory that was supported sufficiently by our data. Nevertheless, in practice, representatives of all three stakeholder groups (the patient, the learner, and the teacher) are present in interventions, and therefore our partial program theories should be considered in conjunction.

토론
Discussion

검토를 통해 예비 전문가가 환자 중심주의를 배워야 할 때와 관련된 다섯 가지 맥락을 분류했습니다.

  • (1) 실제 환자의 전체 질병 궤적에 접근할 수 없음,
  • (2) 환자 중심적 역할 모델이 없음,
  • (3) 환자를 객체로 간주함,
  • (4) 환자와 소통할 시간이 없음,
  • (5) 특정 지식만 정당한 것으로 간주함.

이 다섯 가지 맥락은 중재에서 학습자, 환자, 교사의 구성 요소를 통해 영향을 받습니다. 그 후, 우리는 개입 구성 요소, 맥락, 메커니즘 및 결과를 연결하는 세 가지 부분 프로그램 이론을 개발했습니다. 데이터에서 일상 생활에서 질병에 대한 [정보 제공자로서의 환자의 역할]이 가장 명확했으며, [보호자로서의 교사의 역할]도 마찬가지였습니다. 학습자의 역할과 관련해서는 [의료 전문가로서의 학습자]와 [기록자로서의 학습자] 모두 자주 등장했지만, 많은 논문에서 학습자는 '그냥 일반 학습자'였기 때문에 추가 분석을 하기에는 너무 이질적인 역할로 판명되었습니다. 환자 중심성 결과와 함께 가장 자주 확인된 메커니즘은 모두 관계형 클러스터(내러티브 개발 및 환자와의 참여)에 속하는 반면, 이러한 메커니즘은 주로 일상 생활에서 [질병에 대한 정보 제공자로서 환자]에 관한 논문에서 유추할 수 있는 것이 분명해졌습니다.
In our review, we have classified five different contexts which are relevant when upcoming professionals should learn patient-centredness:

  • (1) No access to the whole illness trajectory of real people;
  • (2) Non-patient-centred role models;
  • (3) Patients are seen as an object;
  • (4) No time to connect with patients;
  • (5) Only certain knowledge is considered legitimate.

These five contexts are influenced through learner, patient and teacher components in the intervention(s). Subsequently, we developed three partial-program theories which connect intervention components, contexts, mechanisms, and outcomes. In our data, the role of the patient as an informant about illness in daily life was most explicit, as was the role of the teacher as a carer. As to the learner roles, both the learner as a medical professional and the learner as a chronicler often occurred, even though in many papers, learners were ‘just regular learners’, which turned out to be a too heterogeneous role for further analysis. It became clear that the mechanisms most frequently identified in combination with patient-centredness outcomes were all in the relational cluster (developing narratives and engagement with patients), while these mechanisms were inferred primarily in papers about patients as an informant about their disease in daily life.

검토 결과, [환자들의 권위 있는 지식 부족]이 학습에 방해가 되는 것으로 나타났습니다. 이와는 대조적으로, 환자 중심주의에 초점을 맞추지 않고 일반적으로 환자로부터의 학습에 관한 논문에서 헨릭슨과 링스테드(2011)는 이러한 부족함이 학습자에게 이점으로 작용한다고 설명했습니다. 여기서 학습자들은 '바보 같은 질문'을 하는 것을 덜 두려워하고 환자 치료 중에 초기 전문가처럼 행동해야 한다는 부담감을 덜 수 있습니다(Henriksen and Ringsted 2011). 검토를 통해 우리는 [배려하는 교사caring teacher]란 [학습자를 지원하는 것]임을 알게 되었습니다. 그러나 Branch 등(2001)의 논문에서는 돌보는 사람이 된다는 것은 '환자의 경험에 대한 감정적 반응을 숨기지 않는 것'이기도 하다고 강조했습니다. 이 논문에서는 배려와 지원을 아끼지 않는 교사는 좋은 추가 기능이 아니라 학습자가 특정 개입을 더 잘 받아들이는 데 도움이 될 수 있다는 점이 분명하게 드러났습니다(Branch 등, 2001). 또한, 우리는 역할 모델이 환자에 대한 교사의 행동뿐만 아니라 학습자에게 피드백을 주는 것(예: 학습자가 환자를 마치 사물처럼 이야기할 때)이라는 것을 발견했습니다. 
In our review, patients’ lack of authoritative knowledge was seen as hindering for learning. In contrast, in their paper on learning – in general, not with a specific focus on patient-centredness – from patients, Henriksen and Ringsted (2011) described this lack as an advantage for learners. Here, they were less afraid to ask ‘stupid questions’ and get rid of the pressure to behave as nascent professionals during patient care (Henriksen and Ringsted 2011). In our review, we learned that a caring teacher is about supporting learners. In the paper by Branch et al. (2001), however, it was emphasised that being a carer is also about ‘not hiding your emotional reactions to patients’ experiences’. In their paper, it was evident that a caring and supportive teacher is not a nice additional extra, but that such a teacher may help to make learners more receptive to certain interventions (Branch et al. 2001). Furthermore, we found that role modelling was not only teachers’ behaviour towards patients but also giving feedback to learners, for example when learners talked about patients as if they were objects.

... 한 레지던트가 "X씨는 백혈구 수치가 높은 '췌장염'입니다."라고 발표하면 저는 "췌장염을 앓고 있는 우리가 돌보고 있는 신사분이 X씨라는 뜻인가요?"라고 끼어들곤 했습니다. 놀랍게도 레지던트들과 학생들도 저만큼이나 충격을 받았고 서로에게 이러한 순간을 지적하기 시작했습니다. [Branch 외. 2001]
…when a resident presented, “Mr. X is the ‘pancreatitis’ with a high white blood cell count,” I would interrupt, saying, “Do you mean Mr. X is the gentleman we are taking care of who suffers from pancreatitis?” To my pleasant surprise, I found that the residents and students were as shocked as myself and began pointing out these moments to each other. [Branch et al. 2001]

우리는 메커니즘을 다음 네 가지 클러스터로 나누어 환자 중심주의의 (차원) 발달이 어떻게 일어나는지 보여주었습니다. 

  • (1) 비교 및 결합, 관점 확장,
  • (2) 내러티브 개발 및 환자와의 참여,
  • (3) 자아실현,
  • (4) 사회화

이 중 일부는 인지적/이성적 라인을 따라 환자 중심주의에 기여하는 반면, 다른 일부는 정서적 연결에 더 중점을 둡니다. 아래에서는 논문에서 확인한 메커니즘을 해석하고 풍부하게 하는 데 도움이 될 수 있는 중간 범위 이론과 관련하여 이러한 클러스터에 대해 논의합니다.
We divided the mechanisms into four clusters,

  • (1) comparing and combining as well as broadening perspectives, 
  • (2) developing narratives and engagement with patients, and
  • (3) self-actualisation, and
  • (4) socialisation, 

to show how the development of (dimensions of) patient-centredness occurs. Some of these contribute to patient-centredness along cognitive/rational lines, while others are more about emotional connections. Below, we discuss these clusters in relation to middle range theories that may help to interpret and enrich the mechanisms we identified in the papers.

부분 프로그램 이론과 관련된 중간 범위 이론
Middle range theories in relation to our partial program theories

['비교하고 결합하고 관점을 넓히기']라는 제목 아래 여러 가지 메커니즘이 클러스터로 묶여 있는데, 이는 [주로 인지적 과정]입니다. [전환 이론]은 이러한 메커니즘이 어떻게 환자 중심이 되는 것을 촉진할 수 있는지 이해하는 데 유용해 보입니다. 이 이론에서는 [이전의 가정에 의문을 제기]하고, [비판적으로 자기 성찰]을 하며, [세상을 보는 방식을 바꾸는 것]이 학습에 필수적이라고 가정합니다(Prout 외. 2014). 전환 이론에서 "활성화 사건"은 학습 과정을 시작하는 것을 설명하는 데 사용되는 개념입니다. 프로그램 이론에서 활성화 이벤트는 학습자, 환자 또는 교사의 개입 요소가 될 수 있습니다. 변혁적 학습을 통해 학습자는 [새로운 의미에 대해 더 열린 마음]을 갖게 되고 이를 [주변 환경의 다른 사람들과 논의]하게 됩니다.
Several of our mechanisms were in a cluster under the heading ‘comparing and combining as well as broadening perspectives’, which were primarily cognitive processes. Transformation theory seems useful for understanding how these mechanisms may promote becoming patient-centred. In this theory, it is assumed that questioning prior assumptions, critical self-reflection and changing how to see the world, is essential for learning (Prout et al. 2014). In transformation theory, an “activating event” is a concept used to describe what starts the learning process. In our program theory, the activating event could be the learner, patient or teacher intervention component. By transformational learning, learners become more open to new meanings and discuss these with others in their environment.

환자 중심이 되기 위한 학습에는 종종 ['내러티브 개발 및 환자와의 참여']로 묶인 메커니즘이 포함됩니다. 이러한 메커니즘은 [사회적 학습 이론]을 나타냅니다. 이러한 메커니즘이 어떻게 환자 중심이 되는 것을 촉진할 수 있는지 이해하는 데 [스토리 이론](Smith and Liehr 2005)이 유용해 보입니다. [스토리 이론]에서는 [의도적 대화, 자기 관계와의 연결, 용이성 창출]이라는 세 가지 개념이 핵심입니다. 검토에서 확인된 개입 구성 요소 중 하나인 [기록자의 역할]"환자에게 가장 중요한 것이 무엇인지" 알아내기 위한 [의도적 대화]에 관한 이 이론을 적용할 수 있는 기회를 제공합니다(Smith and Liehr 2005). "사람들이 [자신과 자신의 상황을 경험하는 방식]은 문화적으로 매개된 사회적 상호작용을 통해 '구성'된다는 생각의 결과로서 이야기를 통해 학습이 일어나기 때문에"(Smith and Liehr 2005) 이 이론의 [세 가지 핵심 개념]은 환자 중심성을 학습하는 데 필수적이라고 주장합니다. 이 리뷰에 포함된 개입에서는 참가자가 과제의 일부로 작성한 성찰 일지에서 이러한 논의를 따르고 있습니다.
Learning to become patient-centred often involves mechanisms clustered under ‘developing narratives and engagement with patients’. These mechanisms indicate social learning theories. Story theory (Smith and Liehr 2005) seems useful for understanding how these mechanisms may promote becoming patient-centred. In story theory, three concepts are key: intentional dialogue, connecting with self-in-relation, and creating ease. The role of a chronicler, one of the intervention components identified in our review, provides opportunities for applying this theory which is about intentional dialogue to find out “what matters most to patients” (Smith and Liehr 2005). We argue that the three key concepts of this theory are essential in learning patient-centredness (Millender 2011) because learning occurs through stories as a result of the idea that “the way people experience themselves and their situation is ‘constructed’ through culturally mediated social interactions” (Smith and Liehr 2005). In the interventions included in this review, this line of arguing is followed in reflective journals written by participants as (part of) their assignments.

상황적 학습 이론(Lave and Wenger 1991)과 같은 다른 사회 학습 이론은 [맥락과 해당 커뮤니티]에 적합한 [가치, 규범 및 관계]에 초점을 맞추기 때문에 관련성이 있습니다(Handley 외. 2006). 이러한 이론은 [환자 중심 학습에 기여하는 메커니즘]을 불러일으키는 데 있어 맥락과 그 중요성을 더 잘 이해하는 데 도움이 될 수 있습니다. 예를 들어, 특정 유형의 지식이 다른 지식보다 더 정당한 것으로 간주되는 맥락이 개입이 학습을 지원하는 방식에 영향을 미친다는 것을 확인했습니다. 또한 역할 모델의 존재를 의미하는 맥락은 참가자가 지식, 기술 및 경험을 습득하는 사회 문화적 관행과 일치합니다(Prout 외. 2014). 
Other social learning theories, such as situated learning theory (Lave and Wenger 1991), are relevant because of their focus on the context and the values, norms and relationships fitting for that community (Handley et al. 2006). These theories can help better understand the contexts and their importance for evoking mechanisms that contribute to learning patient-centredness. For example, we identified that a context in which specific types of knowledge are considered as more legitimate than others affected the way the intervention supported learning. Also, our context which refers to the presence of role models aligns with socio-cultural practices in which participants gain knowledge, skills and experience (Prout et al. 2014).

마지막으로, 검토에서 확인한 메커니즘과 잘 부합하는 사회적 학습 이론은 [역할 이론]입니다. [역할 이론]은

  • "개인이 사회적 환경 내에서 차지하는 특정 위치와 관련하여
    개인이 어떻게 행동할 것으로 기대되는지, 그리고
    다른 사람들이 어떻게 행동할 것으로 기대하는지를 설명하고자 한다"(Richards 2015, 382쪽).

이번 검토에서 우리는 계층 구조 내에서 특정 위치의 중요성을 확인했습니다. 역할 이론은 우리의 결과를 더 잘 이해하는 데 도움이 될 수 있습니다. '좋은 교사'는 감정적이지 않은 전문가가 될 것으로 기대되는 맥락에서 교사가 보호자로서의 역할을 맡으면 긴장이 발생할 수 있습니다. 후자의 경우, 돌봄이 의사의 역할에 속하지 않는 것으로 간주되는 맥락에서 환자 중심주의에 대한 개입이 덜 성공적이었던 우리의 리뷰 [G]에서도 볼 수 있었습니다.
Finally, a social learning theory that aligns well with the mechanisms we identified in our review is role theory. Role theory

  • “seeks to explain how individuals are expected to act and how they expect others to act in reference to particular positions they occupy within the social milieu” (Richards 2015, p. 382).

In our review, we identified the importance of particular positions within a hierarchy. Role theory could help to understand our results better: in a context in which ‘the good teacher’ is expected to be a non-emotional expert, tensions may arise if teachers take up the role as a carer. The latter was also seen in our review [G] where, in a context in which caring was not considered to belong to the role of the physician, an intervention on patient-centredness was less successful.

요약하자면, 환자 중심성을 높이기 위한 중재를 설계하거나 평가할 때는 위의 이론을 보다 면밀히 따져볼 필요가 있습니다. 검토 결과, 많은 논문(32편 중 17편)이 이론을 명시적으로 언급하고 있지만, 이 이론이 개입 설계 또는 평가에 어떻게 도움이 되었는지 설명하는 경우는 드물었습니다.
In sum, when designing or evaluating interventions that were intended to lead to more patient-centredness, closer alignment with the above theories is warranted. In our review, we saw that many (17 of the 32) papers mention a theory explicitly, but that they rarely explain how this theory was instrumental in the design or evaluation of the intervention.

제한점
Limitations

[개입 구성 요소]가 [상황에 긍정적으로 또는 부정적으로 영향을 미치는 방식]이 포함된 논문의 저자에 의해 항상 명확하게 표시되지는 않았습니다. 따라서 연구자로서 저희의 해석이 필요했습니다. 컨텍스트는 서론이나 토론 섹션에서 종종 언급되었지만, 컨텍스트가 어떻게 메커니즘을 불러일으켰는지는 명확하지 않았습니다. 저희는 구성을 뒷받침하기 위해 다양한 관점(방법 섹션에 설명되어 있음)에서 접근했지만, 여전히 포함된 논문의 저자에 의존했습니다. 예를 들어, 문헌에서 학습자는 실제 환자와의 중재와 같은 실제 학습 상황을 즐기는 것으로 알려져 있습니다(Diemers 등. 2007, 2008). 따라서 실제 학습 상황의 경우 동기 부여와 관련된 메커니즘(예: 유용하다고 느끼는 것)이 중요할 것으로 예상했습니다. 그러나 이러한 메커니즘은 자주 주목받지 못했기 때문에 실제 환자를 대상으로 한 학습의 동기와 진정성 사이의 연관성은 더 연구할 필요가 있습니다.
The way intervention components affected the context positively or negatively was not always clearly indicated by the authors of the included papers. This necessitated an interpretation by us as researchers. The context was often mentioned in the introduction or discussion section, but it was less clear how that context evoked mechanisms. We followed an approach from diverse perspectives (described in the methods section) to support our configurations, but we still depended on the authors of the included papers. An example is that in the literature learners are said to enjoy authentic learning situations, such as interventions with real patients (Diemers et al. 2007, 2008). Therefore, we expected that in the case of authentic learning situations, mechanisms linked to motivation would be important (such as feeling useful). However, such mechanisms were not frequently notable and, therefore, the link between motivation and authenticity of learning with real patients deserves further study.

이질적인 개입 세트를 기반으로 세 가지 부분적인 프로그램 이론으로 구성된 프로그램 이론을 개발하려면 타협이 필요합니다. 위에서 언급한 구성을 설정하기 위해 개입 구성 요소와 컨텍스트를 분리하기로 결정했습니다. 실제로는 구성 요소와 컨텍스트가 상호 작용합니다. 예를 들어, [의료 전문가로서 학습자의 역할]은 [교사가 역할 모델로서 하는 일]과 상호작용하며, 이는 Branch의 배경 문서에 나와 있는 것처럼 돌보는 교사가 학습자를 돕고 환자 치료에 책임감을 부여하는 장면에서 잘 드러납니다:
Developing a program theory, composed of three partial program theories, based on a heterogenic set of interventions, asks for compromises. To be able to establish configurations, mentioned above, we have chosen to split up the intervention components and the contexts. In practice, the components and contexts will interact. For example, the role of a learner as a medical professional interacts with what teachers do as role models, which is shown in the background paper by Branch where a caring teacher helps the learner and gives responsibility in the patients’ care:

... 주치의는 학생의 치료 참여를 촉진하고, 실습에 대해 학생과 상호 반영하며, 환자와 초보 학습자를 존중하는 수준에서 사례에 대한 토론을 촉진했습니다. 레지던트와 같이 경험이 많은 학습자가 환자와의 토론을 주도하도록 초대되었을 수도 있습니다. 역할 모델링의 진정한 힘은 [...] 그들이 환자의 인본주의적 치료에 참여할 때 발휘됩니다. [Branch 외. 2001] 
…the attending physician facilitated the student’s participation in the care, reflected mutually with the student on the exercise, and facilitated a discussion of the case at a level that was respectful to the patient and to the novice learner. A learner with more experience, such as a resident, might have been invited to take the lead in the discussion with the patient. The true power of role modelling […] as they participate in the humanistic care of patients. [Branch et al. 2001]

마지막으로, 본 검토에 포함된 논문에서 Regan de Bere와 Nunn(2016)이 확인한 바와 같이, 환자 중심성에 대한 개념적 명확성과 공유된 정의가 부족하다는 점에서 본 검토의 한계를 찾을 수 있습니다. Bleakley와 Bligh(2008)가 이미 지적한 바와 같이, 환자 중심주의를 운영할 때 내재된 복잡성을 고려해야 합니다. RAMESES 가이드라인(2016)에서 권장하는 대로 결과에 대한 명확한 관점을 가지고 분석을 시작할 수는 없었지만, Scholl 등(2014)의 프레임워크를 사용하여 결과를 구조화했습니다. 이 프레임워크는 중재 개발자가 환자 중심주의의 어떤 측면에 도달하고자 하는지를 명확히 하는 데 좋은 지침을 제공할 수 있습니다. 
Finally, a limitation of our review is to be found in the lack of conceptual clarity and shared definitions of patient-centredness, as was also ascertained by Regan de Bere and Nunn (2016), in the papers included in our review. As Bleakley and Bligh (2008) already indicated, the inherent complexity of operationalising patient-centredness has to be taken into account. Although we could not, as recommended in the RAMESES guidelines (2016), start our analysis with a clear perspective on outcomes, we used the framework by Scholl et al. (2014) to structure our outcomes. This framework may also provide good guidance for developers of interventions to clarify what aspects of patient-centredness they aim to reach.

실무로의 번역
Translation into practice

부분적인 프로그램 이론은 특정 맥락에서 특정 중재 설계가 참여자가 (더) 환자 중심적이 되도록 돕는 데 성공할 수 있는 이유 또는 성공하지 못하는 이유를 논의하기 위한 논의의 출발점으로 도움이 될 것입니다. 환자 중심 학습을 목표로 하는 인터랙션을 개발할 때 교육자는 [자신의 상황을 명시적으로 고려]하면 도움이 될 수 있습니다. 연구 결과에서 두 가지 범주의 맥락이 나타난 것은 이와 관련이 있어 보입니다. [역할 모델], [전체 질병 궤적에 대한 접근성], [환자와 소통할 시간이 없는 상황]은 중재 설계를 통해 영향을 미치기 쉬운 것으로 보입니다. 다른 범주의 맥락은 [어떤 지식이 정당한 것으로 간주]되고 [환자를 객체로 보는지]에 관한 것으로, [직장 문화의 특성]과 관련이 있는 경우가 많습니다. 검토 대상 논문 중 일부는 이러한 맥락을 소규모로 국지적으로 변화시킬 수 있었다고 설명했지만, 개입이 환자 중심주의에 더 기여하기 위해서는 큰 변화가 필요한 것으로 보입니다. 본 연구 결과는 학습자의 환자 중심성을 향상시키기 위한 개입을 개발하고자 하는 교육자에게 잠재적인 개입의 성공과 관련된 맥락을 파악하기 위한 출발점을 제공할 수 있습니다. 이를 통해 가장 적절한 접근 방식을 선택하고 긍정적인 학습 결과를 얻을 가능성을 높일 수 있습니다. 
The partial program theories are intended to be helpful as a starting point for discussion, to discuss why a certain intervention design, within a specific context, could or could not be successful in helping participants to become (more) patient-centred. When developing an intervention aimed at learning patient-centredness, educators could benefit by explicitly taking their contexts into account. In doing so, it seems relevant that in our results two categories of contexts appeared. Role models, access to the whole illness trajectory and no time to connect with patients seem to be easier to influence through the design of the intervention. The other category of contexts, about what knowledge is considered legitimate and seeing the patient as an object, often pertains to characteristics of the workplace culture. Even though some of the papers in our review described that they were able to change such contexts on a small and localised scale, a major change seems to be necessary to make interventions contribute to patient-centredness more. Our results may provide educators, who wish to develop interventions to improve learners’ patient-centredness, with starting points for identifying contexts relevant to the success of potential interventions. This may help them to select the most appropriate approach and increase chances to attain positive learning outcomes.

부분적 프로그램 이론은 맥락과 상호작용하는 개입의 다양한 구성 요소가 개입의 결과와 이러한 결과가 어떻게 발생하는지와 관련이 있음을 보여줍니다. 여기서 역할 모델에 대해 더 폭넓게 생각하는 것이 개발자에게 유용한 것 같습니다. [역할 모델로서의 교사]는 [환자 중심적인 행동]을 보여줄 뿐만 아니라 [학습자에게 피드백]을 주고 [자신의 감정적 감정을 보여줍니다]. 환자의 역할과 관련하여 중재 설계자는 환자를 흥미로운 교육 '자료'로만 생각해서는 안 되며(Bleakley and Bligh 2008; Spencer 외. 2000), 교육에서 [인간으로서 환자가 가질 수 있는 부가가치], 즉 교사로서 또는 학습자에게 [질병이 일상 생활에서 무엇을 의미하는지 알려줄 수 있는 사람]으로 인식해야 합니다(Towle 외. 2010). 
Our partial program theories show that different components of the intervention, interacting with contexts, are relevant for the outcome of the intervention and how this outcome comes about. Here, thinking more broadly about role models seems to be useful for developers. The teacher as a role model does not only showcase patient-centred behaviour but also gives feedback to learners and shows his or her emotional feelings. With respect to the patients’ role, designers of interventions should not think about the patient as interesting teaching 'materials' (Bleakley and Bligh 2008; Spencer et al. 2000), but be aware of the added value that the patient as a human being may have in education: as a teacher or as someone who can tell the learner what an illness means in daily life (Towle et al. 2010).

요약하자면, 교육자들에게 우리의 검토는 단순한 '해야 할 일' 목록이 아니라 토론과 성찰을 위한 귀중한 포인트를 제공할 수 있을 것입니다. 우리의 분석에서 컨텍스트는 분명해졌을 뿐만 아니라 변경하기가 상당히 어려운 것으로 나타났습니다(MacLeod와 Frank 2010). 한 논문에서는 환자 중심주의에 기여하기 위해 개입을 도입하는 데 따르는 복잡성을 '이데올로기적 걸림돌'이라고까지 불렀습니다(Regan de Bere와 Nunn 2016). 환자가 학습의 파트너가 아닌 정보의 출처로 위치하는 문화에서 환자 중심주의에 기여하는 중재는 설계, 실행 및 유지에 상당한 노력이 요구됩니다. 
In sum, for educators, our review will potentially provide valuable points for discussion and reflection, but not a simple ‘to do’ list. In our analysis, contexts became not only evident but also appeared to be quite hard to change (MacLeod and Frank 2010). In one of the papers, the complexity of introducing interventions with the aim of contributing to patient-centredness was even called an “ideological stumbling block” (Regan de Bere and Nunn 2016). In a culture in which patients are positioned as a source of information instead of partners in learning, interventions that contribute to patient-centredness will ask for substantial efforts in the design, implementation and sustenance.

연구자에게 권장하는 주요 사항 중 하나는 개입을 실행한 맥락에 대해 보다 명확하게 설명하는 것입니다. 이 검토를 수행하면서 특정 교육 설계가 환자 중심성에 영향을 미치는지 여부에 초점을 맞춘 많은 효과 연구를 발견했습니다. 이러한 연구들은 제한된 방식으로 맥락을 설명합니다. 맥락의 사회문화적 측면과 개입을 설계하고 실행할 때 이러한 측면이 어떤 방식으로 고려되었는지에 대해 더 많이 아는 것이 도움이 될 것입니다. 최근의 논평(Horsley and Regehr 2018)에서도 이러한 명확성을 옹호했습니다.
For researchers, one of our main recommendations would be to be more explicit about the context in which they implemented their intervention. In doing this review, we noticed many effect studies, focusing on whether certain educational designs affect the outcome patient-centredness. Those studies describe the context in a limited manner. It would help to know more about sociocultural aspects of the context and in what way these aspects were considered when designing and implementing the intervention. In a recent commentary (Horsley and Regehr 2018), this clarity was advocated for as well.

또한 환자 중심주의를 배우기 위한 교육과 중재는 ['고유한 인격체로서의 환자'와 같은 명확한 차원]뿐만 아니라 [바쁜 병동에서 이러한 태도를 갖추는 데 필요한 팀워크]에 대해서도 살펴보는 등 환자 중심주의를 보다 폭넓게 바라보는 것이 도움이 될 것이라고 주장합니다.
Additionally, we argue that training and interventions to learn patient-centredness would benefit from a broader look at patient-centredness: not only looking at clear dimensions such as ‘the patient as a unique person’ but also at the teamwork necessary to obtain these attitudes in a busy ward.

이 리뷰에 포함된 논문에서 저자들은 이론을 전혀 언급하지 않거나 경험적 테스트에 적합하지 않은 '거창한' 이론을 언급했습니다. 저희는 프로그램 이론과 함께 논의된 [중간 범위의 이론]들이 연구자들이 개입을 설계하고 평가할 때 다른 이론을 사용하는 데 도움이 될 수 있다고 생각합니다. 예를 들어, [역할 이론]은 학습자가 환자, 학습자 또는 교사 등 특정 역할을 맡은 사람에게 기대하는 바를 생각하는 데 도움이 되기 때문에 환자 중심성을 학습자의 차원에 맞게 중재를 설계하는 데 유용한 접근 방식이 될 수 있다고 주장합니다.
In the papers included in this review, authors either did not refer to theory at all or referred to ‘grand’ theories that hardly lend themselves to empirical testing. We think that the middle-range theories discussed along with our program theory could help researchers to use other theories when designing and evaluating interventions. We argue that role theory, for example, would be a useful approach in designing interventions with the purpose for learning participants’ dimensions of patient-centredness because it helps thinking about what learners expect of people in a specific role, may it be the patient, the learner or the teacher.

결론
Conclusion

실제 환자가 참여하는 중재에 관한 문헌을 현실주의적으로 분석하여 세 가지 프로그램 이론을 개발했습니다. 이러한 이론은 환자, 학습자 및 교사의 역할이 정당한 것으로 간주되는 지식의 종류 또는 전체 질병 궤적에 대한 통찰력과 같은 맥락과 어떻게 상호 작용하는지 더 잘 이해하는 데 도움이 됩니다. 우리는 환자 중심주의가 학습되는 방식을 설명하는 메커니즘의 네 가지 클러스터를 추론했습니다. 이러한 클러스터는 환자 중심주의 학습에 관한 문헌에서 알려진 경우도 있지만 그렇지 않은 경우가 더 많은 다양한 중간 범위 이론과 일치합니다. 이 검토는 향후 연구를 위한 잠재적 영역을 제공하고 환자 중심주의를 목표로 하는 중재 개발자에게 생각할 거리를 제공합니다.
By doing a realist analysis of the literature about interventions in which real patients take part, we developed three program theories. These theories help to understand better how the role of the patients, the learners and that of the teachers interact with contexts such as the kind of knowledge that is considered legitimate or insight in the whole illness trajectory. We inferred four clusters of mechanisms that explain how patient-centredness is learned. These clusters align with different middle-range theories, which sometimes are known from the literature about learning patient-centredness but more often not. This review provides potential areas for future research and gives developers of interventions aimed at patient-centredness food-for-thought.

 

 


 

Med Teach. 2020 Apr;42(4):380-392. doi: 10.1080/0142159X.2019.1695767. Epub 2019 Dec 18.

 

Learning from patients about patient-centredness: A realist review: BEME Guide No. 60

Affiliations collapse

Affiliations

1Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, Utrecht University, Utrecht, The Netherlands.

2Center for Education Development and Research in Health Professions (CEDAR), University of Groningen and University Medical Center Groningen, Groningen, The Netherlands.

3Department of Primary and Interdisciplinary Care, Skills Lab, University of Antwerp, Antwerp, Belgium.

PMID: 31852313

DOI: 10.1080/0142159X.2019.1695767

Abstract

Background: Patient-centred work is an essential part of contemporary medicine. Literature shows that educational interventions contribute to developing patient-centredness, but there is a lack of insight into the associated learning processes.Objective: Through reviewing articles about educational interventions involving patients, we aspire to develop a program theory that describes the processes through which the educational interventions are expected to result in change. The processes will clarify contextual elements (called contexts) and mechanisms connected to learning patient-centredness.Methods: In our realist review, an initial, rough program theory was generated during the scoping phase, we searched for relevant articles in PubMed, PsycINFO, ERIC, CINAHL and Embase for all years before and through 2016. We included observational studies, case reports, interviews, and experimental studies in which the participants were students, residents, doctors, nurses or dentists. The relevance and rigour of the studies were taken into account during analysis. With deductive as well as inductive coding, we extended the rough program theory.Results: In our review, we classified five different contexts which affect how upcoming professionals learn patient-centredness. These aspects are influenced through components in the intervention(s) related to the learner, the teacher, and the patient. We placed the mechanisms together in four clusters - comparing and combining as well as broadening perspectives, developing narratives and engagement with patients, self-actualisation, and socialisation - to show how the development of (dimensions of) patient-centredness occurs. Three partial-program-theories (that together constituting a whole program theory) were developed, which show how different components of interventions within certain contexts will evoke mechanisms that contribute to patient-centredness.Translation into daily practice: These theories may help us better understand how the roles of patients, learners and teachers interact with contexts such as the kind of knowledge that is considered legitimate or insight in the whole illness trajectory. Our partial program theories open up potential areas for future research and interventions that may benefit learners, teachers, and patients.

자기평가: 그 모든 한계에도 불구하고, 왜 우리는 여전히 측정하고 가르치는가? 스코핑 리뷰의 교훈 (Med Teach, 2022)
Self-assessment: With all its limitations, why are we still measuring and teaching it? Lessons from a scoping review
Natasha Yatesa , Suzanne Gougha and Victoria Brazila,b

 

 

소개
Introduction

자기 평가(SA)는 의사가 안전하게 진료하고 평생 학습하는 데 중요하고 심지어 필수적인 것으로 간주됩니다(Boud and Falchikov 1989). 그러나 이러한 가정은 [SA의 복잡하고 잘 정의되지 않은 특성]과 [역량과의 상관관계가 없다는 점]을 강조하면서 광범위하게 반박되었습니다(Eva 외. 2012). 이러한 반박과 SA의 한계에 대한 수십 년간의 증거에도 불구하고 의학교육에서 SA의 오용은 지속되고 있습니다. 우리는 지난 10년간 의학교육에서 자기평가(SA)가 학습을 발전시키는 데 어떻게 적절하게 사용되었는지에 대한 공식적인 범위 검토를 수행하면서 그 정도를 발견했습니다. 우리는 SA가 유효하고 신뢰할 수 있는 결과 측정으로 오용되거나 그 자체로 학습 목표로 추구되는 경우가 많다는 사실에 놀랐습니다. SA의 문서화된 한계를 무시하여 범위 검토에서 제외된 논문을 종합하여 제시합니다. 우리는 성과를 개선하고/하거나 평생 학습자를 개발하는 방식으로 의대생 교육에서 SA를 독점적으로 발전시켜야 한다는 주장에 무게를 더하기를 희망합니다. 현재의 오해와 관행을 계속 유지한다면 의학교육의 중요한 영역에서 수행된 연구가 낭비될 위험이 있습니다.
Self-assessment (SA) is assumed to be important, even necessary, for doctors to practice safely and for life-long learning (Boud and Falchikov 1989). However, this assumption has been extensively rebutted (Eva et al. 2012), with emphasis on the complex, ill-defined nature of SA, and the lack of correlation with competence. Despite this rebuttal and decades of evidence around the limitations of SA, its misuse in Medical Education persists. We discovered the extent of this while undertaking a formal scoping review on how self-assessment (SA) has been used appropriately to advance learning in medical education over the past 10 years. We were surprised to find SA frequently misappropriated as a valid and reliable outcome measure, or sought as a learning goal in and of itself. We present a synthesis of papers excluded from our scoping review because they ignored the documented limitations of SA. We hope to add weight to calls to exclusively progress SA in medical student education in ways that improve performance and/or develop life-long learners. If we continue with current misconceptions and practices, we risk wasting research done in this important area of medical education.

배경
Background


시간이 지남에 따라 자가 평가에 대한 반복적인 요구에도 불구하고 보편적으로 받아들여지는 자가 평가의 정의는 아직 없습니다(Sargeant 2008). 정의는 연구 방법론, 해석 및 결론을 뒷받침하는 데 중요합니다. 무수히 많은 개념화 중에서 설명하는 데 도움이 되는 세 가지 명확한 구분이 있습니다:
There is no universally accepted definition of self-assessment, despite repeated calls for this over time (Sargeant 2008). Definitions are important to underpin research methodologies, interpretation, and conclusions. Within a myriad of conceptualisations, there are three clear distinctions that we have found helpful to explicate:

SA는 가이드형인가, 비가이드형인가? 
Is SA guided or unguided?
[가이드가 없는 SA]는 내부적으로 생성되고 고립되어 있으며 개인주의적입니다. 이를 '비공개 SA' 또는 '내부 피드백'이라고 부르기도 합니다. 이는 검사하기 어려울 수 있고(Andrade 2010) 항상 개입할 수 있는 것은 아니기 때문에 가치가 제한적입니다.  
Unguided SA
 is internally generated, isolated, and individualistic. Sometimes this is referred to as ‘private SA’ or ‘internal feedback.’ This has limited value because it can be difficult to inspect (Andrade 2010) and is not always amenable to intervention.  
  [가이드형 SA]는 동료나 교사와 같은 다른 사람의 피드백을 통해 보다 협력적이고 구조화되어 있지만, 이러한 피드백을 구하고 처리하는 책임은 여전히 개인에게 있습니다. 이는 또한 지시형 또는 정보에 입각한 자기 평가로 이해될 수 있습니다. 우리는 이것이 교육 및 학습의 맥락에서 SA를 고려하는 필수적인 방법이라고 주장하는 사람들의 주장에 동의합니다(Sargeant 2008).
Guided SA
 is more collaborative and structured, with input from others like peers or teachers, although the responsibility for seeking and processing this input still rests on the individual. This may also be understood as directed or informed Self-Assessment. We agree with those who contend that this is an essential way to consider SA in the context of teaching and learning (Sargeant 2008).
SA는 글로벌 성과인가요, 아니면 특정 성과인가요? 
Is SA a global or a specific accomplishment?
[글로벌 SA]는 본질적으로 자신감의 표현입니다('나는 좋은 운전자다').
Global SA
 is essentially a statement of self-confidence (‘I am a good driver’).
  [특정적 SA]특정 영역에 초점을 맞춘 진술입니다('나는 안전하게 후진 평행 주차를 할 수 있다'). 우리는 자신의 전반적인 능력을 과대평가하는 경향이 있지만, 좁고 집중된 과제에 대해 생각하도록 요청받으면 자기 평가가 더 정확해지는 경향이 있습니다(Eva and Regehr 2011).
Specific SA
 is a focussed statement in a particular area (‘I am able to reverse-parallel park safely’). We tend to overestimate our global abilities, but when asked to think about a narrow, focussed task we tend to be more accurate in our self-assessment (Eva and Regehr 2011).
[글로벌 SA]는 대부분 도움이 되지 않지만, [특정 SA]는 학습을 진전시키는 데 유용한 보조 도구가 될 수 있습니다.
While global SA is mostly unhelpful, specific SA can be a valuable aide to progress learning.
SA는 능력인가요, 접근 방식인가요? Is SA an ability or an approach? 어떤 사람들은 SA를 [학습하고 측정할 수 있는 기술이나 능력]으로 간주합니다. 여기서 SA는 최종 목표로 간주되며 외부 표준과 비교하여 측정할 수 있습니다. 개입이 개선된 자기 평가로 이어진다는 것을 입증하면 성공으로 간주할 수 있습니다. 이러한 관점에서 자기 평가는 총체적인 목적을 가진 학습에 대한 평가입니다. 학생이 외부 측정치에 가까울수록 학생의 자기 평가가 더 나은 것으로 간주되며, 이는 학습/성과 향상으로 이어질 것이라는 가정을 하게 됩니다.
Some view SA as a skill or ability, able to be learned and measured. Here SA is seen as the end goal, and measurable compared to an external standard. Demonstrating that intervention leads to improved self-assessment would be considered a success. Through this lens, self-assessment is an assessment of learning, with a summative purpose. The closer the student approximates an external measure, the better their SA is considered to be, and the assumption is that this will then lead to improved learning/performance.

그러나 이는 문제가 있습니다. 학생이 자기 평가를 더 정확하게 하는 법을 배운다고 해서 SA 기술을 더 잘하게 된 것일 뿐, 반드시 앎, 이해 또는 수행 능력이 더 좋아지는 것은 아니며안타깝게도 자기 평가 개선이 성과 향상으로 이어진다는 증거는 거의 없습니다(Davis 외. 2006). 또한 SA를 습득해야 할 기술로 간주하는 경우, 일반적으로 약점은 파악하지만 강점은 강화하지 않는 '격차 기반' 학습 접근 방식에 중점을 둡니다(Eva and Regehr 2005).

However, this is problematic—if students learn to self-assess more accurately, all that we have done is made them better at the skill of SA but not necessarily better at knowing, understanding, or performing—unfortunately, there is little evidence that improved self-calibration will lead to improved performance (Davis et al. 2006). Additionally, if SA is seen as a skill to obtain, the focus is generally a ‘gaps-based’ approach to learning, where weaknesses are identified but strengths are not reinforced (Eva and Regehr 2005).
  SA를 [전략 또는 접근 방식]으로 보는 사람들도 있습니다. [최종 목표]는 [학습, 성과 또는 안전을 개선]하는 것이며 SA는 이를 위한 수단입니다. 이러한 관점을 통해 연구자들은 SA를 교육학, 프로세스 또는 습관으로 논의하며 [평가의 정확성]보다는 [자체 평가의 메커니즘에 초점]을 맞춥니다. SA는 학습을 위해, 그리고 보다 형성적인 목적으로 수행되고 있습니다. 이를 통해 약점만 다루는 것이 아니라 강점 강화(Cheng 외. 2021)를 포함한 다양한 평가와 조치를 취할 수 있습니다.
Others see SA as a strategy or approach. The end goal is to improve learning, performance, or safety, and SA is the means to that end. Through this lens, researchers discuss SA as a pedagogy, process, or even a habit, and focus is on the mechanisms of self-assessing rather than the accuracy of that assessment. SA is being done for learning, and with a more formative purpose. This allows for a spectrum of assessments and actions, including reinforcement of strengths (Cheng et al. 2021) and not just addressing weaknesses.

평생 학습자를 개발하기 위해 우리가 선호하는 개념화는 SA가 학습을 진전시키기 위한 안내적이고 구체적인 전략이라는 것입니다.
In pursuit of developing life-long learners, our preferred conceptualisation is that SA is a guided, specific, strategy to progress learning.

중복되는 개념 및 정의
Overlapping concepts and definitions

자기 평가
Self-evaluation

종종 SA와 같은 의미로 사용되는 이 용어는 주로 자기조절학습(SRL) 문헌에서 학습 주기 내의 특정 하위 과정을 설명하기 위해 사용되었습니다. [자기 평가]는 짐머만 사이클의 [세 번째('성찰') 단계]에서 이루어집니다.
Often used interchangeably with SA, this term has primarily been used in the Self-Regulated Learning (SRL) literature to describe a specific sub-process within the cycle of learning. Self-evaluation occurs in the third (‘Reflection’) phase of Zimmerman’s cycle.

자기 모니터링
Self-monitoring

특정 상황에서 행동할 수 있는 기술/지식을 유지하고 있는지에 대한 가능성을 [순간순간 인식]하는 것입니다. [SRL 문헌에서는 짐머만 사이클의 두 번째('행동') 단계]에서 발생합니다. 자기 모니터링은 '순간적 SA'라고 부를 수 있지만(McConnell 외. 2012), 그 이상의 개념과 기능도 포함합니다.
A moment-by-moment awareness of the likelihood that one maintains the skill/knowledge to act in a particular situation. Within the SRL literature, this occurs in the second (‘Action’) phase of Zimmerman’s cycle. Self-monitoring could be called ‘SA in-the-moment’ (McConnell et al. 2012), although it also includes conceptualisations and functions beyond this.

평가적 판단
Evaluative judgement

'자신과 타인의 업무의 질에 대한 결정을 내릴 수 있는 능력'으로, 주로 임상 환경과 전문성 개발을 탐구할 때 사용됩니다(p471)(Tai 외. 2018).
‘The capability to make decisions about the quality of work of oneself and others,’ is used primarily in clinical settings and when exploring the development of expertise (p471) (Tai et al. 2018).

자기 효능감
Self-efficacy

종종 [과제에 대한 자신감]으로 설명되며, Bandura(1997)는 이를 [특정 과제를 성공적으로 수행할 수 있는 자신의 능력에 대한 역량]으로 설명했습니다.
Often described as task-specific self-confidence; Bandura (1997) described it as SA of one’s competence to successfully undertake a particular task.

조사 방법
Methods

우리는 지난 10년간 의학교육에서 학습을 발전시키기 위해 자기평가(SA)가 어떻게 사용되었는지를 조사하는 범위 검토를 수행했습니다. 우리는 오픈 사이언스 프레임워크에 프로토콜을 등록(NY 2021)하고 Arksey와 O'Malley의 6단계 프레임워크(Arksey and O'Malley 2005)에 따라 진행했습니다. 
We undertook a scoping review exploring how self-assessment (SA) has been used to advance learning in medical education over the past 10 years. We registered (NY 2021) our protocol on the Open Science Framework and were guided by Arksey and O’Malley’s 6-step framework (Arksey and O'Malley 2005).

  • 1단계: '지난 10년 동안(2011년 이후) 의학교육자들은 의대생 교육에서 성과를 개선하고/하거나 평생 학습자를 개발하는 방식으로 SA를 어떻게 개념화하고, 탐구하고, 발전시키고 있는가'라는 연구 질문을 확인했습니다. 
  • 2단계: 2021년 9월 14일에 체계적 문헌고찰 가속기(SRA)의 자동화 도구를 사용하여 관련 연구를 확인했습니다 (Clark, Glasziou 외. 2020). 액세스한 데이터베이스는 Pubmed, Embase, PsychINFO 및 ERIC입니다. 선임 정보 전문가는 검색 및 선별 과정을 지원하는 다양한 자동화된 도구에 익숙합니다. 각 데이터베이스에 대한 전체 검색 전략은 프로토콜(NY 2021)(공개적으로 볼 수 있음)에 게시되어 있으며, SRA 다국어 검색 번역기(Clark, Sanders 외. 2020)를 사용하여 개발되었습니다. SRA 중복 제거기(Clark, Glasziou 외. 2020)를 사용하여 중복 논문을 제거한 후 1151개의 연구가 남았습니다. 
  • 3단계: 모든 저자는 포함 및 제외 기준을 사용하여 관련 연구를 독립적으로 선택하고, SRA Screenatron 도구를 사용하여 제목과 초록을 검토했습니다(Clark, Glasziou 외. 2020). 가이드가 없는 SA를 설명하거나 SA를 글로벌 스킬(일반적인 자신감)로 정의한 연구는 제외했습니다. 또한 SA의 정확성(즉, 학습을 위한 SA가 아닌 학습에 대한 SA)을 파악하는 데 주안점을 두고 학생의 자기평가를 외부 측정에 보정하는 능력을 측정하는 것을 목표로 한 연구도 제외했습니다. 마지막으로, SA를 사용하여 개입의 성공 여부를 평가한 연구는 학습 진도에 초점을 맞춘 것이 아니라 프로그램을 평가하는 데 SA를 사용했기 때문에 제외했습니다. 
  • Step 1: We identified the research question: ‘Over the past 10 years (since 2011), how are medical educators conceptualising, exploring, and progressing SA in medical student education in ways that improve performance and/or develop life-long learners?’
  • Step 2: We identified the relevant studies on 14/09/2021 using automation tools from the Systematic Review Accelerator (SRA) (Clark, Glasziou et al. 2020). Databases accessed were Pubmed, Embase, PsychINFO, and ERIC. A senior information specialist is familiar with various automated tools assisted with the searching and screening process. The full search strategy for each database is published with our protocol (NY 2021) (publicly viewable) and was developed using the SRA Polyglot Search Translator (Clark, Sanders et al. 2020). After duplicate papers were removed using SRA Deduplicator (Clark, Glasziou et al. 2020), 1151 studies remained.
  • Step 3: All authors independently selected relevant studies using inclusion and exclusion criteria, reviewing titles and abstracts using the SRA Screenatron tool (Clark, Glasziou et al. 2020). We excluded studies that described unguided SA or defined SA as a global skill (general self-confidence). We also excluded studies that aimed to measure students’ abilities to calibrate their self-assessment to an external measure, with the primary focus being identifying the accuracy of SA (i.e. SA of learning, rather than SA for learning). Finally, we excluded studies that used SA to evaluate the success of an intervention, because their focus was not on using SA to progress learning, instead, they used SA to evaluate a program.

이 시점에서 저희는 제외된 연구들에서 상당수의 연구자들이 SA의 문서화된 한계를 무시하는 방식으로 SA를 사용하고 있다는 점을 우려하여 연구를 잠시 중단했습니다. 저희는 SA에 대한 지속적인 오해의 범위를 정의하기 위해 제외된 연구들을 비판적으로 분석하는 이례적인 조치를 취했습니다. 그 결과 연구 이해나 접근 방식에 결함이 있다고 판단되는 두 그룹을 발견했습니다: 
At this point, we paused, because we were concerned to note in our excluded studies that significant numbers of researchers are using SA in ways that ignore its’ documented limitations. We took the unorthodox step of critically analysing these excluded studies, aiming to define the extent of persistent misconceptions regarding SA. We identified two groups where we believe research had a flawed understanding or approach:

  • 그룹 1. 학습의 SA에 대한 연구로, 어떤 조건/특성 때문에 학생들이 자가 교정 능력이 떨어지는지 탐구하고, 정확도를 개선하는 방법에 초점을 맞춥니다(따라서 교정의 정확도가 학습의 전제 조건이 아니라는 점을 인정하지 않음).
  • Group 1. Research into SA of learning, exploring what conditions/features lead to students being poor self-calibrators, with a focus on how to improve their accuracy (and therefore no acknowledgement that accuracy of calibration is not a pre-requisite for learning).
  • 그룹 2. 참가자의 SA를 개입/프로그램의 결과 측정치로 사용한 연구. 이 연구에서는 참가자가 학습 또는 개선되었다고 진술하는 것이 프로그램 또는 개입의 효과에 대한 증거라고 가정했습니다.
  • Group 2. Research that used SA of participants as an outcome measure for their intervention/program. The assumption made in this research is that participants stating they learned or improved was evidence of the efficacy of their program or intervention.

 

결과
Results

검색에서 확인된 1151개의 연구 중 821개는 의대생을 포함하지 않았거나, SA를 다루지 않았거나, 영어로 되어 있지 않아 완전히 제외되었고, 27개는 가이드가 없는 SA를 탐구했거나 SA를 글로벌 기술(일반적인 자신감)로 정의했기 때문에 제외되었습니다. 96건은 전체 텍스트 검토에 포함되었고, 207건의 연구는 SA의 문서화된 한계를 무시했기 때문에 제외되었습니다. 
Of the 1151 studies identified in the search, 821 were excluded outright (e.g. did not include medical students, were not looking at SA, or were not in English) and 27 because they explored unguided SA or defined SA as a global skill (general self-confidence). Ninety-six were included for full-text review, leaving 207 studies excluded because they ignored the documented limitations of SA.

나머지 207개 연구는 여기에서 추가 분석에 포함되었습니다. 
These remaining 207 studies are included in our further analysis here.

  • 그룹 1. 연구자들이 학습의 SA를 살펴본 39개의 연구를 확인했습니다. 이 연구는 지적으로 흥미로울 수 있지만(예를 들어 여성은 일반적으로 학습에 대한 자기 평가에서 남성보다 '정확성'이 떨어지는 것으로 나타났습니다), 성과 향상이나 평생 학습에는 거의 유용하지 않습니다. 이 그룹의 연구자들은 학생들이 더 정확하게 자기 평가를 할 수 있도록 가르치는 것을 '해결책'으로 제안하기도 했습니다. 이 중 ¼ 이상이 1등급 의학교육 학술지(즉, 영향력 지수 2.0 이상)에 게재되었습니다(그림 1(인포그래픽) 참조). 
  • Group 1. We identified 39 studies where researchers looked at SA of learning. This research may be intellectually interesting (for example females are usually found to be less ‘accurate’ than males at self-assessment of their learning), but it is of little utility for improving performance or lifelong learning. Sometimes researchers in this group went on to suggest as a ‘solution’ that students be taught to self-assess more accurately. Over ¼ of these were published in tier 1 Medical Education Journals (i.e. with an Impact Factor of >2.0); see Figure 1 (infographic).
  • 그룹 2. 188개는 프로그램 또는 개입의 효과를 평가하기 위한 결과 측정으로 SA를 사용했습니다. 시간이 지남에 따라 이러한 방식으로 SA를 사용하는 출판물이 줄어들고 있다는 징후는 보이지 않습니다(그림 1(인포그래픽) 참조).
  • Group 2. One hundred and sixty-eight used SA as an outcome measure to assess the efficacy of a program or intervention. There is no sign that publications using SA in this way are abating over time—see Figure 1 (infographic).
 

참가자의 지식 및/또는 기술 습득과 같은 [객관적인 결과를 평가하기 위한 척도로 SA를 사용]한 94건의 연구를 확인했습니다. 이 중 63개는 학습/성과에 대한 다른 증거를 찾으려 하지 않고 [SA를 유일한 결과 측정치로 사용]했습니다.
We identified 94 studies that used SA as a measure to evaluate an objective outcome, like knowledge and/or skills acquisition of participants. Of these, 63 used SA as the sole outcome measure with no attempt to seek other evidence of learning/performance.

나머지 31개 연구는 자체 평가 개선도와 다른 결과(예: 시험에서 얻은 점수 또는 전문가 의견)를 모두 측정하고 그 일치 여부를 고려했습니다. 이 중 
A further 31 studies measured both self-assessed improvement and another outcome (e.g. marks achieved on a test, or expert opinion) and considered whether they aligned. Of these:

  • 20개 연구에서 [SA의 원자가valence]가 [객관적 결과의 원자가valence]와 상관관계가 있는 것으로 나타났습니다. 거의 항상 이러한 값은 긍정적인 방향이었습니다(즉, '학생들이 더 많이 배웠고 스스로도 더 많이 배웠다고 평가했습니다.'). 이러한 연구에 참여한 학생들의 지식/기술이 향상되었을 가능성도 있지만, SA가 학습을 과대평가하거나 과소평가했을 가능성도 있습니다.
    Twenty of these found that the valence of SA correlated with the valence of objective outcomes. Almost invariably these valences were in a positive direction (i.e. ‘the students learned more and they also self-assessed as learning more.’) It is possible that students in these studies did improve in their knowledge/skills, but also possible their SA over-estimated or under-estimated their learning.
  • 연구 중 5건은 [SA]와 [다른 결과]를 모두 측정하였고 [정렬에 대해 불분명]했습니다.
    Five of the studies measuring both SA and another outcome were unclear about alignment
  • 6개의 연구에서 자체 평가 결과를 다른 측정치와 비교하여 정량화한 결과, [초보자가 과대 평가]하거나 [전문가가 과소 평가]하는 것이 분명하게 나타났습니다.
    Six studies quantified self-assessed results compared with another measure and clearly found over-estimation by novices, or under-estimation by experts.

SA는 [자신감, 불안감 또는 공감]과 같은 [주관적인 결과를 측정하는 합리적인(때로는 유일한) 방법]입니다. 74건의 연구 중

  • 12건은 교육 개입에 대한 학생들의 자기보고 불안 또는 공감을 측정한 연구였습니다.
  • 나머지 62개 연구는 자체 평가한 자신감을 측정했습니다. 이는 역량에 대한 강력한 검증 없이 자신감이 결과로 나타나는 상황에서 문제가 됩니다.
    • 62개 연구 중 13개 연구는 자신감만을 유일한 결과로 측정했습니다.
    • 나머지 39개는 자신감과 다른 한 가지 결과를 함께 측정했지만, 대개 각 측정값의 원자가 또는 방향을 보고하는 등 상관관계를 시도하지 않았습니다.

변함없이 자신감이 높아지면 성과도 높아졌지만, 학생들이 자신의 성과에 비해 자신감이 지나치게 높거나 낮은지 여부는 조사되지 않았습니다. 추가로 10건의 연구에서는 자신감이 객관적인 성과 측정치와 일치하지 않는 것으로 나타났습니다.
SA is a reasonable (and sometimes the only) way to measure subjective outcomes, like confidence, anxiety, or empathy. We found 74 studies where this occurred, including

  • 12 which measured students’ self-reported anxiety or empathy in response to a teaching intervention.
  • The remaining 62 studies measured self-assessed confidence. This becomes problematic in situations where confidence is an outcome without any robust validation of competence.
    • Thirteen of those 62 studies measured confidence as the only outcome.
    • Another 39 measured confidence plus one other outcome, but usually did not attempt to correlate them, mostly reporting a valence or direction of each measure.

Invariably, as confidence increased so did performance, but what was not explored was whether students were now over-or under-confident compared to their performance? A further 10 studies found that confidence did not align with an objective measure of performance.

요약하면, 지난 10년 동안 많은 저자들은 SA를 학습이 이루어지기 위해 ['정확성'을 향상시켜야 하는 기술]로 간주하거나, [SA를 유효하고 신뢰할 수 있는 결과 측정 도구로 가정]하여 [교육 개입의 '성공'을 측정하는 방법으로 SA를 사용]해 왔습니다. [SA를 오용한 연구]를 발표한 저자(207명)가 [근거 기반 적용 가능성이 있는 연구]를 발표한 저자(96명)보다 두 배 이상 많았습니다. 원래의 범위 검토에서는 SA의 오용 사례를 의도적으로 찾아낸 것이 아니라 성과를 개선하거나 평생 학습자를 개발하는 방식으로 SA를 사용한 사례만 조사했기 때문에 오용 사례의 수는 이보다 훨씬 더 많을 수 있습니다.
In summary, many authors over the past 10 years have viewed SA as a skill that needs to be improved in ‘accuracy’ for learning to take place, or have used SA as a way to measure the ‘success’ of teaching intervention, assuming SA to be a valid and reliable outcome measurement tool. More than twice as many authors (207) have published research that misappropriates SA, than those who published research with the probable evidence-based application (96). The number of misuses may be even higher than this, as our original Scoping review was not deliberately searching for misappropriated use of SA, but rather for its use in ways that improve performance and/or develop life-long learners.

토론
Discussion

제외된 연구에 대한 분석에서 알 수 있듯이 SA의 주제와 사용에 대한 오해가 많습니다이러한 오해가 근거에 기반한 연구보다 두 배 이상 많은 연구가 발표되고 있습니다. 이러한 현상이 발생하는 이유를 살펴보기 위해 의학교육 이외의 교육 및 학습에 대한 비유를 고려해 보겠습니다.

  • 십대 자녀에게 자동차 운전법을 가르치려고 한다고 상상해 보세요. 몇 번의 운전을 가르친 결과, 아이들은 처음부터 끝까지 무사히 운전할 수 있었습니다. 자녀는 이제 고속도로에서 운전하고 싶다고 말하지만, 부모는 자녀의 능력에 대해 확신이 없습니다. 그래서 당신은 자녀가 자신의 한계를 인식하고 아직 고속도로 운전에 적합하지 않다는 것을 깨닫는 데 도움이되기를 바라면서 자녀에게 자가 평가를 요청합니다. 자녀는 잠시 앉아서 생각한 다음 고속도로를 운전할 준비가 되었다고 확신하며 몇 가지 근거에 기반한 이유를 설명합니다. 지금까지 성공적인 여행만 했고, 과속이나 교통 위반 벌금을 받은 적이 없으며, 친구들보다 더 많은 거리를 운전했기 때문에 상대적으로 더 경험이 많다고 느낀다는 것입니다.

Misconceptions around the subject and use of SA abound, as our analysis of excluded studies demonstrates. More than twice as much research is being published with these misconceptions than with evidence-based constructs. To help explore why this is occurring, we would like you to consider an analogy of teaching and learning that sits outside Medical Education:

  • imagine you are trying to teach a teenager how to drive a car. You take them out for a few trips, and they manage to get from start to finish unscathed. They tell you that they now want to drive on the highway, but you are not as confident in their abilities as they are. So, you ask them to self-assess, hoping this will help them recognise their limits and realise they are not fit for highway driving yet. They sit and think for a while and then explain that they are sure that they are ready to drive the highway, giving you some evidence-based reasons: they have had only successful trips so far, they have never received a speeding or traffic infringement fine, and they have driven more distance than their friends have, so feel comparatively more experienced.

크루거와 더닝의 연구에 따르면 이러한 능력에 대한 과대 평가는 정상입니다(Ehrlinger 외. 2008). 당신은 지금까지의 여행이 상대적으로 쉬웠다는 점(조용한 이면도로를 다녔다는 점), 수많은 아차사고(옆자리에 앉아있을 때 구해냈다는 점), 동료와 자신을 비교하는 것은 능력의 기준이 될 수 없다는 점 등을 지적하며 그들의 주장에 반박할 수 있습니다. 이제 이 대화는 그림 2에 제시된 것처럼 여러 가지 방식으로 진행될 수 있습니다:
This over-estimation of their abilities would be normal according to research by Kruger and Dunning (Ehrlinger et al. 2008). You counter their arguments by pointing out the relative ease of the trips so far (you have taken quiet back roads), their numerous near misses (from which you rescued them as you were sitting next to them), and the fact that comparing themselves with peers is not a benchmark for capability, as none of them are competent yet either. This conversation could now go several different ways, as presented in Figure 2:

[시나리오 1과 2]는 예측 가능하지만 원치 않는 반응으로 학습을 방해합니다. 우리는 종종 [인지적 편향]과 [자기 이미지 보호/인상 관리](Evans et al. 2005)와 같은 요인으로 인해 [정확한 자기 평가를 하지 못합니다](Yu et al. 2020). 마찬가지로 저성과자의 SA를 개선하면 자기효능감이 낮아질 위험이 있습니다(Artino 2012). 성과가 낮은 학습자가 자신이 하위 사분위수에 속한다는 사실에 직면하게 되면 개선할 수 있다는 자신감을 잃을 수 있습니다. 이렇게 학습에 대한 정확한 SA를 요구하면 의도치 않게 성과가 악화될 수 있습니다.
Scenarios 1 and 2 would be predictable but unwanted responses, which hamper learning. We frequently cannot self-assess accurately (Yu et al. 2020): things like our cognitive biases and self-image protection/impression management (Evans et al. 2005) get in the way. Likewise, improving the SA of a low performer risks lowering their self-efficacy (Artino 2012). If a poor performer is forced to face the fact that they are in the lower quartile they may lose the confidence that they can improve. This push for accurate SA of their learning then may inadvertently worsen performance.

문제는 청소년이 [시나리오 3의 반응]을 보이도록 도울 수 있을까요? 그렇다면 어떻게? 정확한 자기 평가를 하는 방법을 배우도록 도와주면 그렇게 할 수 있을까요? 
The question is, can we assist the teenager to have a Scenario 3 response? If so, how? Will helping them learn to accurately self-assess enable that to occur?

바로 이 지점에서 [증거에 맞는 방식으로 SA를 개념화]하는 것이 중요해집니다. 청소년이 운전을 잘하기 위해서는 정확한 자기 평가를 배워야 한다는 가정은 면밀한 조사를 통해 입증되지 않았습니다. 요컨대, [SA가 운전 실력 향상으로 가는 길이라고 가정하는 연구]는 [근본적으로 결함]이 있습니다(Colliver 외. 2005). [SA 정확도 향상에 초점을 맞추는 것]은 [운전 능력 향상에 초점을 맞추기]보다는, 청소년들이 [능숙하게 운전할 수 있는 도로에만 청소년들을 투입하는 것]과 같습니다(따라서 SA를 안전한 운전 성능에 맞추는 것). 
This is where it becomes essential to conceptualise SA in a way that fits the evidence. The assumption that this teenager needs to learn to self-assess accurately before they can progress in their driving does not hold up under scrutiny. In short, research that assumes that SA is the route to improved performance is fundamentally flawed (Colliver et al. 2005). Focussing on improving SA accuracy is like only ever putting our teenagers on roads that they can competently drive (therefore matching their SA to a safe driving performance), rather than focussing on improving their driving.

다시 의학교육에 대한 생각으로 돌아와서, 몇 년 전부터 이러한 점을 지적하는 강력하고 분명한 목소리가 있었습니다. 언뜻 역설적으로 보이지만 데이터는 분명합니다. Eva, Regehr, Gruppen(Eva 외. 2012)이 설명한 것처럼, SA를 가르치는 데 많은 노력을 기울이면 학생과 의사가 보다 정확하게 자기 평가를 할 수 있는(즉, SA 기술/능력을 개발할 수 있는) 교육을 성공적으로 수행할 수 있지만, 반드시 성과를 향상시키지는 못합니다. 또한 학생들이 자기 평가 능력을 향상시키는 데 도움을 줄 수는 있지만(Lopez와 Kossack 2007), SA 향상과 성과(Andersen 외. 2015), 통찰력(Ehrlinger 외. 2008), 학습(Colthart 외. 2008) 또는 환자 결과(Al-Kadri 외. 2012) 사이에는 상관관계가 낮다는 것은 잘 알려진 사실입니다. 대신 약한 학생의 성과를 개선하기 위한 증거 기반 방법에 초점을 맞추면 일반적으로 [부산물]로 [자기 평가 능력이 향상]됩니다(Hawkins 외. 2012).  
Returning to thinking about Medical Education, there have been strong and clear voices making these points for some years now. Although it seems paradoxical at first glance, the data is clear. As Eva, Regehr, and Gruppen (Eva et al. 2012) explain, if our efforts go into teaching SA, we may successfully train students and doctors who can more accurately self-assess (i.e. develop their SA skills/abilities), but not necessarily improve their performance. Additionally, it’s well-established that although we can help students to improve their ability to self-assess (Lopez and Kossack 2007), there is a poor correlation between improving SA and their performance (Andersen et al. 2015), insight (Ehrlinger et al. 2008), learning (Colthart et al. 2008), or patient outcomes (Al-Kadri et al. 2012). If instead, we focus on evidence-based ways of improving the performance of weaker students, their ability to self-assess will usually improve as a by-product (Hawkins et al. 2012).

10대 운전자의 시나리오를 다시 생각해본다면, 시간이 지남에 따라 이들을 관찰한다면, 도로에서 더 많은 감독을 받는 시간을 보내면 더 많은 도전에 노출되어 유능한 운전자가 되기 위해 필요한 것이 무엇인지에 대한 미묘한 관점을 얻을 수 있기를 바랍니다(그림 3-시나리오 4). 그러면 (아마도!) 초기의 잘못된 자신감을 비웃고 더 많은 경험이 쌓인 후에야 고속도로에 나설 수 있을 것입니다.
If we reconsider our teenage driver scenario and observe them over time, we would hope that spending some more supervised hours on the road would expose them to increasing challenges, helping them gain a more nuanced perspective of what it takes to be a competent driver (Figure 3—scenario 4). They would then (possibly!) laugh at their earlier misplaced self-confidence and only take to the highway when they were more experienced.

이 마지막 시나리오는 [학습자가 목표로 하는 명확하고 구체적인 표준]이 있고 [다른 사람의 의견에 반응하는 안내형 SA]입니다. 이것이 바로 학습을 위한 SA입니다.
Notice that this final scenario is guided SA where the learner has clear, specific standards to aim for and is responsive to the input of others. It is SA for learning.

왜 그렇게 많은 연구의 초점이 여전히 성과, 학습 또는 안전의 개선 측정과는 별개로 SA를 주요 예상 결과로 개선하는 데 맞춰져 있는지 궁금합니다. 또는 왜 그렇게 많은 연구자들이 SA의 제한된 신뢰성과 타당성을 고려하지 않고 프로그램의 '성공'을 평가하는 도구로 SA를 사용하는 것일까요? 이러한 [SA의 오용]은 특히 초보자나 실력이 약한 학생과 관련하여 문제가 되는데, 이들은 프로그램을 통해 실력이 향상되었다고 생각하기 때문에 이제 전문가가 되었다고 생각할 수 있습니다(Kruger and Dunning 1999). 
We wonder why the focus of so much research is still on improving SA as the primary anticipated outcome, detached from any measure of improvement in performance, learning or safety? Or why so many researchers use SA as a tool to evaluate the ‘success’ of a program, without regard to its limited reliability and validity? This misuse of SA is particularly troubling with regard to novices or weaker students, who may believe that because they have been through a program and feel they have improved, they are now an expert (Kruger and Dunning 1999).

한 가지 가능한 이유는 [SA 데이터가 양적 및 질적 형태로 수집하기 쉽기 때문]에, 한계와 역설에 익숙하지 않은 연구자들은 [SA가 간단하고 유용한 정보를 제공한다고 잘못 생각]하기 때문입니다. 또한 학생들은 자신의 SA를 묻는 질문에 의문을 제기하지 않을 것입니다. 사람들은 직관적으로 자신이 자신을 가장 잘 평가할 수 있는 위치에 있다고 믿기 때문입니다. 인간은 자신을 가장 잘 안다고 생각하는 경향이 있습니다.
One likely reason is that SA data is easy to collect, in both quantitative and qualitative forms, so researchers who are unfamiliar with its limitations and paradoxes, mistakenly think it provides straightforward and useful information. In addition, students are unlikely to question being asked for their own SA: it is intuitive for people to believe that they are in the best position to assess themselves. Humans tend to think we know ourselves best.

보다 구조적인 수준에서는 SA의 개념을 정의하기 어렵고, 그 결과 의미와 적용에 일관성이 없다는 것이 문제의 일부일 수 있습니다. 또 다른 문제는 학문적 대화의 사일로화일 수 있으며, 다른 맥락에서 병렬적으로 논의가 진행되는 것입니다[다른 곳에서 발생한 것처럼(Evans 외. 2021)]. 네트워크 분석은 주제에 대한 학문적 담론의 단절을 드러낼 수 있으며(Manlove 외. 2016), 이는 SA의 향후 연구 분야가 될 수 있습니다. 
At a more structural level, part of the problem is likely the difficulty in defining the concept of SA, with resulting in inconsistency in meaning and application. Another issue may be siloing of academic conversations, with parallel discussions happening in different contexts [as has occurred elsewhere (Evans et al. 2021)]. A  network analysis can reveal disconnects in academic discourses on topics (Manlove et al. 2016) and this may be an area of future research for SA.

마지막으로, 이 연구 논문의 비정통적 접근 방식이 의학교육의 다른 주제에도 도움이 될지 궁금합니다. SA를 도움이 되지 않는 방식으로 사용한 연구의 정도를 정량화함으로써 [의학교육자들이 SA를 측정 도구로 사용하는 것을 고려할 때 잠시 멈칫할 수 있기를 바랍니다]. SA가 부적절하게 사용되면 연구가 낭비될 수 있습니다. 의료 연구 전반에서 연구 낭비를 제한해야 한다는 요구가 증가하고 있으며(Chalmers and Glasziou 2009), 이에 대한 책임은 자금 제공자, 규제 기관, 학술지, 학술 기관, 연구자 등 다섯 가지 주요 이해관계자에게 있습니다(Moher 외. 2016). 
Finally, we wonder if our unorthodox approach in this research paper would benefit other topics in Medical Education? By quantifying the extent of research that has used SA in unhelpful ways, we hope medical educators will take a pause when considering using SA as a measurement tool. If SA is used inappropriately, their research may be wasted. Within Healthcare research more broadly there have been increasing calls to limit research waste (Chalmers and Glasziou 2009) with responsibility for this laying at the feet of five main stakeholders—funders, regulators, journals, academic institutions, and researchers (Moher et al. 2016).

이를 통해 의학교육은 우선순위가 가장 높은 주제를 연구하고, 적절한 연구 설계와 방법론을 보장하며, 연구 결과를 필요로 하는 사람들에게 접근 가능한 방식으로 연구 결과를 보고하고 배포하는 것을 최적화하는 등의 고려 사항을 통해 학문으로서의 이점을 얻을 수 있습니다. 체계적 문헌고찰과 범위 검토에서 '제외된' 연구에는 향후 연구 질문에 대한 해답이 있을 수 있는데, 이는 향후 반복되지 않기를 바라는 단점이 포함되어 있기 때문입니다. 연구 제외 이유를 체계적으로 평가함으로써 이 논문이 자체 평가에 대한 이해를 높이고 향후 이 분야의 연구를 개선하는 데 기여했기를 바랍니다.
Learning from this, Medical Education as a discipline may benefit from the following considerations: researching the highest priority topics; ensuring appropriate research designs and methodologies; and optimising reporting and dissemination of findings in accessible ways, to the people who need to see them. It may be that answers to future research questions can be found in the ‘excluded’ studies in our Systematic and Scoping reviews, as these contain the shortcomings which we want to ensure are not replicated in the future. By systematically assessing the reasons for excluding research, we hope this paper has contributed to a deeper understanding of Self-Assessment and improved future research in this area.

결론
Conclusion

SA를 탐구하는 교육자가 SA에 대한 명확한 정의를 가지고 있을 때, 이는 학생의 학습 방식에 대한 이용 가능한 증거에 부합합니다: SA는 가이드가 있고(가이드가 없는 것이 아니라), 구체적이며(포괄적인 것이 아니라), 능력이 아닌 과정으로 개념화하는 것이 가장 도움이 됩니다. 우리는 의학교육자와 학자들이 [결과 측정으로서의 자기평가의 한계]와 [자기평가가 추구해야 할 목표라는 증거가 부족하다]는 점을 이해할 것을 촉구합니다. 앞으로 나아가는 데 도움이 되는 한 걸음은 개입/프로그램의 '성공'에 대한 의미 있는 결과 측정으로 참여자 SA를 사용하는 것을 중단하는 것입니다. SA는 평생 학습 과정의 중요한 부분이지만, 학생들이 외부 측정치(학습에 대한 자기 평가)로 SA를 보정하는 데 어려움을 겪는 이유에 대한 연구는 유용한 목적이 없습니다. 마찬가지로, 정확도/보정을 개선하려는 시도 역시 잘못된 방향의 노력입니다. 프로젝트를 설계하는 연구자, 학술 기관, 자금 제공자, 저널은 더 이상 이러한 결함이 있는 방식으로 SA를 사용하는 것을 받아들이지 않는 것이 좋습니다. 그 대신 학생들의 [학습을 위한 자기평가]를 발전시키는 데 집중해야 합니다. 
When educators exploring SA have a clear definition of SA, it fits the available evidence on how students learn: SA is most helpfully conceptualised as a guided (not unguided), specific (not global), process (not ability). We call on medical educators and scholars to understand the limitations of self-assessment as an outcome measure, and the lack of evidence for it being a goal to strive for. A helpful step forward would be for us to stop using participant SA as a meaningful outcome measure of the ‘success’ of an intervention/program. While SA is an important part of the life-long learning process, research into reasons why students are poor at calibrating their SA with external measures (self-assessment of learning) serves no useful purpose. Similarly, attempts to improve their accuracy/calibration is misdirected effort. Researchers designing projects, academic institutions, funders, and journals would benefit from no longer accepting the use of SA in these flawed ways. Efforts should instead focus on progressing students’ self-assessment for learning.

우리는 SA의 [정의, 한계, 역설을 명확히 하는 것]이 의과대학 내 교육과 학습에 의미 있는 적용을 가능하게 할 것이라고 믿습니다. 이러한 한계와 역설을 인정하지 않는 연구의 범위에 대한 우리의 탐구가 향후 이 분야의 연구 낭비를 줄이는 데 도움이 되기를 바랍니다. 마지막으로, 의학교육 연구의 다른 분야에서도 마찬가지로 범위 설정 및 체계적 문헌고찰에서 '제외된' 연구를 검토하여 학문으로서 개선할 수 있는 부분에 대한 인사이트를 얻을 수 있지 않을까 생각합니다. 
We believe that clarifying definitions, limitations, and the paradoxes of SA will afford meaningful application to teaching and learning within medical schools. We hope that our exploration of the extent of research that fails to acknowledge these limitations and paradoxes will help reduce future wastage of research in this area. Finally, we wonder if other areas of medical education research would similarly benefit from examining ‘excluded’ studies from scoping and systematic reviews, to gain insights into what where we could improve as a discipline.

 


 

Med Teach. 2022 Nov;44(11):1296-1302. doi: 10.1080/0142159X.2022.2093704. Epub 2022 Jul 4.

 

Self-assessment: With all its limitations, why are we still measuring and teaching it? Lessons from a scoping review

Affiliations collapse

1Faculty of Health Sciences and Medicine, Bond University, Gold Coast, Australia.

2Translational Simulation Collaborative, Gold Coast, Australia.

PMID: 35786121

DOI: 10.1080/0142159X.2022.2093704

Abstract

Introduction: Self-Assessment (SA) is often assumed to be essential for learning, however, this assumption has been extensively rebutted. Research shows SA has significant limitations, including its lack of correlation with competence.

Method: We undertook a scoping review of SA in medical education (2011-2021) and surprisingly discovered substantial research where SA was erroneously assumed to be a valid measure of successful learning, or a skill needing to be taught. Although the initial intent of our scoping review was to explore where SA is being effectively used to advance lifelong learning, we paused to explore the extent of the problem of its misuse and resulting in wasted research, by examining excluded studies.

Results: From 1151 articles, we identified 207 which ignored the documented limitations of SA. Thirty-nine studies explored SA of learning. This research has limited utility: increasing the accuracy of SA does not improve performance or lifelong learning. One hundred and sixty-eight used SA as an outcome measure to assess a program or intervention, including 63 where self-assessed improvement in knowledge/skills was the sole measure. SA of self-confidence was measured in 62 studies. When confidence was compared with an objective measure of performance, both invariably increased, but confidence did not always align with competence when measured.

Discussion: Many researchers mistakenly assume the ability to accurately self-assess is essential for learning, so focus on teaching self-calibration rather than evidence-based methods of advancing learning. Other researchers incorrectly suppose that self-reported improvements in knowledge/skills provide evidence of the efficacy of a program/intervention. This is particularly troubling with regards to novices, who may believe that because they self-assess to have improved/gained confidence, they are now competent.

Conclusion: Our findings highlight the significant volume of research being done where SA is misunderstood and/or misused as a measurement. We posit reasons that such research continues to take place and suggest solutions moving forward.

Keywords: Self-assessment; life-long learning; medical education; research wastage.

가이드라인: 의학교육에서 임상술기 직접관찰의 할 것, 하지 말 것, 모르는 것 (Perspect Med Educ, 2017)
Guidelines: The do’s, don’ts and don’t knows of direct observation of clinical skills in medical education
Jennifer R. Kogan1 · Rose Hatala2 · Karen E.Hauer3 · Eric Holmboe4

 

 

소개
Introduction

임상 술기를 직접 관찰하는 것은 역량 기반 의학교육의 핵심 평가 전략이지만, 모든 졸업생이 필수 영역에서 역량을 갖출 수 있도록 하는 것은 항상 의료 전문직 교육에 필수적이었습니다[1, 2]. 이 가이드라인에서는 다음과 같은 역량에 대한 정의를 사용합니다: '의학교육 또는 실습의 정의된 단계에서 특정 맥락에서 모든 영역에서 요구되는 능력을 보유하는 것[1]'. 이제 교육 프로그램과 전문과목은 관찰 및 평가할 수 있는 필수 역량, 역량 구성 요소, 발달 이정표, 성과 수준 및 위탁 가능한 전문 활동(EPA)을 정의했습니다. 그 결과, 학습자(의대생, 대학원 또는 대학원 수련의)가 의미 있고 진정성 있고 현실적인 환자 치료 및 임상 활동에 참여하는 동안 감독자가 관찰하는 직접 관찰이 점점 더 강조되는 평가 방법[3, 4]이 되고 있습니다[4, 5]. [직접 참관]은 의학교육 연락 위원회, 의학전문대학원 교육 인증위원회, 영국 파운데이션 프로그램과 같은 의학교육 인증 기관에서 요구합니다[6,7,8]. 그러나 그 중요성에도 불구하고 임상 술기에 대한 직접 관찰은 드물고 관찰의 질이 떨어질 수 있습니다[9,10,11]. 양질의 직접 관찰이 부족하면 학습에 중대한 영향을 미칩니다. 형성적 관점에서 학습자는 임상 술기 개발을 지원하기 위한 피드백을 받지 못합니다. 또한 학습자의 역량과 궁극적으로 환자에게 제공되는 치료의 질에 대한 종합적인 평가도 위태롭습니다.  
While direct observation of clinical skills is a key assessment strategy in competency-based medical education, it has always been essential to health professions education to ensure that all graduates are competent in essential domains [1, 2]. For the purposes of these guidelines, we use the following definition of competent: ‘Possessing the required abilities in all domains in a certain context at a defined stage of medical education or practice [1].’ Training programs and specialties have now defined required competencies, competency components, developmental milestones, performance levels and entrustable professional activities (EPAs) that can be observed and assessed. As a result, direct observation is an increasingly emphasized assessment method [3, 4] in which learners (medical students, graduate or postgraduate trainees) are observed by a supervisor while engaging in meaningful, authentic, realistic patient care and clinical activities [4, 5]. Direct observation is required by medical education accrediting bodies such as the Liaison Committee on Medical Education, the Accreditation Council of Graduate Medical Education and the UK Foundation Program [6,7,8]. However, despite its importance, direct observation of clinical skills is infrequent and the quality of observation may be poor [9,10,11]. Lack of high quality direct observation has significant implications for learning. From a formative perspective, learners do not receive feedback to support the development of their clinical skills. Also at stake is the summative assessment of learners’ competence and ultimately the quality of care provided to patients.

이 백서에서 제안하는 가이드라인은 임상 술기 직접 관찰에 관한 문헌의 종합을 기반으로 하며, 학습자 감독자와 의학교육 임상 교육 프로그램을 담당하는 교육 리더 모두에게 실질적인 권장 사항을 제공합니다. 이 백서의 목적은

  • 1) 일선 교사, 학습자 및 교육 리더가 직접 관찰의 질과 빈도를 개선하도록 돕고,
  • 2) 직접 관찰에 대한 현재의 관점을 공유하며,
  • 3) 이 분야를 발전시키기 위한 향후 연구 의제에 정보를 제공할 수 있는 이해의 격차를 파악하는 것입니다.

The guidelines proposed in this paper are based on a synthesis of the literature on direct observation of clinical skills and provide practical recommendations for both supervisors of learners and the educational leaders responsible for medical education clinical training programs. The objectives of this paper are to

  • 1) help frontline teachers, learners and educational leaders improve the quality and frequency of direct observation;
  • 2) share current perspectives about direct observation; and
  • 3) identify gaps in understanding that could inform future research agendas to move the field forward.

방법
Methods

이 지침은 직접 관찰에 대한 연구 경험이 있고 임상 환경에서 학부(의대생) 및 대학원/대학원(레지던트/펠로우) 학습자를 가르치고, 관찰하고, 피드백을 제공한 실무 경험이 있는 2개국 4명의 의학교육자의 전문가 의견과 함께 기존 증거에 대한 서술적 검토[12]로 이루어졌습니다. 반복적인 프로세스를 통해 가이드라인을 개발했습니다. 특히 병력 청취, 신체 검사, 상담 및 시술 기술 관찰과 같이 학습자가 환자 및 그 가족과 상호작용하는 모습을 직접 관찰하는 것으로 범위를 제한했습니다. 고품질의 직접 관찰을 촉진하고 보장하는 권장 사항을 만들기 위해 일선 교사/감독자, 학습자, 교육 리더 및 상황을 구성하는 기관에 초점을 맞추었습니다.

This is a narrative review [12] of the existing evidence coupled with the expert opinion of four medical educators from two countries who have research experience in direct observation and who have practical experience teaching, observing, and providing feedback to undergraduate (medical student) and graduate/postgraduate (resident/fellow) learners in the clinical setting. We developed the guidelines using an iterative process. We limited the paper’s scope to direct observation of learners interacting with patients and their families, particularly observation of history taking, physical exam, counselling and procedural skills. To create recommendations that promote and assure high quality direct observation, we focused on the frontline teachers/supervisors, learners, educational leaders, and the institutions that constitute the context. 

형성 평가와 총괄 평가 모두에 사용되는 직접 관찰을 다루었습니다. 평가의 단계는 연속적이지만,

  • [형성 평가]는 학습자 성취도에 대한 증거를 도출하고 해석하여 교사와 학습자가 수업의 다음 단계에 대한 결정을 내리는 데 사용하는 저부담의 평가로 정의하고,
  • [총괄 평가]는 행정적 결정의 주요 목적(예: 진도 진행 여부, 졸업 여부 등)을 위해 학습자를 평가하도록 고안된 고부담의 평가로 정의합니다[13].

다음은 제외했습니다

  • 1) 모의 진료, 비디오 녹화 진료 및 기타 기술(예: 프레젠테이션 기술, 전문가 간 팀 기술 등)에 대한 관찰,
  • 2) 실무 의사에 초점을 맞춘 직접 관찰,
  • 3) 다른 형태의 작업장 기반 평가(예: 차트 감사)

직접 관찰의 중요한 측면은 관찰 후 학습자에게 피드백을 제공하는 것이지만, 피드백 가이드라인이 이미 발표되었기 때문에 피드백에 초점을 맞춘 가이드라인의 수를 제한하기로 합의했습니다[14].
We addressed direct observation used for both formative and summative assessment. Although the stakes of assessment are a continuum, we define

  • formative assessment as lower-stakes assessment where evidence about learner achievement is elicited, interpreted and used by teachers and learners to make decisions about next steps in instruction, while
  • summative assessment is a higher-stakes assessment designed to evaluate the learner for the primary purpose of an administrative decision (i. e. progress or not, graduate or not, etc.) [13].

We excluded

  • 1) observation of simulated encounters, video recorded encounters, and other skills (e. g. presentation skills, inter-professional team skills, etc.);
  • 2) direct observation focused on practising physicians; and
  • 3) other forms of workplace-based assessment (e. g. chart audit).

Although an important aspect of direct observation is feedback to learners after observation, we agreed to limit the number of guidelines focused on feedback because a feedback guideline has already been published [14].

이러한 매개변수를 정의한 후 각 저자는 아래에 정의된 대로 해야 할 일, 하지 말아야 할 일 및 모르는 일 목록을 독립적으로 생성했습니다. 특히 '모르겠다'는 답변이 있을 경우 교육 관행을 바꿀 수 있는 항목에 초점을 맞추었습니다. 일련의 반복적인 토론을 통해 [해야 할 일, 하지 말아야 할 일, 모르는 일] 목록에 합의할 때까지 목록을 검토하고 토론하고 다듬었습니다. 그런 다음 항목은 4명의 저자가 나누어 담당했으며, 각 저자는 할당된 항목에 대한 찬성과 반대의 증거를 식별할 책임이 있었습니다. 주로 임상 술기의 직접 관찰에 초점을 맞춘 근거를 찾았지만, 근거가 부족한 경우 다른 평가 양식과 관련된 근거도 고려했습니다. 그런 다음 모든 저자가 증거 요약을 공유했습니다. 필요한 경우 근거에 따라 항목을 재분류하고 상충되는 근거가 있는 항목은 '잘 모름' 범주로 이동했습니다. 그룹 합의를 통해 이전 지침의 강도 지표를 사용하여 각 지침을 뒷받침하는 근거의 강도를 결정했습니다([14], 표 1). 직접 관찰이 아닌 평가 방식을 추정하여 얻은 증거는 보통 이상의 지지를 받지 못했습니다. 
With these parameters defined, each author then independently generated a list of Do’s, Don’ts and Don’t Knows as defined below. We focused on Don’t Knows which, if answered, might change educational practice. Through a series of iterative discussions, the lists were reviewed, discussed and refined until we had agreed upon the list of Do’s, Don’ts and Don’t Knows. The items were then divided amongst the four authors; each author was responsible for identifying the evidence for and against assigned items. We primarily sought evidence explicitly focused on direct observation of clinical skills; however, where evidence was lacking, we also considered evidence associated with other assessment modalities. Summaries of evidence were then shared amongst all authors. We re-categorized items when needed based on evidence and moved any item for which there was conflicting evidence to the Don’t Know category. We used group consensus to determine the strength of evidence supporting each guideline using the indicators of strength from prior guidelines ([14]; Table 1). We did not give a guideline higher than moderate support when evidence came from extrapolation of assessment modalities other than direct observation.

결과
Results

원래 목록에는 개별 감독자, 학습자, 교육 프로그램을 담당하는 교육 리더의 세 그룹에 초점을 맞춘 지침이 있었습니다. 이 초기 목록에는 67개(해야 할 일 35개, 하지 말아야 할 일 16개, 잘 모르겠다 16개)의 항목이 있었습니다. 그룹 토론을 통해 유사하거나 중복되는 항목을 합쳐 33개로 줄였고, 중요하지 않다고 판단된 항목은 2개만 삭제했습니다. 교육 프로그램을 담당하는 교육 리더를 위한 지침에 학습자 중심의 항목을 포함시켜 중복 항목을 줄이고, 교육 리더가 학습자가 직접 관찰하고 피드백을 학습 전략의 일부로 통합하도록 활성화하는 학습 문화를 조성하는 것이 얼마나 중요한지 강조하기로 했습니다.
Our original lists had guidelines focused on three groups: individual supervisors, learners, and educational leaders responsible for training programs. This initial list of Do’s, Don’ts and Don’t Knows numbered 67 (35 Do’s, 16 Don’ts, 16 Don’t Knows). We reduced this to the 33 presented by combining similar and redundant items, with only two being dropped as unimportant based on group discussion. We decided to embed items focused on learners within the guidelines for educational leaders responsible for training programs to reduce redundancy and to emphasize how important it is for educational leaders to create a learning culture that activates learners to seek direct observation and incorporate feedback as part of their learning strategies.

증거를 검토한 결과, 원래 '해야 할 일'로 정의된 4개 항목이 '모름'으로 이동되었습니다. 해야 할 일, 하지 말아야 할 일 및 잘 모름의 최종 목록은 개별 감독자에 초점을 맞춘 지침(표 2)과 교육 프로그램을 담당하는 교육 리더에 초점을 맞춘 지침(표 3)의 두 섹션으로 나뉩니다. 이 원고의 나머지 부분에서는 각 지침을 뒷받침하는 주요 근거와 이용 가능한 문헌에 근거한 지침의 강점을 제공합니다.
After review of the evidence, four items originally defined as a Do were moved to a Don’t Know. The final list of Do’s, Don’ts and Don’t Knows is divided into two sections: guidelines that focus on individual supervisors (Table 2) and guidelines that focus on educational leaders responsible for training programs (Table 3). The remainder of this manuscript provides the key evidence to support each guideline and the strength of the guideline based on available literature.

 

 

직접 관찰을 수행하는 개별 임상 감독자를 위한 근거가 포함된 가이드라인
Guidelines with supporting evidence for individual clinical supervisors doing direct observation

개별 슈퍼바이저가 해야 할 일
Do’s for individual supervisors

지침 1. 실제 임상 현장에서 실제 임상 업무를 관찰합니다.
Guideline 1. Do observe authentic clinical work in actual clinical encounters.

직접 관찰은 임상 현장에서 이루어지는 평가로서 임상 역량 평가를 위한 밀러 피라미드의 최상단에 있는 '해야 한다'의 평가를 지원합니다[15, 16]. 교육과 평가의 목표는 임상 환경에서 감독 없이 실습할 수 있는 의사를 배출하는 것이므로, 학습자는 임상 역량을 입증해야 하는 환경에서 관찰되어야 합니다. 실제 임상 상황은 시뮬레이션이나 역할극보다 더 복잡하고 미묘하며 다양한 맥락을 수반하는 경우가 많으므로 실제 임상 진료를 직접 관찰하면 이러한 복잡성을 탐색하는 데 필요한 임상 기술을 관찰할 수 있습니다[17].
Direct observation, as an assessment that occurs in the workplace, supports the assessment of ‘does’ at the top of Miller’s pyramid for assessing clinical competence [15, 16]. Because the goal of training and assessment is to produce physicians who can practise in the clinical setting unsupervised, learners should be observed in the setting in which they need to demonstrate clinical competence. Actual clinical encounters are often more complex and nuanced than simulations or role plays and involve variable context; direct observation of actual clinical care enables observation of the clinical skills required to navigate this complexity [17].

학습자와 교사는 임상 활동 참여를 통한 실습이 학습의 핵심임을 인식하고 있습니다[18,19,20]. [진정성]은 상황별 학습의 핵심 요소이며, 학습이 실제와 가까울수록 기술을 더 빠르고 효과적으로 학습할 수 있습니다[21, 22]. 또한 학습자는 실제 환자와의 만남과 그 환경이 시뮬레이션된 만남보다 더 자연스럽고 유익하며 흥미진진하다고 느끼며, 시뮬레이션된 만남보다 실제 만남에 대해 더 많은 준비를 하고 자율 학습에 대한 더 강한 동기를 표현할 수 있습니다[23]. 학습자는 시간이 지남에 따라 의미 있는 임상 진료에 참여하는 것을 관찰한 후 발생하는 평가와 피드백을 중요하게 생각합니다[24,25,26]. [진정한 만남의 예]로는 임상팀이 이미 병력을 확보한 환자에 대해 학습자가 병력을 작성하는 것을 지켜보는 것보다, 학습자가 초기 병력을 작성하는 것을 지켜보는 것을 들 수 있습니다. 
Learners and teachers recognize that hands-on-learning via participation in clinical activities is central to learning [18,19,20]. Authenticity is a key aspect in contextual learning; the closer the learning is to real life, the more quickly and effectively skills can be learned [21, 22]. Learners also find real patient encounters and the setting in which they occur more natural, instructive and exciting than simulated encounters; they may prepare themselves more for real versus simulated encounters and express a stronger motivation for self-study [23]. Learners value the assessment and feedback that occurs after being observed participating in meaningful clinical care over time [24,25,26]. An example of an authentic encounter would be watching a learner take an initial history rather than watching the learner take a history on a patient from whom the clinical team had already obtained a history.

감독자는 학습자를 실제 상황에서 관찰하려고 노력할 수 있지만, 저자의 경험에 따르면 학습자는 관찰을 받을 때 비실제적인 실습(예: 환자 병력을 기록할 때 전자 건강 기록에 입력하지 않거나 더 집중적인 검사가 적절한데도 종합적인 신체 검사를 하는 등)을 기본값으로 설정할 수 있습니다. 관찰자 효과가 성과에 미치는 영향에 대해서는 논란의 여지가 있지만(호손 효과라고도 함)[11, 27], 관찰자는 학습자가 실제 업무 행동에 대한 피드백을 받을 수 있도록 학습자가 '평소에 하던 대로' 하도록 격려해야 합니다. 관찰자는 [호손 효과]에 대한 두려움을 임상 환경에서 학습자를 관찰하지 않는 이유로 사용해서는 안 됩니다[지침 18 참조].
Although supervisors may try to observe learners in authentic situations, it is the authors’ experience that learners may default to inauthentic practice when being observed (for example, not typing in the electronic health record when taking a patient history or doing a comprehensive physical exam when a more focused exam is appropriate). While the impact of observer effects on performance is controversial (known as the Hawthorne effect) [11, 27], observers should encourage learners to ‘do what they would normally do’ so that learners can receive feedback on their actual work behaviours. Observers should not use fear of the Hawthorne effect as a reason not to observe learners in the clinical setting [see Guideline 18].

지침 2. 관찰하기 전평가의 후과 및 성과를 포함하여 목표를 논의하고 기대치를 설정하여 학습자를 준비시킵니다.
Guideline 2.
Do prepare the learner prior to observation by discussing goals and setting expectations, including the consequences and outcomes of the assessment.

목표 설정에는 학습자와 감독자 간의 협상이 포함되어야 하며, 가능한 경우 직접 관찰에서는 학습자가 가장 필요하다고 느끼는 것에 초점을 맞추어야 합니다. 학습자의 목표는 어떤 활동에 참여할지, 그리고 그러한 활동에 대한 접근 방식에 대해 학습자가 선택할 수 있도록 동기를 부여합니다. 잘 수행하고 '잘 보이는 것'보다는 [학습과 개선을 지향하는 목표]는 학습자가 직접 관찰에 수반될 수 있는 피드백과 교육을 더 잘 수용할 수 있도록 합니다[28, 29]. 학습자가 직접 관찰을 언제, 무엇을 위해 수행할지 자율적으로 결정하면 관찰에 대한 동기가 강화되고 수행 목표에서 학습 목표로 초점을 전환할 수 있습니다[30, 31]. 교사는 학습자의 목표를 묻고 이를 해결하기 위해 교육 및 관찰의 초점을 조정함으로써 이러한 자율성을 촉진할 수 있습니다. 예를 들어, 동일한 임상 상황 내에서 감독자는 학습자가 관찰 및 피드백의 초점(병력 기록, 의사소통 또는 환자 관리)을 선택할 수 있도록 함으로써 학습자를 위한 직접 관찰의 관련성을 높일 수 있습니다. 학습자의 목표는 프로그램 목표, 역량(예: 마일스톤) 및 특정 개인의 필요와 일치해야 합니다[지침 17 참조]. 모든 수준의 학습자에게 목표를 설정하도록 요청하면 어려움을 겪는 학습자에게 초점을 맞추기보다는 모든 학습자에게 개선의 중요성을 정상화하는 데 도움이 됩니다. 관찰자와 학습자 간의 협력적 접근 방식은 아래에 설명된 자기조절 학습 주기의 첫 번째 단계인 학습 계획을 촉진합니다[32]. 학습자는 특정 개인화된 목표를 식별하고 이를 위해 노력하라는 요청을 받아들이고, 그렇게 함으로써 학습에 대한 책임감을 심어줍니다[31]. 
Setting goals should involve a negotiation between the learner and supervisor and, where possible, direct observation should include a focus on what learners feel they most need. Learners’ goals motivate their choices about what activities to engage in and their approach to those activities. Goals oriented toward learning and improvement rather than performing well and ‘looking good’ better enable learners to embrace the feedback and teaching that can accompany direct observation [28, 29]. Learners’ autonomy to determine when and for what direct observation will be performed can enhance their motivation to be observed and shifts their focus from performance goals to learning goals [30, 31]. Teachers can foster this autonomy by soliciting learners’ goals and adapting the focus of their teaching and observation to address them. For example, within the same clinical encounter, a supervisor can increase the relevance of direct observation for the learner by allowing the learner to select the focus of observation and feedback-history taking, communication, or patient management. A learner’s goals should align with program objectives, competencies (e. g. milestones) and specific individual needs [see Guideline 17]. Asking learners at all levels to set goals helps normalize the importance of improvement for all learners rather than focusing on struggling learners. A collaborative approach between the observer and learner fosters the planning of learning, the first step in the self-regulated learning cycle described below [32]. Learners are receptive to being asked to identify and work towards specific personalized goals, and doing so instills accountability for their learning [31].

관찰하기 전에 관찰자는 평가의 후과에 대해서도 학습자와 논의해야 합니다. 관찰이 높은 수준의 평가가 아닌 피드백을 위해 사용되는 경우 이를 명확히 하는 것이 중요합니다. 학습자는 직접 관찰을 통해 얻을 수 있는 형성적 학습 기회의 이점을 인식하지 못하는 경우가 많으므로 이점을 설명하는 것이 도움이 될 수 있습니다[33]. 
Prior to observation, observers should also discuss with the learner the consequences of the assessment. It is important to clarify when the observation is being used for feedback as opposed to high-stakes assessment. Learners often do not recognize the benefits of the formative learning opportunities afforded by direct observation, and hence explaining the benefits may be helpful [33].

가이드라인 3. 학습자의 자기조절 학습 능력배양합니다.
Guideline 3. 
Do cultivate learners’ skills in self-regulated learning.

학습을 향상시키기 위한 직접 관찰을 위해 학습자는 개별 목표를 달성하기 위해 받은 피드백의 유용성을 극대화하는 전략을 사용할 준비가 되어 있어야 합니다. 자신의 학습 요구와 지식 및 성과 향상에 필요한 행동에 대한 인식은 직접 관찰의 가치를 최적화합니다. 자기조절 학습은 다음의 지속적인 주기를 설명합니다[32].  

  • 1) 학습을 위한 계획 수립,
  • 2) 활동 중 자기 모니터링 및 학습과 성과를 최적화하기 위해 필요한 조정,
  • 3) 활동 후 목표 달성 여부 또는 어려움의 위치와 이유에 대한 성찰

직접 관찰의 맥락에서의 예는 그림 1에 나와 있습니다. 직접 관찰의 맥락에서 발생하는 것처럼 활동 중에 소량의 특정 피드백을 제공하면 자기조절 학습이 극대화됩니다[34]. 훈련생은 피드백을 구함으로써 스스로 평가한 성과를 보강하는 정도가 다양합니다[35]. [직접 관찰과 피드백을 결합]하면 학습자가 받는 피드백의 양을 늘림으로써 이러한 문제를 극복하는 데 도움이 될 수 있습니다[프로그램 지침 18 참조]. 

For direct observation to enhance learning, the learner should be prepared to use strategies that maximize the usefulness of feedback received to achieve individual goals. Awareness of one’s learning needs and actions needed to improve one’s knowledge and performance optimize the value of being directly observed. Self-regulated learning describes an ongoing cycle of

  • 1) planning for one’s learning;
  • 2) self-monitoring during an activity and making needed adjustments to optimize learning and performance; and
  • 3) reflecting after an activity about whether a goal was achieved or where and why difficulties were encountered [32].

An example in the context of direct observation is shown in Fig. 1. Self-regulated learning is maximized with provision of small, specific amounts of feedback during an activity [34] as occurs in the context of direct observation. Trainees vary in the degree to which they augment their self-assessed performance by seeking feedback [35]. Direct observation combined with feedback can help overcome this challenge by increasing the amount of feedback learners receive [see Program Guideline 18].

가이드라인 4. 중요한 임상 술기는 대리 정보를 사용하지 말고 직접 관찰을 통해 평가합니다.
Guideline 4. 
Do assess important clinical skills via direct observation rather than using proxy information.

감독자는 평가할 술기를 직접 관찰해야 합니다. 실제로 감독자는 학습자의 임상 술기를 평가할 때 대리 정보를 기반으로 하는 경우가 많습니다. 예를 들어, 감독자는 학습자가 환자를 진찰하는 것을 듣고 병력 및 신체 검사 기술을 추론하거나 학습자와 팀과의 상호 작용을 기반으로 환자와의 대인관계 기술을 추론하는 경우가 많습니다[36]. 직접 관찰은 임상 수행 평가의 질, 의미, 신뢰성 및 타당성을 향상시킵니다[37]. 감독자와 학습자는 직접 관찰에 기반한 평가를 효과적인 평가자의 가장 중요한 특성 중 하나로 간주합니다[38]. 또한 학습자는 직접 관찰에 기반한 교육생에 대한 직접적인 정보에 근거할 때 교육 중 평가가 가치 있고 정확하며 신뢰할 수 있다고 생각할 가능성이 더 높습니다[39]. 예를 들어, 로테이션이 끝날 때 평가할 술기인 경우, 감독자는 로테이션이 진행되는 동안 학습자가 병력을 작성하는 것을 여러 번 직접 관찰해야 합니다. 
Supervisors should directly observe skills they will be asked to assess. In reality, supervisors often base their assessment of a learner’s clinical skills on proxy information. For example, supervisors often infer history and physical exam skills after listening to a learner present a patient or infer interpersonal skills with patients based on learner interactions with the team [36]. Direct observation improves the quality, meaningfulness, reliability and validity of clinical performance ratings [37]. Supervisors and learners consider assessment based on direct observation to be one of the most important characteristics of effective assessors [38]. Learners are also more likely to find in-training assessments valuable, accurate and credible when they are grounded in first-hand information of the trainee based on direct observation [39]. For example, if history taking is a skill that will be assessed at the end of a rotation, supervisors should directly observe a learner taking a history multiple times over the rotation.

가이드라인 5. 실습을 방해하지 않고 관찰합니다.
Guideline 5. 
Do observe without interrupting the encounter.


관찰자는 가능한 한 학습자가 방해받지 않고 면담을 진행할 수 있도록 해야 합니다. 학습자는 자율성과 점진적인 독립성을 중요하게 생각합니다[40, 41]. 많은 학습자는 이미 직접 관찰이 학습, 자율성 및 환자와의 관계에 방해가 된다고 느끼고 있으며, 방해는 이러한 우려를 더욱 악화시킵니다[42, 43]. 학습자가 환자 치료에 참여하는 동안 방해하면 학습자의 구두 사례 발표(임상 추론에 대한 직접 관찰의 예)를 중단한 감독자에 대한 연구에서 볼 수 있듯이 중요한 정보가 누락될 수 있습니다[44]. 또한 평가자는 종종 자신의 존재가 학습자와 환자의 관계를 손상시킬 수 있다고 걱정합니다. 관찰자는 환자가 학습자를 우선적으로 바라볼 수 있도록 환자의 주변 시야에 위치하여 직접 관찰하는 동안 방해 요소를 최소화할 수 있습니다. 이러한 위치에서도 관찰자는 [학습자와 환자의 얼굴을 모두 볼 수 있어야 비언어적 단서를 식별]할 수 있습니다. 또한 관찰자는 학습자가 심각한 오류를 범하지 않는 한 학습자와 환자의 상호 작용을 방해하지 않음으로써 관찰자의 존재감을 최소화할 수 있습니다. 관찰자는 과도한 움직임이나 소음(예: 펜 두드리기)과 같이 주의를 산만하게 하는 방해 요소를 피해야 합니다. 
Observers should enable learners to conduct encounters uninterrupted whenever possible. Learners value autonomy and progressive independence [40, 41]. Many learners already feel that direct observation interferes with learning, their autonomy and their relationships with patients, and interruptions exacerbate these concerns [42, 43]. Interrupting learners as they are involved in patient care can lead to the omission of important information as shown in a study of supervisors who interrupted learners’ oral case presentations (an example of direct observation of clinical reasoning) [44]. Additionally, assessors often worry that their presence in the room might undermine the learner-patient relationship. Observers can minimize intrusion during direct observation by situating themselves in the patient’s peripheral vision so that the patient preferentially looks at the learner. This positioning should still allow the observer to see both the learner’s and patient’s faces to identify non-verbal cues. Observers can also minimize their presence by not interrupting the learner-patient interaction unless the learner makes egregious errors. Observers should avoid distracting interruptions such as excessive movement or noises (e. g. pen tapping).

가이드라인 6. 인지적 편견인상 형성 및 암묵적 편견이 관찰 중에 도출된 추론에 영향을 미칠 수 있음을 인식합니다.
Guideline 6. 
Do recognize that cognitive bias, impression formation and implicit bias can influence inferences drawn during observation.

직접 관찰을 통해 도출된 평가의 타당성에는 여러 가지 위협이 있습니다. 평가자는 학습자를 관찰하기 시작하는 순간부터 즉각적인 인상을 형성하며(종종 적은 정보에 근거하여), 종종 빠르게(몇 분 내에) 수행 능력을 판단할 수 있다고 생각합니다[45, 46]. 이러한 빠른 판단, 즉 인상은 개인이 사람의 성격이나 행동에 대한 정보를 인지하고, 조직화하고, 통합하는 데 도움이 됩니다[47]. 인상 형성 문헌에 따르면 이러한 초기 판단이나 추론은 무의식적으로 빠르게 이루어지며 향후 상호 작용, 사람에 대해 기억하는 것, 미래 행동에 대해 예측하는 것에 영향을 미칠 수 있습니다[48]. 또한 학습자의 역량에 대한 판단은 다른 학습자와의 상대적 비교(대조 효과)에 의해 영향을 받을 수 있습니다[49]. 예를 들어, 실력이 부족한 학습자를 관찰한 후 한계 실력의 학습자를 관찰한 감독자는 이전에 실력이 뛰어난 학습자를 관찰했을 때보다 한계 실력의 학습자에 대해 더 호의적인 인상을 가질 수 있습니다. 관찰자는 이러한 편견을 인식하고 관찰된 행동에 근거하여 판단할 수 있도록 충분히 오래 관찰해야 합니다. 감독자는 [추론이 높은 인상]보다는 [추론이 낮고 관찰 가능한 행동]에 초점을 맞춰야 합니다. 예를 들어 학습자가 팔짱을 끼고 서서 나쁜 소식을 전하는 경우, 관찰 가능한 행동은 학습자가 팔짱을 끼고 서 있다는 것입니다. 이러한 행동은 상황에 대한 공감 부족 또는 불편함을 나타낸다는 추론 인상이 높습니다. 관찰자는 자신의 높은 수준의 추론이 정확하다고 가정해서는 안 됩니다. 오히려 비언어적 의사소통의 일부로서 팔짱을 끼는 것이 무엇을 의미하는지 학습자와 함께 탐구해야 합니다[가이드라인 16 참조]. 
There are multiple threats to the validity of assessments derived from direct observation. Assessors develop immediate impressions from the moment they begin observing learners (often based on little information) and often feel they can make a performance judgment quickly (within a few minutes) [45, 46]. These quick judgments, or impressions, help individuals perceive, organize and integrate information about a person’s personality or behaviour [47]. Impression formation literature suggests that these initial judgments or inferences occur rapidly and unconsciously and can influence future interactions, what is remembered about a person and what is predicted about their future behaviours [48]. Furthermore, judgments about a learner’s competence may be influenced by relative comparisons to other learners (contrast effects) [49]. For example, a supervisor who observes a learner with poor skills and then observes a learner with marginal skills may have a more favourable impression of the learner with marginal skills than if they had previously observed a learner with excellent skills. Observers should be aware of these biases and observe long enough so that judgments are based on observed behaviours. Supervisors should focus on low inference, observable behaviours rather than high inference impressions. For example, if a learner is delivering bad news while standing up with crossed arms, the observable behaviour is that the learner is standing with crossed arms. The high inference impression is that this behaviour represents a lack of empathy or discomfort with the situation. Observers should not assume their high-level inference is accurate. Rather they should explore with the learner what crossed arms can mean as part of non-verbal communication [see Guideline 16].

가이드라인 7. 관찰 후 피드백은 관찰 가능한 행동에 초점을 맞춰 제공하세요.
Guideline 7. 
Do provide feedback after observation focusing on observable behaviours.

직접 관찰 후 피드백은 이전에 발표된 모범 사례를 따라야 합니다[14]. 직접 관찰은 행동 계획과 관련된 시기적절한 행동 기반 피드백을 동반할 때 학습자가 더 잘 받아들일 수 있습니다[50]. 직접 관찰 후 피드백은 학습자의 즉각적인 우려 사항을 해결하고, 구체적이고 가시적이며, 학습자가 개선을 위해 앞으로 다르게 수행해야 할 사항을 이해하는 데 도움이 되는 정보를 제공할 때 가장 의미가 있습니다[31, 51]. 긍정적인 피드백은 학습자의 자신감을 향상시켜 학습자가 더 많은 관찰과 피드백을 구하도록 유도하기 때문에 학습자가 잘한 것을 설명하는 것은 중요합니다[31]. 피드백은 직접 대면하여 제공할 때 가장 효과적이므로 감독자는 대면 토론 없이 단순히 평가 양식에 피드백을 문서화하는 것을 피해야 합니다. 
Feedback after direct observation should follow previously published best practices [14]. Direct observation is more acceptable to learners when it is accompanied by timely, behaviourally based feedback associated with an action plan [50]. Feedback after direct observation is most meaningful when it addresses a learner’s immediate concerns, is specific and tangible, and offers information that helps the learner understand what needs to be done differently going forward to improve [31, 51]. Describing what the learner did well is important because positive feedback seems to improve learner confidence which, in turn, prompts the learner to seek more observation and feedback [31]. Feedback is most effective when it is given in person; supervisors should avoid simply documenting feedback on an assessment form without an in-person discussion.

지침 8. 학습자가 피드백을 통합할 수 있도록 종단적으로 관찰합니다.
Guideline 8. 
Do observe longitudinally to facilitate learners’ integration of feedback.

교수자가 학습자를 시간이 지남에 따라 반복적으로 관찰하면 학습이 촉진되며, 이를 통해 전문성 개발에 대한 더 나은 그림을 그릴 수 있습니다. 학습자는 자신의 성과를 되돌아보고 [종적 관계]에서 학습 목표 및 목표 달성에 대해 감독자와 논의할 수 있을 때 매우 만족합니다[25, 31]. 종적 관계는 학습자가 자신의 학습 진행 상황을 목격하고 학습자로서의 더 넓은 관점에서 피드백을 제공할 수 있는 기회를 제공합니다[25]. 지속적인 관찰은 감독자가 학습자의 역량과 한계를 평가하는 데 도움이 될 수 있으며, 이를 통해 학습자에게 앞으로 얼마나 많은 감독이 필요한지 알 수 있습니다[52]. 유능하게 임상 활동을 수행하는 것이 관찰된 학습자에게 다음 임상 활동에서 더 큰 독립성을 가지고 이러한 활동을 수행할 수 있는 권리가 부여되면 자율성이 강화됩니다[53]. 숙련된 임상 교사는 개별 학습자의 목표와 필요에 맞게 교육을 조정하는 기술을 습득하며, 직접 관찰은 학습자 중심의 교육 및 감독 접근 방식에서 중요한 구성 요소입니다[54]. 동일한 감독자가 학습자를 종적으로 관찰할 수 없는 경우, 여러 교수진이 프로그램 수준에서 이러한 관찰 순서를 수행하는 것이 중요합니다.
Learning is facilitated by faculty observing a learner repeatedly over time, which also enables a better picture of professional development to emerge. Learners appreciate when they can reflect on their performance and, working in a longitudinal relationship, discuss learning goals and the achievement of those goals with a supervisor [25, 31]. Longitudinal relationships afford learners the opportunity to have someone witness their learning progression and provide feedback in the context of a broader view of them as a learner [25]. Ongoing observation can help supervisors assess a learner’s capabilities and limitations, thereby informing how much supervision the learner needs going forward [52]. Autonomy is reinforced when learners who are observed performing clinical activities with competence are granted the right to perform these activities with greater independence in subsequent encounters [53]. Experienced clinical teachers gain skill in tailoring their teaching to an individual learner’s goals and needs; direct observation is a critical component of this learner-centred approach to teaching and supervision [54]. If the same supervisor cannot observe a learner longitudinally, it is important that this sequence of observations occurs at the programmatic level by multiple faculty.

가이드라인 9. 많은 학습자가 직접 관찰을 거부한다는 점을 인식하고 이러한 주저함을 극복할 수 있는 전략을 준비합니다.
Guideline 9. 
Do recognize that many learners resist direct observation and be prepared with strategies to try to overcome their hesitation.

일부 학습자는 직접 관찰이 유용하다고 생각하지만[24, 55], 많은 학습자는 직접 관찰을 (주로 사용되는 평가 도구와 무관하게) '체크 박스 연습' 또는 커리큘럼상의 의무로 간주합니다[33, 56]. 학습자는 여러 가지 이유로 직접 관찰을 거부할 수 있습니다. 학습자는 직접 관찰이 [불안을 유발하고 불편하며 스트레스가 많고 인위적이라고 생각]할 수 있습니다[31, 43, 50, 57, 58]. 학습자의 저항은 교수자가 자신을 관찰하기에는 너무 바쁘고[43], 관찰할 시간이 있는 교수자를 찾는 데 어려움을 겪을 것이라는 믿음에서 비롯될 수도 있습니다[59]. 많은 학습자는 [직접 관찰이 피드백이 아닌 높은 수준의 평가에만 사용될 때] 교육적 가치가 거의 없다고 생각하며[60], 교육 및 개선 계획이 포함된 피드백 없이는 직접 관찰이 유용하지 않다고 생각합니다[59]. 한 연구에서는 미니 CEX의 일환으로 100회 이상의 피드백 세션을 오디오 녹음한 결과 교수진이 학습자와 함께 실행 계획을 세우는 데 거의 도움이 되지 않는 것으로 나타났습니다[61]. 학습자는 [교육 도구로서의 직접 관찰]과 [평가 방법으로서의 직접 관찰] 사이에 충돌이 있다고 인식합니다[57, 60]. 많은 학습자는 직접 관찰이 학습, 자율성, 효율성 및 환자와의 관계에 방해가 된다고 생각합니다[42, 43]. 또한 학습자는 자신의 학습을 촉진하기 위해 어려운 상황을 독립적으로 처리하는 것을 중요하게 생각합니다[62, 63]. 

Although some learners find direct observation useful [24, 55], many view it (largely independent of the assessment tool used) as a ‘tick-box exercise’ or a curricular obligation [33, 56]. Learners may resist direct observation for multiple reasons. They can find direct observation anxiety-provoking, uncomfortable, stressful and artificial [31, 43, 50, 57, 58]. Learners’ resistance may also stem from their belief that faculty are too busy to observe them [43] and that they will struggle to find faculty who have time to observe [59]. Many learners (correctly) believe direct observation has little educational value when it is only used for high-stakes assessments rather than feedback [60]; they do not find direct observation useful without feedback that includes teaching and planning for improvement [59]. One study audiotaped over a hundred feedback sessions as part of the mini-CEX and found faculty rarely helped to create an action plan with learners [61]. Learners perceive a conflict between direct observation as an educational tool and as an assessment method [57, 60]. Many learners feel that direct observation interferes with learning, autonomy, efficiency, and relationships with their patients [42, 43]. Furthermore, learners value handling difficult situations independently to promote their own learning [62, 63].

감독자는 직접 관찰에 대한 학습자의 [저항을 줄이기 위한 전략]을 사용할 수 있습니다. 학습자는 관찰을 수행하는 개인과 종적인 관계를 맺을 때 직접 관찰 과정에 참여할 가능성이 더 높습니다. 학습자는 감독자가 자신에게 투자하고, 자신을 존중하며, 자신의 성장과 발달에 관심을 갖고 있다고 느낄 때 더 잘 받아들입니다[31, 64]. 학습자는 일반적으로 관찰이 정기적으로 이루어질 때 직접 관찰에 더 익숙해지기 때문에 관찰은 자주 이루어져야 합니다[58]. [로테이션을 시작할 때 학습 및 기술 개발을 위한 직접 관찰의 역할에 대해 논의]하면 직접 관찰의 양이 늘어납니다[43]. 감독자는 학습자에게 직접 관찰할 수 있음을 알려야 합니다. 감독자는 직접 관찰이 피드백 및 개발을 위해 사용되는 경우와 더 높은 수준의 평가를 위해 사용되는 경우를 명시하여 학습자에게 관찰의 위험도를 명확히 알려야 합니다. 감독자는 학습자가 총괄 평가를 위한 직접 관찰보다 [형성 목적의 직접 관찰]을 더 긍정적으로 여긴다는 점을 기억해야 합니다[65]. 또한 학습자는 개인화된 학습 목표에 초점을 맞추고[31] 효과적이고 질 높은 피드백이 뒤따를 때 직접 관찰을 더 중요하게 여기고 참여할 가능성이 높습니다.
Supervisors can employ strategies to decrease learners’ resistance to direct observation. Learners are more likely to engage in the process of direct observation when they have a longitudinal relationship with the individual doing the observations. Learners are more receptive when they feel a supervisor is invested in them, respects them, and cares about their growth and development [31, 64]. Observation should occur frequently because learners generally become more comfortable with direct observation when it occurs regularly [58]. Discussing the role of direct observation for learning and skill development at the beginning of a rotation increases the amount of direct observation [43]. Supervisors should let learners know they are available for direct observation. Supervisors should make the stakes of the observation clear to learners, indicating when direct observation is being used for feedback and development versus for higher-stakes assessments. Supervisors should remember learners regard direct observation for formative purposes more positively than direct observation for summative assessment [65]. Additionally, learners are more likely to value and engage in direct observation when it focuses on their personalized learning goals [31] and when effective, high quality feedback follows.

개별 감독자가 하지 말아야 할 사항
Don’ts for individual supervisors

가이드라인 10. 피드백을 정량적 평가로 제한하지 마세요.
Guideline 10. Don’t limit feedback to quantitative ratings.

직접 관찰한 내러티브 코멘트는 학습자에게 풍부한 피드백을 제공합니다. 수치 평가가 포함된 평가 양식을 사용하는 경우 학습자에게 서술형 피드백도 제공하는 것이 중요합니다. 많은 직접 관찰 평가 도구는 평가자가 학습자의 수행을 설명하기 위해 숫자 등급을 선택하도록 유도합니다[66]. 그러나 수행 점수를 의미 있게 해석하려면 평가자의 추론에 대한 통찰력을 제공하는 서술적 코멘트가 필요합니다. 내러티브 코멘트는 역량 성취도에 대한 신뢰할 수 있고 방어 가능한 의사 결정을 지원할 수 있습니다[67]. 또한 내러티브 피드백은 건설적인 방식으로 제공될 경우 교육생이 수행 능력의 강점과 약점을 정확하게 파악하고 역량 개발을 안내하는 데 도움이 될 수 있습니다[46]. 직접 관찰 자체에 대한 증거가 부족하고 정량적 평가가 성적과 동일하지 않지만, 다른 평가 문헌에 따르면 학습자가 성적이나 코멘트와 함께 성적만 받았을 때 학습 이득을 보이지 않는다고 합니다. 학습자는 성적에 집중하고 코멘트를 무시하기 때문에 코멘트가 포함된 성적을 받을 때 학습 이득이 발생하지 않는다는 가설이 있습니다[68,69,70]. 반면, 성적 없이 코멘트만 받은 학습자는 큰 학습 이득을 보입니다[68,69,70]. [서술적 피드백이 없는 성적]은 학습자에게 개선을 자극할 수 있는 충분한 정보와 동기를 제공하지 못합니다[26]. 합격/불합격 등급이 특정 수치 등급보다 학생에게 더 잘 받아들여질 수 있지만[51], 전체 등급을 사용하면 피드백에 대한 수용도가 떨어질 수도 있습니다[71]. 직접 관찰 후 학습자와 평점을 공유하는 것의 장단점은 알려져 있지 않지만, 형성적 평가에 직접 관찰을 사용하는 경우 학습자가 강점 영역(잘 수행한 기술)과 개선이 필요한 기술을 설명하는 서술형 피드백을 받는 것이 중요합니다.
Narrative comments from direct observations provide rich feedback to learners. When using an assessment form with numerical ratings, it is important to also provide learners with narrative feedback. Many direct observation assessment tools prompt evaluators to select numerical ratings to describe a learner’s performance [66]. However, meaningful interpretation of performance scores requires narrative comments that provide insight into raters’ reasoning. Narrative comments can support credible and defensible decision making about competence achievement [67]. Moreover, narrative feedback, if given in a constructive way, can help trainees accurately identify strengths and weaknesses in their performance and guide their competence development [46]. Though evidence is lacking in direct observation per se and quantitative ratings are not the same as grades, other assessment literature suggests that learners do not show learning gains when they receive just grades or grades with comments. It is hypothesized that learning gains do not occur when students receive grades with comments because learners focus on the grade and ignore the comments [68,69,70]. In contrast, learners who receive only comments (without grades) show large learning gains [68,69,70]. Grades without narrative feedback fail to provide learners with sufficient information and motivation to stimulate improvement [26]. The use of an overall rating may also reduce acceptance of feedback [51] although a Pass/Fail rating may be better received by students than a specific numerical rating [71]. Although the pros and cons of sharing a rating with a learner after direct observation are not known, it is important that learners receive narrative feedback that describes areas of strength (skills performed well) and skills requiring improvement when direct observation is being used for formative assessment.

가이드라인 11. 학습자와 환자 모두의 허락을 구하고 준비하지 않은 상태에서 환자 앞에서 피드백을 제공하지 마십시오.
Guideline 11.
Don’t give feedback in front of the patient without seeking permission from and preparing both the learner and the patient.

감독자가 환자 앞에서 직접 관찰한 후 학습자에게 피드백을 제공하려는 경우, 사전에 학습자와 환자의 허락을 구하는 것이 중요합니다. 피드백은 일반적으로 조용하고 사적인 장소에서 제공되며, 환자 앞에서 피드백을 제공하면 학습자와 환자의 관계가 손상될 수 있으므로 이러한 허락은 특히 중요합니다. 허가를 구하지 않았거나 허가를 받지 않은 경우 학습자는 환자 앞에서 피드백을 받아서는 안 됩니다. 그러나 예외적으로 환자가 안전하고 효과적이며 환자 중심의 치료를 받지 못하는 경우에는 즉시 중단해야 하며, 이러한 상황에서는 이러한 중단이 피드백의 한 형태임을 인식하고 학습자를 지지하고 비하하지 않는 방식으로 즉각 중단해야 합니다.  
If a supervisor plans to provide feedback to a learner after direct observation in front of a patient, it is important to seek the learner’s and patient’s permission in advance. This permission is particularly important since feedback is typically given in a quiet, private place, and feedback given in front of the patient may undermine the learner-patient relationship. If permission has not been sought or granted, the learner should not receive feedback in front of the patient. The exception, however, is when a patient is not getting safe, effective, patient-centred care; in this situation, immediate interruption is warranted (in a manner that supports and does not belittle the learner), recognizing that this interruption is a form of feedback.

병상 교육은 학습자에게 효과적이고 흥미를 유발할 수 있지만[72, 73] 일부 학습자는 환자 앞에서 교육하는 것이 환자와의 치료 동맹을 약화시키고 긴장된 분위기를 조성하며 질문할 수 있는 능력을 제한한다고 느낍니다[73, 74]. 그러나 환자 중심주의 시대에는 피드백에서 환자 목소리의 역할과 중요성이 증가할 수 있습니다. 실제로 이전 연구에 따르면 많은 환자가 자신의 치료에 대해 논의할 때 의료진이 병상에 있기를 원한다고 합니다[75]. 직접 관찰의 맥락에서 환자와 치료 및 교육적 제휴를 가장 잘 구축하는 방법은 추가적인 주의가 필요합니다. 
Although bedside teaching can be effective and engaging for learners, [72, 73] some learners feel that teaching in front of the patient undermines the patient’s therapeutic alliance with them, creates a tense atmosphere, and limits the ability to ask questions [73, 74]. However, in the era of patient-centredness, the role and importance of the patient voice in feedback may increase. In fact, older studies suggest many patients want the team at the bedside when discussing their care [75]. How to best create a therapeutic and educational alliance with patients in the context of direct observation requires additional attention.

개별 감독자의 경우 모름
Don’t Knows for individual supervisors

지침 12. 직접 관찰하는 동안 인지 부하가 미치는 영향은 무엇이며 이를 완화하기 위한 접근 방식은 무엇입니까?
Guideline 12. 
What is the impact of cognitive load during direct observation and what are approaches to mitigate it?

평가자는 학습자를 관찰하고 평가하는 동시에 환자를 진단하고 돌보려고 노력하면서 상당한 인지 부하를 경험할 수 있습니다[76]. [지각 부하]가 관찰자의 주의력을 압도하거나 초과할 수 있습니다. 이러한 과부하는 한 자극에 집중하면 다른 자극에 대한 지각이 손상되는 '부주의성 실명'을 유발할 수 있습니다[76]. 예를 들어, 학습자의 임상적 추론에 집중하는 동시에 환자를 진단하려고 하면 감독자가 학습자의 의사소통 기술에 주의를 기울이는 데 방해가 될 수 있습니다. 평가자가 평가해야 하는 차원 수가 증가하면 평가의 질이 떨어집니다[77]. 경험이 많은 관찰자는 학습자와 환자에 대한 휴리스틱, 스키마 또는 수행 스크립트를 개발하여 정보를 처리함으로써 관찰 능력을 향상시킵니다[45, 76]. 또한 고도로 숙련된 교수진은 더 강력한 스키마 및 스크립트와 관련된 노력이 감소하기 때문에 인지 부하를 줄이면서 강점과 약점을 감지할 수 있습니다[78]. [평가 도구 설계]도 인지 부하에 영향을 미칠 수 있습니다. 예를 들어, Byrne과 동료들은 인지 부하를 측정하기 위해 검증된 도구를 사용하여 마취를 유도하는 수련의에 대한 객관적인 구조화된 임상 시험을 위해 주관적인 평가 척도보다 20개 이상의 항목 체크리스트를 작성하도록 요청받았을 때 교수진이 더 큰 인지 부하를 경험하는 것으로 나타났습니다[79]. 시뮬레이션이 아닌 상황에서 직접 관찰하는 동안 인지적 부하가 미치는 영향과 관찰자가 중요한 요소만 평가하도록 평가 양식을 구성하여 평가할 항목 수를 제한하는 방법을 결정하기 위해서는 더 많은 연구가 필요합니다. 
An assessor can experience substantial cognitive load observing and assessing a learner while simultaneously trying to diagnose and care for the patient [76]. Perceptual load may overwhelm or exceed the observer’s attentional capacities. This overload can cause ‘inattentional blindness,’ where focusing on one stimulus impairs perception of other stimuli [76]. For example, focusing on a learner’s clinical reasoning while simultaneously trying to diagnose the patient may interfere with the supervisor’s ability to attend to the learner’s communication skills. As the number of dimensions raters are asked to assess increases, the quality of ratings decreases [77]. More experienced observers develop heuristics, schemas or performance scripts about learners and patients to process information and thereby increase observational capacity [45, 76]. More highly skilled faculty may also be able to detect strengths and weaknesses with reduced cognitive load because of the reduced effort associated with more robust schemes and scripts [78]. Assessment instrument design may also influence cognitive load. For example, Byrne and colleagues, using a validated instrument to measure cognitive load, showed that faculty experienced greater cognitive load when they were asked to complete a 20 plus item checklist versus a subjective rating scale for an objective structured clinical examination of a trainee inducing anaesthesia [79]. More research is needed to determine the impact of cognitive load during direct observation in non-simulated encounters and how to structure assessment forms so that observers are only asked to assess critical elements, thereby limiting the number of items to be rated.

지침 13. 다양한 기술을 직접 관찰하기 위한 최적의 시간길이는 얼마입니까?
Guideline 13. 
What is the optimal duration for direct observation of different skills?

최근 직접 관찰 및 피드백 관련 문헌의 대부분은 바쁜 업무 환경에서 효율성을 높이기 위해 직접 관찰 시간을 짧고 집중적으로 유지하는 데 초점을 맞추고 있습니다[80]. 짧은 관찰은 환자를 만나는 시간이 짧은 임상 전문과목에 적합하지만, 다른 전문과목의 경우 긴 관찰을 통해서만 알 수 있는 진료의 관련 측면을 짧은 관찰로 놓칠 수 있습니다. 직접 관찰 및 피드백을 위한 시급한 질문 중 하나는 다양한 전문과목, 학습자 및 술기에 대한 최적의 면담 시간을 결정하는 것입니다. 최적의 면담 시간은 환자의 요구, 관찰 대상 과제, 학습자의 역량, 교수자의 과제에 대한 친숙도 등 여러 변수를 반영해야 할 것입니다[78, 81].
Much of the recent direct observation and feedback literature has focused on keeping direct observation short and focused to promote efficiency in a busy workplace [80]. While short observations make sense for clinical specialties that have short patient encounters, for other specialties relevant aspects of practice that are only apparent with a longer observation may be missed with brief observations. One of the pressing questions for direct observation and feedback is to determine the optimal duration of encounters for various specialties, learners and skills. The optimal duration of an encounter will likely need to reflect multiple variables including the patient’s needs, the task being observed, the learner’s competence and the faculty’s familiarity with the task [78, 81].

교육자/교육 지도자를 위한 근거가 있는 가이드라인
Guidelines with supporting evidence for educators/educational leaders

교육 지도자를 위한 지침
Do’s for educational leaders

지침 14. 관련 임상 기술과 전문성을 기준으로 참관자를 선정합니다.
Guideline 14. 
Do select observers based on their relevant clinical skills and expertise.

프로그램 디렉터와 같은 교육 리더는 관련 임상 기술과 교육 전문성을 기준으로 참관자를 선정해야 합니다. 공정하고 신뢰할 수 있는 평가를 위해서는 콘텐츠 전문성(모범 술기가 어떤 것인지에 대한 지식과 이를 평가할 수 있는 능력)이 전제 조건입니다[82]. 그러나 평가자는 종종 내용 전문성이 부족하다고 느끼는 술기를 직접 관찰하도록 요청받으며, 평가자는 체크리스트를 사용하여 자신의 임상 기술 부족을 보완할 수 있다고 생각하지 않습니다[83]. 또한 감독자 자신의 임상 기술이 학습자를 평가하는 방식에 영향을 미칠 수 있습니다[78]. 평가자의 특성이 평가자 자신의 역량 결핍의 결과인 경우[84], 평가자가 관찰 및 피드백 시 자신을 표준으로 삼는 경우 학습자도 동일한 결핍 또는 역량 장애를 습득할 수 있습니다[78,85,86,87,88,89,90]. 교수진은 종종 자신을 학습자 수행을 평가하는 기준(즉, 참조 프레임)으로 사용하기 때문에[82], 임상 술기 전문성을 기반으로 평가자를 선정하거나 평가자가 자신을 참조 프레임으로 사용하지 않고 유능하고 전문적인 수행을 인정할 수 있도록 평가자 교육을 제공하는 것이 중요합니다.  
Educational leaders, such as program directors, should select observers based on their relevant clinical skills and educational expertise. Content expertise (knowledge of what exemplar skill looks like and having the ability to assess it) is a prerequisite for fair, credible assessment [82]. However, assessors are often asked to directly observe skills for which they feel they lack content expertise, and assessors do not believe using a checklist can make up for a lack of their own clinical skill [83]. Additionally, a supervisor’s own clinical skills may influence how they assess a learner [78]. When assessors’ idiosyncrasy is the result of deficiencies in their own competencies [84] and when assessors use themselves as the gold standard during observation and feedback, learners may acquire the same deficiencies or dyscompetencies [78, 85,86,87,88,89,90]. Because faculty often use themselves as the standard by which they assess learner performance (i. e. frame of reference), [82] it is important to select assessors based on their clinical skills expertise or provide assessor training so assessors can recognize competent and expert performance without using themselves as a frame of reference.

프로그램 수준에서는 특정 술기를 평가할 수 있는 [전문성을 갖춘 개인]에게 필요한 관찰 유형을 조정하는 것이 현명합니다. 예를 들어, 프로그램 디렉터는 심장 전문의에게 학습자의 심장 검사를 관찰하도록 요청하고 완화 치료 의사에게 학습자의 치료 목표에 대한 논의를 관찰하도록 요청할 수 있습니다. 학습자가 평가할 특정 술기에 대한 콘텐츠 전문 지식과 임상적 통찰력을 갖춘 평가자를 사용하는 것도 중요한데, 학습자는 이러한 개인의 피드백을 신뢰할 수 있고 신뢰할 수 있다고 생각할 가능성이 높기 때문입니다[20, 64]. 전문성이 부족한 경우 교수진의 역량 부족을 교정하도록 돕는 것이 중요합니다[91]. 평가와 관련된 교수진 개발은 이론적으로 교수진 자신의 임상 기술을 향상시키는 동시에 관찰 기술을 향상시키는 '일거양득'이 될 수 있습니다[91]. 
At a programmatic level, it is prudent to align the types of observations needed to individuals who have the expertise to assess that particular skill. For example, a program director might ask cardiologists to observe learners’ cardiac exams and ask palliative care physicians to observe learners’ goals of care discussions. Using assessors with content expertise and clinical acumen in the specific skill(s) being assessed is also important because learners are more likely to find feedback from these individuals credible and trustworthy [20, 64]. When expertise is lacking, it is important to help faculty correct their dyscompetency [91]. Faculty development around assessment can theoretically become a ‘two-for-one’—improving the faculty’s own clinical skills while concomitantly improving their observation skills [91].

평가자는 임상 술기 전문성 외에도 [다양한 교육 수준에서 학습자에게 기대할 수 있는 사항에 대한 지식]이 있어야 합니다[83]. 평가자는 교수 및 교육에 전념하고, 학습자의 성장을 촉진하는 데 투자하며, 학습자의 폭넓은 정체성과 경험에 관심을 갖고, 학습자를 신뢰하고 존중하며 돌볼 의향이 있어야 합니다[64][지침 28 참조].
In addition to clinical skills expertise, assessors also must have knowledge of what to expect of learners at different training levels [83]. Assessors must be committed to teaching and education, invested in promoting learner growth, interested in learners’ broader identity and experience, and willing to trust, respect and care for learners [64] [see Guideline 28].

가이드라인 15. 가능하면 직접 관찰을 위해 새로운 도구를 만들기보다는 기존의 타당성 근거가 있는 평가 도구를 사용합니다.
Guideline 15. 
Do use an assessment tool with existing validity evidence, when possible, rather than creating a new tool for direct observation.

직접 관찰을 기반으로 학습자의 수행을 평가하는 데 도움이 되는 많은 도구가 존재합니다[66, 92]. 교육자는 새로운 도구를 만들기보다는 가능하면 타당성 근거가 있는 기존 도구를 사용해야 합니다[93]. 교육자의 목적에 맞는 도구가 존재하지 않는 경우, 기존 도구를 수정하거나 새로운 도구를 만드는 방법이 있습니다. 직접 관찰을 위해 새로운 도구를 만들거나 기존 도구를 수정하려면 타당도 근거 축적을 포함하여 도구 설계 및 평가 지침을 따라야 합니다[94]. 필요한 타당도 증거의 양은 낮은 수준의 형성 평가보다 높은 수준의 총합 평가에 사용되는 도구의 경우 더 많을 것입니다. 
Many tools exist to guide the assessment of learners’ performance based on direct observation [66, 92]. Rather than creating new tools, educators should, when possible, use existing tools for which validity evidence exists [93]. When a tool does not exist for an educator’s purpose, options are to adapt an existing tool or create a new one. Creating a new tool or modifying an existing tool for direct observation should entail following guidelines for instrument design and evaluation, including accumulating validity evidence [94]. The amount of validity evidence needed will be greater for tools used for high-stakes summative assessments than for lower-stakes formative assessments.

도구 설계는 평가자 응답의 신뢰도를 최적화하는 데 도움이 될 수 있습니다. 도구의 앵커 또는 응답 옵션은 수행을 평가하는 방법에 대한 지침을 제공할 수 있습니다. 예를 들어, 행동 앵커 또는 발달 수행의 스펙트럼을 따라 행동을 설명하는 이정표로 정의된 앵커는 평가자의 일관성을 향상시킬 수 있습니다[95]. 학습자에게 필요한 감독 정도 또는 감독자가 느끼는 신뢰도에 대한 감독자의 인상을 묻는 척도는 감독자가 생각하는 방식과 더 잘 일치할 수 있습니다[96]. 전체적인 인상은 긴 체크리스트보다 평가자 간에 성과를 더 안정적으로 포착할 수 있습니다[97]. 특정 [체크리스트]와 [글로벌 인상] 중 어떤 것을 선택할지, 그리고 그 사이의 모든 것을 선택할지는 주로 평가의 목적에 따라 달라집니다. 예를 들어, [피드백]이 주요 목표인 경우 (학습자가 세분화되고 구체적인 피드백을 받지 못한다면) [총체적 평가]는 거의 유용하지 않습니다. 어떤 도구를 선택하든, 도구가 서술적 코멘트를 위한 충분한 공간을 제공하는 것이 중요합니다[71][가이드라인 16 및 30 참조]. 
Tool design can help optimize the reliability of raters’ responses. The anchors or response options on a tool can provide some guidance about how to rate a performance; for example, behavioural anchors or anchors defined as milestones that describe the behaviour along a spectrum of developmental performance can improve rater consistency [95]. Scales that query the supervisor’s impressions about the degree of supervision the learner needs or the degree of trust the supervisor feels may align better with how supervisors think [96]. A global impression may better capture performance reliably across raters than a longer checklist [97]. The choice between the spectrum of specific checklists to global impressions, and everything in between, depends primarily on the purpose of the assessment. For example, if feedback is a primary goal, holistic ratings possess little utility if learners do not receive granular, specific feedback. Regardless of the tool selected, it is important for tools to provide ample space for narrative comments [71] [see Guideline 16 and 30].

중요한 것은 도구의 타당도는 궁극적으로 도구의 사용자와 도구가 사용되는 맥락에 달려 있다는 것입니다. 평가자(예: 교수진)가 직접 관찰하는 것이 바로 도구라고 주장할 수 있습니다. 따라서 프로그램 책임자는 도구를 사용할 관찰자를 교육하는 것보다 도구를 설계하는 데 너무 많은 시간을 할애하는 경우가 많다는 점을 인식해야 합니다[지침 20 참조].
Importantly, validity ultimately resides in the user of the instrument and the context in which the instrument is used. One could argue that the assessors (e. g. faculty), in direct observation, are the instrument. Therefore, program directors should recognize that too much time is often spent designing tools rather than training the observers who will use them [see Guideline 20].

지침 16. 관찰자에게 직접 관찰하는 방법공유된 정신 모델과 공통 평가 기준을 채택하고 피드백을 제공하는 방법을 교육합니다.
Guideline 16. 
Do train observers how to conduct direct observation, adopt a shared mental model and common standards for assessment, and provide feedback.

감독자가 학습자를 관찰한 후 내리는 평가는 매우 다양하며, 같은 상황을 관찰하는 감독자라도 사용하는 도구에 따라 평가와 평가가 달라질 수 있습니다. 관찰자가 수행의 다양한 측면에 초점을 맞추고 우선순위를 정하고 수행을 판단하는 데 다른 기준을 적용하기 때문에 변동성이 발생합니다[46, 82, 98, 99]. 평가자는 또한 역량에 대한 서로 다른 정의를 사용합니다[82, 98]. 관찰자가 성과를 판단하는 데 사용하는 기준은 종종 경험적이고 특이한 방식으로 도출되며, 일반적으로 최근 경험의 영향을 받고[49, 82, 100], 첫인상에 크게 의존할 수 있습니다[48]. 평가자는 자신의 훈련과 수년간의 임상 및 교육 관행의 결과로 특이성을 개발합니다. 이러한 특성이 강력한 임상적 증거와 모범 사례에 근거한 경우 반드시 도움이 되지 않는 것은 아닙니다. 예를 들어, 평가자는 환자 중심 면담의 전문가로서 면담의 다른 측면을 배제하고 관찰하는 동안 그러한 행동과 기술을 크게 강조할 수 있습니다[47]. 뛰어난 성과 또는 매우 취약한 성과를 식별하는 것은 간단한 것으로 간주되지만, '회색 영역'의 성과에 대한 결정은 더 어렵습니다 [83].  

The assessments supervisors make after observing learners with patients are highly variable, and supervisors observing the same encounter assess and rate the encounter differently regardless of the tool used. Variability results from observers focusing on and prioritizing different aspects of performance and applying different criteria to judge performance [46, 82, 98, 99]. Assessors also use different definitions of competence [82, 98]. The criteria observers use to judge performance are often experientially and idiosyncratically derived, are commonly influenced by recent experiences, [49, 82, 100] and can be heavily based on first impressions [48]. Assessors develop idiosyncrasies as a result of their own training and years of their own clinical and teaching practices. Such idiosyncrasies are not necessarily unhelpful if based on strong clinical evidence and best practices. For example, an assessor may be an expert in patient-centred interviewing and heavily emphasize such behaviours and skills during observation to the exclusion of other aspects of the encounter [47]. While identifying outstanding or very weak performance is considered straightforward, decisions about performance in ‘the grey area’ are more challenging [83].

[평가자 교육]은 이러한 직접 관찰의 한계를 극복하는 데 도움이 될 수 있지만 완전히 없애지는 못합니다.
Rater training can help overcome but not eliminate these limitations of direct observation.

[수행 차원 교육]은 참가자들이 관찰되는 [수행의 측면]과 [수행 평가 준거]에 대한 이해를 공유하는 평가자 교육 접근 방식입니다[9]. 예를 들어, 슈퍼바이저는 환자에게 약물 복용 시작에 대해 상담할 때 중요한 기술이 무엇인지 논의할 수 있습니다. 대부분의 평가자는 판단의 [스캐폴드 또는 중추 역할을 하는 프레임워크]를 환영합니다[83]. [성과 차원 교육]을 받은 수퍼바이저는 이 과정을 통해 평가 기준에 대한 [공유된 정신 모델]을 제공받음으로써 보다 표준화되고 체계적이며 포괄적이고 구체적인 관찰을 할 수 있고, 이전에는 주의를 기울이지 않았던 기술에 주의를 기울이며, 구체적인 피드백을 제공하는 자기 효능감을 향상시킬 수 있었다고 설명합니다[91]. 
Performance dimension training is a rater training approach in which participants come to a shared understanding of the aspects of performance being observed and criteria for rating performance [9]. For example, supervisors might discuss what are the important skills when counselling a patient about starting a medication. Most assessors welcome a framework to serve as a scaffold or backbone for their judgments [83]. Supervisors who have done performance dimension training describe how the process provides them with a shared mental model about assessment criteria that enables them to make more standardized, systematic, comprehensive, specific observations, pay attention to skills they previously did not attend to, and improve their self-efficacy giving specific feedback [91].

[기준 프레임 교육]평가자에게 [평가 차원과 관련된 적절한 기준을 제공]함으로써 관찰 및 평가 중에 성과에 대한 공통 개념화(즉, 기준 프레임)를 사용하도록 평가자를 교육함으로써 [성과 차원 교육]을 기반으로 합니다[101]. 비의료 성과 평가 문헌의 체계적 검토 및 메타 분석에 따르면 [기준 프레임 교육]은 중간 정도의 효과 크기로 평가 정확도를 크게 향상시키는 것으로 나타났습니다[101, 102]. 의학 분야에서 Holmboe 등은 표준화된 레지던트 및 환자와 함께 실제 실습을 포함한 8시간의 [기준 프레임 훈련 세션]이 개입 8개월 후 직접 관찰에서 관용을 약간 줄이고 정확도를 향상시킨다는 것을 보여주었습니다[9]. 그러나 짧은 평가자 교육(예: 반나절 워크숍)은 평가자 간 신뢰도를 개선하는 것으로 나타나지 않았습니다[103].
Frame of reference training builds upon performance dimension training by teaching raters to use a common conceptualization (i. e., frame of reference) of performance during observation and assessment by providing raters with appropriate standards pertaining to the rated dimensions [101]. A systematic review and meta-analysis from the non-medical performance appraisal literature demonstrated that frame of reference training significantly improved rating accuracy with a moderate effect size [101, 102]. In medicine, Holmboe et al. showed that an 8‑hour frame of reference training session that included live practice with standardized residents and patients modestly reduced leniency and improved accuracy in direct observation 8 months after the intervention [9]. However, brief rater training (e. g. half day workshop) has not been shown to improve inter-rater reliability [103].

교수진 개발 프로그램 책임자는 참가자가 교육 및 임상 업무에 평가자 교육을 적용할 수 있도록 하는 방법을 계획해야 합니다. 전략에는 참가자가 인지한 필요와 관련된 자료와 업무 맥락에서 적용 가능한 형식을 만드는 것이 포함됩니다[82, 104]. 교육 효과를 위한 [효과적인 교수진 개발]의 주요 특징에는 체험 학습, 피드백 제공, 효과적인 동료 및 동료 관계, 의도적인 커뮤니티 구축, 종단적 프로그램 설계 등이 포함됩니다[105, 106]. 동료 평가와 협업을 위해 서로 의지하는 교육자 간부 집단인 실무 커뮤니티를 개발하는 데 초점을 맞춘 교수진 개발은 평가자 교육에서 특히 중요하며 참가자들로부터 긍정적인 반응을 얻고 있습니다[91]. 그룹 교육은 교수진 개발의 중점을 개인에서 직접적인 관찰과 피드백에 투자하는 교육자 커뮤니티로 옮기는 것이 중요하다는 점을 강조합니다. 평가자는 직접 관찰 후 피드백을 제공할 때, 특히 건설적인 피드백을 제공할 때 긴장을 경험하기 때문에[107], 평가자 교육에는 [직접 관찰 후 효과적인 피드백 제공에 대한 교육]도 포함되어야 합니다[지침 32 참조]. 평가자 교육이 중요하지만, 평가자 교육에 대한 해답이 없는 질문이 여전히 많이 남아 있습니다[가이드라인 31 참조].

Directors of faculty development programs should plan how to ensure that participants apply the rater training in their educational and clinical work. Strategies include making the material relevant to participants’ perceived needs and the format applicable within their work context [82, 104]. Key features of effective faculty development for teaching effectiveness also include the use of experiential learning, provision of feedback, effective peer and colleague relationships, intentional community building and longitudinal program design [105, 106]. Faculty development that focuses on developing communities of practice, a cadre of educators who look to each other for peer review and collaboration, is particularly important in rater training and is received positively by participants [91]. Group training highlights the importance of moving the emphasis of faculty development away from the individual to a community of educators invested in direct observation and feedback. Because assessors experience tension giving feedback after direct observation, particularly when it comes to giving constructive feedback [107], assessor training should also incorporate teaching on giving effective feedback after direct observation [see Guideline 32]. While rater training is important, a number of unanswered questions about rater training still remain [see Guideline 31].

지침 17. 직접 관찰이 프로그램 목표 및 역량(예: 마일스톤)과 일치하는지 확인합니다.
Guideline 17. 
Do ensure direct observation aligns with program objectives and competencies (e. g. milestones).

명확하게 표현된 프로그램 목표와 목적은 [직접 관찰의 목적을 정의하기 위한 단계]를 설정합니다[93]. 정의된 평가 프레임워크는 학습자와 감독자의 교육 목표에 대한 이해를 조정하고 평가에 사용할 도구의 선택을 안내합니다.

  • 프로그램 디렉터는 관찰할 실습의 구성 요소를 정의하는 분석적 접근법('분해하기')을 사용하여 목표와 목적을 정의할 수 있으며, 이를 통해 세부 체크리스트를 작성할 수 있습니다[108].
  • 또한 [종합적 접근법]을 사용하여 유능하고 신뢰할 수 있는 실습에 필요한 업무 활동을 정의할 수 있으며, 이를 통해 위탁 등급과 같은 보다 총체적인 척도를 적용할 수 있습니다[109].

프로그램 디렉터는 감독자와 학습자가 직접 관찰을 위해 학습자 목표를 논의할 때 프로그램에서 사용되는 프로그램 목표, 역량, 마일스톤 및 EPA를 참조하도록 권장할 수 있습니다.
Clearly articulated program goals and objectives set the stage for defining the purposes of direct observation [93]. A defined framework for assessment aligns learners’ and supervisors’ understandings of educational goals and guides selection of tools to use for assessment.

  • Program directors may define goals and objectives using an analytic approach (‘to break apart’) defining the components of practice to be observed, from which detailed checklists can be created [108].
  • synthetic approach can also be used to define the work activities required for competent, trustworthy practice, from which more holistic scales such as ratings of entrustment can be applied [109].

Program directors can encourage supervisors and learners to refer to program objectives, competencies, milestones and EPAs used in the program when discussing learner goals for direct observation.

지침 18. 학습자가 진정성 있게 연습하도록 유도하고 피드백을 환영하는 문화를 조성합니다.
Guideline 18. 
Do establish a culture that invites learners to practice authentically and welcome feedback.

대부분의 학습자는 성적과 고난도 시험에 중점을 두는 예비 대학 또는 학부 문화에서 의과대학에 입학합니다. 의과대학에서는 여전히 성적과 시험이 학습자의 행동을 크게 좌우할 수 있으며, 학습자는 여전히 낮은 비중의 평가를 학습 기회라기보다는 극복해야 할 총체적인 장애물로 인식할 수 있습니다. 학습자는 종종 학습 또는 성과에 대한 기대치에 대해 여러 가지 상충되는 메시지를 감지합니다[60, 110]. 그렇다면 현재 상황을 어떻게 [학습자 중심]으로 바꿀 수 있을까요? 
Most learners enter medical school from either pre-university or undergraduate cultures heavily steeped in grades and high-stakes tests. In medical school, grades and tests can still drive substantial learner behaviour, and learners may still perceive low-stakes assessments as summative obstacles to be surmounted rather than as learning opportunities. Learners often detect multiple conflicting messages about expectations for learning or performance [60, 110]. How then can the current situation be changed to be more learner centred?

프로그램은 학습자에게 언제, 어디서 학습 문화가 낮은 위험도의 실습 기회를 제공하는지 명시적으로 파악하고, 학습자가 직접 관찰을 학습 활동으로 받아들일 수 있는 문화를 조성해야 합니다. 임상 교육 환경에서 학습자는 유능해 보이고 높은 점수를 받아야 한다는 압박감을 최소화하는 감독자가 학습 목표에서 다루는 술기를 수행하는 것을 관찰할 기회가 필요합니다[111]. 학습 방향(학습자 또는 숙달 목표 대 수행 목표)은 학습 결과에 영향을 미칩니다[112].

  • [숙달 지향적 학습자]는 학습을 위해 노력하고, 피드백을 요청하며, 도전을 수용하고, 개선을 축하합니다. 반대로
  • [성과 지향적 학습자]는 유능해 보이고 실패를 피할 수 있는 기회를 찾습니다.
  • 같은 기술이나 과제를 연습하고 재시도할 수 있고 노력과 개선에 대해 보상하는 문화는 숙달 지향성을 촉진합니다.

Programs should explicitly identify for learners when and where the learning culture offers low-stakes opportunities for practice, and programs should foster a culture that enables learners to embrace direct observation as a learning activity. In the clinical training environment, learners need opportunities to be observed performing the skills addressed in their learning goals by supervisors who minimize perceived pressures to appear competent and earn high marks [111]. Orientations to learning (learner or mastery goals versus performance goals) influence learning outcomes [112].

  • A mastery-oriented learner strives to learn, invites feedback, embraces challenges, and celebrates improvement.
  • Conversely, a performance-oriented learner seeks opportunities to appear competent and avoid failure.
  • A culture that enables practice and re-attempting the same skill or task, and rewards effort and improvement, promotes a mastery orientation. 

[학습을 희생하면서 성적, 완벽함 또는 정답을 강조하는 문화]는 학습자가 직접 관찰을 피하는 것처럼 실패를 피하기 위해 적극적으로 노력하는 부적응적인 '성과 회피' 목표를 조장할 수 있습니다[28]. 프로그램은 직접 관찰자가 정확성보다는 연습과 노력에 가치를 두는 커뮤니케이션 관행을 사용하도록 장려해야 합니다. 낮은 수준의 환경에서 직접 관찰 및 피드백을 수행하는 교사의 역할과 평가자의 역할을 분리하면 학습자가 이러한 구분을 명확히 알 수 있습니다. 또한 프로그램은 학습자에게 학습에 대한 개인적 선택권을 부여하고 학습자와 감독자 간의 종적 관계를 보장함으로써 학습자가 피드백을 잘 받아들이는 문화가 조성되도록 해야 합니다[14, 26]. 
A culture that emphasizes grades, perfection or being correct at the expense of learning can promote maladaptive ‘performance-avoid’ goals in which learners actively work to avoid failure, as in avoiding being directly observed [28]. Programs should encourage direct observers to use communication practices that signal the value placed on practice and effort rather than just on correctness. Separating the role of teacher who conducts direct observation and feedback in low-stakes settings from the role of assessor makes these distinctions explicit for learners. Programs should also ensure their culture promotes learner receptivity to feedback by giving learners personal agency over their learning and ensuring longitudinal relationships between learners and their supervisors [14, 26].

가이드라인 19. 직접 관찰을 가능하게 하거나 방해하는 시스템 요인에 주의를 기울이십시오.
Guideline 19. 
Do pay attention to systems factors that enable or inhibit direct observation.

[의료 실습 환경의 구조와 문화]는 직접 관찰에 대한 가치를 뒷받침할 수 있습니다. 수련생은 지도의가 언제, 어떤 활동을 관찰하는지에 주의를 기울이고 이를 바탕으로 어떤 교육 및 임상 활동을 중요하게 여기는지 유추합니다[42]. 
The structure and culture of the medical training environment can support the value placed on direct observation. Trainees pay attention to when and for what activities their supervisors observe them and infer, based on this, which educational and clinical activities are valued [42].

교육 환경 내에서 환자 중심 치료에 초점을 맞추면 [마이크로 시스템] 내에서 학습자와 감독자가 공유하는 임상 실습 과제를 통해 일상적인 임상 치료에 교육을 포함할 수 있습니다 [113]. 직접 관찰 과정에 대한 교수진의 동의는 학습을 위한 직접 관찰의 중요성에 대한 교육과 교수진이 병상에서 학습자와 함께 시간을 보낼 수 있는 일정 구조를 통해 얻을 수 있습니다[93]. 환자 간호를 수행할 때 직접 관찰을 수행하도록 교수진을 교육하면 이 작업이 효율적이고 질 높은 간호 및 교육에 필수적인 것으로 간주됩니다[114]. 이 교육 전략에 대한 환자와 가족의 선호도는 이 전략이 치료에 도움이 된다고 인식하고 있으며, 그 결과 임상의의 환자 만족도가 더 높아질 수 있음을 시사합니다[115, 116]. 
A focus on patient-centred care within a training environment embeds teaching in routine clinical care through clinical practice tasks shared between learners and supervisors within microsystems [113]. Faculty buy-in to the process of direct observation can be earned through education about the importance of direct observation for learning and through schedule structures that enable faculty time with learners at the bedside [93]. Training faculty to conduct direct observation as they conduct patient care frames this task as integral to efficient, high quality care and education [114]. Patient and family preferences for this educational strategy suggest that they perceive it as beneficial to their care, and clinicians can enjoy greater patient satisfaction as a result [115, 116].

교육 리더는 직접 관찰을 제한하는 [시스템 장벽]을 해결해야 합니다.

  • 직접 관찰과 피드백을 위한 시간 부족은 직접 관찰을 가로막는 가장 흔한 장벽 중 하나입니다. 프로그램은 교육 및 환자 치료 시스템(예: 감독자:학습자 비율, 환자 센서스)이 직접 관찰 및 피드백을 위한 시간을 허용하도록 보장해야 합니다.
  • 교육 병원의 더 큰 임상 진료 환경에 주의를 기울이면 직접 관찰을 촉진하기 위해 해결해야 할 추가적인 장벽을 발견할 수 있습니다.
  • 현재의 수련 환경은 전자 건강 기록을 사용하여 컴퓨터로 빠르게 작업을 완료하는 데 집중하는 경우가 너무 많으며, 환자와 상호 작용하거나 교육 활동에 소요되는 수련의의 시간은 소수에 불과합니다[117, 118].
  • 드물지 않게, 감독자-학습자 쌍이 자주 바뀌면 학습자가 피드백을 통합하고 개선 사항을 입증하기 위해 동일한 감독자가 오랜 시간 동안 관찰하기 어렵습니다[119].
  • 프로그램 디렉터는 학습 환경에 대한 감독자와 학습자의 인식, 건설적인 피드백을 주고받는 능력, 학습자에 대한 판단의 공정성을 신뢰하는 능력을 향상시키는 종적 관계를 제공하는 커리큘럼 구조를 고려해야 합니다[26, 120, 121].
  • 대학원 의학 교육에서 외래 연속성 경험을 재설계하면 이러한 직접 관찰 및 피드백을 위한 종단적 기회를 촉진할 수 있습니다[122]. 

Educational leaders must address the systems barriers that limit direct observation.

  • Lack of time for direct observation and feedback is one of the most common barriers to direct observation. Programs need to ensure that educational and patient care systems (e. g. supervisor:learner ratios, patient census) allow time for direct observation and feedback.
  • Attention to the larger environment of clinical care at teaching hospitals can uncover additional barriers that should be addressed to facilitate direct observation.
  • The current training environment is too often characterized by a fast-paced focus on completing work at computers using the electronic health record, with a minority of trainee time spent interacting with patients or in educational activities [117118].
  • Not uncommonly, frequent shifts in supervisor-learner pairings make it difficult for learners to be observed by the same supervisor over time in order to incorporate feedback and demonstrate improvement [119].
  • Program directors should consider curricular structures that afford longitudinal relationships that enhance supervisors’ and learners’ perceptions of the learning environment, the ability to give and receive constructive feedback and trust the fairness of judgments about learners [26120121].
  • Redesign of the ambulatory continuity experience in graduate medical education shows promise to foster these longitudinal opportunities for direct observation and feedback [122].

프로그램에 집중하지 않기
Don’ts focused on program

지침 20. 직접 관찰에 적합한 도구를 선택한다고 해서 평가자 교육이 필요 없다고 가정하지 마십시오.
Guideline 20. 
Don’t assume that selecting the right tool for direct observation obviates the need for rater training.

직접 관찰용 도구 사용자는 잘 설계된 도구는 평가자가 사용법을 모두 이해할 수 있을 정도로 명확할 것이라고 잘못 생각할 수 있습니다. 그러나 앞서 설명한 바와 같이, 어떤 도구를 선택하든 평가자는 도구를 사용하여 직접 관찰하고 관찰 내용을 기록하도록 교육받아야 합니다. 실제 측정 도구는 도구가 아니라 교수 감독자입니다. 
Users of tools for direct observation may erroneously assume that a well-designed tool will be clear enough to raters that they will all understand how to use it. However, as described previously, regardless of the tool selected, observers should be trained to conduct direct observations and record their observations using the tool. The actual measurement instrument is the faculty supervisor, not the tool.

지침 21. 학습자에게 직접 관찰을 요청하는 책임을 전적으로 학습자에게만 지우지 마세요.
Guideline 21. 
Don’t put the responsibility solely on the learner to ask for direct observation.

학습자와 감독자는 직접 관찰 및 피드백이 이루어질 수 있도록 함께 책임을 져야 합니다. 학습자는 실제 임상 수행에 초점을 맞춘 의미 있는 피드백을 원하지만, [31, 39] 일반적으로 직접 관찰을 학습에 유용한 것으로 평가하는 것과 자율적이고 효율적으로 수행하기를 원하는 것 사이에서 긴장을 경험합니다[42]. 직접 관찰, [자율성, 효율성]이라는 두 가지 목표를 동시에 인정하면서 직접 관찰이 일상적인 활동의 일부가 되는 교육 문화로 바꾸면 학습자의 부담을 완화할 수 있습니다. 직접 관찰을 요청하는 학습자의 책임을 제거하거나 줄이고, 부분적으로 교수진과 프로그램의 책임으로 만드는 것은 이 학습 활동에 대한 [공동 책임]을 촉진할 것입니다. 
Learners and their supervisors should together take responsibility for ensuring that direct observation and feedback occur. While learners desire meaningful feedback focused on authentic clinical performance, [31, 39] they commonly experience tension between valuing direct observation as useful to learning and wanting to be autonomous and efficient [42]. Changing the educational culture to one where direct observation is a customary part of daily activities, with acknowledgement of the simultaneous goals of direct observation, autonomy and efficiency, may ease the burden on learners. Removing or reducing responsibility from the learner to ask for direct observation and making it, in part, the responsibility of faculty and the program will promote shared accountability for this learning activity.

지침 22. 교수자가 교사와 평가자 사이에서 느끼는 긴장을 과소평가하지 마십시오.
Guideline 22. 
Don’t underestimate faculty tension between being both a teacher and assessor.

20년 전 마이클 J. 고든은 교수자가 교사(학습자에게 지도 제공)와 평가자(학습자가 성과 기준을 충족하는지 교육 프로그램에 보고)로서 겪는 갈등에 대해 설명했습니다[123]. 현재 역량 기반 의학교육의 시대에는 교수진이 [직접 관찰한 상황을 보고해야 하는 요구 사항이 증가함]에 따라 이러한 긴장이 지속되고 있습니다. Gordon의 해결책은 역량 기반 의학교육의 많은 발전 사항을 반영합니다. 학습자 중심의 두 가지 시스템을 개발하여 일선 교수진이 학습자에게 피드백과 지도를 제공하는 시스템과 교수자 중심의 시스템으로 최소한의 역량을 유지하지 못하는 학습자를 모니터링하거나 선별하여 추가 의사 결정 및 평가를 전문 표준위원회에 넘기는 것입니다[124, 125]. 프로그램은 역량 기반 의학교육에서 직접 관찰을 사용할 때 교수진의 이중적 입장에 민감해야 하며, 이러한 역할 갈등을 최소화하는 패러다임을 고려해야 합니다[126, 127].

Two decades ago Michael J. Gordon described the conflict faculty experience being both teacher (providing guidance to the learner) and high-stakes assessor (reporting to the training program if the learner is meeting performance standards) [123]. In the current era of competency-based medical education, with increased requirements on faculty to report direct observation encounters, this tension persists. Gordon’s solution mirrors many of the developments of competency-based medical education: develop two systems, one that is learner-oriented to provide learners with feedback and guidance by the frontline faculty, and one that is faculty-oriented, to monitor or screen for learners not maintaining minimal competence, and for whom further decision making and assessment would be passed to a professional standards committee [124, 125]. Programs will need to be sensitive to the duality of faculty’s position when using direct observation in competency-based medical education and consider paradigms that minimize this role conflict [126, 127].

가이드라인 23. 직접 관찰에 대한 학습 문화를 방해할 수 있으므로 모든 직접 관찰을 높은 위험도로 만들지 마십시오.
Guideline 23. 
Don’t make all direct observations high stakes; this will interfere with the learning culture around direct observation.

대부분의 직접 관찰은 높은 위험도가 아니라 교수자가 학습자의 일상 업무에 대한 지침을 제공할 수 있도록 진정한 환자 중심 진료에 대한 낮은 위험도의 평가로 수행해야 합니다. 직접 관찰의 장점 중 하나는 학습자가 실제 임상 업무에 어떻게 접근하는지 볼 수 있는 기회이지만, 관찰 행위로 인해 성과가 변경되면 이점이 상쇄될 수 있습니다[128]. 학습자는 자신의 성과가 변경될 정도로 관찰과 관련된 '이해관계'에 매우 민감합니다[11, 31]. 직접 관찰에 대한 레지던트의 인식에 대한 질적 연구에서 레지던트들은 관찰자를 기쁘게 하기 위해, 그리고 성과가 채점된다고 생각하기 때문에 임상 스타일을 바꾼다고 보고했습니다. 직접 관찰은 학습자의 목표를 환자 중심 진료에서 성과 중심 진료로 전환했습니다[11]. 또 다른 연구에서 레지던트들은 직접 관찰에 대한 '이해관계'가 없기 때문에(관찰과 관련된 대화는 관찰자와 학습자 사이에만 이루어짐) 임상 수행의 진정성을 높이는 데 도움이 된다고 인식했습니다[31].
Most direct observations should not be high stakes, but rather serve as low-stakes assessments of authentic patient-centred care that enable faculty to provide guidance on the learner’s daily work. One of the benefits of direct observation is the opportunity to see how learners approach their authentic clinical work, but the benefits may be offset if the act of observation alters performance [128]. Learners are acutely sensitive to the ‘stakes’ involved in observation to the point that their performance is altered [11, 31]. In a qualitative study of residents’ perceptions of direct observation, residents reported changing their clinical style to please the observer and because they assumed the performance was being graded. The direct observation shifted the learner’s goals from patient-centred care to performance-centred care [11]. In another study, residents perceived that the absence of any ‘stakes’ of the direct observation (the conversations around the observations remained solely between observer and learner) facilitated the authenticity of their clinical performance [31].

지침 24. 직접 관찰을 사용하여 중요한 종합 결정을 내릴 때는 너무 짧은 시간 동안 너무 적은 수의 평가자에 의한 너무 적은 수의 직접 관찰에 근거하여 결정하지 말고 직접 관찰 데이터에만 의존하지 마십시오.
Guideline 24. 
When using direct observation for high-stakes summative decisions, don’t base decisions on too few direct observations by too few raters over too short a time and don’t rely on direct observation data alone.

단일 임상 성과에 대한 단일 평가에는 잘 설명된 한계가 있습니다:

  • 1) 평가는 단 한 명의 평가자의 인상을 포착하고
  • 2) 임상 수행은 단일 콘텐츠 영역으로 제한되는 반면 학습자는 평가의 내용, 환자 및 맥락에 따라 다르게 수행합니다[129, 130].

평가의 일반화 가능성을 높이려면 다양한 콘텐츠(예: 진단 및 술기) 및 맥락에서 학습자의 수행을 관찰하는 평가자의 수를 늘리는 것이 중요합니다[131, 132]. 또한 특정 시점의 학습자의 임상 수행은 감정 상태, 동기 부여 또는 피로와 같은 외부 요인에 의해 영향을 받을 수 있습니다. 따라서 일정 기간 동안 관찰한 내용을 캡처하면 보다 안정적으로 성과를 측정할 수 있습니다.
A single assessment of a single clinical performance has well-described limitations: 1) the assessment captures the impression of only a single rater and 2) clinical performance is limited to a single content area whereas learners will perform differentially depending on the content, patient, and context of the assessment [129, 130]. To improve the generalizability of assessments, it is important to increase the number of raters observing the learner’s performance across a spectrum of content (i. e. diagnoses and skills) and contexts [131, 132]. Furthermore, a learner’s clinical performance at any given moment may be influenced by external factors such as their emotional state, motivation, or fatigue. Thus, capturing observations over a period of time allows a more stable measure of performance.

평가 프로그램에서 여러 평가 도구(예: 의학 지식 테스트, 모의 상황, 직접 관찰)에서 수집한 정보를 결합하면 단일 평가 도구보다 학습자의 역량을 더 균형 있게 평가할 수 있습니다[133, 134]. 역량은 다차원적이며, 단일 평가 도구로는 모든 차원을 하나의 형식으로 평가할 수 없습니다[130, 135]. 이는 평가 도구를 뒷받침하는 타당성 논거를 검토할 때 분명하게 드러나는데, 단일 평가 도구의 경우 항상 논거의 강점과 약점이 존재합니다[136, 137]. 역량에 관한 의사결정에 도움이 되는 최상의 증거를 제공하는 도구를 신중하게 선택하는 것이 중요합니다[136, 137]. 예를 들어, 기술 숙련도를 평가하는 것이 목표인 경우, 프로그램은 시술의 적응증, 금기 사항 및 합병증에 대한 지식 테스트와 시뮬레이션 실험실의 파트별 트레이너를 사용한 직접 관찰 및 실제 임상 환경에서의 직접 관찰(환자와의 의사소통과 더불어 학습자의 기술 숙련도를 평가할 수 있는 곳)을 결합할 수 있습니다.
Combining information gathered from multiple assessment tools (e. g. tests of medical knowledge, simulated encounters, direct observations) in a program of assessment will provide a more well-rounded evaluation of the learner’s competence than any single assessment tool [133, 134]. Competence is multidimensional, and no single assessment tool can assess all dimensions in one format [130, 135]. This is apparent when examining the validity arguments supporting assessment tools; for any single assessment tool, there are always strengths and weaknesses in the argument [136, 137]. It is important to carefully choose the tools that provide the best evidence to aid decisions regarding competence [136, 137]. For example, if the goal is to assess a technical skill, a program may combine a knowledge test of the indications, contraindications and complications of the procedure with direct observation using part-task trainers in the simulation laboratory with direct observation in the real clinical setting (where the learner’s technical skills can be assessed in addition to their communication with the patient).

모름
Don’t Knows

가이드라인 25. 학습자의 독립성과 효율성이라는 가치를 훼손하지 않으면서 학습자가 관찰을 요청하도록 동기를 부여하는 프로그램은 무엇입니까?
Guideline 25. 
How do programs motivate learners to ask to be observed without undermining learners’ values of independence and efficiency?

직접적인 관찰/피드백과 자율성/효율성을 동시에 중시하는 학습 문화를 조성하는 데 따르는 어려움에 대해 논의되었지만, 이 문제에 대한 해결책은 명확하지 않습니다. 잠재적인 접근 방식으로는 교수진을 대상으로 짧은 만남에 대한 직접 관찰을 장려하고(따라서 효율성에 미치는 영향을 최소화), 교수진의 일상 업무에 직접 관찰의 근거를 두는 것이 있습니다[66, 129]. 그러나 이러한 전략은 직접 관찰에 더 많은 시간이 필요한 역량(예: 전문적 행동, 협업 기술 등)과는 반대로 짧은 관찰이 가능한 특정 업무에 직접 관찰을 집중하는 의도하지 않은 효과를 가져올 수 있습니다[60]. 또 다른 해결책은 환자 회진, 인수인계, 병원 또는 외래 환자 퇴원, 클리닉 프리셉팅 등과 같은 일상 업무에 직접 관찰을 포함시키는 것입니다. 기존 활동을 활용하면 일부 전문 분야 및 프로그램에서 발생하는 것처럼 학습자가 직접 관찰을 요청해야 하는 부담을 줄일 수 있습니다. 예를 들어, 맥마스터의 응급의학과 레지던트 프로그램에서는 관찰 영역을 공식화하여 각 응급실 교대 근무 시 감독자가 직접 관찰하는 시스템을 활용하고 있습니다[138]. 학습자가 관찰을 요청하도록 동기를 부여하는 추가 접근 방식을 식별하는 것이 중요합니다. 

While the difficulties in having a learning culture that simultaneously values direct observation/feedback and autonomy/efficiency have been discussed, solutions to this problem are less clear. Potential approaches might be to target faculty to encourage direct observation of short encounters (thus minimizing the impact on efficiency) and to ground the direct observation in the faculty’s daily work [66, 129]. However, these strategies can have the unintended effect of focusing direct observation on specific tasks which are amenable to short observations as opposed to competencies that require more time for direct observation (e. g. professional behaviour, collaboration skills etc.) [60]. Another solution may be to make direct observation part of daily work, such as patient rounds, hand-offs, discharge from a hospital or outpatient facility, clinic precepting and so forth. Leveraging existing activities reduces the burden of learners having to ask for direct observation, as occurs in some specialties and programs. For example, McMaster’s emergency medicine residency program has a system that capitalizes on supervisors’ direct observation during each emergency department shift by formalizing the domains for observation [138]. It will be important to identify additional approaches that motivate learners to ask for observation.

가이드라인 26. 전문과목은 어떻게 직접 관찰의 초점을 환자가 중요하게 여기는 임상 진료의 중요한 측면으로 확대할 수 있습니까?
Guideline 26. 
How can specialties expand the focus of direct observation to important aspects of clinical practice valued by patients?

환자와 의사는 임상 진료의 측면의 상대적 중요도에 대해 의견이 다릅니다. 예를 들어, 환자는 건강 관련 정보의 효과적인 전달의 중요성을 더 높게 평가합니다[139]. 평가자가 가장 중요하게 생각하거나 가장 편한 것만 관찰하는 경우, 직접 관찰의 초점을 임상 진료의 모든 중요한 측면으로 어떻게 확장할 수 있을까요[42]? 역량 기반 의료 교육에서 프로그램은 특정 술기를 강조하는 로테이션을 활용하여 관찰 빈도가 낮은 영역으로 직접 관찰의 초점을 확대할 수 있습니다(예: 일반 내과 입원 환자에서 관절 검사를 관찰하는 대신 류마티스내과 로테이션을 사용하여 학습자의 근골격 검사 술기를 직접 관찰하는 것). 분명한 것은 모든 사람이 모든 것을 관찰하기를 기대하는 것은 실패한 접근 방식이라는 것입니다. 각 전문과목별로 직접 관찰의 초점을 확대하여 환자가 중요하게 여기는 임상 치료의 측면을 포괄하는 방법을 배우기 위한 연구가 필요합니다. 많은 교수진 개발 프로그램은 교수진 개인만을 대상으로 하며, 조직 내 프로세스나 문화적 변화를 목표로 하는 프로그램은 상대적으로 적습니다[106]. 따라서 이러한 유형의 프로세스 변화를 목표로 하는 교육 리더를 위한 교수진 개발이 필요할 것으로 보입니다.
Patients and physicians disagree about the relative importance of aspects of clinical care; for example, patients more strongly rate the importance of effective communication of health-related information [139]. If assessors only observe what they most value or are most comfortable with, how can the focus of direct observation be expanded to all important aspects of clinical practice [42]? In competency-based medical education, programs may take advantage of rotations that emphasize specific skills to expand the focus of direct observation to less frequently observed domains (e. g. using a rheumatology rotation to directly observe learners’ musculoskeletal exam skills as opposed to trying to observe joint exams on a general medicine inpatient service). What seems apparent is that expecting everyone to observe everything is an approach that has failed. Research is needed to learn how to expand the focus of direct observation for each specialty to encompass aspects of clinical care valued by patients. Many faculty development programs target only the individual faculty member, and relatively few target processes within organizations or cultural change [106]. As such, faculty development for educational leaders that target these types of process change is likely needed.

지침 27. 프로그램은 어떻게 위험 부담이 크고 빈번하지 않은 직접 관찰 평가 문화를 위험 부담이 적고 형성적이며 학습자 중심적인 문화로 바꿀 수 있는가?
Guideline 27. 
How can programs change a high-stakes, infrequent direct observation assessment culture to a low-stakes, formative, learner-centred culture?

형성적 평가에 직접 관찰을 집중하는 것의 중요성은 이미 설명했습니다. 그러나 높은 수준의 빈번한 직접 관찰 평가 문화를 낮은 부담의, 형성적이고, 학습자 중심의 직접 관찰 문화로 바꾸는 데 도움이 되는 추가적인 접근 방식이 여전히 필요합니다[140]. 평가 빈도를 높이고, 학습자가 평가를 요청할 수 있도록 권한을 부여하고[141, 142], 학습자에게 피드백 및 코칭을 위한 평가가 중요하다는 점을 강조하는 전략과 그 영향을 탐구하는 연구가 필요합니다[143,144,145]. 평가 프로그램의 설계, 모니터링 및 지속적인 개선에 학습자를 효과적으로 참여시키는 방법도 추가 연구가 필요합니다[146].
The importance of focusing direct observation on formative assessment has been described. However, additional approaches are still needed that help change a high-stakes, infrequent direct observation assessment culture to a low-stakes, formative, learner-centred direct observation culture [140]. Studies should explore strategies for and impacts of increasing assessment frequency, empowering learners to seek assessment, [141, 142] and emphasizing to learners that assessment for feedback and coaching are important [143,144,145]. How a program effectively involves its learners in designing, monitoring and providing ongoing improvement of the assessment program also merits additional study [146].

직접 관찰은 형성 평가에 중점을 두어야 하지만, 궁극적으로 모든 교육 프로그램은 승진 및 전환과 관련하여 중대한 결정을 내려야 합니다. 연구에 따르면 프로그램에 더 정확한 평가 정보가 있을수록 더 정확하고 더 나은 정보를 바탕으로 고위험 결정을 내릴 수 있는 것으로 나타났습니다[134, 135]. 고위험 결정을 내리기 위해 여러 관찰을 사용하는 것 외에[140], 프로그램이 고위험 결정을 내리기 위해 여러 개의 낮은 위험도 직접 관찰 평가를 가장 잘 사용할 수 있는 방법은 정확히 밝혀지지 않았습니다. 또한, 평가가 궁극적으로 고위험 의사결정을 위해 집계될 때 프로그램이 평가가 낮은 위험도로 인식되도록(즉, 하나의 낮은 위험도 관찰이 고위험 의사결정을 유도하지 않도록) 어떻게 보장할 수 있을까요? 
Although direct observation should be focused on formative assessment, ultimately all training programs must make a high-stakes decision regarding promotion and transition. Research has shown that the more accurate assessment information a program has, the more accurate and better informed the high-stakes decisions are [134, 135]. Other than using multiple observations to make a high-stakes decision [140], it is not clear exactly how programs can best use multiple low-stakes direct observation assessments to make high-stakes decisions. Additionally, how do programs ensure that assessments are perceived as low stakes (i. e. that no one low-stakes observation will drive a high-stakes decision) when assessments ultimately will be aggregated for higher-stakes decisions?

그룹 프로세스는 프로그램 평가의 새로운 필수 구성 요소로 떠오르고 있습니다. 일부 국가에서는 임상 역량 위원회라고 불리는 이러한 그룹이 이제 의학 대학원 교육의 필수 구성 요소로 자리 잡았습니다[147]. 질적 및 양적 직접 관찰을 통해 얻은 강력한 데이터는 그룹 판단 프로세스에 매우 유용하여 진도에 대한 의사 결정을 개선할 수 있습니다[148, 149]. 직접 관찰은 역량 및 위탁에 관한 의사 결정을 개선하기 위해 그룹 프로세스를 사용할 수 있는 평가 프로그램에 중요한 입력으로 작용해야 하지만, 데이터를 가장 잘 집계하는 방법은 아직 명확하지 않습니다.
Group process is an emerging essential component of programmatic assessment. In some countries these groups, called clinical competency committees, are now a required component of graduate medical education [147]. Robust data from direct observations, both qualitative and quantitative, can be highly useful for the group judgmental process to improve decisions about progression [148, 149]. While direct observation should serve as a critical input into a program of assessment that may use group process to enhance decision making regarding competence and entrustment, how to best aggregate data is still unclear.

지침 28. 소수의 핵심 교수진을 직접 관찰을 수행하는 '마스터 교육자'로 육성하면 어떤 이점이 있나요?
Guideline 28. 
What, if any, benefits are there to developing a small number of core faculty as ‘master educators’ who conduct direct observations?

많은 프로그램에서 직접 관찰이 부족한 문제에 대한 한 가지 잠재적 해결책은 환자 치료에 대한 책임을 동시에 맡지 않고 [직접 관찰을 수행하는 것이 주된 역할인 핵심 평가자 그룹]과 병행하는 시스템을 개발하는 것입니다. 교수진이 직접 관찰에 대한 교육과 보수를 지원받은 새로운 피드백 프로그램에서 레지던트들은 임상 기술, 학습자로서의 발전, 정서적 웰빙 측면에서 혜택을 받았습니다[31]. 이러한 접근 방식은 교수진 개발 노력을 직접 관찰하고 피드백하는 기술을 개발할 소수의 교육자에게 집중할 수 있게 해줍니다. 이러한 교육자 간부는 보다 구체적이고 맞춤화된 피드백을 제공할 가능성이 높으며, 이들의 관찰은 일일 임상 감독자의 통찰력을 보완하여 학습자의 교육 경험을 잠재적으로 향상시킬 수 있습니다[150]. 이러한 접근 방식은 또한 일일 임상 감독자의 시간 제약과 바쁜 업무에 대한 해결책을 제공할 수도 있습니다. 이 접근 방식의 구조, 이점 및 비용에 대해서는 연구가 필요합니다.

One potential solution to the lack of direct observation in many programs may be to develop a parallel system with a core group of assessors whose primary role is to conduct direct observations without simultaneous responsibility for patient care. In a novel feedback program where faculty were supported with training and remuneration for their direct observations, residents benefited in terms of their clinical skills, development as learners and emotional well-being [31]. Such an approach would allow faculty development efforts to focus on a smaller cadre of educators who would develop skills in direct observation and feedback. A cadre of such educators would likely provide more specific and tailored feedback and their observations would complement the insights of the daily clinical supervisors, thus potentially enhancing learners’ educational experience [150]. Such an approach might also provide a work-around to the time constraints and busyness of the daily clinical supervisors. The structure, benefits and costs of this approach requires study.

지침 29. 위임 기반 척도가 특히 절차에 기반하지 않은 전문 분야의 경우 구성적 척도를 달성하는 데 가장 적합한 접근 방식입니까?
Guideline 29. 
Are entrustment-based scales the best available approach to achieve construct aligned scales, particularly for non-procedurally-based specialties?

아직 결론은 나지 않았지만, 위임 척도가 불만족에서 우수 또는 미흡에서 우수와 같은 형용사적 앵커를 사용하는 오래된 척도보다 더 우수할 수 있다는 연구가 증가하고 있습니다[151, 152]. 이러한 구형 척도는 교수자에게 척도에 따른 의미 있는 정의와 학습자의 성과를 다른 학습자 또는 다른 표준과 비교해야 하는지 여부를 제공하지 않는 '잘못 정렬된 척도 구성'의 예입니다. 정렬된 척도 구성은 교수진의 교육적 정신 구성에 '정렬'되는 서술적 설명어가 있거나없는 앵커를 사용합니다. (종종 감독 수준을 기반으로 하는) 위임은 교수자가 학습자에 대해 내려야 하는 결정 유형(예: 반응적 감독에 대한 신뢰 또는 신뢰하지 않음)과 더 잘 일치합니다.

  • Crossley와 동료들은 영국 수련생을 대상으로 파운데이션 기간에 실시한 미니 임상 평가 연습에서 발달 기술 척도를 사용하여 기존의 미니-CEX보다 조정된 척도를 사용하여 더 나은 신뢰도와 수용 가능성을 발견했습니다[96].
  • Regehr와 동료들은 교수진에게 일반적인 척도 대신 표준화된 설명적 내러티브를 사용하도록 요청하면 레지던트 그룹 간의 성과를 더 잘 변별할 수 있다는 사실을 발견했습니다[153].
  • 다른 연구자들도 수퍼비전 수준을 척도 앵커로 사용하는 관찰 도구의 신뢰도와 수용성이 더 우수하다는 사실을 발견했습니다[152, 154].

While the verdict is still out, there is growing research that entrustment scales may be better than older scales that use adjectival anchors such as unsatisfactory to superior or poor to excellent [151, 152]. These older scales are examples of ‘construct misaligned scales’ that do not provide faculty with meaningful definitions along the scale and whether learner performance is to be compared with other learners or another standard. Construct aligned scales use anchors with or without narrative descriptors that ‘align’ with the educational mental construct of the faculty. Entrustment, often based on level of supervision, is better aligned with the type of decisions a faculty member has to make about a learner (e. g. to trust or not trust for reactive supervision).

  • Crossley and colleagues, using a developmental descriptive scale on the mini-clinical evaluation exercise grounded in the Foundation years for UK trainees, found better reliability and acceptability using the aligned scale than the traditional mini-CEX [96].
  • Regehr and colleagues found asking faculty to use standardized, descriptive narratives versus typical scales led to better discrimination of performance among a group of residents [153].
  • Other investigators have also found better reliability and acceptability for observation instruments that use supervision levels as scale anchors [152154].

따라서 위탁 척도는 여전히 주의가 필요하지만 유망한 발전으로 보입니다. 신뢰도는 타당도의 한 측면일 뿐이며, 다른 도구에 적용되는 것과 동일한 문제가 위탁 척도에서도 여전히 문제가 될 수 있습니다. 예를 들어, 평가자는 수퍼비전에 대해 매우 다른 견해를 가지고 있을 수 있습니다(즉, 공유된 정신 모델이 부족할 수 있습니다). Teman과 동료들은 교수진이 수술실에서 레지던트를 감독하는 방식이 다양하다는 사실을 발견했으며, 레지던트에게 필요한 감독 유형을 가장 잘 결정하기 위해서는 교수진 개발이 필요하다고 주장했습니다[155, 156]. 절차적 즈위쉬 척도는 다양한 수준의 수퍼비전과 상관관계가 있는 수퍼바이저 행동에 대한 강력한 설명자를 제공합니다[108, 154, 158]. 위임 척도를 사용한 연구는 주로 절차적 전문 분야에 초점을 맞추었으며, 절차적 기반이 아닌 전문 분야에서의 유용성을 이해하려면 더 많은 연구가 필요합니다. 또한, 위탁 척도와 행동 고정 척도(학습자의 행동에 초점을 맞춘 척도)의 상대적 장점을 파악하기 위한 연구가 필요합니다. 
Thus entrustment scales appear to be a promising development though caution is still needed. Reliability is only one aspect of validity, and the same problems that apply to other tools can still be a problem with entrustment scales. For example, assessors can possess very different views of supervision (i. e. lack a shared mental model). Teman and colleagues found that faculty vary in how they supervise residents in the operating room; they argued that faculty development is needed to best determine what type of supervision a resident needs [155, 156]. The procedural Zwisch scale provides robust descriptors of supervisor behaviours that correlate with different levels of supervision [108, 154, 158]. Studies using entrustment scales have largely focused on procedural specialties; more research is needed to understand their utility in more non-procedural based specialties. Research is also needed to determine the relative merits of behaviourally anchored scales (focused on what a learner does) versus entrustment scales.

가이드라인 30. 관찰 데이터를 '즉시' 기록할 수 있도록 테크놀로지을 사용하는 가장 좋은 접근 방식은 무엇인가요?
Guideline 30. 
What are the best approaches to use technology to enable ‘on the fly’ recording of observational data?


기술은 평가 또는 정성적 의견과 같은 관찰 데이터를 즉시 기록하는 데 도움이 될 수 있습니다. 직접 관찰에 대한 경험적 의학 교육 연구의 대부분은 평가 도구가 제공되는 형식보다 평가 도구에 더 중점을 두었습니다[66]. 그러나 임상 진료가 종이 기반에서 전자 플랫폼으로 진화하고 있다는 점을 고려할 때, 직접 관찰의 기록, 작성 및 제출이 휴대용 장치 또는 기타 전자 플랫폼을 사용하여 촉진될 수 있다는 것은 직관적으로 이해가 됩니다. 이 영역에서 수행된 몇 가지 연구에서 전자적 접근 방식의 타당성과 사용자 만족도를 문서화했지만, 공유 목표 개발을 촉진하고 관찰 품질을 지원하며 관찰을 수집 및 종합하기 위해 전자 플랫폼을 최적화하는 방법을 이해하려면 더 많은 연구가 필요합니다[159,160,161,162]. 
Technology can facilitate the immediate recording of observational data such as ratings or qualitative comments. Much of the empirical medical education research on direct observation has focused on the assessment tool more than the format in which the tool is delivered [66]. However, given the evolution of clinical care from paper-based to electronic platforms, it makes intuitive sense that the recording, completion and submission of direct observations may be facilitated by using handheld devices or other electronic platforms. The few studies done in this realm have documented the feasibility of and user satisfaction with an electronic approach, but more research is necessary to understand how to optimize electronic platforms both to promote the development of shared goals, support observation quality and collect and synthesize observations [159,160,161,162].

가이드라인 31. 관찰의 질과 학습자 피드백을 개선하기 위한 최상의 교수자 개발 접근법 및 실행 전략은 무엇인가요?
Guideline 31. 
What are the best faculty development approaches and implementation strategies to improve observation quality and learner feedback?

이미 설명한 바와 같이, 평가자 인지에 대한 최근 연구는 평가자 특이성, 다양한 기준 프레임, 인지적 편향, 암묵적 편향 및 인상 형성 등 직접 관찰에 영향을 미치는 주요 요인에 대한 몇 가지 통찰력을 제공합니다[46, 47, 82, 98, 99]. 그러나 이것이 교수진 개발에 대한 접근 방식에 어떤 정보를 제공할 수 있는지는 잘 알려져 있지 않습니다. 예를 들어, 교육 리더가 평가자가 자신의 특이한 성향을 인식하도록 돕고 학습자가 다양한 평가자로부터 충분한 종단 샘플링을 받아 핵심 역량의 모든 측면을 관찰할 수 있도록 보장하는 것의 가치 또는 영향은 무엇일까요? 평가자가 강력한 공유 기준 기반 멘탈 모델을 개발하거나 위탁 기반 척도[96, 151, 152, 157]를 사용하도록 하면 평가자의 특이성 및 인지적 편향(예: 대조 효과)이 감소할 수 있습니까? [지침 16 참조].  
As already described, recent research in rater cognition provides some insights on key factors that affect direct observation: assessor idiosyncrasy, variable frames of reference, cognitive bias, implicit bias and impression formation [46, 47, 82, 98, 99]. However, how this can inform approaches to faculty development is not well understood. For example, what would be the value or impact of educational leaders helping assessors recognize their idiosyncratic tendency and ensuring learners receive sufficient longitudinal sampling from a variety of assessors to ensure all aspects of key competencies are observed? Would assessor idiosyncrasy and cognitive bias (e. g. contrast effect) be reduced by having assessors develop robust shared criterion-based mental models or use entrustment-based scales [96, 151, 152, 157]? [See Guideline 16].

성과 차원 훈련 및 기준 프레임 훈련의 원칙에 기반한 보다 집중적인 평가자 훈련은 평가자의 관용을 줄이고, 평가 정확도를 개선하며, 직접 관찰 및 피드백을 통해 자기 평가의 편안함을 개선하지만[9, 91], 평가자 훈련이 학습자에 대한 관찰, 평가 또는 피드백의 질을 향상시키는지에 대한 연구는 구체적으로 조사되지 않았습니다.
While more intensive rater training based on the principles of performance dimension training and frame of reference training decreases rater leniency, improves rating accuracy, and improves self-assessed comfort with direct observation and feedback [9, 91], studies have not specifically explored whether rater training improves the quality of observation, assessment or feedback to learners.

평가자 훈련의 최적 구조와 기간도 불분명합니다[9, 103, 157]. 직접 관찰은 복잡한 기술이며 일회성이 아닌 지속적인 교육과 연습이 필요할 가능성이 높습니다. 그러나 직접 관찰, 평가 및 피드백의 질을 향상시키는 데 가장 효과적인 평가자 교육 구조를 결정하기 위해서는 연구가 필요합니다. 초기 교육은 어느 정도의 시간이 필요할까요? 어떤 유형의 종단적 교육 또는 스킬 리프레시가 필요한가요? 얼마나 자주 실시해야 하나요? 평가자에게 평가의 질이나 서술에 대한 피드백을 제공하면 어떤 이점이 있나요? 기존 연구에 따르면 종일 교육은 효과가 미미하다는 점을 고려할 때, 집중 교육의 타당성을 판단하는 것이 중요합니다.
The optimal structure and duration of assessor training is also unclear [9, 103, 157]. Direct observation is a complex skill and likely requires ongoing, not just one-time, training and practice. However, studies are needed to determine what rater training structures are most effective to improve the quality of direct observation, assessment and feedback. Just how long does initial training need to be? What type of longitudinal training or skill refreshing is needed? How often should it occur? What is the benefit of providing assessors feedback on the quality of their ratings or their narratives? Given that existing studies show full day training has only modest effects, it will be important to determine the feasibility of intensive training.

가이드라인 32. 환자 또는 다른 의료진 구성원의 직접 관찰과 피드백을 직접 관찰 접근법에 어떻게 통합해야 합니까?
Guideline 32. 
How should direct observation and feedback by patients or other members of the health care team be incorporated into direct observation approaches?

환자 및 기타 의료 전문가는 학습자 수행의 다양한 측면을 일상적으로 관찰하며 감독자의 피드백을 보완하는 피드백을 제공할 수 있습니다. 환자의 관점과 경험을 포함하지 않고 환자 중심 치료를 가르치고 평가하는 것은 매우 어렵습니다. 전문가 간 팀워크의 중요성을 고려할 때, 다른 팀원의 평가에 대해서도 마찬가지입니다. 환자 경험 설문조사와 다중 출처 피드백 도구(환자 설문조사 포함)는 이제 환자와 의료 전문가의 관찰과 경험을 포착하는 데 일반적으로 사용됩니다[163, 164]. 다중 소스 피드백은 적절하게 구현될 경우 유용한 정보를 제공하고 행동을 변화시키는 데 효과적일 수 있습니다[165, 166]. 환자로부터 적시에 피드백을 받는 것이 도움이 될지는 아직 알려지지 않았습니다. Concato와 Feinstein은 방문이 끝날 때 환자에게 세 가지 질문을 하면 병원과 개별 의사에게 풍부한 피드백이 제공된다는 것을 보여주었습니다[167]. 방문이 끝날 때 환자에게 '오늘 필요한 모든 것을 얻었습니까?"라고 간단히 묻는 환자 중심 기법은 의미 있는 피드백으로 이어질 수 있으며, 직접 관찰 시 '환자가 안전하고 효과적이며 환자 중심적인 치료를 받았습니까'를 주요 참조 프레임으로 사용하는 개념과 잘 부합할 수 있습니다[52]. 이 두 가지 기법이 유용할 수 있지만, 환자와 전문가 간 팀을 더 높은 수준의 평가에 활용하기 위해서는 더 많은 연구가 필요합니다. 연구에 따르면 다중 출처 피드백(전문가 간 그룹의 관찰을 나타내는)을 고위험 평가에 일상적으로 사용해서는 안 된다고 강력히 제안합니다[163]. 
Patients and other health professionals routinely observe various aspects of learner performance and can provide feedback that complements supervisors’ feedback. It is very hard to teach and assess patient-centred care without involving the perspective and experiences of the patient. Given the importance of inter-professional teamwork, the same can be said regarding assessments from other team members. Patient experience surveys and multi-source feedback instruments (which may include a patient survey) are now commonly used to capture the observations and experiences of patients and health professionals [163, 164]. Multisource feedback, when implemented properly, can be effective in providing useful information and changing behaviour [165, 166]. What is not known is whether just-in-time feedback from patients would be helpful. Concato and Feinstein showed that asking the patient three questions at the end of the visit yielded rich feedback for the clinic and the individual physicians [167]. A patient-centred technique that simply asks the patient at the end of the visit ‘did you get everything you needed today?’ may lead to meaningful feedback and aligns well with the concept of using ‘did the patient get safe, effective, patient-centred’ as the primary frame of reference during direct observation [52]. While these two techniques might be of benefit, more research is needed before using patients and the inter-professional team for higher-stakes assessment. Research strongly suggests that multi-source feedback (representing the observations of an inter-professional group) should not be routinely used for higher-stakes assessment [163].

가이드라인 33. 직접 관찰이 학습자 및 환자 결과에 영향을 줍니까?
Guideline 33. 
Does direct observation influence learner and patient outcomes?

의학교육에서 직접 관찰의 핵심적인 역할에도 불구하고, 직접 관찰이 학습자 및 환자 결과를 개선한다는 것을 입증하는 결과 데이터는 거의 존재하지 않습니다. 임상 및 절차 역량은 안전하고 효과적인 환자 중심 진료의 기본입니다. 직접 관찰 평가가 학습자 결과를 개선하고 따라서 환자 결과를 개선한다는 증거는 부족하지만, 논리와 간접적인 증거는 존재합니다. 의도적인 실습과 코칭은 기술 향상과 전문성 개발을 지원합니다[168]. 의료 전문가들 사이에서 더 나은 의사소통 기술이 더 나은 환자 결과와 관련이 있다는 증거는 높은 수준의 역량을 보장하기 위해 그러한 기술을 관찰하고 피드백을 제공하는 것의 중요성을 강력하게 뒷받침합니다[169]. 반대로 앞서 지적한 바와 같이, 전 과정에서 직접 관찰은 드물고 의사의 역량에는 격차가 존재합니다[170,171,172]. 따라서 직접 관찰이 중요하지 않다고 결론을 내리는 것은 비논리적이지만, 학습자와 환자 결과에 미치는 영향을 극대화하는 최선의 방법을 결정하기 위해서는 훨씬 더 많은 연구가 필요합니다. 
Despite the central role of direct observation in medical education, few outcome data exist to demonstrate that direct observation improves learner and patient outcomes. Clinical and procedural competencies are foundational to safe, effective, patient-centred care. While evidence is lacking to show that direct observation assessments improve learner outcomes, and therefore patient outcomes, logic and indirect evidence do exist. Deliberate practice and coaching support skill improvement and the development of expertise [168]. The evidence that better communication skills among health professionals is associated with better patient outcomes strongly supports the importance of observing and providing feedback about such skills to ensure high levels of competence [169]. Conversely, as pointed out earlier, direct observation is infrequent across the continuum and there are gaps in practising physicians’ competencies [170,171,172]. Thus, it would be illogical to conclude direct observation is not important, but much more work is needed to determine the best methods that maximize the impact on learner and patient outcomes.

요약
Summary

역량 기반 의학교육의 시대에 그 중요성이 더욱 강조되고 있는 오랜 평가 전략인 임상 술기 직접 관찰에 초점을 맞춘 가이드라인 목록을 작성했습니다. 이 작업은 다양한 관점을 대표하는 광범위한 문헌을 종합하고, 반복적인 작업을 거쳤으며, 현재 문헌에 대한 우리의 합의를 나타냅니다. 체계적인 검토가 아니었기 때문에 가이드라인에 정보를 제공할 수 있는 연구를 놓쳤을 수 있습니다. 저자들이 북미 출신이어서 관점과 권장사항의 일반화 가능성이 제한될 수 있습니다. 각 가이드라인의 강점을 결정하기 위해 그룹 합의를 사용했지만, 근거 강도에 대한 해석은 주관적이었습니다.
We have compiled a list of guidelines focused on direct observation of clinical skills, a longstanding assessment strategy whose importance has heightened in the era of competency-based medical education. This work synthesizes a wide body of literature representing multiple viewpoints, was iterative, and represents our consensus of the current literature. Because this was not a systematic review, we may have missed studies that could inform the guidelines. Authors were from North America, potentially limiting generalizability of viewpoints and recommendations. Although we used group consensus to determine the strengths of each guideline, our interpretation of evidence strength was subjective.

결론
Conclusions

이 가이드라인은 보건의료 전문직 교육에서 직접 관찰의 양과 질을 높이는 데 도움이 되도록 고안되었습니다. 직접 관찰을 개선하려면 감독자 개인과 학습자뿐만 아니라 그들이 일하고 훈련하는 조직과 문화에도 초점을 맞춰야 합니다. 수퍼바이저와 학습자 모두에게 직접 관찰에 참여하도록 동기를 부여하고 직접 관찰(및 그에 따른 피드백)이 실현 가능하고 가치 있으며 효과적인 지원 교육 시스템과 문화를 조성하는 전략과 개입을 식별하기 위해 해야 할 일이 많이 남아 있습니다. 이러한 접근 방식의 설계는 자기조절 학습과 평가자의 인지에 대한 이해의 증가와 같은 개념에 기반해야 합니다. 이러한 전략을 설계, 보급 및 평가하려면 문화 변화라는 매우 어려운 작업에 참여할 준비가 된 교육 리더에 대한 투자가 필요합니다. 우리가 아는 한, 맥락, 문화 및 시스템을 고려하면서 동시에 교육 리더, 감독자 및 학습자에게 초점을 맞추어 임상 술기에 대한 직접 관찰을 개선하기 위한 이러한 다각적이고 포괄적인 접근 방식은 설명된 바가 없습니다. 학습자와 감독자가 직접 관찰을 사용하여 진행 상황을 평가하고 더 높은 수준의 평가에 정보를 제공할 수 있도록 권한을 부여하면 교육 시스템 전체가 학습자의 역량을 향상하고 환자 치료를 개선할 수 있습니다.

These guidelines are designed to help increase the amount and quality of direct observation in health professions education. Improving direct observation will require focus not just on the individual supervisors and their learners but also on the organizations and cultures in which they work and train. Much work remains to be done to identify strategies and interventions that motivate both supervisors and learners to engage in direct observation and that create a supportive educational system and culture in which direct observation (and the feedback that follows) is feasible, valued and effective. The design of these approaches should be informed by concepts such as self-regulated learning and the growing understanding of rater cognition. Designing, disseminating and evaluating such strategies will require an investment in educational leaders prepared to engage in the very difficult work of culture change. To our knowledge, such a multifaceted, comprehensive approach to improving direct observation of clinical skills by simultaneously focusing on educational leaders, supervisors, and learners, while considering the context, culture and system has not been described. Empowering learners and their supervisors to use direct observation to assess progress and inform higher-stakes assessments enables the educational system as a whole to improve learners’ capabilities and enhance the care of patients.

 


Perspect Med Educ. 2017 Oct;6(5):286-305. doi: 10.1007/s40037-017-0376-7.

Guidelines: The do's, don'ts and don't knows of direct observation of clinical skills in medical education

Affiliations

1Perelman School of Medicine at the University of Pennsylvania, Philadelphia, PA, USA. Jennifer.kogan@uphs.upenn.edu.

2University of British Columbia, Vancouver, British Columbia, Canada.

3University of California San Francisco, San Francisco, CA, USA.

4Accreditation Council of Graduate Medical Education, Chicago, IL, USA.

PMID: 28956293

PMCID: PMC5630537

DOI: 10.1007/s40037-017-0376-7

Free PMC article

Abstract

Introduction: Direct observation of clinical skills is a key assessment strategy in competency-based medical education. The guidelines presented in this paper synthesize the literature on direct observation of clinical skills. The goal is to provide a practical list of Do's, Don'ts and Don't Knows about direct observation for supervisors who teach learners in the clinical setting and for educational leaders who are responsible for clinical training programs.

Methods: We built consensus through an iterative approach in which each author, based on their medical education and research knowledge and expertise, independently developed a list of Do's, Don'ts, and Don't Knows about direct observation of clinical skills. Lists were compiled, discussed and revised. We then sought and compiled evidence to support each guideline and determine the strength of each guideline.

Results: A final set of 33 Do's, Don'ts and Don't Knows is presented along with a summary of evidence for each guideline. Guidelines focus on two groups: individual supervisors and the educational leaders responsible for clinical training programs. Guidelines address recommendations for how to focus direct observation, select an assessment tool, promote high quality assessments, conduct rater training, and create a learning culture conducive to direct observation.

Conclusions: High frequency, high quality direct observation of clinical skills can be challenging. These guidelines offer important evidence-based Do's and Don'ts that can help improve the frequency and quality of direct observation. Improving direct observation requires focus not just on individual supervisors and their learners, but also on the organizations and cultures in which they work and train. Additional research to address the Don't Knows can help educators realize the full potential of direct observation in competency-based education.

Keywords: Assessment; Clinical Skills; Competence; Direct Observation; Workplace Based Assessment.

관리추론과 환자-의사 상호작용: 공동 의사결정과 모의 외래환자로부터의 통찰(Med Teach, 2023)
Management reasoning and patient-clinician interactions: Insights from shared decision-making and simulated outpatient encounters (Med Teach. 2023)
David A. Cooka , Ian G. Hargravesb , Christopher R. Stephensonc and Steven J. Durningd

소개
Introduction

의학은 환자의 상태를 진단하는 것보다 환자의 필요를 관리하는 것이 더 중요하지만, 방대한 임상 추론 문헌은 대부분 관리를 무시하여 환자 치료에 해를 끼칩니다(Cook 외. 2018). 최근 연구에서는 환자-임상의 상호작용의 중요한 역할을 포함하여 관리 추론의 중요한 특징이 강조되었습니다(Cook 외. 2018, 2019, 2022b). 실제로 관리 추론은 '전적으로 특정 임상의 내부가 아닌 개인 간의 공간에서' 이루어집니다. (Cook 외. 2022b) 그러나 관리 추론의 이러한 상호 작용 및 의사소통 측면에 대한 우리의 이해는 아직 불완전하게 개발되어 있습니다. 이 연구의 목적은 [관리 추론에서 환자-임상의 상호 작용]에 대한 이해를 넓히는 것입니다. 
Medicine is more about managing patient needs than diagnosing their conditions, but the vast clinical reasoning literature largely ignores management, to the detriment of patient care (Cook et al. 2018). Recent work has highlighted important characteristics of management reasoning (Cook et al. 2018, 2019, 2022b), including the crucial role of patient-clinician interactions. Indeed, management reasoning takes place substantially ‘in the space between individuals rather than entirely within a given clinician.’ (Cook et al. 2022b) However, our understanding of this interactional and communicative aspect of management reasoning remains incompletely developed. The purpose of this study is to expand our understanding of patient-clinician interactions in management reasoning.

관리 추론이 진단 추론보다 더 중요하다는 것은 틀림없는 사실이지만, 현재까지 거의 모든 임상 추론 연구와 이론은 후자에 초점을 맞추고 있습니다(Elstein 외. 1978; Eva and Norman, 2005; Norman, 2005; Norman 외. 2007, 2017). 최근에야 관리 추론이 명시적으로 구분되고 논의되기 시작했습니다(Cook et al. 2018, 2019). [관리 추론]은 '임상의가 임상 정보(병력, 검사 결과, 검사 결과), 선호도, 의학 지식, 맥락(상황) 요소를 통합하여 치료, 추가 검사, 후속 진료, 제한된 자원의 할당 등 개별 환자의 관리에 대한 결정을 내리는 인지적 과정'으로 정의되었습니다(Cook 외. 2019).
Management reasoning is arguably more important than diagnostic reasoning, yet nearly all clinical reasoning research and theory to-date have focused on the latter (Elstein et al. 1978; Eva and Norman, 2005; Norman, 2005; Norman et al. 2007, 2017). Only recently has management reasoning been explicitly distinguished and discussed (Cook et al. 2018, 2019). Management reasoning has been defined as ‘the cognitive processes by which clinicians integrate clinical information (history, exam findings, and test results), preferences, medical knowledge, and contextual (situational) factors to make decisions about the management of an individual patient, including decisions about treatment, further testing, follow-up encounters, and allocation of limited resources’ (Cook et al. 2019).

[임상적 추론]은 일반적으로 개별 임상의의 인지적 활동으로 해석됩니다. 이는 진단 추론의 경우 대부분 사실일 수 있지만, 관리 추론은 [공유된 숙고 과정]을 수반합니다(Cook 외. 2019). 최근의 경험적 연구에서 우리는 관리 추론의 주요 특징을 파악하기 위해 의사-환자 간 만남을 시뮬레이션한 동영상을 검토했으며(Cook 외. 2022b), 관리 추론이 한 사람 내에서만 발생하는 것이 아니라 [사람들이 상호 작용할 때 발생]한다는 사실을 확인했습니다. 그러나 우리는 관리 추론에서 환자-임상의 상호 작용에 대한 이해가 불완전하다는 것을 인식하고 있으며, 이 현상에 대한 이해를 심화시키고 교육, 평가, 임상 실습 및 향후 연구에 대한 시사점을 식별해야한다고 느낍니다. 저희는 공유 의사결정(SDM)이 이 문제를 밝히는 데 도움이 될 수 있다고 가정합니다. 공유 의사결정은 환자와 임상의가 함께 협력적으로 의사결정 또는 추론하는 방법에 대한 이론적, 실제적 프레임워크를 제공하기 때문입니다.
Clinical reasoning is commonly construed as a cognitive activity within individual clinicians. While this may be largely true of diagnostic reasoning, management reasoning entails a shared deliberative process (Cook et al. 2019). In a recent empirical study we reviewed videos of simulated physician-patient encounters to identify key features of management reasoning (Cook et al. 2022b), and confirmed that management reasoning occurs as people interact rather than solely within one person. We recognize, however, our incomplete comprehension of patient-clinician interactions in management reasoning, and feel prompted to deepen our understanding of this phenomenon and identify implications for teaching, assessment, clinical practice, and future research. We postulate that shared decision-making (SDM) could help illuminate this issue, as SDM offers theoretical and practical frameworks about how patients and clinicians collaboratively make decisions, or reason, together.

SDM은 [환자의 자율성]과 [환자 중심 치료]를 강화하기 위해 널리 장려되고 있습니다. SDM의 다양한 모델은 환자-임상의 상호 작용의 다양한 측면을 강조합니다. 대부분의 모델은 치료 옵션 중에서 선택할 때와 같이 의사 결정에 환자를 참여시키는 데 중점을 둡니다(Makoul and Clayman, 2006; Bomhof-Roordink 외. 2019). 

  • 미국 의료 연구 및 품질 관리국 'SHARE' 모델은 임상의가 
    • 명시적으로 환자의 참여를 구하고
    • 옵션을 탐색하도록 돕고
    • 환자의 가치와 선호도를 평가하고, 
    • 결정에 도달하고, 
    • 결정을 평가할 것을 권장합니다(미국 의료 연구 및 품질 관리국, 2020). 
  • 이와 유사하게, 엘윈의 '세 가지 대화' 모델은 다음을 강조합니다(엘윈 외. 2012, 2017). 
    • 환자를 의사 결정에 참여시키고(팀 대화), 
    • 각 옵션에 대해 논의하며(옵션 대화), 
    • 선호도를 통합하는 것(결정 대화)
  • 최근에는 많은 임상 상황에서 욕구를 명확히 하거나, 문제를 해결하거나, 인간성과 정체성의 실존적 문제를 해결하는 것만큼 옵션 중에서 선택하는 것이 중요하지 않다는 점을 인식한 '목적 지향적 SDM'(Purposeful SDM)이라는 새로운 SDM 모델이 제안되었습니다(표 1)(Hargraves et al. 2019). 

SDM is widely promoted to enhance patient autonomy and patient-centered care. Different models of SDM emphasize different aspects of patient-clinician interactions. Most models focus on involving patients in decision-making (Makoul and Clayman, 2006; Bomhof-Roordink et al. 2019), as when choosing among treatment options.

  • The Agency for Healthcare Research and Quality ‘SHARE’ model encourages clinicians to
    • explicitly Seek the patient’s participation; 
    • Help them explore options; 
    • Assess their values and preferences; 
    • Reach a decision; and 
    • Evaluate the decision (Agency for Healthcare Research and Quality, 2020).
  • Similarly, Elwyn’s ‘Three Talk’ model emphasizes engaging patients in decision-making (team talk), discussing each option (option talk), and integrating preferences (decision talk) (Elwyn et al. 2012, 2017).
  • Recently, a new model of SDM has been proposed—‘Purposeful SDM’ (Hargraves et al. 2019)—that recognizes that many clinical situations do not entail choosing among options so much as clarifying desires, solving problems, or addressing existential issues of humanity and identity (Table 1) (Hargraves et al. 2019). 

이러한 관점에서 볼 때, [SDM의 주요 목적]은 환자를 의사 결정에 '참여'시키는 것이 아니라, [환자의 특정 문제에 적합한 협력 방법을 수립하는 것]입니다. [목적 지향적 SDM]은 [참여 중심]이 아닌 [문제 중심]입니다.
Viewed through this lens, the primary purpose of SDM is not to “involve” patients in decision-making, but to establish methods for working together that are appropriate for this patient’s particular problems. Purposeful SDM is problem-focused rather than involvement-focused.

이 연구에서는 이러한 SDM 모델('참여 중심', '문제 중심')과 함께 환자와 임상의가 함께 관리 결정을 내리는 시뮬레이션 동영상을 활용하여 성공적인 또는 실패한 관리 추론에 기여하는 환자-임상의 상호 작용 요소를 조명했습니다.
In this study we drew on videos of simulated patients and clinicians as they made management decisions together, along with these SDM models (‘involvement-focused,’ ‘problem-focused’), to illuminate elements of patient-clinician interactions that contribute to successful or unsuccessful management reasoning.

방법
Method

저희는 환자와 임상의의 모의 만남을 담은 10개의 비디오를 검토하고 [두 가지 개념적 렌즈]를 통해 관찰한 내용을 체계적으로 해석했습니다. 이 동영상은 이전 연구(Cook 외. 2022b)에서 사용한 것과 동일한 동영상이지만, 새롭게 수집한 데이터와 새로운 프레임워크를 사용하여 고유한 스토리를 전달했습니다.
We reviewed 10 videos of simulated patient-clinician encounters and systematically interpreted our observations through two distinct conceptual lenses. These were the same videos used in our previous study (Cook et al. 2022b), but we used newly-collected data and novel frameworks to tell a unique story.

비디오 선택 및 코딩
Video selection and coding

각 비디오(45초~7분 길이)는 [레지던트 의사와 환자 간의 모의 외래 환자 추적 관찰]을 묘사했습니다. 평가자 교육 연구에 사용된 비디오 라이브러리에서 5가지 의학적 문제(고혈압, 고지혈증, 섬유근육통, 당뇨병, 유두상 갑상선암)에 대해 각각 2개의 비디오를 선택했습니다(Cook et al. 2009, Cook and Beckman, 2009). 이러한 경험을 통해 학습을 극대화하기 위해 52명의 내과 의사('극단적인 그룹'의 의도적 표본 추출)의 미니 CEX '상담' 평점을 기준으로 각 문제별로 가장 낮은 점수와 가장 높은 점수를 받은 동영상을 선정했습니다.

  • 3개 문제(6개 동영상)의 경우 레지던트-배우가 열악한 연기와 우수한 연기를 번갈아 연기하도록 코칭을 받았고, 환자-배우에게는 대본 없이 진정성 있는 답변을 하도록 지시했습니다.
  • 나머지 4개의 비디오는 대본이 작성되었습니다.

Each video (45 s to 7 min long) portrayed a simulated outpatient follow-up encounter between a resident physician and a patient. From a library of videos used in a study of rater training (Cook et al. 2009; Cook and Beckman, 2009) we selected 2 videos for each of 5 medical problems (hypertension, hyperlipidemia, fibromyalgia, diabetes mellitus, and papillary thyroid cancer). To maximize our learning from these encounters, we selected the lowest-scoring and highest-scoring videos for each problem, based on mini-CEX ‘counseling’ ratings from 52 internal medicine physicians (‘extreme groups’ purposive sampling).

  • For 3 problems (6 videos) the resident-actor was coached to alternately portray poor or superior performance, and the patient-actor was instructed to respond authentically; there was no written script.
  • The other 4 videos were scripted.

비디오는 2단계에 걸쳐 검토되었습니다. 1단계(2020년 11월)에서는 4명의 연구자(DAC, CRS, SJD, LDG)가 개방형 프롬프트가 포함된 코딩 양식(부록 1, 보충 자료)을 사용하여 각 비디오를 최소 2회 독립적으로 검토했습니다: 

  • '이 만남에서 관리 추론의 어떤 특징이 분명하게 드러나는가? [환자, 제공자, 시스템 선호도, 제약, 가치의 우선순위, 의사소통 및 공유된 의사결정(환자 선호도 확인, 통합) 등 5가지 특징 목록]'을 포함한 5가지 특징 목록으로 구성했습니다. 

논의는 2021년 10월부터 2022년 1월까지 계속되었습니다.
We reviewed videos in 2 phases. In phase 1 (November 2020), 4 investigators (DAC, CRS, SJD, LDG) independently reviewed each video at least twice using a coding form with open-ended prompts (Appendix 1, Supplementary material), including:

  • ‘What features of management reasoning are evident in this encounter? [list of 5 features including Prioritization of patient, provider, and system preferences, constraints, and values; and Communication and shared decision-making (ascertainment of, integration with patient preferences)].’

Discussions continued October 2021 to January 2022.

데이터 분석 중에 커뮤니케이션과 SDM은 추가 조사가 필요한 것으로 나타났으며, 2단계(2022년 2월)에서 이를 추진했습니다. 먼저 Purposeful SDM 모델의 주요 개발자를 포함하여 SDM에 대한 폭넓은 경험을 가진 조사자(IGH)를 팀에 추가했습니다(Hargraves 외. 2019). 그는 처음에는 1단계 조사 결과에 대한 지식 없이 독립적으로 동일한 동영상을 검토한 다음, 조사자 DAC 및 SJD와 함께 동일한 동영상을 검토했습니다. 이 조사자들은 처음에는 독립적으로, 그다음에는 그룹으로 각 비디오를 검토하면서 참여 중심 모델과 문제 중심 모델을 차례로 명시적으로 고려했습니다. 각 영상에 대해 각 모델의 주요 특징이 반영되었는지 여부를 코딩했습니다(부록 2, 보충 자료). 참여 중심 모델의 주요 특징은 쓰리 토크 모델(OPTION-5(Barr 외. 2015))에 기반한 도구와 최근 검토(Bomhof-Roordink 외. 2019)를 통해 확인했습니다. 문제 중심 모델의 경우, 목적 중심 SDM(Hargraves et al. 2019)의 개념을 사용했습니다(표 1). 개방형 프롬프트를 사용하여 추가 관찰 사항을 문서화했습니다.
During data analysis, communication and SDM stood out as warranting further investigation, which we pursued in phase 2 (February 2022). We first added to our team an investigator (IGH) with extensive experience in SDM including being a principal developer of the Purposeful SDM model (Hargraves et al. 2019). He reviewed the same videos, first independently and without knowledge of the phase 1 findings, and then in conjunction with investigators DAC and SJD. These investigators reviewed each video, first independently and then as a group, explicitly considering in turn the involvement-focused and problem-focused models of SDM. For each video we coded whether key features of each model were reflected in the encounter (Appendix 2, Supplementary material). We identified key features of the involvement-focused model from an instrument grounded in the Three Talk model (the OPTION-5 (Barr et al. 2015)) and from a recent review (Bomhof-Roordink et al. 2019). For the problem-focused model we used concepts from Purposeful SDM (Hargraves et al. 2019) (Table 1). We documented additional observations using open-ended prompts.

데이터 분석 및 모델 구축
Data analysis and model building

우리는 4단계로 진행되는 지속적인 비교 질적 분석 접근법(Glaser, 1965)을 사용했습니다.

  • 1단계에서는 모든 원시 관찰을 요약하고 검토하여 1단계의 초기 관찰을 분석했습니다(특정 SDM 모델 없음). 그런 다음 여러 차례의 음성 및 전자 매개 대화를 통해 중복을 해결하고 이러한 관찰을 반복적으로 재구성, 재개념화, 정교화 및 구체화했습니다. 
    We used a constant comparative qualitative analysis approach (Glaser, 1965), proceeding in 4 stages. In stage 1, we analyzed the initial observations from phase 1 (no specific SDM model) by summarizing and reviewing all raw observations. We then engaged in multiple voice and electronic-mediated conversations to resolve redundancies and iteratively reorganize, reconceptualize, elaborate, and refine these observations.
  • 환자-임상의 상호작용에 대한 추가 인사이트를 얻을 수 있는 기회를 인식하고 풍부한 새로운 데이터를 수집했습니다(위의 2단계, IGH의 독립 검토 및 그룹의 공동 검토 포함). 이러한 데이터를 분석하는 동안(2단계) 우리는 각 비디오에 대해 차례로 논의하면서 개별 코딩을 공유하고 각 모델의 렌즈를 통해 이러한 결과를 해석하면서 추가 인사이트를 공동으로 정교화했습니다.
    Recognizing the opportunity to glean further insights regarding patient-clinician interactions, we collected abundant new data (phase 2 above, including IGH's independent review and the group’s conjoint review). During analysis of these data (stage 2) we discussed each video in turn, sharing our individual codings and jointly elaborating further insights as we interpreted these findings through the lens of each respective model.
  • 다음 단계(3단계)에서는 관리 추론과 환자-임상의 상호작용 간의 연관성을 명시적으로 탐색했습니다. 1단계와 2단계에서 얻은 결과를 사용하여 이전에 보고된 12가지 관리 추론 기능(Cook 외. 2022b) 각각과 참여 중심 및 문제 중심 SDM 모델 간의 상호 영향을 체계적으로 고려했습니다.
    Next (stage 3) we explicitly explored the connections between management reasoning and patient-clinician interactions. Using our findings from stages 1 and 2, we systematically considered the reciprocal implications between each of 12 previously-reported management reasoning features (Cook et al. 2022b) and the involvement-focused and problem-focused SDM models.
  • 마지막으로 4단계에서는 축 분석을 통해 다양한 기능과 모델을 비교하여 몇 가지 잠정적인 모범 사례를 식별했습니다.
    Finally, in stage 4 we used axial analysis to cut across the various features and models, to identify several tentative best practices.

모든 검토자가 주요 결과에 대해 완전한 합의를 도출했습니다. 부록 1~3에는 모든 관찰 및 분석에 대한 자세한 문서 추적이 포함되어 있습니다(보충 자료).
All reviewers came to full consensus on key findings. Appendices 1–3 contain a detailed paper trail of all observations and analyses (Supplementary material).

반사성
Reflexivity

우리의 배경은 관찰과 분석에 분명 영향을 미쳤습니다. 세 명의 조사자(DAC, SJD, CRS)는 현직 의사(일반 내과 전문의)이며, 두 명의 조사자는 교육/인지(SJD) 및 디자인, 커뮤니케이션, SDM(IGH) 분야의 박사 학위를 보유하고 있습니다. 세 명의 연구자(DAC, SJD, CRS)가 12가지 기능의 경영 추론 모델을 설명하는 데 협력했으며, 한 명(IGH)은 목적 지향적 SDM 모델을 개발했습니다(Hargraves et al. 2019).
Our background surely influenced our observations and analyses. Three investigators (DAC, SJD, CRS) are practicing physicians (general internists) and 2 have PhDs with backgrounds in education/cognition (SJD) and design, communication, and SDM (IGH). Three investigators (DAC, SJD, CRS) collaborated in describing the 12-feature model of management reasoning and one (IGH) developed the purposeful SDM model (Hargraves et al. 2019).

결과
Results

개념적 모델 없이 관찰(1단계)
Observations without conceptual model (stage 1)

먼저 특정 개념적 모델 없이 관찰한 내용을 보고하고, 환자-임상의 상호작용의 질이 특히 낮거나 추론 과정 또는 치료 계획에 해로운 것으로 보이는 사례(및 그에 따른 시사점, 결과 또는 문제)를 강조했습니다. 또한, 아무리 우수한 동영상이라도 평범한 수준에 불과하다고 판단되는 경우라도 우수한 성과를 보인 사례도 찾았습니다. 표 2에는 각 비디오와 눈에 띄는 인상이 설명되어 있으며, 부록 1에는 관찰 및 분석에 대한 자세한 내용이 나와 있습니다(보충 자료).
We first report observations made without a specific conceptual model, highlighting instances (and corresponding implications, consequences, or problems) in which the patient-clinician interaction was of particularly poor quality or seemed detrimental to the reasoning process or care plan. We also sought instances of superior performance, although in our judgment even the best videos were only mediocre. Table 2 describes each video and salient impressions; Appendix 1 details our observations and analyses (Supplementary Material).

의사는 환자의 자율성이나 선택을 허용하거나 장려하지 않는 경우가 많았습니다. 이는 대놓고 드러나는 경우는 거의 없었습니다(예: 단 하나의 옵션만 제시된 동영상은 단 한 편뿐이었음). 그러나 [미묘한 표현]으로 자율성을 침해하는 경우가 종종 있었습니다. 한 조사자는 의사들이 '환자의 의사를 뭉겠다'고 썼습니다. 예를 들어, 새로 갑상선 유두암 진단을 받은 환자에게 의사는 환자에게 수술 상담을 최우선 순위로 삼으라고 주장했습니다('이 암은 가장 중요한 암입니다. 이 암은 당신에게 가장 중요한 것이어야 합니다'), 환자가 업무 마감일과 충돌할 수 있다는 우려를 표명했음에도 불구하고. 고혈압, 고지혈증, 당뇨병에 대한 약물 치료 시작을 보여주는 동영상에서는 일반적으로 한 가지 치료 옵션이 권장되는 것으로 제시되고 환자는 이에 동의할 것으로 예상(설득)되었습니다. 이러한 문제는 치료 옵션을 식별할 때뿐만 아니라 치료 목표를 정의할 때도 나타났습니다. 예를 들어, 당뇨병 환자의 경우 의사는 국가 가이드라인에 정의된 치료 목표를 선언했지만 환자에게 이러한 목표에 동의하는지 묻지 않았습니다. 
Physicians often failed to allow or encourage patient autonomy or choice. This was rarely overt (i.e. there was only one video in which only a single option was presented). However, subtle statements frequently subverted autonomy. As one investigator wrote, physicians ‘steamrolled over the patient’s wishes.’ For example, with a patient with newly-diagnosed papillary thyroid cancer, the physician insisted that the patient make surgical consultation a top priority (‘This cancer is the most important thing. It has to be the most important thing for you’), even after the patient expressed concerns that this would conflict with work deadlines. In videos showing initiation of drugs for hypertension, hyperlipidemia, or diabetes, a single treatment option was typically presented as recommended, and the patient was expected (convinced) to acquiesce. This problem appeared not only in identifying treatment options; it also showed up in defining goals of care. For example, in a patient with diabetes, the physician declared treatment targets as defined by national guidelines, but never asked the patient if she agreed with these targets.

의사는 일반적으로 환자에게 관리 옵션 목록을 제공했지만, 환자에게 이러한 옵션 중 하나를 선택하도록 명시적으로 권유하는 경우는 드물었습니다. 상담의 여러 단계에서 어려움이 드러났습니다.

  • 상담의 목표(해결해야 할 문제)를 정의할 때 새로 섬유근육통을 진단받은 환자는 진단의 정확성을 다시 확인하고자 했지만, 의사는 곧바로 치료 옵션으로 대화를 이끌었습니다.
  • 고혈압에 대한 초기 약물 치료를 논의할 때 환자에게 세 가지 옵션이 주어졌지만 의사는 환자의 의견을 묻지 않고 환자의 의학적 문제에 맞춰 가장 적합한 옵션을 선택했습니다.
  • 치료 목표를 고려할 때 의사는 가이드라인에 따른 당화혈색소(A1c) 목표치를 제시하고 그보다 조금 더 높은 목표를 제시하는 환자의 제안을 무시했습니다.

이와 대조적으로 한 동영상에서는 의사가 고지혈증 환자에게 "어떤 약이 가장 좋을지 결정할 수 있도록 도와달라"고 요청했습니다. 
Although physicians usually offered patients a list of management options, only infrequently did they explicitly invite patients to participate in choosing among these options. Difficulties surfaced at various stages in the encounter.

  • When defining the goal of the encounter (problem to be addressed), a patient with newly-diagnosed fibromyalgia wanted to revisit the accuracy of the diagnosis, yet the physician steered the conversation immediately to treatment options.
  • When discussing initial drug treatment for hypertension, a patient was given three options but the physician selected her top choice (tailored to the patient’s medical problems) without soliciting the patient’s input.
  • When considering goals of therapy, a physician invoked a guideline-aligned A1c target and subsequently brushed aside the patient’s suggestion of a slightly higher target.

By contrast, in one video the physician invited a patient with hyperlipidemia, ‘I’d like you to help me decide what would be the best medication for you.’

의사는 종종 그 만남에서 [해결해야 할 핵심 문제를 정의하는 데 실패]했습니다. 위의 갑상선암과 섬유근육통의 예는 이 점을 잘 보여줍니다. 환자를 참여시키지 않음으로써 환자에게 가장 중요한 문제(상충되는 개인적 우선순위 또는 진단의 정확성에 대한 우려)를 고려하지 않고 궁극적으로 의사에게 가장 중요한 문제(치료)에 집중하게 됩니다. 이 문제는 치료 옵션을 논의할 때 다른 만남에서도 나타났습니다. 예를 들어, 의사는 일반적으로 치료 효과에 초점을 맞추는 반면 환자는 부작용과 비용에 대해 인내심을 발휘했습니다.
Physicians often fell short in defining the core problem to be addressed in the encounter. The examples above of thyroid cancer and fibromyalgia illustrate this point: By failing to involve the patient, they ultimately focused on the problem most salient to the physician (treatment) without considering the problem most salient to the patient (juggling conflicting personal priorities or concerns about the accuracy of diagnosis). This issue surfaced in other encounters when discussing treatment options. For example, physicians commonly focused on treatment efficacy while patients perseverated on side effects and cost.

의사들은 때때로 [근본적인 우려, 두려움, 질문]을 탐색하고 해결하지 못했습니다.

  • 한 고혈압 환자는 '약을 먹어본 적이 없어요. 아스피린도 거의 먹지 않아요. 그런데 이제 두 가지 약을 먹으라고 하네요'라고 말했습니다. 의사는 약이 저렴하고 매우 안전하다고 짧게 대답한 후 '그러니 걱정하지 않으셔도 됩니다'라고 말하며 대화를 끝냈습니다.
  • 고지혈증 환자에게 의사는 다른 옵션은 제시하지 않고 스타틴 계열의 약을 추천하면서 '부작용이 거의 없다'고 언급했습니다. 환자가 '괜찮지만 얼마나 심각한가요?"라고 묻자 의사는 "콜레스테롤을 효과적으로 낮추는 한 심각하지 않습니다."라고 대답했습니다. 장기적으로는 정말 큰 차이를 만들 수 있습니다."라고 대답했습니다. 그 후 의사는 이 문제에 대해 더 자세히 알아보지 않고 처방전을 작성했습니다.

Physicians occasionally failed to explore and address underlying concerns, fears, and questions.

  • A patient with hypertension expressed, ‘I’ve never had to take anything. I barely ever take an aspirin. Now you’re talking about putting me on two drugs.’ The physician responded with a brief assurance that the medication was inexpensive and quite safe, and then terminated the conversation with, ‘So that shouldn’t concern you.’
  • For a patient with hyperlipidemia the physician recommended a statin drug (without offering other options) and noted ‘few side effects.’ When the patient asked, ‘I guess that’s okay, but how serious is this?’ the physician responded, ‘Well, it’s not serious as long as we effectively lower your cholesterol. It can really make a big difference in the long run.’ The physician then moved forward with writing a prescription without further exploring this concern.

의사 결정의 [맥락을 적절히 탐색하지 않는 것]은 어디에나 있었습니다. 의사들은 환자에게 선호도, 가치관, 자원, 제약 조건에 대해 직접 질문하는 경우가 거의 없었으며, 설령 질문을 하더라도 짧고 피상적인 수준에 그쳤습니다. 의사들은 상황을 완전히 파악하기 위해 추가 질문을 하는 경우는 거의 없었습니다. 의료팀이나 시스템의 자원과 제약을 탐색하는 의사의 사례는 훨씬 더 적었습니다.
Failure to adequately explore the context of decision-making was ubiquitous. Physicians rarely asked the patients directly about their preferences, values, resources, and constraints, and when they did the inquiry was brief and superficial. Physicians never followed with additional probing questions to fully explore the situation. We saw even fewer examples of physicians exploring the resources and constraints of the health care team or system.

좀 더 넓게 보면, 이러한 실패와 다른 실패는 환자가 있는 곳에서 환자를 만나지 못하는 것, 즉 의사의 의제와 환자의 의제 사이의 단절로 통합됩니다. 위의 예 외에도 당뇨병 환자의 다른 동영상에서 의사는 약물 치료를 강화하라는 권고를 완화할 수 있는 생활 습관 조치에 관한 과거의 논의를 기억하지 못했습니다.
Viewed more broadly, these and other failures coalesce into a failure to meet the patient where they are—a disconnect between the physician’s agenda and the patient’s. In addition to the examples above, in a different video of a patient with diabetes, the physician failed to remember past discussions regarding lifestyle measures that might have mitigated recommendations to intensify drug treatment.

거의 모든 동영상에서 의사는 [환자와 함께 환자의 욕구, 우선순위, 동반 질환, 자원 등을 통합하는 데 실패]했습니다. 많은 동영상에서 의사는 매우 제한된 범위의 옵션을 제시했습니다. 다른 영상에서는 의사가 여러 가지 옵션을 제시했지만 환자의 고유한 요구 사항을 고려하여 명시적으로 우선순위를 정하지 않았습니다. 이러한 행동은 비개인적인(맞춤화되지 않은 일반적인) 계획으로 이어졌습니다.
In nearly every video, the physician failed to work jointly with the patient to integrate their desires, priorities, comorbidities, resources, etc. In many videos the physician presented a very limited range of options. In others, the physician offered several options but did not explicitly prioritize these considering the patient’s unique needs. Such actions resulted in an impersonal (untailored, generic) plan.

마지막으로, 임상의와 환자 간의 [관계를 구축하는 데 전반적으로 실패]한 것으로 나타났습니다. 어떤 경우에는 '증기 압연' 욕구와 잊혀진 대화의 경우처럼 의도하지는 않았지만 적극적으로 (비록 의도하지는 않았지만) 이러한 관계가 훼손되었습니다. 다른 영상에서는 서두르는 대화, 고민을 털어놓지 않거나 무시하는 말투 등 좀 더 미묘했습니다. 이는 관계에 상처를 입혔다기보다는 소홀히 한 것을 반영합니다.
Finally, we noted a general failure to build the clinician-patient relationship. In some instances, this relationship was actively (albeit unintentionally) undermined—as in the cases of ‘steamrolled’ desires and forgotten conversations. In other videos it was more subtle—a hurried conversation, failure to explore concerns, or dismissive tone of voice. These reflect a neglect of rather than injury to the relationship.

SDM의 렌즈를 통한 조사(2단계)
Examination through lens of SDM (stage 2)

위의 결과를 바탕으로 상호 보완적인 인사이트를 발견하기 위해 두 가지 SDM 렌즈(모델)를 사용하여 이러한 동영상을 체계적으로 재검토했습니다(표 2). 부록 2에는 방법, 관찰 및 분석에 대한 자세한 내용이 나와 있습니다(보충 자료).
Building on the above results, we systematically re-examined these videos (Table 2) using 2 lenses (models) of SDM in hope of discovering complementary insights; Appendix 2 details our methods, observations, and analyses (Supplementary Material).

참여 중심 모델
Involvement-focused model

[참여 중심 모델]은 [SDM의 전통적인 개념]으로, 환자가 의사 결정에 적절히 참여하여 자율성을 존중하고 가치에 기반한 의사 결정을 내릴 수 있도록 돕는 것이 SDM의 주된 이유입니다. 이러한 접근 방식의 대표적 모델인 [Three Talk 모델](Elwyn 외. 2017)은 SDM 회의에서 환자가 [팀 토크, 옵션 토크, 결정 토크]의 3단계 대화에 참여하도록 합니다. 우리는 이 모델을 사용하여 참여 중심 렌즈를 통해 해석된 관찰을 구조화했습니다.
The involvement-focused model represents a traditional concept of SDM wherein the predominant reason for SDM is to ensure that patients are adequately involved in decision-making, to respect their autonomy and help them make value-informed decisions. Representative of this approach, the Three Talk model (Elwyn et al. 2017) involves patients at 3 stages of conversation in SDM encounters: team talk, option talk, and decision talk. We used this model to structure observations interpreted through the involvement-focused lens.

팀 대화와 관련된 문제
Issues related to team talk

[팀 대화]는 환자를 의사 결정에 참여하도록 유도하고 협력 관계를 구축합니다. 위에서 언급한 바와 같이, 의사들이 환자를 의사 결정에 참여시키고 동등하게 협력하도록 초대하는 경우는 거의 없었습니다. 또한, 많은 경우 대화가 오히려 [환자의 자율성을 떨어]뜨리고 [의사와 환자 간의 권력 차이를 확대]하는 것으로 나타났습니다. 갑상선암 환자(환자에게 즉시 외과 의사를 만나라고 주장), 섬유근육통 환자(진단 재검토 요청 무시), 당뇨병 환자(치료 강화를 연기하자는 환자의 제안 묵살) 등이 그 예입니다.
Team talk invites the patient to participate in decision-making and establishes a collaborative relationship. As noted above, only rarely did physicians invite patients to engage in decision-making and collaborate as equals. Moreover, we noted that in many instances the conversation actually diminished patient autonomy and escalated the power differential between physician and patient. Examples include the patients with thyroid cancer (insisting the patient see a surgeon immediately), fibromyalgia (ignoring requests to revisit the diagnosis), and diabetes (dismissing the patient’s proposal to defer treatment intensification).

옵션 대화와 관련된 문제
Issues related to option talk

옵션 대화에서는 의사가 제시한 옵션의 장단점에 대해 논의합니다. 거의 모든 만남에서 의사는 거의 옵션을 제공하지 않았으며 각 옵션에 대한 정보도 제한적으로만 제공했습니다(위 및 표 2 참조).
Option talk discusses the pros and cons of the options identified by the physician. In nearly every encounter, physicians offered very few options and provided only limited information about each option (see above and Table 2).

또한 의사들은 [아무것도 하지 않는 옵션]을 제공한 적이 없었습니다. 그들은 항상 환자에게 조치가 필요하고 환자가 원한다고 가정했습니다.
Moreover, physicians never offered the option of doing nothing. They always assumed action was needed and desired by patient.

이러한 [제한된 옵션 목록]은 대안, 부작용, 비용, 일반적으로 차선책이 선호되는 상황(예: 비용 또는 편의성 기준) 또는 즉각적인 조치가 없을 때의 실제 위험에 대한 불완전한 지식 등 [임상의의 지식 격차]를 반영하는 것일 수 있다는 것을 감지했습니다. 지식 격차는 또한 환자의 질문에 대한 불완전한 답변(부분적인 답변 또는 질문과 관련이 있지만 직접적으로 다루지 않는 답변)의 기저에 있는 것으로 보였습니다. 지식 격차가 흔해 보였지만, 실제 지식 오류는 이 동영상에서 드물게 나타났습니다.
We sensed that such constrained option lists could reflect clinician knowledge gaps—incomplete knowledge of alternatives, side effects, costs, situations in which a typically-suboptimal option might emerge as preferred (e.g. based on cost or convenience), or the true risk of no immediate action. Knowledge gaps also seemed to underlie incomplete responses to patient questions (partial answers or answers related to but not directly addressing the question). Although knowledge gaps seemed common, actual knowledge errors were infrequent in these videos.

디시전 토크와 관련된 문제
Issues related to decision talk

[의사 결정 대화]는 환자의 선호도를 파악하고 통합하여 결정을 내리는 과정입니다. 의사는 일반적으로 환자의 선호도를 구하거나 환자를 진지하게 대하지 않았습니다. 예를 들어 당화혈색소(A1c)가 목표치보다 약간 높은 당뇨병 환자에게 메트포르민 용량을 늘려야 하며 생활습관 개선만으로는 충분하지 않다는 말을 들었다면, 이는 목표 설정과 치료 결정 모두에 환자가 관여하는 것에 위배됩니다. 마찬가지로, 유두상 갑상선암 환자는 우려를 표명했지만 대놓고 무시당했습니다.
Decision talk identifies and integrates the patient’s preferences culminating in a decision. Physicians typically did not seek the patient’s preferences or earnestly entertain them. For example, a patient with diabetes and an A1c slightly above goal was told she needed to increase the dose of metformin and that lifestyle measures would be inadequate; this violates involvement in both establishing targets and deciding on treatment. Likewise, the patient with papillary thyroid cancer expressed concerns that were overtly brushed aside.

문제 중심 모델
Problem-focused model

[문제 중심 모델](목적 중심 SDM, 표 1)에서 SDM의 목적은 [환자가 겪고 있는 문제(좁게 또는 넓게 정의)를 해결]하기 위한 것입니다.

  • 참여 중심 모델이 대화 유형과 자율성에 중점을 두는 반면,
  • 문제 중심 접근 방식은 환자와 의사가 이 문제에 대한 적절한 대응책을 함께 발견할 수 있는 방법에 중점을 둡니다. 이러한 단계에는 다음이 포함됩니다.
    • 문제 특성 파악,
    • 가능한 해결책 제안,
    • 여러 가능성 중 판단 방법 결정,
    • 치료 계획 개발

이러한 요소(문제, 가능성, 판단 접근 방식, 치료 계획) 중 어느 것도 대화 전에 완전히 정의된 것은 없습니다. 대화의 목적은 각 요소를 발견하고, 활용하고, 조정하여 잘 형성된 대응 가능한 문제 해결책을 만들기 위해 '함께 노력'하는 것입니다. 이와 대조적으로, 많은 [참여 중심 모델]은 문제와 가능성(옵션)을 미리 알고 있다고 가정합니다. 우리는 문제 중심 렌즈를 통해 관찰한 내용을 해석하면서 추가적인 인사이트를 발견했습니다.
In the problem-focused model (Purposeful SDM, Table 1), the reason for SDM is to address a problem (narrowly or broadly defined) that the patient is experiencing.

  • Whereas the involvement-focused model concentrates on types of talk and autonomy,
  • a problem-focused approach concentrates on how the patient and physician can discover together an appropriate response to this problem. These steps include:
    • characterize the problem,
    • propose possible solutions,
    • determine how to judge among the possibilities, and
    • develop a care plan.

None of these elements (the problem, possibilities, judgement approach, or care plan) are fully defined prior to the conversation. The purpose of conversation is to ‘endeavor together’ to discover, use, and adapt each element to craft a well-formed, responsive problem resolution. By contrast, many involvement-focused models assume that the problem and possibilities (options) are known up-front. We identified additional insights as we interpreted our observations through the problem-focused lens.

문제 특성화와 관련된 문제
Issues related to characterizing problems

환자와 임상의 사이의 숙의 과정에는 [문제 상황의 인간적 의미를 이해]하는 것이 포함됩니다. 이 동영상에서 의사들은 일반적으로 문제에 대한 매우 좁은 개념을 전달했으며, 종종 '높은 LDL = 스타틴' 또는 '갑상선암 = 수술 상담'과 같은 특정 해결책과 연결시켰습니다. 환자나 상황보다는 검사 결과, 조직 검사 또는 질병을 관리하는 경향이 있었습니다.
Part of deliberation between patient and clinician involves developing an understanding of the human significance of the problematic situation. In these videos, physicians typically conveyed a very narrow conception of the problem, often linking it with a specific solution: ‘high LDL = statin’ or ‘thyroid cancer = surgical consultation.’ They tended to manage the test result, biopsy, or disease, rather than the patient or situation.

이 과정에서 [상황의 진정한 의미], 즉 문제와 잠재적 해결책이 [환자의 건강과 삶 전반에 미치는 영향]을 전달하지 못했습니다. 이러한 실패에는 건강에 미치는 영향보다는 다음이 포함될 수 있습니다.

  • 중간 목표(예: 심혈관 질환 위험보다는 당화혈색소 또는 혈압 목표)에 초점을 맞추거나,
  • 질병의 영향을 최소화하거나 과장하거나,
  • 문제와 해결책이 불일치하는 경우(예: 심각한 문제와 사소한 해결책 또는 그 반대의 경우)

예를 들어, 유두상 갑상선암은 비교적 무해하여 몇 주만 지연해도 문제가 거의 발생하지 않는데도 의사는 긴급한 수술 의뢰를 고집했습니다. 이러한 실패는 의사(및 상황에 대한 잘못된 이해)나 대화 중 잘못된 단계로 인해 발생할 수 있습니다. 결론적으로, 대부분의 '공유' 의사 결정은 공식적이고 의사의 치료 목표에 초점을 맞춘 것이었습니다.
In doing so, they failed to convey the true meaning of the situation—the impact (of both the problem and potential solutions) on the patient’s health and life in general. This failure could involve

  • focusing on interim objectives rather than health impact (e.g. A1c or blood pressure targets rather than risk of cardiovascular events),
  • minimizing or overstating the impact of illness, or
  • mismatching the problem and solution (e.g. a serious problem and a trifling solution, or vice versa).

For example, papillary thyroid cancer is relatively innocuous, such that a few-week delay would rarely cause problems, yet the physician insisted on urgent surgical referral. Such failures can arise within the physician (and their [mis]understanding of the situation), or from mis-steps in the conversation. As a corollary, most ‘shared’ decision-making was formulaic and focused on physician goals of care.

예외적으로, 의사와 환자가 합의된 관심사(예: 문제)를 해결하기 위해 함께 노력한 몇 가지 사례를 확인했습니다. 예를 들어, 다른 갑상선암 동영상에서 의사는 환자가 외과 의사를 만나야 한다는 일방적인 결정을 내렸지만, 환자가 스케줄에 대한 상충을 명확히 밝히자 대화는 외과 의사와의 상담을 촉진하는 방법을 모색하는 공동 문제 해결 심의 과정으로 전환되었습니다. 환자는 초기 결정이 내려졌음에도 불구하고 궁극적으로 만족하는 모습을 보였습니다. 이는 공동 심의를 거치지 않은 '동일한' 시나리오에서 의제와 우선순위가 충돌한 것과는 극명한 대조를 이룹니다.
As an exception, we identified a few instances in which the physician and patient worked together to address an area of agreed-upon concern (i.e. a problem). For example, in the other thyroid cancer video the physician made a unilateral decision that the patient needed to see a surgeon, but when the patient articulated scheduling conflicts the conversation transformed into a joint problem-solving deliberation exploring how to facilitate the surgeon consultation. The patient ultimately appeared satisfied even though the initial decision was thrust upon him. This contrasts sharply with the ‘same’ scenario without joint deliberation that led to a clash of agendas and priorities.

가능성 제안과 관련된 문제
Issues related to proposing possibilities

심의의 또 다른 측면은 문제에 대한 [가능한 대응을 식별]하는 것입니다. 환자와 의사는 함께 협력하여 [가능한 해결책을 제안하거나, 만들거나, 발견]할 수 있습니다. 위에서 언급한 바와 같이, 이 동영상에서 의사들은 거의 항상 하나의 실행 가능한 옵션을 제시하고 다른 대안(있는 경우)은 열등한 것으로 묘사했습니다. 까다로운 수술 상담을 해결한 대화(위)는 공동으로 협력하여 대안을 찾아낸 좋은 예입니다.
Another aspect of deliberation involves identifying possible responses to a problem. Working together, patients and physicians may propose, create, or discover possible solutions. As noted above, in these videos physicians nearly always offered a single viable option, portraying alternatives (if any) as inferior. The conversation that resolved the challenging surgical consultation (above) is a good example of working jointly to identify alternative possibilities.

가능성 판단과 관련된 문제
Issues related to judging possibilities

가능성이 발생하면 그 중에서 [판단하는 방법(예: 옵션 비교, 협상, 문제 해결, 통찰력 개발)을 수립]해야 합니다. 이상적으로는 환자와 의사가 함께 이 작업을 수행합니다. 이 비디오에서 의사가 환자와 적극적으로 협력하여 서로의 관심사를 해결하는 경우는 드물었습니다. 오히려 거의 항상 의사가 진료실에 들어가기 전에 결정을 내린 다음 환자가 이 계획에 동의하도록 설득하는 데 시간을 소비하는 것처럼 보였습니다. 이러한 계획은 대개 합리적이며 가이드라인에 부합하는 것이었지만 환자의 의견은 반영되지 않았습니다. 한 조사자는 '추론은 임상의의 머릿속이 아닌 교과서에서 일어났다'고 기록했습니다.

As possibilities arise, methods of judging among them (e.g. weighing options, negotiating, problem solving, developing insight) should be established. Ideally, patients and physicians do this together. In these videos, physicians infrequently worked actively with patients to solve a problem of mutual concern. Rather, it nearly always appeared that the physician had made the decision before walking into the room, and then spent time convincing the patient to agree with this plan. Such plans were usually well-reasoned and in alignment with guidelines, but devoid of patient input. As one investigator documented, ‘Reasoning happened in the textbook (not in the clinician’s head).’

우리는 치료 목표(예: 의사가 당뇨병, 고지혈증 또는 고혈압에 대한 가이드라인 기반 치료 목표에 대해 논의하지 않고 선택)와 치료 접근 방식(예: 약물, 약물 용량, 수술 상담)에 대한 대화에서 이러한 단점을 인식했습니다. 또한 의사가 섬유근육통 진단에 대한 환자의 불확실성을 무시하고 치료 계획의 윤곽을 미리 제시하는 경우와 같이 진료 흐름에서도 이러한 문제를 확인할 수 있었습니다. 이러한 행동은 거의 항상 의사와 환자 사이에 마찰을 일으켰고, 의사는 자신의 해결책이 좋다고 환자를 설득하려고 노력했습니다.
We recognized this shortcoming in conversations about both goals of treatment (e.g. the physician selecting [rather than discussing] guideline-based treatment targets for diabetes, hyperlipidemia, or hypertension) and treatment approach (e.g. drug, drug dose, surgical consultation). We also saw this in the encounter flow, as when the physician ignored the patient’s uncertainty with her diagnosis of fibromyalgia and forged ahead to outline a care plan. Such actions nearly always resulted in friction between physician and patient, with the physician trying to convince the patient that their solution was good.

케어 플랜 개발과 관련된 문제
Issues related to developing care plans

공유된 심의 프로세스는 문제에 대응하고 지적, 실제적, 정서적으로 [합당한 치료 계획을 수립]하기 위해 노력합니다. 지적, 실용적, 정서적 일관성을 갖춘 대응 계획의 예는 거의 없었습니다. 대신 일반적으로 치료 계획 내에서 [단절] 또는 [잔류 갈등]이 관찰되었습니다.

  • 갑상선암 동영상은 문제의 심각성(상담의 시급성이 낮음)과 제안된 해결책의 영향(환자의 삶에 큰 지장을 초래함) 사이의 갈등을 보여줍니다.
  • 두 섬유근육통 동영상의 치료 계획은 체계적이지 않고 불완전하며 환자의 경험과 단절되어 있어 환자가 실질적인 다음 단계에 대해 당황하고 좌절하며 방향을 잃게 만들었습니다.
  • 다른 영상에서는 환자의 질문에 대한 답변이 없거나 치료 목표가 강요될 때 지적 단절이 빈번하게 발생하고, 환자의 우려를 무시하거나 자율성을 무시할 때 감정적 갈등이 발생하는 것을 볼 수 있었습니다.

The shared deliberative process works towards a care plan that responds to the problem and makes sense intellectually, practically, and emotionally. We saw few examples of responsive plans with intellectual, practical, and emotional coherence. Instead, we typically observed a disconnect or residual conflict within the care plan.

  • The thyroid cancer video illustrates a conflict between the severity of the problem (low urgency for consultation) and the impact of the proposed solution (large disruption of the patient’s life).
  • The care plans in both fibromyalgia videos were poorly organized, incomplete, and disconnected from the patient’s experience, leaving the patient perplexed, frustrated, and disoriented regarding practical next steps.
  • In other videos we saw frequent intellectual disconnects when patient questions remained unanswered or treatment targets were imposed, and emotional conflicts when patient concerns were brushed aside or autonomy had been disregarded.

또한 [명시적인 결정이 없는 경우]도 많이 보았습니다. 의사는 일반적으로 여러 가지 제안을 하고 선호하는 옵션을 언급하기도 했지만, 환자에게 이 계획에 대한 동의를 요청하거나 받지 않았습니다. 이러한 실패는 단순히 의사 소통 능력이 부족해서일 수도 있지만, 일부 영상에서는 의사가 환자의 동의를 얻지 못했다는 인식 때문에 의도적으로 동의를 구하지 않는 것처럼 보이기도 했습니다. 그 외에는 최선의 접근 방식에 대한 불확실성 때문인 경우가 더 많았습니다.
Additionally, we saw many instances in which there was no explicit decision. The physician had typically made a number of suggestions, and might have even stated a preferred option, but never asked for or received agreement for this plan from the patient. Although such failures could reflect merely poor communication skills, in some videos the physician seemed to deliberately avoid asking for agreement, due perhaps to awareness that the conversation had failed to garner the patient’s buy-in. Less often, vacillation seemed due to uncertainty in the best approach.

 

환자-임상의 상호 작용 및 관리 추론 통합(3단계)
Integrating patient-clinician interactions and management reasoning (stage 3)

마지막으로, 관리 추론에서 커뮤니케이션과 환자-임상의 상호작용의 역할을 이해하고자 하는 본 연구의 목표를 달성하기 위해, 이전에 확인된 관리 추론의 12가지 특징(Cook 외. 2022b)과 두 SDM 모델 간의 상호 영향을 체계적으로 고려함으로써 2단계 결과와 관리 추론의 틀을 통합했습니다(부록 3에서는 관찰 및 분석에 대한 자세한 내용(보충 자료)을 확인할 수 있습니다). 이 분석에서 얻은 수많은 인사이트(표 3) 중 주목할 만한 결과는 다음과 같습니다:
Finally, to achieve our study’s goal of understanding the role of communication and patient-clinician interaction in management reasoning, we integrated stage 2 findings and the framework of management reasoning by systematically contemplating the reciprocal implications between 12 previously-identified features of management reasoning (Cook et al. 2022b) and the two SDM models; Appendix 3 details our observations and analyses (Supplementary Material). Among the numerous insights from this analysis (Table 3), notable findings include:

 

  • [문제 중심 모델]에서 선호도는 환자가 아닌 [상황의 특징]입니다. 문제가 있는 상황은 바람직하지 않으며, 가능한 해결책은 다양한 제약 조건 내에서 인간의 범위 내에서 상황에 대응하는 성공 여부에 따라 바람직(선호) 또는 그렇지 않은 것으로 결정됩니다. [관리 추론]은 문제 상황에 대한 바람직한 해결책을 모색합니다.
  • In the problem-focused model, preferences are a feature of the situation not the patient. Problematic situations are undesirable, and possible solutions are desirable (preferred) or otherwise according to their success in responding to the human scope of the situation within various constraints. Management reasoning seeks a preferred solution to a problematic situation.
  • [환자, 의사, 상황적 제약 사이의 역동적인 상호 작용]은 거의 모든 대면 진료에서 보통 또는 높은 수준입니다. 문제가 쉽게 파악되거나 쉽게 해결될 수 있었다면 환자는 집에 머물렀을 것입니다. 임상의는 [일이 예상대로 진행되지 않을 때 개입]합니다.
  • The dynamic interplay among patient, physician, and situational constraints is moderate or high for virtually all face-to-face encounters. If the problem were readily characterized or easily resolved, the patient would have stayed home. Clinicians get involved when things do not go as expected.
  • 의사들은 [충분한 정보를 바탕으로 상호 합의]하여 선택하기보다는 [잠정적이거나 임시적인 용어('일단 약을 시작하고 마음에 들지 않으면 바꾸면 된다')로 '결정']을 내리는 경우가 많습니다. 물론 '관리 계획의 지속적인 모니터링과 조정'은 관리 추론의 핵심 기능이지만(Cook et al., 2019), 의사들은 종종 [질문을 미루고, 우려를 제쳐두고, 대립을 피하기] 위해 이러한 [전제에 지나치게 의존]하는 것처럼 보였습니다('캔을 걷어차 버리자').
  • Physicians often couched “decisions” in tentative or temporary terms (‘Let’s start the medication and if you don’t like it then we can change’) rather than mutually agreeing upon a well-informed choice. Of course, ‘ongoing monitoring and adjustment of the management plan’ is a key feature of management reasoning (Cook et al., 2019); but physicians often seemed to rely excessively on this premise to defer questions, brush aside concerns, and avoid confrontation (‘kick the can down the road’).
  • [언어]는 [미묘하지만 중요한 방식으로 문제를 형성하고 변화]시킬 수 있습니다. 예를 들어, 고지혈증을 'LDL을 낮춰야 한다' 또는 '심장마비 위험을 줄여야 한다'는 식으로 표현하면 문제의 성격과 해결책을 협상하는 방식이 크게 달라질 수 있습니다.
  • Language can shape and change the problem in subtle but important ways. For example, framing hyperlipidemia as needing to ‘lower your LDL’ or ‘reduce your risk of a heart attack’ could lead to dramatically different problem characterizations and solution negotiations.
  • 우리는 이전에 정의한 관리 스크립트('합리적인 관리 계획의 개발을 용이하게 하기 위해 관리 옵션과 임상의 업무를 시간적 또는 논리적 순서로 표현하고 연결하는 사전 편집된 개념적 지식 구조'(Cook et al., 2022b))와 관련된 [두 가지 새로운 인사이트]에 주목하고 나중에 자세히 설명했습니다(Cook et al., 2022a).
  • We noted 2 novel insights related to management scripts, which we previously defined (‘precompiled conceptual knowledge structures that represent and connect management options and clinician tasks in a temporal or logical sequence to facilitate development of a rational management plan’ (Cook et al., 2022b)) and later elaborated in detail (Cook et al., 2022a).
    • 첫째, 의사의 [관리 스크립트]는 [의사의 요구](예: 효율적인 만남 또는 신속한 결정 촉진)와 [환자의 요구](문제 이해, 질문에 대한 답변, 개별화된 결정 지원)의 [우선순위]에 따라 연속체를 따라 달라집니다.
    • First, physicians’ management scripts vary along a continuum in prioritizing physician needs (e.g. promoting an efficient encounter or expedited decision) vs patient needs (understanding problems, answering questions, supporting individualized decisions).
    • 둘째, [환자]는 의사와의 상호작용을 안내하는 [자신만의 인지적 스크립트]를 가지고 있습니다. 환자와 의사의 스크립트는 모두 유사한 범주(문제, 옵션, 선호도, 교육 요구, 대인관계, 진료 흐름(Cook et al., 2022a))로 구성될 가능성이 높지만, 각 범주에 대해 인스턴스화된 세부 사항은 상당히 다를 수 있습니다.
    • Second, patients have their own cognitive script that guides their interactions with the physician. Both patient and physician scripts likely comprise similar categories (problems, options, preferences, education needs, interpersonal interactions, and encounter flow (Cook et al., 2022a)); but the details instantiated for each category could differ substantially.
  • 문제를 해결할 때 의사는 [일반적으로 이미 알고 있는 지식에 의존]하게 되는데, 이 지식이 당면한 문제를 해결하는 데 항상 적절하거나 충분하지 않을 수 있습니다. 부주의적이든(모르는 것을 알기 어렵기 때문) 의도적이든(지식 격차를 인정하거나 추가 정보를 찾는 것을 피하기 위해서) 이러한 지식 격차는 문제 중심 SDM을 방해합니다.
  • When grappling with a problem, physicians will typically rely on what they already know—which may not always be relevant or sufficient to resolve the problem at hand. Whether inadvertent (it is difficult to know what one does not know) or conscious (to avoid admitting a knowledge gap or seeking additional information), such knowledge gaps will interfere with problem-focused SDM.
  • 환자-의사 관계의 특성은 모델에 따라 다릅니다.
    • [참여 중심 모델]에서는 공감과 신뢰의 효과적인 관계가 효과적인 SDM의 매개체(전구체)입니다. 이와 대조적으로
    • [문제 중심 모델]에서는 효과적인 관계가 SDM의 산물입니다(즉, 의사와 환자가 문제를 해결하기 위해 함께 노력하면서 관계가 발전합니다).
  • The nature of patient-physician relationships differs by model.
    • In the involvement-focused model an effective relationship of empathy and trust is a mediator of (precursor to) effective SDM. By contrast,
    • in the problem-focused model an effective relationship is the product of SDM (i.e. the relationship develops as the physician and patient work together to solve problems).
  • [설득]도 모델에 따라 다르게 볼 수 있습니다
    • [참여 중심 모델]에서는 설득은 [나쁜 것으로 간주]된다. 권위, 권력 차이, 바람직하지 않은 행동을 하도록 영향력을 행사하는 것이고, 이러한 위험 때문에 환자를 의사 결정에 참여시켜 자율성을 보존해야 한다고 생각합니다. 
    • [문제 중심 모델]에서 설득은 좋은 것이고, 사실상 목표이다. 모든 이해관계자가 이 상황에서 최선의 행동 방침을 상호 설득할 때까지 함께 노력하는 것이지만, 과도한 권력 기반 설득은 이 방법을 왜곡하는 것입니다.
  • Persuasion can also be viewed differently, depending on the model.
    • In the involvement-focused model, persuasion is bad, connoting authority, power differential, and influence to do (undesirable) actions; this risk drives the need to preserve autonomy by involving patients in decision-making.
    • In the problem-focused model, persuasion is good—indeed, the goal: stakeholders work together until all are mutually persuaded that a course of action is best in this circumstance; undue power-based persuasion is a corruption of this method.

 

제안된 모범 사례(4단계)
Proposed best practices (stage 4)

이러한 결과를 바탕으로 우리는 관리 추론에서 환자-임상의 상호작용에 대한 몇 가지 잠정적인 모범 사례를 확인했으며, 이는 상자 1에 나열되어 있습니다.
Drawing on these findings we identified several tentative best practices for patient-clinician interactions in management reasoning, listed in Box 1.

토론
Discussion

이 연구는 관리 추론이 전적으로 [임상의의 머릿속]이 아닌 [개인 사이의 공간]에서 발생한다는 이전 연구(Cook 외. 2022b)의 관찰을 더욱 입증합니다. 두 가지 다른 SDM 렌즈를 통해 시뮬레이션된 환자-임상의 만남을 분석하여 관리 추론에서 의사소통과 환자-임상의 상호 작용의 상호 보완적인 측면을 조명했습니다. SDM의 [참여 중심 모델]은 환자의 자율성을 보호하고, 정보를 명확하게 교환하며, 최종 결정이 환자의 선호도에 부합하도록 관리 추론에 환자를 참여시키는 것이 중요하다는 점을 강조했습니다. [문제 중심 모델]은 공동의 노력에서 소통과 상호 작용이 어떻게 올바른 의사 결정을 촉진하는 데 도움이 되는지 밝혀냈습니다. 
This study further probes the observation from an earlier study (Cook et al. 2022b) that management reasoning occurs in the space between individuals and not entirely in the clinician’s head. Analysis of simulated patient-clinician encounters through 2 distinct SDM lenses illuminated complementary aspects of communication and patient-clinician interaction in management reasoning. The involvement-focused model of SDM highlighted the importance of involving patients in management reasoning so that patient autonomy is preserved, information is clearly exchanged, and final decisions align with the patient’s preferences. The problem-focused model revealed how communication and interaction in a shared endeavor serve to facilitate well-formed decisions.

이러한 모델을 함께 사용하면 [커뮤니케이션 실패에 대한 순진한(분석 1단계) 관찰을 이해]하는 데 도움이 됩니다. 예를 들어, 의사가 환자를 만나지 못했거나 선택권을 제공하지 않은 경우, 모델에 따라 다른 이유로 인해 특히 중요하다는 것을 이제 알 수 있습니다.

  • [참여 중심 관점]에서 볼 때, 이는
    • 환자와 임상의 사이의 거리를 악화시키고,
    • 종종 의사소통을 차단하고,
    • 불신을 키우고,
    • 자율성을 무시하고,
    • 만남을 의지의 대결로 바꾸고,
    • 환자를 복종적인 역할로 강요하고,
    • 지속적인 관리에 필요한 파트너십을 해치고,
    • 환자가 임상의의 조언을 따라야 한다는 확신을 갖지 못하게 만들었습니다.
  • [문제 중심의 관점]에서 볼 때, 이와 같은 관찰된 실패는 임상의가 다음과 같은 것을 의미합니다. 
    • 문제가 있는 곳에서 환자를 직접 만나지 않았기 때문에 문제의 중요성/의미에 대해 자세히 설명하거나 해결하지 못했습니다, 
    • 문제를 해결하기 위해 환자-임상의 팀의 잠재력을 충분히 개발하지 못했습니다, 
    • 임상의가 선험적으로 결정한 것보다 상황에 더 잘 대응할 수 있는 심의 옵션을 제시하지 않았습니다, 
    • 종종 만남에서 모든 심의를 제거했습니다 (옵션이 논의되지 않을 때 얼마나 많은 '추론'또는 '의사 결정'이 발생하는지에 대한 의문을 제기합니다). 
    • 불확실한 가치의 치료 계획을 초래했습니다.

 

Together, these models help make sense of the naïve (analysis stage 1) observations of communicative failures. For instance, we now recognize that when physicians failed to meet the patient where they were or offered no choices, this was especially significant—for reasons that differed depending on the model.

  • From the involvement-focused perspective this
    • exacerbated the distance between patient and clinicians, and
    • often shut down communication,
    • bred mistrust,
    • disrespected autonomy,
    • turned the encounter into a contest of wills,
    • forced the patient into a subservient role,
    • harmed the partnership needed for ongoing management, and
    • left the patient unconvinced that they should follow the clinician’s advice.
  • From the problem-focused perspective these same observed failures meant that the clinician
    • didn’t meet the patient where their problem lay (and hence the full significance/meaning of the problem was never elaborated or addressed),
    • didn’t develop the full potential of the patient-clinician team to address the problem,
    • didn’t bring forward options for deliberation that might better respond to the situation than those determined by the clinician a priori,
    • often removed all deliberation from the encounter (raising the question of how much ‘reasoning’ or ‘decision-making’ occurs when no options are discussed), and
    • resulted in a care plan of uncertain value.

제한 사항
Limitations

이 연구에는 한계가 있습니다.

  • 첫째, 우리가 관찰한 동영상은 실제와 같이 보이도록 의도되었지만 실제 임상의와 환자 간의 상호작용은 하나도 없었고 4편은 대본으로 제작되었습니다. 이는 동영상의 맥락(성인 외래 환자 일차 진료만), 수행 수준(절반은 의도적으로 열악한 것으로 선택되었고 모범적인 것은 없음), 원래 목적(평가자 교육 연구(Cook 외. 2009; Cook 및 Beckman, 2009)) 및 총 수와 함께 임상적 다양성을 종합적으로 제한했습니다. 현재의 결론이 적절히 뒷받침된다고 생각하지만, 다른 관점(예: 대본에 없는, 다른 임상 상황, 다른 수준의 임상의 수행 또는 교육)을 제공하는 다른 사례에 대한 분석을 통해 추가적인 인사이트가 나올 수 있을 것으로 예상합니다. 일부 독자는 다른 목적으로 제작된 동영상의 용도 변경을 무시할 수도 있지만, 이미 언급한 맥락적 제약에 비추어 볼 때 이는 사소한 문제라고 생각합니다.
  • 둘째, 많은 추론은 실패한 사례나 바람직한 행동을 취하지 않은 채 방치된 사례에 대한 관찰을 바탕으로 이루어졌으며, 이러한 추론은 당연히 관찰자의 관점에 따라 달라질 수 있습니다.
  • 셋째, 본 연구의 저자들은 기본 개념 프레임워크의 개발자(Cook 외. 2018; Hargraves 외. 2019)이자 관찰자 및 분석가로서 본 연구의 관찰자 및 분석가로서 참여했으며, 이는 강점이지만 확증 편향의 위험도 있습니다. 또 다른 강점으로는 이러한 동영상을 사용하여 이전에 발표한 연구를 확장하면서
    • (a) SDM 및 정성적 데이터 분석에 상당한 전문성을 갖춘 새로운 연구자(IGH)를 팀에 추가하고,
    • (b) 모든 동영상을 개별적으로 또는 그룹으로 재검토하여 풍부하고 새로운 데이터를 수집하고,
    • (c) 상호 보완적인 개념적 렌즈를 사용하여 이러한 데이터를 분석하고,
    • (d) 이러한 분석을 지속하여 SDM과 경영 추론을 통찰력 있게 통합하고 여러 모범 사례를 식별했다는 점을 들 수 있습니다.
  • 요컨대, 이러한 강점은 제안된 결론을 충분히 뒷받침하지만, 이러한 한계는 추가적인 인사이트를 발견해야 함을 시사합니다.

This study has limitations.

  • First, the videos we observed were intended to be realistic, but none were actual clinician-patient interactions and 4 were scripted. This, together with the videos’ context (only adult outpatient primary care), performance levels (half were intentionally selected as poor, and none were exemplary), original purpose (a study of rater training (Cook et al. 2009; Cook and Beckman, 2009)), and total number, collectively constrained the clinical variation. While we believe the present conclusions are appropriately supported, we surmise that additional insights could emerge from analysis of other encounters offering alternative perspectives (e.g. unscripted, different clinical contexts, or different levels of clinician performance or training). Some readers might also discount the repurposing of videos created for another purpose, however we believe this constitutes a minor concern relative to the contextual constraints already noted.
  • Second, many inferences were based on observations of failures or presumably desirable actions left undone; such inferences naturally depend on the perspectives of those observing.
  • Third, we authors were both the developers of the guiding conceptual frameworks (Cook et al. 2018; Hargraves et al. 2019) and the observers and analyzers in this study; while this is a strength, there is some risk of confirmation bias. Additional strengths include that as we extended our previously-published work using these videos, we
    • (a) added to our team a new investigator (IGH) with substantial expertise in SDM and qualitative data analysis;
    • (b) collected abundant, rich new data by re-reviewing all videos both independently and as a group;
    • (c) analyzed these data using complementary conceptual lenses; and
    • (d) continued this analysis to a achieve an insightful integration of SDM with management reasoning and identify several best practices.
  • In short, these strengths provide ample support for the conclusions proposed, but these limitations suggest that additional insights remain to be discovered.

시사점
Implications

이번 연구 결과는 교육, 임상 실무 및 향후 연구에 중요한 [시사점]을 제공합니다. 가장 중요한 것은 관리 추론이 임상의의 머릿속이 아닌 [개인 간의 공간에서 일어나는 의도적인 상호 작용]을 포함한다는 점입니다. [환자와 임상의]는 [서로 다른 관점, 지식 기반, 경험, 우선순위에서 '추론'을 하지만, 한 팀으로서 '추론']을 해야 합니다. 이는 관리 추론을 가르치고 평가하는 방식에 중요한 시사점을 줍니다. 관리 추론의

  • [일부 요소]는 [임상의 내부]에서 발생하는 것처럼 가르치고 평가할 수 있지만,
  • [환자와 임상의의 상호 작용]도 가르치고 평가해야 하며,
  • [어느 시점에는 전체 전체를 종합]해야 합니다.

Our findings have important implications for education, clinical practice, and future research. Most importantly, we affirm that management reasoning involves a deliberative interaction occurring in the space between individuals and not within a clinician’s mind. Both the patient and clinician are ‘reasoning’—from different perspectives, knowledge bases, experiences, and priorities, but hopefully as a team. This has notable implications for how we teach and assess management reasoning.

  • While some elements of management reasoning can be taught and assessed as though occurring within the clinician,
  • the patient-clinician interaction must also be taught and assessed;
  • and at some point the entire whole must be assembled.

관리 추론과 SDM은 모두 [상호 합의된 공유 문제], 예를 들어 진단 라벨이나 목표치를 벗어난 검사실 수치보다는 [생명이나 건강에 대한 상호 위협에 초점]을 맞출 때 가장 효과적입니다. [공유된 문제]

  • 공동의 문제 해결('함께 노력하기')을 유도하고,
  • 사회적 조정과 협력(의사소통적 행동)을 촉진하며,
  • 갈등과 과도한 설득의 필요성을 피할 수 있습니다.

이러한 관점에서 볼 때 관리 추론은 문제 해결을 위해 고심하는 것입니다.
Management reasoning and SDM both work best when focused on a shared, mutually-agreed problem—a mutual threat to life or health rather than, for example, a diagnostic label or off-target lab value. A shared problem

  • engenders joint problem-solving (‘endeavoring together’),
  • promotes social coordination and cooperation (communicative action (Bohman and Rehg 2017)), and
  • avoids conflict and the need for undue persuasion.

Viewed this way, management reasoning is deliberating to resolve the problem.

우리는 임상의와 환자라는 [두 가지 관리 스크립트]가 작용한다는 새로운 통찰력을 강조합니다. 우리는 이전에 [임상의의 관리 스크립트]에 대해 자세히 설명했습니다. (Cook 외. 2022a) [관리 추론의 환자 측면]은 지금까지 무시되어 왔습니다. 또한 일부 스크립트는 주로 임상의에게 도움이 되는 반면(즉, 진료 효율 개선), 다른 스크립트는 환자의 대응 결정을 돕는 데 더 중점을 둡니다(연속선상에 있음). 대부분의 스크립트는 의식적인 생각 없이 자발적으로 개발되며(Custers 2015), [환자 중심 스크립트를 촉진하는 방법]은 아직 불분명합니다. SDM과 관련된 관리 스크립트에 대한 추가 연구가 필요합니다.
We underscore the novel insight that there are two management scripts in play—the clinician’s and the patient’s. We previously elaborated on the clinician’s management script; (Cook et al. 2022a) the patient’s side of management reasoning has thus-far been neglected. We also note that some scripts primarily help the clinician (i.e. to improve encounter efficiency), while others (along a continuum) are more directed at helping the patient make a responsive decision. Most scripts develop spontaneously (Custers 2015)—without conscious thought—and it remains unclear how to promote patient-centered scripts. Further research on management scripts as they relate to SDM is needed.

특히 SDM과 관련된 관리 추론에 대한 연구가 초기 단계에 있다는 점을 강조합니다. 위에서 언급한 것 외에도 이 연구에서 제기된 중요한 질문은 다음과 같습니다: 공유된 숙고 과정으로서 관리 추론을 어떻게 평가할 수 있는가? (임상 진단 추론을 평가하기 위한 수많은 도구가 존재하지만, 관리 추론에 초점을 맞춘 도구는 없습니다(Daniel 외. 2019). 마찬가지로, [참여 중심의 SDM]을 평가하기 위한 수많은 도구가 존재하지만, [대응적 문제 해결]에 초점을 맞춘 도구는 없습니다(Hartasanchez 외. 2021)). '좋은' 의학적 결정이란 무엇인가요? (보편적인 정의는 없으며, 환자 대 임상의, 임상 지침 대 실제 경험, 전문의 대 일차 진료, 참여 중심 SDM 대 문제 중심 SDM 등 다양한 관점에 따라 답이 달라지며, 이는 교육 및 평가에 중요한 영향을 미칩니다).
We emphasize the nascent state of research in management reasoning, especially in relation to SDM. Important questions emerging from this study, beyond those noted above, include: How can we assess management reasoning as a shared deliberative process? (Numerous tools exist to assess clinical diagnostic reasoning; none focus on management reasoning (Daniel et al. 2019). Similarly, numerous tools exist to assess involvement-focused SDM; none focus on responsive problem-solving (Hartasanchez et al. 2021)). What is a ‘good’ medical decision? (There is no universal definition; the answer varies for different perspectives—patient vs clinician, clinical guideline vs practical experience, specialist vs primary care, involvement-focused SDM vs problem-focused SDM, etc.—with important implications for training and assessment).

마지막으로, 연구 결과에 따르면 [좋은 의사소통]은 [단순히 좋은 매너]가 아니며, [명목상 환자를 '참여'시키는 암기식 행동]이 최선의 결정에 도달하는 데 적합하지도 않습니다. 의사소통에는 목적이 있으며, 대화가 의사소통의 목적과 동떨어져 있거나 관련된 사람들이 서로 다른 목표를 추구한다면 추론과 의사 결정에 상당한 장애가 발생할 것입니다(Hargraves 외. 2019; Bohman and Rehg 2017). 임상의는 [환자와 임상의가 함께 할 수 있는 의사소통]이 무엇인지 인식하고, 진료 목적에 따른 잠재력을 깨닫고, 환자별 및 상황별 요구에 민첩하게 부응할 수 있도록 의사소통의 내용과 방법을 조정해야 합니다.
Finally, our findings suggest that good communication is not simply good manners, nor that rote behaviors to nominally ‘involve’ patients are adequate for reaching the best decision. Communication serves a purpose, and if talking is detached from the purpose of communication, or if those involved seek disparate goals, then reasoning and decision-making will be significantly impaired (Hargraves et al. 2019; Bohman and Rehg 2017). Clinicians must (learn to) recognize what communication enables patients and clinicians to do together, realize that potential in achieving an encounter-specific purpose, and adjust what and how they communicate to nimbly meet patient-specific and situation-specific demands.

 


Med Teach. 2023 Feb 10;1-13. doi: 10.1080/0142159X.2023.2170776. Online ahead of print.

Management reasoning and patient-clinician interactions: Insights from shared decision-making and simulated outpatient encounters

Affiliations collapse

1Office of Applied Scholarship and Education Science, Mayo Clinic College of Medicine and Science; and Division of General Internal Medicine, Mayo Clinic, Rochester, MN, USA.

2Mayo Clinic National Shared Decision Making Resource Center, Mayo Clinic, Rochester, MN, USA.

3Division of General Internal Medicine, Mayo Clinic, Rochester, MN, USA.

4Center for Health Professions Education, Uniformed Services University of the Health Sciences, Bethesda, MD, USA.

PMID: 36763491

DOI: 10.1080/0142159X.2023.2170776

Abstract

Purpose: To expand understanding of patient-clinician interactions in management reasoning.

Methods: We reviewed 10 videos of simulated patient-clinician encounters to identify instances of problematic and successful communication, then reviewed the videos again through the lens of two models of shared decision-making (SDM): an 'involvement-focused' model and a 'problem-focused' model. Using constant comparative qualitative analysis we explored the connections between these patient-clinician interactions and management reasoning.

Results: Problems in patient-clinician interactions included failures to: encourage patient autonomy; invite the patient's involvement in decision-making; convey the health impact of the problem; explore and address concerns and questions; explore the context of decision-making (including patient preferences); meet the patient where they are; integrate situational preferences and priorities; offer >1 viable option; work with the patient to solve a problem of mutual concern; explicitly agree to a final care plan; and build the patient-clinician relationship. Clinicians' 'management scripts' varied along a continuum of prioritizing clinician vs patient needs. Patients also have their own cognitive scripts that guide their interactions with clinicians. The involvement-focused and problem-focused SDM models illuminated distinct, complementary issues.

Conclusions: Management reasoning is a deliberative interaction occurring in the space between individuals. Juxtaposing management reasoning alongside SDM generated numerous insights.

Keywords: Decision Making; clinical decision-making; communication; diagnostic reasoning; shared; therapeutic reasoning.

질적연구에서 포화를 위한 표본 수: 실증 시험의 체계적 문헌고찰(Soc Sci Med. 2022)
Sample sizes for saturation in qualitative research: A systematic review of empirical tests
Monique Hennink a,*, Bonnie N. Kaiser b

 

 

1. 서론
1. Introduction

[포화]는 질적 연구에서 목적적 표본의 적절성을 평가하기 위한 가장 일반적인 지침 원칙입니다(Morse, 1995, 2015; Sandelowski, 1995). 그러나 포화 평가에 대한 지침과 포화에 도달하는 데 필요한 표본 크기는 모호했습니다. 최근까지 포화는 다양한 유형의 정성적 데이터로 경험적으로 평가되지 않았습니다. 포화에 대한 경험적 평가에 대한 관심이 높아지면서 이 주제에 대한 연구가 많이 이루어졌고, 이를 종합하고 이를 통해 무엇을 배울 수 있는지 파악할 수 있는 적절한 시기가 되었습니다. 이 체계적인 검토에서는 정성적 연구에서 포화를 경험적으로 평가하는 연구를 식별하고, 포화에 필요한 표본 크기, 포화를 평가하는 데 사용되는 전략 및 이러한 연구에서 도출할 수 있는 지침을 확인하고자 했습니다.  
Saturation is the most common guiding principle for assessing the adequacy of purposive samples in qualitative research (Morse, 1995, 2015; Sandelowski, 1995). However, guidance on assessing saturation and the sample sizes needed to reach saturation have been vague. Until recently, saturation had not been empirically assessed with different types of qualitative data. A growing interest in empirical assessment of saturation has now generated a body of research on the topic, making it an opportune time to synthesize it and identify what we can learn from it. This systematic review sought to identify studies that empirically assess saturation in qualitative research, to identify sample sizes needed for saturation, strategies used to assess saturation, and guidance we can draw from these studies.

포화의 개념은 Glaser와 Strauss(1967)에 의해 ['이론적 포화']로 개발되었으며, 질적 연구에 대한 영향력 있는 근거 이론 접근법의 일부였습니다. 근거 이론은 사회 현상을 설명하기 위해 텍스트 데이터로부터 사회학 이론을 개발하는 데 중점을 둡니다. 이 접근법에서 [이론적 포화]란 "이론적 구성에 대한 더 많은 데이터를 수집해도 새로운 속성이 드러나지 않고, 새로운 근거 이론에 대한 더 이상의 이론적 통찰을 얻지 못하는 지점"을 의미합니다(Bryant와 Charmaz, 2007, 611쪽). 따라서 데이터 수집에서 [중요한 이슈나 통찰력이 데이터에서 모두 소진되는 시점]은 이론을 구성하는 개념적 범주가 '포화 상태'에 이르렀음을 의미하며, 이는 새로운 이론이 포괄적이고 데이터에 근거한 이론이 될 수 있음을 의미합니다. 이론적 포화 상태는 샘플링, 데이터 수집, 데이터 분석이 동시에 이루어지는 반복적인 과정에도 내재되어 있으며(Sandelowski, 1995), 데이터는 포화 상태가 될 때까지 샘플링에 지속적으로 정보를 제공합니다. 
The concept of saturation was developed by Glaser and Strauss (1967) as ‘theoretical saturation’ and was part of their influential grounded theory approach to qualitative research. Grounded theory focuses on developing sociological theory from textual data to explain social phenomena. Within this approach, theoretical saturation refers to “the point at which gathering more data about a theoretical construct reveals no new properties, nor yields any further theoretical insights about the emerging grounded theory” (Bryant and Charmaz, 2007 p.611). Thus, it is the point in data collection when all important issues or insights are exhausted from data, which signifies that the conceptual categories that comprise the theory are ‘saturated’, so that the emerging theory is comprehensive and well-grounded in data. Theoretical saturation is also embedded in an iterative process of concurrently sampling, collecting data, and analyzing data (Sandelowski, 1995), whereby data continuously inform sampling until saturation.

대부분의 질적 연구는 근거 이론 접근법을 따르지 않지만, 포화 개념은 다른 질적 연구 접근법에서 널리 사용되며, 일반적으로 '데이터 포화' 또는 '주제 포화'라고 불립니다(Hennink et al., 2017). 이러한 포화의 광범위한 적용은 이론적 포화에서와 같이 이론을 개발하기 위한 데이터의 적절성보다는 표본 크기를 평가하는 데 더 중점을 둡니다. 더 넓은 맥락에서 사용되는 포화는 데이터 수집에서 [추가적인 문제나 인사이트]가 발견되지 않고 [데이터가 반복]되기 시작하여 더 이상의 데이터 수집이 [중복되는 시점]을 의미하며, 이는 적절한 표본 크기에 도달했음을 의미합니다. 포화는 표본이 연구 대상 현상에 적합한지, 즉 수집된 데이터가 연구 대상 이슈의 다양성, 깊이, 미묘한 차이를 포착하고 있는지, 따라서 [내용 타당성]을 입증할 수 있는지를 나타내는 중요한 지표입니다(Francis et al., 2010). 포화에 도달하는 것은 데이터 수집을 강력하고 타당하게 만드는 데 도움이 되는 질적 연구의 중요한 구성 요소가 되었습니다(O'Reilly and Parker, 2013). 또한 포화는 "저자가 리뷰어와 독자에게 제공하는 질적 엄격성에 대한 가장 빈번한 보증"입니다(Morse, 2015, 587쪽). 이 리뷰에서는 [근거 이론]의 매개변수 외부에서 사용될 때 포화를 위한 적절한 표본 크기에 대해 알려진 바가 적기 때문에 더 넓은 맥락에서 포화에 초점을 맞춥니다.
Although most qualitative research does not follow a grounded theory approach, the concept of saturation is widely used in other approaches to qualitative research, where it is typically called ‘data saturation’ or ‘thematic saturation’ (Hennink et al., 2017). This broader application of saturation focuses more on assessing sample size rather than the adequacy of data to develop theory (as in theoretical saturation). When used in the broader context, saturation refers to the point in data collection when no additional issues or insights are identified and data begin to repeat so that further data collection is redundant, signifying that an adequate sample size is reached. Saturation is an important indicator that a sample is adequate for the phenomenon studied – that data collected have captured the diversity, depth, and nuances of the issues studied – and thereby demonstrates content validity (Francis et al., 2010). Reaching saturation has become a critical component of qualitative research that helps make data collection robust and valid (O’Reilly and Parker, 2013). Moreover, saturation is “the most frequently touted guarantee of qualitative rigor offered by authors to reviewers and readers" (Morse, 2015, p. 587). In this review, we focus on saturation in the broader context, since less is known about adequate sample sizes for saturation when used outside of the parameters of grounded theory.

질적 표본의 엄밀성을 뒷받침하는 포화의 중요성에도 불구하고, 발표된 질적 연구에서 표본 크기가 정당화되는 방식에 대한 투명성이 일관되게 부족합니다(Morse, 1995; Guest 외., 2006; Kerr 외., 2010; Carlsen and Glenton, 2011; Hennink 외., 2017). [포화는 적절한 표본 크기에 대한 가장 일반적인 근거로 인용]되지만(Morse, 1995, 2015), 포화를 평가한 방법과 그 근거에 대한 자세한 내용은 질적 연구에서 거의 찾아볼 수 없습니다. Vasileiou 등(2018)은 15년 동안 건강 관련 저널에서 [심층 인터뷰를 사용한 질적 연구]를 체계적으로 검토한 결과, 대다수의 논문에서 표본 크기에 대한 정당성을 제공하지 않는다는 사실을 발견했습니다. 정당성이 제시된 경우 55%의 논문에서 포화가 인용되었지만, 포화에 대한 주장은 "연구 자체에서 수행된 절차와 관련하여 입증되지 않았으며"(12쪽), 당면한 연구와 거리가 먼 다른 문헌의 추가 인용만 제시되었습니다. 
Despite the importance of saturation to support the rigor of qualitative samples, there is a consistent lack of transparency in how sample sizes are justified in published qualitative research (Morse, 1995; Guest et al., 2006; Kerr et al., 2010; Carlsen and Glenton, 2011; Hennink et al., 2017). Although saturation is the most commonly cited justification for an adequate sample size (Morse, 1995, 2015), details of how saturation was assessed and the grounds on which it was determined are largely absent in qualitative studies. Vasileiou et al. (2018) conducted a systematic review of qualitative studies using in-depth interviews in health-related journals over a 15-year period and found the vast majority of articles provided no justification for their sample size. Where justifications were given, saturation was cited in 55% of articles; however, claims of saturation were “never substantiated in relation to procedures conducted in the study itself” (p. 12); only further citations of other literature were given that moved away from the study at hand.

마찬가지로 [포커스 그룹 토론]을 사용한 220개의 연구를 체계적으로 검토한 결과(Carlsen and Glenton, 2011), 83%가 표본 크기를 정당화하기 위해 포화를 사용했지만, 포화에 대한 근거 없는 주장이나 미리 정해진 표본 크기를 사용하면서 포화를 달성했다는 언급 등 포화를 달성한 방법에 대한 피상적인 보고만 제공한 것으로 밝혀졌습니다. 또 다른 연구(Francis 외, 2010)에서는 16개월에 걸쳐 사회과학 및 의학 저널의 논문을 검토한 결과 대부분의 논문이 포화 상태에 도달했다고 주장하지만 포화 상태를 정의, 달성 또는 정당화하는 방법에 대해 명확하게 설명하지 않는 것으로 나타났습니다. Marshall 등(2013)도 83개의 [질적 연구를 검토]한 결과 포화가 어떤 연구에서도 설명되지 않았다는 사실을 발견했습니다. 포화가 어떻게 평가되거나 결정되었는지에 대한 연구 기반의 설명 없이 포화를 주장하는 것에 대한 우려가 커지고 있습니다. 포화 상태에 도달했다는 근거 없는 주장은 포화 상태라는 개념의 가치를 훼손합니다. 부분적으로 이러한 투명성 부족은 포화를 평가하는 방법에 대한 공개된 지침이 없기 때문일 수 있습니다(Morse, 1995; Guest et al., 2006). 이 검토에서는 경험적 연구에서 포화를 평가하는 데 사용되는 전략을 파악하여 질적 연구에서 포화를 보고할 때 투명성을 높일 수 있는 방법을 모색하고자 합니다. 

Similarly, a systematic review of 220 studies using focus group discussions (Carlsen and Glenton, 2011) found that 83% used saturation to justify their sample size; however, they provided only superficial reporting of how it was achieved, including unsubstantiated claims of saturation and references to achieving saturation while still using a predetermined sample size. Another study (Francis et al., 2010) reviewed articles in the journal Social Science and Medicine over 16 months and found most articles claimed they had reached saturation but provided no clarity on how saturation was defined, achieved, or justified. Marshall et al. (2013) also reviewed 83 qualitative studies and found saturation was not explained in any study. There are increasing concerns over claims of saturation without study-based explanations of how it was assessed or determined. Unsubstantiated claims of reaching saturation undermine the value of the concept. In part, this lack of transparency may reflect the absence of published guidance on how to assess saturation (Morse, 1995; Guest et al., 2006). In this review, we seek to identify the strategies used to assess saturation in empirical research, which may encourage greater transparency in reporting saturation in qualitative studies.

또한, 다양한 질적 방법에서 포화에 도달하는 데 필요한 특정 표본 크기에 대한 지침은 방법론 문헌에 없거나 모호하며, 근거에 기반하지 않은 일반적인 '경험 법칙'만 제공합니다(Morse, 1995; Guest 외, 2006; Kerr 외, 2010; Bryman, 2012; Hennink 외, 2019). 포화를 경험적으로 평가하는 연구가 이러한 격차를 메우기 시작하면서, 질적 연구에서 포화에 대한 표본 크기에 대해 절실히 필요한 경험적 기반 지침을 제공할 수 있게 되었습니다. 
In addition, guidance on specific sample sizes needed to reach saturation in different qualitative methods has been absent or vague in the methodological literature, providing only general “rules of thumb” that are rarely evidence-based (Morse, 1995; Guest et al., 2006; Kerr et al., 2010; Bryman, 2012; Hennink et al., 2019). As research empirically assessing saturation begins to fill this gap, it allows us to provide much-needed empirically based guidance on sample sizes for saturation in qualitative research.

이 체계적 검토에서는 [정성적 데이터의 포화를 평가하는 실증적 연구를 종합]하고자 합니다. 특히, 포화를 평가하는 데 사용되는 전략을 문서화하고, 다양한 질적 방법을 사용하여 포화에 도달하는 데 필요한 표본 크기를 식별하며, 질적 연구를 위한 표본 크기에 대한 지침을 제시하고자 합니다. 저희가 알기로는 포화에 대한 경험적 연구에 대한 체계적인 검토는 이번이 처음이며, 따라서 질적 연구를 검토하는 연구자, 학술지, 학술지 심사자, 윤리 검토 위원회 및 자금 지원 기관에 유용한 리소스를 제공합니다. 연구자는 연구 제안서 및 프로토콜에서 적절한 표본 크기를 추정할 때 이 결과를 참조할 수 있으며, 이를 통해 연구 자원을 보다 효율적으로 사용하고 제안된 표본 크기에 대한 명확한 근거를 마련할 수 있습니다. 마찬가지로, 연구 결과는 질적 연구에 적합한 표본 크기에 대한 증거 기반의 기대치를 제공하여 연구를 검토하고 자금을 지원하는 사람들에게 지침을 제공할 수 있습니다.
In this systematic review, we aim to synthesize empirical studies that assess saturation in qualitative data. In particular, we aim to document strategies used to assess saturation, identify sample sizes needed to reach saturation using different qualitative methods, and suggest guidance on sample sizes for qualitative research. To our knowledge, this is the first systematic review on empirical studies of saturation and therefore provides a valuable resource for researchers, academic journals, journal reviewers, ethical review boards, and funding agencies that review qualitative research. Researchers can refer to our results when estimating an appropriate sample size in research proposals and protocols, which may lead to more efficient use of research resources and clearer justifications for proposed sample sizes. Similarly, our results may provide evidence-based expectations regarding adequate sample sizes for qualitative research to guide those who review and fund research.

2. 조사 방법
2. Methods

저희는 체계적 문헌고찰을 수행하고 보고할 때 체계적 문헌고찰 및 메타분석에 대한 선호 보고 항목(PRISMA) 가이드라인을 따랐습니다(Moher et al., 2009). 그림 1은 식별, 선별 및 포함된 논문 수를 보여줍니다. 데이터베이스 검색과 인용 검색을 포함한 2단계 검색 프로세스를 사용했습니다. 

We followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines in conducting and reporting our systematic review (Moher et al., 2009). Fig. 1 shows the number of articles identified, screened, and included. We used a two-stage search process, including database searches and citation searches.

먼저 PubMed, Embase, Sociological Abstracts, CINAHL 등 4개의 데이터베이스를 사용하여 제목, 초록, 키워드/색인에 '포화'와 다음 용어 중 하나가 포함된 논문 또는 서적 챕터를 검색했습니다: "인터뷰", "포커스 그룹", "질적" 또는 "주제"(전체 검색어는 부록 표 참조). 검색 결과는 영어 및 인간 연구로 제한되었습니다. 데이터베이스 검색은 2019년 1월 31일부터 2월 1일까지 수행되었으며 2020년 7월 10일에 업데이트되었습니다. 두 저자는 모든 논문 제목, 초록, 필요한 경우 전문을 독립적으로 선별하여 적격성을 결정했습니다. 불일치하는 부분은 논의하고 합의를 통해 해결했습니다. 연구에 포함되려면,

  • a) 질적 연구에서 포화를 평가하기 위해 경험적 데이터를 사용하거나 가상의 데이터를 사용하여 포화를 결정하는 통계 모델을 사용해야 하고,
  • b) 근거 이론을 벗어난 포화에 초점을 맞춰야 하며,
  • c) 저널 논문 또는 책 챕터로 출판되어야 하고,
  • d) 영어로 이용 가능해야 합니다.

데이터베이스 검색을 통해 16개의 논문이 포함되었습니다. 
First, we used four databases -- PubMed, Embase, Sociological Abstracts, and CINAHL -- to search for articles or book chapters that included “saturation” and one of the following terms in the title, abstract, or key words/index: “interview,” “focus group,” “qualitative,” or “thematic” (see Supplemental Table for full search terms). Search results were limited to English-language and human studies. Database searches were conducted on January 31 – February 1, 2019 and updated July 10, 2020. Both authors independently screened all article titles, abstracts, and, where needed, full texts to determine eligibility. Discrepancies were discussed and resolved by consensus. To be eligible for inclusion, studies needed to:

  • a) use empirical data to assess saturation in qualitative research or use a statistical model to determine saturation using hypothetical data,
  • b) focus on saturation outside of grounded theory,
  • c) be published in journal articles or book chapters, and
  • d) be available in English.

Sixteen articles were included from database searches.

둘째, 포함된 논문의 참고문헌 목록을 검토하고 Google Scholar의 '인용 출처' 검색 옵션을 사용하여 인용 검색을 수행하여 포함 기준을 충족하는 추가 기록을 식별했습니다. Google Scholar에서 250개 이상의 인용 논문이 있는 연구의 경우, 인용 논문 내에서 "포화"를 검색하여 처음 250개의 결과(관련성 순으로 정렬됨)를 검토했습니다. 이 단계에서 추가로 7개의 논문이 포함되었습니다.
Second, we conducted citation searches by reviewing the reference lists of included articles and using the “cited by” search option in Google Scholar to identify further records meeting the inclusion criteria. For studies with more than 250 citing articles on Google Scholar, we searched within citing articles for “saturation” and reviewed the first 250 results (which are ordered by relevance). An additional seven articles were included during this step.

23개의 적격 논문에서 다음과 같은 정보를 추출했습니다:

  • a) 논문에 대한 메타 데이터(저자, 저널, 연도),
  • b) 사용된 데이터에 대한 정보(가설적 대 경험적, 인터뷰, 포커스 그룹 토론 등), 연구 목적, 표본 크기, 연구 집단(동종, 이종), 데이터 수집이 반복적이었는지 여부,
  • c) 정의, 목표, 데이터 무작위 배정, 포화 평가 전략, 포화를 위한 표본 크기 및 달성한 포화 수준(예, 코드의 90%) 및
  • d) 추가 정보(제한 사항, 제안된 포화 매개변수).

두 저자는 6개의 논문에서 독립적으로 데이터를 추출하고 결과를 논의했습니다. 이는 명확성 부족이나 중복성 등 데이터 추출 범주에 문제가 있는지 확인하고 두 저자 간의 신뢰성을 확립하기 위해 수행되었습니다. 그런 다음 나머지 각 논문은 두 저자 중 한 명이 데이터 추출을 수행했습니다.
We extracted the following information from the 23 eligible articles:

  • a) meta-data about the article (author, journal, year),
  • b) information about data used (hypothetical vs. empirical; interviews, focus group discussions, etc.), research objective, sample size, study population (homogenous, heterogenous), and whether data collection was iterative,
  • c) information about saturation, including: definition, goal, data randomization, strategy to assess saturation, sample size for saturation, and level of saturation achieved (e.g., 90% of codes), and
  • d) additional information (limitations, any parameters of saturation suggested).

Both authors independently extracted data from 6 articles and discussed results. This was done to identify any issues with the data extraction categories, such as lack of clarity or redundancy, as well as to establish reliability between the two authors. Each remaining article then underwent data extraction by one of the two authors.

경험적 데이터를 사용하여 포화를 평가한 연구와 통계적 모델을 사용한 연구의 결과를 별도로 분석했습니다. 인터뷰 또는 포커스 그룹 토론과 같은 정성적 방법을 통해 포화에 대한 샘플 크기를 분석했습니다. 연구 모집단의 동질성 및 데이터 무작위 추출을 통해 포화를 비교하여 패턴을 파악했습니다. 
We analyzed results separately for studies using empirical data to assess saturation versus those using statistical models. We analyzed sample sizes for saturation by qualitative method: interviews or focus group discussions. We conducted comparisons of saturation by homogeneity of the study population and randomization of data to identify any patterns.

3. 결과
3. Results

체계적인 검토를 통해 질적 연구의 포화를 평가한 23편의 논문을 확인했습니다. 모든 논문은 2006년부터 2020년 사이에 출판되었으며, 대다수(87%, 20/23)가 2014년 이후에 출판되었습니다. 연구 방법론 저널(43%, 10/23)과 사회과학(6/23) 또는 주제별 저널(7/23)(예: 공학, 컴퓨팅, 자연자원)에 게재된 논문이 많았습니다. 경험적 데이터를 사용하여 포화를 평가하는 논문(표 1, 17개 논문)과 통계적 모델링을 사용하여 포화를 예측하는 논문(표 2, 6개 논문)으로 분류했습니다. 이러한 접근 방식과 결과는 비교할 수 없으므로 아래에서 각각을 개별적으로 보고합니다. 

Our systematic review identified 23 articles assessing saturation for qualitative research. All articles were published between 2006 and 2020, with the majority (87%, 20/23) published since 2014. Many articles were published in research methodology journals (43%, 10/23) and others in social science (6/23) or topical journals (7/23) (e.g., engineering, computing, natural resources). We categorized the articles into those assessing saturation using empirical data (Table 1, 17 articles) and those using statistical modeling to predict saturation (Table 2, 6 articles). Since these approaches and results are not comparable, we report each separately below.

 

 

3.1. 포화

평가에 대한 접근 방식
3.1. Approaches to assessing saturation

3.1.1. 경험적 기반 테스트
3.1.1. Empirically based tests

표 1에는 경험적 데이터를 사용하여 포화를 평가한 17개의 논문이 요약되어 있습니다. 일부 논문에서는 여러 데이터 세트를 사용하여 포화를 평가하고 각각의 결과를 개별적으로 보고했기 때문에 표 1에는 17개 논문에서 23개의 테스트가 나와 있습니다(참고: 이러한 연구는 실험적 테스트를 수행하지 않았지만, 간결성을 위해 '테스트'라는 용어를 사용하여 통계적 모델링이 아닌 경험적 데이터를 사용하여 포화를 평가하는 개별 연구를 지칭합니다). 대부분의 논문은 심층 인터뷰(10/17) 또는 포커스 그룹 토론(4/17)에서 얻은 데이터를 사용했으며, 두 가지 유형의 데이터를 모두 사용한 논문은 두 편, 자유 목록 데이터를 사용한 논문은 한 편(Weller et al., 2018)이었습니다. 자유 목록 데이터는 인터뷰 및 포커스 그룹 토론에서 나온 자유로운 서술형 데이터와 비교할 수 없기 때문에 분석에서 Weller 등의 논문은 제외했습니다. 따라서 모든 논문을 설명할 때는 분모 16을 사용하고, 경험적 데이터로 모든 테스트의 데이터 세트와 결과를 설명할 때는 분모 22를 사용합니다.
Table 1 summarizes 17 articles that assess saturation using empirical data. Some articles used multiple datasets to assess saturation and report the results of each separately; therefore, Table 1 shows 23 tests from 17 articles (NB: while these studies were not conducting experimental tests, we use the term ‘test’ for brevity to refer to individual studies using empirical data, as opposed to statistical modeling, to assess saturation). Most articles used data from in-depth interviews (10/17) or focus group discussions (4/17); two articles used both types of data, and one article (Weller et al., 2018) used free list data. We excluded the article by Weller et al. in our analysis because free list data are not comparable to free-flowing narrative data from interviews and focus group discussions. We therefore use the denominator of 16 when describing all articles and 22 when describing the datasets and results of all tests with empirical data.

테스트에 사용된 각 데이터 세트의 원래 연구 목적은 다양했지만, 대부분의 연구(14/16)는 특정 건강 상태(예: 겸상 적혈구 질환, 다발성 경화증, 파제트병), 의료 서비스 또는 개입(예: 유전자 검사, 폭력 예방, 라이프스타일 개입, 환자 유지)의 [경험]과 같은 [건강 문제]에 초점을 맞췄습니다. 이러한 연구 목표는 많은 질적 건강 연구의 전형적인 목표입니다. 사용된 데이터 세트의 표본 크기는 14~132건의 인터뷰와 1~40개의 포커스 그룹으로 다양했습니다. 한 데이터셋(Francis et al., 2010)을 제외한 모든 데이터셋은 포화를 평가하는 데 필요한 표본보다 훨씬 더 큰 표본을 가지고 있어 포화를 평가하는 데 효과적이었습니다. Francis 등(2010)은 사용된 두 데이터 세트 모두에서 연구의 표본 크기와 정확히 일치하는 포화에 도달했다고 보고했습니다. 대부분의 데이터세트(18/22)는 특정 질병(예: HIV, 류마티스 관절염, 겸상 적혈구)을 앓고 있거나 특정 인구 집단(예: 남성 간호사, 의대생, 남아시아 성인, 아프리카계 미국인 남성)의 환자 등 [동질적인 연구 모집단]을 가지고 있었습니다. 나머지 데이터 세트에는 미국 전역의 20~72세 남성 또는 14~18세 청소년과 같이 이질적인 샘플이 더 많았습니다. 
The original research objective for each dataset used in the tests varied, but most studies (14/16) focused on health issues, such as experiences of a specific health condition (e.g., sickle cell disease, multiple sclerosis, Paget's disease), health service, or intervention (e.g., genetic screening, violence prevention, lifestyle interventions, patient retention). These research objectives are typical of much qualitative heath research. The sample size of the datasets used varied from 14 to 132 interviews and 1 to 40 focus groups. All datasets except one (Francis et al., 2010) had a sample that was much larger than the sample ultimately needed for saturation, making them effective for assessing saturation. Francis et al. (2010) report saturation was reached at exactly the sample size of the study for both datasets used. Most datasets (18/22) had a homogenous study population, such as patients with a specific disease (e.g., HIV, rheumatoid arthritis, sickle cell) or from a specific demographic group (e.g., male nurses, medical students, South Asian adults, African American men). The remaining datasets had more heterogeneous samples, such as men aged 20–72 across the US or youths aged 14–18.

저자들은 포화의 목표를 [개별 코드 또는 카테고리의 포화]라는 두 가지 방식으로 설명했습니다. 용어는 기사마다 다르지만,

  • [코드]는 일반적으로 데이터의 개별 이슈, 주제 또는 항목으로 설명되었고
  • [카테고리]는 이슈의 상위 그룹(예: 더 넓은 주제, 메타 주제, 개념)을 나타냅니다.

44%(7/16)의 논문이 코드의 포화를, 31%(5/16)의 논문이 카테고리의 포화를, 25%의 논문이 두 가지 모두를 명시했습니다.
Authors described their goal of saturation in two ways, either as saturation of individual codes or categories. Although terminology varied across articles,

  • codes were typically described as individual issues, topics, or items in data, while
  • categories represented higher-order groupings of issues (e.g., broader themes, meta-themes, concepts).

Forty-four percent (7/16) of articles sought saturation of codes, 31% (5/16) saturation of categories, and 25% stated both.

포화가 정의된 경우, 저자들은 비슷한 정의를 사용했습니다. 전반적으로 포화 상태는 [데이터에서 관련성이 있는 새로운 코드 및/또는 범주가 거의 또는 전혀 발견되지 않는 시점]으로, [연구 현상, 차원, 뉘앙스 또는 가변성에 대한 더 이상의 이해나 기여 없이] [문제가 반복되기 시작하는 시점]으로 설명되었습니다.

  • 일부 논문에서는 두세 번의 연속 인터뷰 또는 포커스 그룹에서 새로운 이슈가 발견되지 않은 후에야 포화 상태를 확인해야 한다고 명시하거나(Coenen 등, 2012; Francis 등, 2010; Morse 등, 2014),
  • 두 명의 연구자에 의해 포화 상태를 결정해야 한다고 명시하기도 했습니다(Morse 등, 2014).
  • 절반 이상의 논문(56%, 9/16)은 포화에 영향을 줄 수 있는 인터뷰 순서를 고려하여 분석 데이터의 순서를 무작위로 지정했습니다.
  • 일부는 무작위 인터뷰 순서와 실제 인터뷰가 진행된 순서 사이의 포화를 비교했고,
  • 다른 일부는 여러 무작위 데이터 순서에 걸쳐 포화를 계산하여 평균을 확인했습니다.

Where saturation was defined, authors used similar definitions. Overall, saturation was described as the point at which little or no relevant new codes and/or categories were found in data, when issues begin to be repeated with no further understanding or contribution to the study phenomenon, its dimensions, nuances, or variability.

  • Some articles further specified that saturation should be confirmed only after no new issues were found in two or three consecutive interviews or focus groups (Coenen et al., 2012Francis et al., 2010Morse et al., 2014) or
  • that it was determined by two researchers (Morse et al., 2014).
  • Over half of articles (56%, 9/16) randomized the order of data for analysis to account for interview order, which might influence saturation.
  • Some compared saturation between the randomized order of interviews and the actual order in which interviews were conducted,
  • while others calculated saturation across multiple randomized orderings of data to identify an average.

[포화를 평가하기 위해 다양한 전략]이 사용되었습니다. 이러한 전략은 표 1에 분류되어 있으며 표 3에 설명된 범주에 따라 분류되어 있습니다.

  • 대부분의 기사(75%, 12/16)는 포화를 평가하기 위해 [단일 전략을 사용]했습니다.
  • 모든 논문은 코드 빈도수, 비교 방법, 중지 기준, 고차 그룹화 등 어떤 형태로든 [코드 빈도수를 사용]하여 포화를 평가했으며,
  • 4편의 논문은 코드 빈도수 외에 다른 접근법을 사용하여 각각의 포화를 비교했습니다(Hennink 외, 2017, 2019; Constantinou 외, 2017; Hagaman과 Wutich, 2017).
  • 많은 논문(37%, 6/16)은 포화를 평가하기 위해 코드 빈도 수만을 사용했는데, 이는 새로운 코드의 빈도가 감소하여 포화에 도달할 때까지 연속된 녹취록 또는 녹취록 세트의 코드를 계산하는 방식이었습니다.
  • 3건(18%, 3/16)의 논문에서는 배치 비교, 중단 기준, 고차 코드 그룹을 계산(개별 코드가 아닌 메타 주제 또는 코드 범주 등)하는 등 코드 빈도 수에 특정 추가 요소를 추가했습니다.
  • 세 논문(Hennink 외., 2017, 2019; Nascimento 외., 2018)에서는 코드 빈도 수와는 완전히 다른 접근 방식인 '코드 의미'를 사용하여 포화를 평가했습니다. 이 접근 방식은 문제, 차원 및 뉘앙스가 완전히 식별되고 이해되었는지를 평가하여 데이터의 문제를 완전히 이해하는 데 중점을 두어 포화에 도달했음을 나타내는 지표로 삼았습니다.
  • 그 후 두 논문(Hennink 외., 2017, Hennink 외., 2019)에서 이 접근 방식을 사용한 포화를 코드 빈도 접근 방식과 비교했습니다.

Various strategies were used to assess saturation. These are categorized in Table 1 and the categories described in Table 3.

  • Most articles (75%, 12/16) used a single strategy to assess saturation.
  • All articles used some form of code frequency counts to assess saturation (including code frequency counts, comparative method, stopping criterion, higher-order groupings), and four articles used another approach in addition to code frequency counts and compared saturation for each (Hennink et al., 20172019Constantinou et al., 2017Hagaman and Wutich, 2017).
  • Many articles (37% 6/16) used only code frequency counts to assess saturation, which involved counting codes in successive transcripts or sets of transcripts until the frequency of new codes diminishes, signaling saturation is reached.
  • Three articles (18%, 3/16) added specific additional elements to code frequency counts, such as batch comparisons, a stopping criterion, or counting higher-order groupings of codes, such as meta-themes or categories of codes rather than individual codes.
  • Three articles (Hennink et al., 20172019Nascimento et al., 2018) used ‘code meaning’ to assess saturation, an entirely different approach from code frequency counts. This approach focused on reaching a full understanding of issues in data as the indicator that saturation is reached, by assessing whether the issue, its dimensions, and nuances are fully identified and understood.
  • Two articles (Hennink et al., 2017Hennink et al., 2019) then compared saturation using this approach with the code frequency approach.

 

3.1.2. 통계적 모델
3.1.2. Statistical models

표 2는 포화를 추정하기 위해 통계적 모델링을 사용한 6개의 논문을 요약한 것입니다. 이 논문들은 위에 요약된 것과는 다른 접근법을 사용했는데, 포화에 필요한 표본 크기를 추정하는 [공식을 개발]하여 데이터 수집 전에 연구 설계에 정보를 제공하기 위해 사용할 수 있습니다.

  • 몇몇 공식은 모집단 내 테마의 유병률 또는 데이터에서 원하는 테마의 인스턴스 수와 같은 유사한 매개변수를 기반으로 한 반면(Fugard and Potts, 2015; Galvin, 2015; Lowe 외, 2018),
  • 로그 정규 분포(Rowlands 외, 2016; Van Rijnsoever, 2017) 또는 집합 이론(Fofana 외, 2020)을 사용한 공식도 있었습니다.
  • 이러한 연구 중 다수는 무작위 표본을 가정한 반면, 무작위 표본과 의도적 표본을 모두 고려한 연구도 있었습니다(Van Rijnsoever, 2017).
  • 대부분의 공식은 인터뷰 데이터를 위해 개발되었으며, 두 개의 논문은 인터뷰, 포커스 그룹, 문서, 문헌 조사 등 다양한 형태의 데이터에 대한 포화 추정에 대해 논의했습니다. 그런 다음 공식의 절반을 경험적 데이터 세트에 적용했습니다.


Table 2 summarizes six articles that used statistical modeling to estimate saturation. These articles used a different approach than those summarized above: they developed a formula to estimate the sample size needed for saturation, which may be used prior to data collection to inform study design.

  • Several formulas were based on similar parameters, such as prevalence of a theme in a population or the desired instances of a theme in data (Fugard and Potts, 2015Galvin, 2015Lowe et al., 2018),
  • while others used a lognormal distribution (Rowlands et al., 2016Van Rijnsoever, 2017) or set theory (Fofana et al., 2020).
  • Many of these studies assumed a random sample, while one accounts for both random and purposive samples (Van Rijnsoever, 2017).
  • Most formulas were developed for interview data, while two articles discussed estimating saturation for various forms of data, including interviews, focus groups, documents, and literature surveys. Half of the formulas were then applied to empirical datasets.

3.2. 포화를 위한 샘플 크기
3.2. Sample size for saturation

그림 2는 [심층 인터뷰 데이터]를 사용한 경험적 기반 테스트의 포화에 대한 샘플 크기를 보여줍니다. 테스트에 사용된 각 데이터 세트(n = 16)에 대한 결과는 별도의 데이터 포인트로 표시됩니다. 서로 다른 표본 크기로 결과가 보고된 경우, 보고된 가장 낮은 표본 크기부터 가장 높은 표본 크기까지 선으로 표시되며, 이 범위에 영향을 미치는 매개변수가 표시됩니다. 저자가 서로 다른 포화 수준을 보고한 경우, 연구 간의 비교 가능성을 위해 90%에 가장 가까운 포화가 표시됩니다. 결과에 따르면 포화에 대한 다양한 접근 방식을 사용한 16개의 테스트에서 포화의 표본 크기는 5개에서 24개의 인터뷰 사이였습니다.

  • 가장 작은 포화 표본 크기는 5개의 인터뷰로, 설문조사 결과를 뒷받침하기 위한 동질적인 연구 모집단을 대상으로 광범위한 범주에서 포화를 구한 연구(Constantinou et al., 2017)였습니다. 이러한 연구 특성을 종합하면 5번의 인터뷰로 포화 상태에 도달한 것을 설명할 수 있습니다.
  • 포화에 대한 가장 큰 표본 크기는 4개국의 메타주제에 대한 포화를 추구한 20~40개(Hagaman and Wutich, 2017)와 연구 질문의 중심이 아닌 코드를 포함한 코드의 의미에 대한 포화를 추구한 24개(Hennink et al., 2017)였습니다.이러한 포화 목표를 달성하려면 더 많은 데이터가 필요하며, 이는 포화에 대한 더 높은 표본 크기를 뒷받침할 수 있습니다.

Fig. 2 shows sample sizes for saturation from empirically based tests using in-depth interview data. The results for each dataset used in the tests (n = 16) are shown as separate data points. Where results are reported at different sample sizes, this is depicted with a line from the lowest to highest sample size reported, and the parameters influencing this range are noted. Where authors report different levels of saturation, saturation closest to 90% is shown for comparability across studies. Results show that across 16 tests using various approaches to saturation, the sample size for saturation ranges between 5 and 24 interviews.

  • The lowest sample size for saturation was 5 interviews (Constantinou et al., 2017), in a study with a homogenous study population that was intended to support survey findings and where saturation was sought in broad categories. Together, these study characteristics may explain reaching saturation at 5 interviews.
  • The highest sample sizes for saturation were 20–40 (Hagaman and Wutich, 2017), where saturation of meta-themes across four countries was sought, and 24 (Hennink et al., 2017), where saturation was sought in the meaning of codes, including codes less central to the research question. These saturation goals require more data, which may support the higher sample sizes found for saturation.

이러한 이상값을 제외하면, 대부분의 데이터 세트는 포화를 평가하기 위해 다양한 접근 방식을 사용했음에도 불구하고 9~17개의 인터뷰, 평균 12~13개의 인터뷰 사이에 포화에 도달했습니다. 대부분의 연구는 비교적 [동질적인 연구 모집단]을 사용했으며 코드, 카테고리 또는 그 조합의 포화 목표가 다양했습니다. 이질적인 샘플을 사용한 연구는 단 3건에 불과했습니다. 이 중 두 연구는 평균보다 큰 표본 크기(16개 인터뷰)에서 포화에 도달했고, 한 연구는 작은 표본 크기(9개 인터뷰)에서 포화에 도달했습니다. 따라서 이 특성에 따른 포화 패턴은 발견되지 않았습니다. 마찬가지로, 대부분의 테스트에서 실제 인터뷰 순서대로 데이터를 분석할 때 무작위 순서와 포화를 비교하지 않았기 때문에 데이터 순서에 따른 포화의 패턴을 식별하기 어려웠습니다. 비교를 수행한 연구에서는 무작위 인터뷰 순서와 실제 인터뷰 순서에서 포화에 차이가 없거나 샘플 크기가 약간 더 큰 것으로 나타났습니다. 무작위 추출을 사용한 연구와 표본 크기의 전체 스펙트럼을 다루지 않은 연구 모두 검토 대상에 포함되지 않았습니다.

Excluding these outliers, most datasets reached saturation between 9 and 17 interviews, with a mean of 12–13 interviews, despite using different approaches to assess saturation. Most of these studies had a relatively homogenous study population and varied in their saturation goal of codes, categories, or a combination. Only three studies used a heterogeneous sample. Two of these studies reached saturation at a larger sample size than the mean (at 16 interviews), and one reached saturation at a smaller sample size (at 9 interviews). Therefore, we found no pattern in saturation by this characteristic. Similarly, it was difficult to identify any pattern of saturation by the order of data, since most tests did not compare saturation when analyzing data in the actual interview order with the randomized order. Those that did make a comparison found no difference or a slightly higher sample size for saturation in the random versus actual order of interviews. Both studies that used randomization and those that did not cover the full spectrum of sample sizes seen in our review.

그림 3은 [포커스 그룹 토론 데이터]를 사용한 6개의 경험적 테스트에서 포화에 대한 샘플 크기를 보여줍니다. 비교 가능성을 위해 다양한 수준의 포화가 보고된 경우 90%에 가장 가까운 포화가 그림에 표시되어 있습니다. 6개의 테스트 모두에서 1~16개의 포커스 그룹 사이에서 포화에 도달했습니다. 두 테스트는 이상치이므로 다른 테스트와 비교할 수 없습니다.

  • 가장 낮은 수치인 Hancock 등(2016)은 단일 비동기 온라인 포커스 그룹의 포화에 대해 보고하며, 포화는 요일 및 참여자별로 보고됩니다.
  • 상위 연구에서는 Morse 등(2014)이 16개 그룹에서 포화에 도달했다고 보고했지만, 코드나 테마가 아닌 공간적 위치에 초점을 맞추었기 때문에 포화에 대한 표본 크기가 더 클 수 있습니다.
  • 나머지 4개의 테스트는 유사한 포화 정의를 사용했으며, 평균 5~6개의 포커스 그룹으로 4~8개의 포커스 그룹 사이에서 포화에 도달했습니다.

대부분의 테스트(4/6)는 연구 모집단이 균질했지만 포화를 평가하는 접근 방식과 코드 또는 카테고리의 포화 목표가 다양했습니다. 이질적인 샘플을 사용한 두 테스트에서는 모두 평균 그룹 수보다 많은 샘플 크기(7개 및 17개 그룹)에서 포화에 도달했습니다.

Fig. 3 shows the sample size for saturation from six empirical tests using data from focus group discussions. For comparability, where various levels of saturation are reported, those closest to 90% are shown in the figure. Across all six tests, saturation was reached between 1 and 16 focus groups. Two tests are outliers and thus not comparable to others.

  • At the lower end, Hancock et al. (2016) report on saturation in a single asynchronous, online focus group, and saturation is reported by day and participant.
  • At the higher end, Morse et al. (2014) report reaching saturation at 16 groups; however, they focus on spatial locations rather than codes or themes, which may account for the higher sample size for saturation.
  • The remaining four tests used similar definitions of saturation and reached saturation between 4 and 8 focus groups, with a mean of 5–6 groups.

Most tests (4/6) had a homogenous study population but varied in their approach to assessing saturation and the saturation goal of codes or categories. In the two tests using heterogeneous samples, both reached saturation at sample sizes above the mean number of groups (at 7 and 17 groups).

 

경험적 데이터에 적용된 포화에 대한 통계 모델을 개발한 연구에서도 포화에 대한 표본 크기는 위와 유사했습니다(표 2). 예를 들어, Rowlands 등(2016)은 로그 정규 분포를 사용하여 [세 가지 인터뷰 데이터 세트]에서 포화를 추정했으며, 그 결과 95% 신뢰 수준에서 포화에 대한 표본 크기는 10개, 10개, 13개로 나타났습니다. (2020)은 [집합 이론과 부분 최소 제곱 회귀]를 사용하여 12개의 인터뷰 데이터 세트에 적용했을 때 포화를 추정했습니다.

In studies that developed statistical models for saturation that were applied to empirical data, the sample sizes for saturation were similar to those above (Table 2). For example, Rowlands et al. (2016) used the lognormal distribution to estimate saturation in three datasets of interviews, and results found the sample sizes for saturation at 95% confidence to be 10, 10, and 13 interviews. Fofana et al. (2020) used set theory and partial least squares regression to estimate saturation at 12 interviews when applied to an empirical dataset.

4. 토론
4. Discussion

이 체계적 문헌고찰은 포화를 평가하는 실증 연구를 식별하고, 포화에 필요한 표본 크기, 포화를 평가하는 데 사용되는 전략, 그리고 이러한 연구에서 도출할 수 있는 지침을 파악하고자 했습니다. 포화를 실증적으로 평가한 23개의 연구를 확인했으며, 이 중 80%가 2014년 이후에 발표된 연구였습니다. 포화를 평가하는 데는 경험적 데이터를 사용한 연구와 통계적 모델을 사용한 연구라는 두 가지 접근 방식이 있었습니다.
This systematic review sought to identify empirical studies that assess saturation, to identify sample sizes needed for saturation, strategies used to assess saturation, and guidelines we can draw from these studies. We identified 23 studies that empirically assessed saturation, with 80% published since 2014. We identified two different approaches to assess saturation: studies that used empirical data and those that used statistical models.

포화를 평가하는 한 가지 접근 방식은 데이터를 수집하기 전에 포화에 대한 표본 크기를 추정하기 위한 통계 모델을 개발하는 데 중점을 두었습니다. 포화를 선험적으로 추정하려는 노력에는 박수를 보내지만, 개발된 많은 공식은 [정성적 연구 수행에 부합하지 않는 암묵적 가정을 기반]으로 하기 때문에 그 유용성이 크게 제한됩니다. 이러한 연구 중 상당수는 [무작위 표본]을 사용하여 전체 인구에서 특정 주제의 유병률 또는 데이터에서 원하는 주제의 인스턴스를 파악하는 등 [확률 기반 가정]을 사용합니다. 게다가 연구자들은 연구를 수행하기 전에 이러한 매개변수를 알지 못하는 경우가 많으며, 항목의 유병률은 질적 연구의 중요한 초점이 아닙니다. [통계 공식]은 정량적 연구자에게 익숙한 검정력 계산과 유사하게 보일 수 있기 때문에, 정성적 연구와 가정이 일치하지 않을 경우 [궁극적으로 달성할 수 없는 과학적 신뢰성]을 오해의 소지가 있는 베일로 만들 수 있다고 생각합니다. 이러한 접근 방식에 대한 우려를 감안하여 더 이상 고려하지 않습니다.
One approach to assessing saturation focused on developing statistical models to estimate sample sizes for saturation prior to data collection. While we applaud efforts to estimate saturation a priori, many of the formulas developed are based on implicit assumptions that do not align with the conduct of qualitative research, thereby significantly limiting their utility. Many of these studies use probability-based assumptions, such as having a random sample and knowing the prevalence of a theme in the broader population or the desired instances of a theme in data. Moreover, researchers often do not know these parameters prior to conducting a study, nor is prevalence of items an important focus of qualitative research. Since a statistical formula may be seen as akin to a power calculation familiar to quantitative researchers, we feel that this may provide a misleading veil of scientific authenticity that ultimately cannot be achieved given the misalignment of assumptions with qualitative research. Given our concerns about these approaches, we do not consider them further.

포화를 평가하는 두 번째 접근 방식은 [경험적 데이터를 사용]했습니다. 심층 인터뷰 데이터를 사용한 16개의 포화 테스트에서 모두 25개 미만의 인터뷰, 더 구체적으로는 [이상값을 제외한 9~17개의 인터뷰]에서 포화에 도달했습니다. 포화를 평가하기 위해 [다양한 접근 방식, 다양한 데이터 세트, 다양한 포화 목표(코드 대 카테고리), 동질적 및 이질적인 연구 모집단을 사용]했음에도 불구하고 연구는 여전히 좁은 범위의 인터뷰 내에서 포화에 도달했습니다. 이는 다양한 접근 방식에 걸쳐 강력한 외부 신뢰도를 보여줍니다. 모든 테스트에서 평균 12~13개의 인터뷰가 포화 상태에 도달했는데, 이는 포화 상태를 경험적으로 평가한 최초의 연구 중 하나인 Guest 외(2006)의 연구 결과(12개의 인터뷰에서 포화 상태에 도달했다고 보고)와 매우 유사합니다. [연구 모집단의 동질성, 무작위 배정 사용, 포화 목표 등] [연구 특성]에 따른 포화에 대한 명확한 패턴을 발견하지 못했는데, 이는 이러한 매개변수를 실제로 평가한 연구가 거의 없었기 때문입니다. 

A second approach to assess saturation used empirical data. In all 16 tests of saturation with data from in-depth interviews, saturation was reached in under 25 interviews, more specifically between 9 and 17 interviews excluding outliers. Despite using different approaches to assess saturation, different datasets, varying saturation goals (codes vs categories), and homogenous and heterogeneous study populations, studies still reached saturation within a narrow range of interviews. This demonstrates strong external reliability across the different approaches. Across all tests, an average of 12–13 interviews reached saturation, which is remarkably similar to findings from Guest et al. (2006), one of the first studies to empirically assess saturation, which reported saturation at 12 interviews. We found no clear pattern in saturation by study characteristics, such as homogeneity of the study population, use of randomization, or saturation goal, largely because few studies actually assessed these parameters in their approach.

포커스 그룹 토론 데이터를 사용한 6개의 테스트에서 포화에 도달한 그룹은 4~8개 그룹으로, 비슷한 좁은 범위였습니다. 인구통계학적 계층화, 이질적인 샘플, 더 넓은 포화 목표(예: 코드 의미, 모든 주제 대 주요 주제)를 사용한 연구에서는 포화에 도달하기 위해 더 많은 그룹이 필요했습니다. 그러나 포커스 그룹 데이터를 비교한 추가 연구 없이 이러한 특성의 영향에 대한 결론을 내리는 것은 신중해야 합니다. 전반적으로 이러한 결과는 다양한 정성적 방법의 포화에 대한 표본 크기에 대한 절실히 필요한 경험적 증거를 제공합니다. 특정 범위의 인터뷰 또는 포커스 그룹 내에서 포화가 수렴하더라도, 아래에서 설명하는 바와 같이 이러한 방법을 사용하는 모든 질적 연구의 일반적인 표본 크기로 사용하거나 잘못 설계되거나 실행된 질적 연구를 정당화하기 위해 이러한 결과를 사용하지 않도록 주의해야 합니다. 대신, 이러한 결과를 질적 연구의 표본 크기를 추정할 때 특정 연구 특성과 함께 고려할 수 있는 지침으로 사용하는 것이 좋습니다. 

In six tests using data from focus group discussions, saturation was reached by 4–8 groups, a similarly narrow range. Studies using demographic stratification, heterogeneous samples, and broader saturation goals (e.g., code meaning, all themes vs main themes) needed more groups to reach saturation. However, we are cautious about drawing conclusions regarding the influences of these characteristics without more studies with focus group data to compare. Overall, these findings provide much-needed empirical evidence of sample sizes for saturation for different qualitative methods. Despite convergence of saturation within a specific range of interviews or focus groups, we caution not to use these findings as generic sample sizes for any qualitative study using these methods, or to justify poorly designed or executed qualitative studies, as we discuss below. Instead, we recommend using these results as guidance to consider alongside the specific study characteristics when estimating the sample size for a qualitative study.

4.1. 연구에 대한 시사점
4.1. Implications for research

체계적 문헌고찰의 결과는 몇 가지 중요한 시사점을 제공합니다. 여기서는 경험적 기반 연구의 시사점에만 초점을 맞춥니다. 이러한 결과는 인터뷰와 포커스 그룹 토론을 사용할 때 포화를 위한 적절한 표본 크기에 관한 경험적 지침을 제공하며, 이는 질적 연구 제안서를 개발할 때 유용할 수 있습니다. 검토 대상에 포함된 대부분의 경험적 기반 연구는 연구 모집단이 균질하고 연구 목표가 집중되어 있었기 때문에 이러한 결과를 다른 유형의 표본이나 더 광범위한 목표를 사용하는 연구에 자신 있게 추정할 수는 없습니다. 따라서 [이러한 결과를 출발점]으로 삼아 인터뷰 또는 포커스 그룹의 잠재적 범위를 파악한 다음 연구 특성(예: 연구 목표, 연구 대상 현상의 성격 및 복잡성, 도구 구조, 샘플링 전략, 표본의 계층화, 연구자의 질적 연구 경험, 포화 목표 및 추구하는 포화 정도)을 고려하여 표본 크기를 구체화하는 것이 좋습니다(Baker and Edwards, 2012; Galvin, 2015; Morse, 1995; 연구 변수를 사용하여 포화를 추정하는 데 대한 자세한 논의는 Hennink et al., 2017 참조). 이러한 고려사항은 각 특정 연구에 보다 맞춤화된 표본 크기로 이어질 뿐만 아니라 제안된 표본 크기에 대한 보다 명확한 정당성을 제공하여 엄밀성을 더합니다. 
The results of our systematic review have several important implications. We focus here only on implications of empirically based studies. These results provide empirical guidance regarding adequate samples sizes for saturation when using interviews and focus group discussions, which can be useful when developing qualitative research proposals. The majority of empirically based studies in our review had a homogenous study population and focused research objectives, so these results cannot be confidently extrapolated to studies using different types of samples or broader goals. Therefore, we recommend using these results as a starting point to identify a potential range of interviews or focus groups then refining the sample size by considering the study characteristics (e.g., study goals, nature and complexity of phenomenon studied, instrument structure, sampling strategy, stratification of sample, researcher's experience in qualitative research, saturation goal, and degree of saturation sought) (Baker and Edwards, 2012; Galvin, 2015; Morse, 1995; see Hennink et al., 2017 for fuller discussion on using study parameters to estimate saturation). These considerations will not only lead to a more tailored sample size for each particular study but also provide clearer justification for the proposed sample size, thereby adding rigor.

또한, 본 연구 결과는 연구자들에게 질적 표본 크기가 '너무 작다'는 일반적인 비판을 반박할 수 있는 강력한 경험적 증거를 제공하며, 이는 일반적으로 이러한 주장에 대한 근거가 제시되지 않음에도 불구하고 비효율적이라는 것을 암시합니다. 우리의 연구 결과는 '작은' 표본 크기가 질적 연구에 효과적이라는 것을 입증하고, 질적 연구에서 적절한 표본 크기의 기준이 되는 포화 상태에 도달할 수 있기 때문에 왜 효과적인지 그 이유를 보여줄 수 있습니다. 또한, 다양한 질적 방법에서 포화 상태에 도달할 수 있는 다양한 표본 크기를 제공함(예: 9~17명의 인터뷰 또는 4~8명의 포커스 그룹)으로써 '작은' 표본이 실제로 어떤 것인지 보여줍니다. [질적 연구를 위한 표본 크기에 대한 일반적인 조언]은 일반적으로 이보다 더 큰 표본 크기를 제안하기 때문에 이 점이 중요합니다. 
Our results also provide researchers with strong empirical evidence to refute the common critique that qualitative sample sizes are ‘too small’, implying that they are ineffective, although no evidence is usually given for these claims. Our results can be used to demonstrate that ‘small’ sample sizes are effective for qualitative research and to show why they are effective – because they are able to reach saturation, the long-held benchmark for an adequate sample size in qualitative research. Furthermore, our results show what a ‘small’ sample actually is, by providing a range of sample sizes for saturation in different qualitative methods (e.g., 9–17 interviews or 4–8 focus groups). This is important because general advice on sample sizes for qualitative research usually suggest higher sample sizes than this.

질적 연구 방법론에 관한 [교과서를 검토]한 결과, 5~60개의 인터뷰(Guest 외, 2006; Constantinou 외, 2017; Hagaman and Wutich, 2017), 2~40개의 포커스 그룹(Guest 외, 2016) 등 표본 크기 권장 사항이 매우 다양하다는 것을 알 수 있었습니다. 더 중요한 것은 이러한 권장 사항 중 어느 것도 경험적 근거가 없다는 것입니다. 적절한 주의 사항과 함께 근거에 기반한 표본 크기 권장 사항을 제공하는 것이 중요합니다.

  • 필요 이상으로 큰 질적 표본은 연구비 낭비, 연구 참여자의 과도한 부담, 데이터 낭비 등 윤리적 문제를 야기하며(Carlsen and Glenton, 2011; Francis 등, 2010),
  • 표본이 너무 작아 포화 상태에 도달하면 연구 결과의 타당성이 떨어집니다(Hennink 등, 2017).

따라서 본 연구 결과는 질적 연구 지침서의 가이드라인의 일부로 포함될 수 있는 포화에 대한 경험적 기반 표본 크기를 제공합니다. 
Reviews of textbooks on qualitative research methodology found that sample size recommendations vary widely, for example 5–60 interviews (Guest et al., 2006; Constantinou et al., 2017; Hagaman and Wutich, 2017) and 2 to 40 focus groups (Guest et al., 2016). More importantly, none of these recommendations is empirically based. Providing evidence-based sample size recommendations, with appropriate caveats, is important.

  • Qualitative samples that are larger than needed raise ethical issues, such as wasting research funds, overburdening study participants, and leading to wasted data (Carlsen and Glenton, 2011Francis et al., 2010),
  • while samples that are too small to reach saturation reduce the validity of study findings (Hennink et al., 2017).

Our results thus provide empirically based samples sizes for saturation that could be included as part of the guidelines in instructional textbooks on qualitative research.

또한 바실레이우 외(2018)는 일부 질적 연구자조차도 자신의 표본 크기가 '작다'고 표현했지만, 이는 "후회나 사과 담론에 따른 한계로 해석된다"(12쪽)는 점을 발견했습니다. 이러한 저자들은 [실증주의를 지향하는 독자들의 우려]에 부응하기 위해 글을 썼을지 모르지만, 질적 근거를 들어 '작은' 표본을 옹호하는 사람은 거의 없었습니다. 우리는 연구자들이 [실증주의 패러다임의 (대부분 부적절한) 우려와 수치적 기대에 대응하기]보다는 [질적 연구의 원칙을 사용하여 표본 크기를 보다 자신 있게 정당화할 수 있도록 우리의 결과를 반영할 것을 권장]합니다. 질적 연구의 표본 크기는 데이터의 적절성에 따라 결정되므로, 효과적인 표본 크기는 숫자(n)보다는 연구 대상 현상에 대한 풍부하고 미묘한 설명을 제공할 수 있는 데이터의 능력에 더 중점을 둡니다. 궁극적으로, 질적 연구를 위한 표본 크기를 결정하고 정당화하는 것은 포화에 영향을 미치는 연구 특성과 분리할 수 없습니다. 우리의 결과는 "소규모 표본에서 엄격하게 수집된 질적 데이터는 사람들의 경험의 전체 차원을 실질적으로 나타낼 수 있다"(Young and Casey, 2019, 12쪽)는 다른 연구 결과를 반영하므로 질적 연구의 엄격성을 평가할 때 이를 한계로 보거나 제시해서는 안 됩니다.
Furthermore, Vasileiou et al. (2018) found that even some qualitative researchers characterized their own sample size as ‘small’, but this was “construed as a limitation couched in a discourse of regret or apology” (p. 12). Although these authors may be writing to the concerns of more positivist-oriented readers, few defended their ‘small’ sample on qualitative grounds. We encourage researchers to reflect on our results to more confidently justify their sample sizes using the principles of qualitative research rather than responding to the (mostly inappropriate) concerns of a more dominant positivist paradigm and their numerical expectations. Sample sizes in qualitative research are guided by data adequacy, so an effective sample size is less about numbers (n's) and more about the ability of data to provide a rich and nuanced account of the phenomenon studied. Ultimately, determining and justifying sample sizes for qualitative research cannot be detached from the study characteristics that influence saturation. Our results echo others, that “rigorously collected qualitative data from small samples can substantially represent the full dimensionality of people's experiences” (Young and Casey, 2019, p.12) and therefore should not be viewed or presented as a limitation when evaluating the rigor of qualitative research.

또한, 이번 연구 결과는 [질적 연구 검토자]에게 [포화에 맞는 효과적인 표본 크기에 대한 경험적 지침]을 제공합니다. 이는 '작은' 표본 크기에 대해 질적 연구를 비판하는 일상적인 관행에 다시 초점을 맞추는 데 도움이 될 수 있으며, 검토자는 대신 연구자에게 다음과 같은 질문을 통해 표본 크기에 대한 보다 명확한 근거를 제시하도록 요청할 수 있습니다:

  • "귀하와 같은 동질적인 연구 모집단에서는 일반적으로 25명 미만으로 포화 상태에 도달할 수 있는데, 왜 40명의 인터뷰 표본을 사용했습니까?"라고 질문할 수 있습니다.

우리는 일반적으로 질적 연구에 효과적인 표본 크기를 결정할 때 수치적 지침만을 사용하는 것을 지지하지는 않지만, 이러한 유형의 질문은 포화에 대한 이용 가능한 경험적 증거를 사용하여 연구자가 표본 크기를 정당화하고 각 개별 연구의 특성을 보다 투명하게 사용하도록 도전하는 보다 정보에 입각한 비판을 반영합니다. 따라서 우리는 질적 연구자들이 표본 크기에 대해 보다 충분한 근거를 제시하고, 질적 연구 검토자들이 이러한 결과를 적용하여 질적 연구에 대한 표본 크기에 대해 보다 효과적인 비평을 제공할 것을 촉구합니다. 이를 통해 질적 연구에 대한 보고와 비평의 질을 개선하고 '작은' 표본 크기에 대한 근거 없는 비판에서 벗어날 수 있습니다.
Our results also provide empirical guidance on effective sample sizes for saturation for reviewers of qualitative research. This may help to refocus the routine practice of criticizing qualitative research for ‘small’ sample sizes so that reviewers may instead ask researchers to provide more explicit justifications for their sample size by asking, for example:

  • “why do you have a sample of 40 interviews, when saturation can typically be reached in less than 25 with a homogenous study population such as yours?”

Although, we generally do not support using only numerical guidance in determining an effective sample size for qualitative research, these types of questions reflect a more informed critique that uses available empirical evidence on saturation to challenge researchers to be more transparent in justifying their sample sizes and using the characteristics of each individual study to do so. We therefore encourage qualitative researchers to provide fuller justifications of their sample sizes and urge reviewers of qualitative studies to apply these findings to provide more effective critiques of sample sizes for qualitative research. This may improve the quality of reporting and critiquing qualitative research and move away from often unsubstantiated critiques of ‘small’ sample sizes.

또한 이번 연구 결과는 [코드 빈도 수]와 [코드 의미 평가의 여러 변형]을 포함하여 [포화를 평가하는 5가지 접근 방식]을 종합적으로 보여줍니다. 이제 질적 연구자들은 데이터 수집 중에 포화를 평가할 수 있는 다양한 전략을 사용할 수 있습니다. 정성적 연구에 대한 수많은 검토에 따르면 포화는 종종 표본 크기를 정당화하기 위해 사용되지만, 포화를 평가하거나 결정하는 방법에 대한 투명성이 압도적으로 부족하다는 사실이 밝혀졌습니다(Carlsen and Glenton, 2011; Francis 외., 2010; Marshall 외., 2013; Vasileiou 외., 2018). 특히 포화가 질적 연구에서 품질의 지표로 환영받는다는 점을 고려할 때 이러한 투명성 부족은 우려스러운 부분입니다. 이는 포화가 연구자들에 의해 거의 근거가 없고 질적 연구 검토자들에 의해 도전받지 않는 엄격함의 외관을 제공하기 위해 "엄격함의 덮개"(Constantinou 외., 2017, 2페이지)로 사용되고 있음을 시사합니다
Our results also synthesize five distinct approaches to assess saturation, including several variations of code frequency counts and assessing code meaning. Qualitative researchers now have an array of strategies to assess saturation during data collection. Numerous reviews of qualitative studies have found that saturation is often used to justify a sample size, but there was an overwhelming lack of transparency in how it was assessed or determined (Carlsen and Glenton, 2011; Francis et al., 2010; Marshall et al., 2013; Vasileiou et al., 2018). This lack of transparency is concerning, particularly given that saturation is hailed as an indicator of quality in qualitative research. It suggests that saturation is being used as a “mantle of rigor” (Constantinou et al., 2017, p. 2) to provide the appearance of rigor that is largely unsubstantiated by researchers and left unchallenged by reviewers of qualitative studies.

이러한 투명성 부족은 어느 정도 [포화 평가에 대한 지침이 없기 때문]일 수 있습니다. 이번 검토에서는 질적 연구자가 포화 평가 방법, 포화에 도달했는지 여부 또는 연구에서 포화가 어느 정도 달성되었는지를 보다 투명하게 보고하기 위해 사용할 수 있는 다양한 전략을 종합했습니다. 이제 연구자는 [포화를 평가하는 전략]과 [포화를 결정한 기준(예: 중지 기준, 누적 빈도 그래프, 코드 비율, 코드 의미)]을 지정할 수 있습니다. 이러한 투명성 강화는 개별 연구의 엄격성뿐만 아니라 질적 연구 전반의 품질에도 분명한 이점이 있습니다. 포화에 대한 투명성이 높아지면 연구의 재현성이 향상되고 포화를 보고하는 방법에 대한 기대치가 높아지며, '포화까지 데이터를 수집했다'와 같이 [일반적이고 근거가 없는 문구]를 사용하는 것에서 벗어날 수 있습니다. 

To some extent, this lack of transparency may reflect the absence of guidance on assessing saturation. Our review has synthesized a range of strategies that can be used by qualitative researchers to become more transparent in reporting how saturation was assessed, whether it was reached, or the extent to which it was achieved in a study. Researchers can now specify a strategy for assessing saturation and the criteria on which it was determined (e.g., a stopping criterion, cumulative frequency graphs, percentage of codes, code meaning). Such greater transparency has clear benefits for the rigor of individual studies but also for the quality of qualitative research as a whole. Greater transparency regarding saturation improves reproducibility of the research and raises expectations on how to report saturation, all of which move away from using generic and unsupported statements such as ‘data were collected until saturation’.

또한 [질적 연구를 발표하는 저널]은 포화에 대한 투명한 보고를 장려하는 데 중요한 역할을 합니다. Vasileiou 등(2018)은 [표본 크기의 정당성 보고]가 [특정 학술지와 일치]한다는 사실을 발견했으며, 이는 학술지 요건이 포화 보고의 투명성 강화 규범을 장려할 수 있음을 시사합니다. 학술지 심사자는 연구자에게 다음 등의 질문을 할 수 있습니다. 

  • '포화를 어떻게 평가했습니까?',
  • '포화에 도달했다는 것을 어떻게 알 수 있습니까? ' 또는
  • '핵심 코드, 범주, 의미 등에서 어느 정도 포화 상태에 도달했습니까?'

이러한 요청은 포화에 대한 보다 투명하고 미묘하며 엄격한 보고가 필요하다는 신호입니다. 이는 단순한 체크리스트 요건을 넘어서는 것으로, 포화가 어떻게 결정되었는지에 대한 연구별 세부 정보 없이 포화에 도달했다는 모호한 보고를 지속시킬 수 있습니다.

In addition, journals publishing qualitative research play a critical role in encouraging transparent reporting of saturation. Vasileiou et al. (2018) found that reporting of sample size justifications aligned with particular academic journals, suggesting that journal requirements may encourage norms of greater transparency in reporting saturation. Journal reviewers may also encourage transparency by asking researchers, for example:

  • ‘how did you assess saturation ?’,
  • ‘how do you know you reached saturation ? ’ , or
  • ‘to what extent was saturation reached – in core codes, categories, meaning etc.?'.

Such requests signal that more transparent, nuanced, and rigorous reporting of saturation is expected. This should go beyond simple check-list requirements, which may simply perpetuate vague reporting that saturation was reached without study-specific details on how it was determined.

본 연구에는 몇 가지 잠재적인 한계가 있습니다. 영어로 출판되고 근거 이론의 인식론적 접근 방식을 벗어난 연구만 포함했으며, 특정 질적 방법에 대한 검색어를 제한적으로 사용했지만 일반적인 방법도 포함했습니다. 이러한 기준 때문에 이미 발표된 다른 포화 테스트는 제외되었을 수 있지만, 검색 기준은 해당 주제에 대한 중요한 연구를 포괄할 수 있을 만큼 충분히 광범위하다고 생각합니다. 검토에서 확인된 논문은 압도적으로 건강 연구에 초점을 맞추고 있으며 포화에 대한 유사한 개념화를 가지고 있습니다. 따라서 연구 결과를 비교하기는 쉽지만, 포화를 다르게 개념화하는 다른 분야에는 이러한 결과를 적용하지 못할 수도 있습니다.
Our study has some potential limitations. We included only studies that were published in English and were outside the epistemological approach of grounded theory, and we used limited search terms for specific qualitative methods but included common methods. While these criteria may have excluded other published tests of saturation, we believe our search criteria were broad enough to capture a significant body of research on the topic. Articles identified in our review focus overwhelmingly on health research and have similar conceptualizations of saturation. While this makes the studies more comparable, these results may not be applicable to other disciplines that may conceptualize saturation differently.

5. 결론
5. Conclusion

포화는 질적 연구에서 표본 크기를 결정할 때 엄격함의 초석으로 간주되지만, 근거 이론을 제외하고는 포화 운영에 대한 지침이 거의 없습니다. 이 체계적 고찰에서는 질적 연구에서 포화를 경험적으로 평가한 연구, 포화를 평가하는 문서화된 접근법, 포화를 위한 표본 크기를 확인한 연구를 확인했습니다. 특히 연구 모집단이 비교적 균질하고 목표가 좁게 정의된 연구에서 좁은 범위의 인터뷰(9~17개) 또는 포커스 그룹 토론(4~8개)으로도 포화를 달성할 수 있음을 보여주는 다양한 포화 평가 접근법을 설명합니다. 체계적인 검토를 통해 포화에 대한 표본 크기를 확인했지만, 특정 매개변수가 포화에 어떤 영향을 미치는지 결정하기 위한 경험적 기반 연구는 거의 발견되지 않았습니다. 연구 목표, 연구 모집단의 특성, 사용된 샘플링 전략(예: 귀납적 샘플링과 고정 샘플링), 데이터 유형, 포화 목표 및 기타 영향과 같은 특정 매개변수가 포화에 미치는 영향에 대한 추가 연구가 필요합니다.

Saturation is considered the cornerstone of rigor in determining sample sizes in qualitative research, yet there is little guidance on its operationalization outside of grounded theory. In this systematic review, we identified studies that empirically assessed saturation in qualitative research, documented approaches to assess saturation, and identified sample sizes for saturation. We describe an array of approaches to assess saturation that demonstrate saturation can be achieved in a narrow range of interviews (9–17) or focus group discussions (4–8), particularly in studies with relatively homogenous study populations and narrowly defined objectives. Although our systematic review identified sample sizes for saturation, we found little empirically based research to determine how specific parameters influence saturation. Further research is needed on how specific parameters influence saturation, such as the study goal, nature of the study population, sampling strategy used (i.e. inductive vs fixed sampling), type of data, saturation goal, and other influences.


 

Soc Sci Med. 2022 Jan;292:114523. doi: 10.1016/j.socscimed.2021.114523. Epub 2021 Nov 2.

Sample sizes for saturation in qualitative research: A systematic review of empirical tests

Affiliations expand

PMID: 34785096

DOI: 10.1016/j.socscimed.2021.114523

Abstract

Objective: To review empirical studies that assess saturation in qualitative research in order to identify sample sizes for saturation, strategies used to assess saturation, and guidance we can draw from these studies.

Methods: We conducted a systematic review of four databases to identify studies empirically assessing sample sizes for saturation in qualitative research, supplemented by searching citing articles and reference lists.

Results: We identified 23 articles that used empirical data (n = 17) or statistical modeling (n = 6) to assess saturation. Studies using empirical data reached saturation within a narrow range of interviews (9-17) or focus group discussions (4-8), particularly those with relatively homogenous study populations and narrowly defined objectives. Most studies had a relatively homogenous study population and assessed code saturation; the few outliers (e.g., multi-country research, meta-themes, "code meaning" saturation) needed larger samples for saturation.

Conclusions: Despite varied research topics and approaches to assessing saturation, studies converged on a relatively consistent sample size for saturation for commonly used qualitative research methods. However, these findings apply to certain types of studies (e.g., those with homogenous study populations). These results provide strong empirical guidance on effective sample sizes for qualitative research, which can be used in conjunction with the characteristics of individual studies to estimate an appropriate sample size prior to data collection. This synthesis also provides an important resource for researchers, academic journals, journal reviewers, ethical review boards, and funding agencies to facilitate greater transparency in justifying and reporting sample sizes in qualitative research. Future empirical research is needed to explore how various parameters affect sample sizes for saturation.

Keywords: Focus group discussions; Interviews; Qualitative research; Sample size; Saturation.

아너(Honors)를 위하여: 임상실습 평가와 성적에 관한 학생의 인식 다기관 연구(Acad Med, 2019)
In Pursuit of Honors: A Multi-Institutional Study of Students’ Perceptions of Clerkship Evaluation and Grading
Justin L. Bullock, MPH, Cindy J. Lai, MD, Tai Lockspeiser, MD, MHPE, Patricia S. O’Sullivan, EdD, Paul Aronowitz, MD, Deborah Dellmore, MD, Cha-Chi Fung, PhD, Christopher Knight, MD, and Karen E. Hauer, MD, PhD 

 

[임상 실습]을 준비하려면 학생들은 광범위하고 빠르게 확장되는 기술과 지식을 습득해야 합니다.1 동시에, 학생들은 특히 특정 전문과목에서 레지던트 자리를 차지하기 위한 경쟁이 치열해지고 있습니다.2,3 이러한 요구는 함께 [부담스러운 임상 학습 환경]을 조성하여 학습자에게 악영향을 미칠 수 있습니다.4 학생들의 스트레스를 유발하는 중요한 요인 중 하나는 [임상실습 성적]입니다.5,6 성적은 학생과 의과대학에 중요한 피드백을 제공하며, 레지던트 프로그램은 레지던트 선발 시 핵심 임상실습 성적에 의존합니다.7-9 성적 배정은 일반적으로 [시험 점수]와 [감독 교수 및 레지던트의 종합 평가]에 의해 결정됩니다. 그럼에도 불구하고 학생과 교육자 모두 성적의 공정성과 정확성에 의문을 제기합니다.4 "평가가 학습을 이끈다"는 교육자의 격언에 비추어 볼 때, 현재 평가 시스템에 대한 부정적인 인식은 학생의 동기 부여, 학습 행동, 성과에 부정적인 영향을 미칠 수 있습니다.10
Preparing for clinical practice requires students to acquire broad and rapidly expanding skills and knowledge.1 Simultaneously, students face increasing competition for residency positions, particularly in certain specialties.2,3 Together, these demands create a taxing clinical learning environment, which may adversely affect learners.4 One significant contributor to student stress is clerkship grading.5,6 Grades provide important feedback to students and medical schools, and residency programs rely on core clerkship grades in resident selection.7–9 Grade assignments are typically informed by examination scores and summative evaluations from supervising faculty and residents. Still, students and educators alike question the fairness and accuracy of grades.4 Drawing from the educator’s adage “assessment drives learning,” negative perceptions of the current assessment system may adversely affect students’ motivation, learning behaviors, and performance.10

임상실습 평가 및 채점에 대한 학생의 우려는 [다양한 요인]으로 인해 발생할 수 있습니다.

  • 감독자는 평가 척도를 다양하게 해석하고 최고 성과에 대한 [공유된 정신 모델이 부족]할 수 있습니다.11-13
  • 학생은 자신을 평가할 때 [감독자가 무엇을 중요하게 생각하는지 불확실]하게 느낄 수 있습니다.14
  • 공정한 평가 시스템은 학생이 학습하고 [학습을 입증할 수 있는 충분한 기회]가 필요하고 [평가 및 채점에 투명한 기준을 사용]하며 [공평]해야 합니다.15,16
    • 한 의과대학에서 실시한 한 연구에 따르면 학생의 38%만이 임상실습 평가가 공정하다고 생각하는 것으로 나타났습니다.17
  • 수퍼바이저가 [수련생을 직접 관찰하지 않고 역량을 평가]하기 때문에 학생들은 평가의 정확성을 의심할 수 있습니다.18,19
  • [편견] 또한 정확성을 위협하고 성적에 대한 회의론을 불러일으킵니다. 의학계에서 소외된 인종 또는 민족(UIM) 출신의 학생은 최고 성적을 받고 아너 소사이어티에 선발될 가능성이 낮습니다.20-22

Students’ concerns around clerkship evaluations and grading may arise from a variety of factors.

  • Supervisors variably interpret assessment scales and may lack a shared mental model of top performance.11–13 
  • Students can feel uncertain about what supervisors value when evaluating them.14 
  • A fair assessment system requires sufficient opportunities for students to learn and demonstrate learning, uses transparent criteria for evaluation and grading, and is equitable.15,16 
    • One study at a single medical school found that only 38% of students felt that clerkship evaluation was fair.17 
  • Students may doubt the accuracy of their evaluations because supervisors evaluate trainees on competencies despite infrequent direct observation of those trainees.18,19 
  • Bias also threatens accuracy and raises skepticism around grades. Students from racial or ethnic groups underrepresented in medicine (UIM) are less likely to earn top grades and honor society selection.20–22

모든 학생은 임상실습 환경이 학습에 미치는 영향에 취약할 수 있습니다.

  • [숙달 지향적 환경]은 학생들이 도전을 추구하고 장애물에 직면했을 때 성공하는 학습에 대한 적응적 접근 방식을 촉진합니다.23
  • 반대로 [성과 지향적 환경]에는
    • 학생들이 [겉으로 유능해 보일 수 있는 과제를 수행]하면 보상하는 "성과 접근 방식"과,
    • 학생들이 [무능해 보일 수 있는 도전적인 상황을 회피]하게 만드는 "성과 회피 방식"이 포함됩니다.
  • [숙달 중심]의 [합격/불합격 전임상 학습 환경]에서 [성과 중심]의 [단계별 채점 임상 학습 환경]으로 전환하면 학생이 숙달 중심의 행동을 경시하고 학습에 부정적인 영향을 미칠 수 있습니다.24
  • [성과 중심]의 학습 문화는 학생의 정보 유지력과 만족도를 떨어뜨리고 소진을 증가시킬 수 있습니다.23,25

All students can be susceptible to influences of the clerkship environment on their learning.

  • A mastery-oriented environment fosters adaptive approaches to learning in which students seek challenges and thrive when facing obstacles.23 
  • Conversely, performance-oriented environments include
    • “performance approach,” which rewards students for performing tasks that they know will make them appear competent,
    • and “performance avoid,” which encourages students to avoid challenging situations that could make them appear incompetent.
  • The transition from a more mastery-oriented pass/fail preclinical learning environment to a more performance-oriented tiered grading clinical learning environment may cause students to deemphasize mastery-oriented behaviors and negatively affect learning.24 
  • A performance-oriented learning culture can decrease students’ retention of information and satisfaction and increase burnout.23,25

[UIM 학생과 비 UIM 학생 간의 성적 격차]는 [평가자의 편견을 넘어 임상실습 학습 환경의 다른 요인에 대한 고려]를 촉구하며, 이는 UIM 학생의 성과 저하에 고유하게 기여할 수 있습니다.21,26 [낙인찍힌 집단]의 취약한 구성원(예: 일반적으로 UIM인 인종/민족 학생)이 [자신이 속한 집단에 대한 낮은 기대치에 부합할 것을 걱정]할 때 [고정관념 위협]을 경험하게 됩니다. [고정관념 위협]은 인지 부하를 증가시키고 습득한 기술과 역량을 발휘하지 못하게 함으로써 집단 간 성과 차이를 악화시킵니다.27-29 인종, 성별, 나이와 관련된 고정관념 위협은 널리 연구되어 왔지만, 의대생 사이에서 고정관념 위협의 영향을 조사한 문헌은 부족합니다.28-32
Grading disparities between UIM and non-UIM students prompt consideration of other forces in the clerkship learning environment, beyond evaluator bias, which may uniquely contribute to poorer UIM student performance.21,26 When vulnerable members of stigmatized groups (e.g., students from races/ethnicities typically UIM) worry that they will conform to lower expectations for their group, they experience stereotype threat. Stereotype threat exacerbates group differences in performance by increasing cognitive load and inhibiting the display of acquired skills and competencies.27–29 While stereotype threats relating to race, gender, and age have been widely explored, a dearth of literature examines effects of stereotype threat amongst medical students.28–32

본 연구는

  • (1) 임상실습 평가 및 채점의 공정성과 정확성에 대한 학생들의 인식을 조사하고,
  • (2) 임상실습 학습 환경에 대한 학생들의 인식을 조사하고,
  • (3) 이러한 인식과 학생의 성취도 사이의 관계를 평가하기 위해 설계되었습니다.

We designed this study to

  • (1) examine students’ perceptions of the fairness and accuracy of clerkship evaluation and grading,
  • (2) examine students’ perceptions of the clerkship learning environment, and
  • (3) assess the relationship between these perceptions and students’ achievement.

방법
Method

설계
Design

이 연구는 여러 기관을 대상으로 한 횡단면 설문조사 연구입니다.
This is a multi-institutional, cross-sectional survey study.

설정
Setting

연구 기관은 서부의 다양한 지리적 위치와 공립/사립 현황을 대표하는 서부 교육 문제 그룹에 속한 미국 학교 6곳을 편의 표본으로 선정했습니다(표 1). 초대받은 학교 중 참여를 거부한 학교는 없었습니다. 6개 기관 심의위원회 모두 이 연구를 승인했습니다. 모든 학교는 학생들에게 가정의학과, 내과, 산부인과, 소아과, 정신과, 외과 임상실습을 이수하도록 요구했습니다. 일부는 추가적으로 요구되는 임상실습이 있었습니다. 이 연구에서 "우등"은 각 학교에서 달성할 수 있는 가장 높은 서클러십 성적을 의미합니다. 전국 의과대학과 마찬가지로 학교마다 우등상을 받을 수 있는 학생 비율, 종단 통합 서클럭의 존재 여부, 성적 부여 방식이 다양했습니다.33
Study institutions were a convenience sample of 6 U.S. schools in the Western Group on Educational Affairs, representing diverse western geographical locations and public/private status (Table 1). No invited schools declined participation. All 6 institutional review boards approved the study. All schools required students to complete family medicine, internal medicine, obstetrics–gynecology, pediatrics, psychiatry, and surgery clerkships (see Supplemental Digital Appendix 1 at https://links.lww.com/ACADMED/A720). Some had additional required clerkships. In this study, “honors” refers to the highest clerkship grade achievable at each school. Consistent with medical schools nationally, schools varied in the percentage of students allowed to receive honors, presence of longitudinal integrated clerkships, and method of grade assignments.33

참여 학생
Participating students

참여 대상은 핵심 임상실습 연도가 끝나는 모든 의대생이었습니다. 5개 학교에서 학생들은 해당 학교의 수석 조사자가 서명한 전자 설문조사 플랫폼에 대한 개별 이메일 링크를 받았습니다. 학교별 규칙에 따라 이메일 초대는 여섯 번째 학교의 학급 목록 서버로 보내야 했습니다. 응답하지 않은 응답자에게는 매주 최대 3회의 리마인더가 전송되었습니다. 설문조사는 공개 후 30일 동안 활성화되었습니다. 설문조사가 완료되면 참가자는 외부 웹사이트를 통해 이메일 주소를 제출하여 10달러 전자 기프트 카드를 받을 수 있었습니다. 데이터 수집 후, 연구에 관여하지 않은 데이터 분석가가 개인 식별 정보를 제거하고 참가자에게 무작위 식별 번호를 할당했습니다. 학생이 인구 통계 섹션을 작성하지 않았거나 임상실습을 3개 미만으로 완료한 경우 설문조사에서 제외되었습니다.
Eligible participants were all medical students at the end of the core clerkship year. At 5 schools, students received an individualized email link to an electronic survey platform (www.qualtrics.com), signed by the lead investigator of that school. School-specific rules required that the email invitation go to the sixth school’s class listserv. Nonrespondents received up to 3 weekly reminders. The survey was active for 30 days after release. Upon completion, participants could submit their email address via an outside website to receive a $10 electronic gift card. After data collection, a data analyst not otherwise involved in the study removed identifying information and assigned participants random identification numbers. Surveys were excluded if the student did not complete the demographics section or completed fewer than 3 clerkships.

이론적 모델 및 설문조사 개발
Theoretical model and survey development

설문조사 개발 지침에 따라 설문조사를 개발했습니다.34 두 명의 저자(J.L.B., K.E.H.)가 문헌을 검토하여 학생들의 임상실습 성적에 대한 인식을 둘러싼 주요 이론, 증거 및 격차를 파악했습니다. 한 학교(캘리포니아대학교 샌프란시스코 캠퍼스[UCSF])에서는 의과대학 학장들과 함께 임상실습 채점에 관한 학생 타운홀을 개최했습니다. 문헌 검토와 타운홀 피드백을 바탕으로, 저희는 학생 평가의 공정성과 정확성, 학생의 동기 부여와 노력, 피드백에 대한 학생의 인식, 학생의 학습 환경, 학생의 성취 결과에 기여하는 요인에 대한 [학생의 인식에 대한 모델]을 개발했습니다(그림 1). 이 모델을 사용하여 2개의 연구 학교(UCSF, 콜로라도대학교 의과대학)에서 23명의 학생이 서면으로 또는 4개의 포커스 그룹 중 1개 그룹에서 피드백을 제공한 설문조사 항목을 개발하고 파일럿 테스트를 거쳤습니다. 최종 설문조사에는 적응 학습 척도(PALS) 및 고정관념 취약성 척도(SVS) 매뉴얼의 문항도 포함되었습니다.28,35 PALS 숙달, 수행 접근 방식, 수행 회피 교실 목표 구조 척도 및 SVS 고정관념 위협 항목을 "임상실습"을 참조하도록 수정했습니다. 파일럿 학생들에게 혼란을 줄 수 있는 이중 부정적 표현으로 인해 원래의 SVS 항목 3개를 제거했습니다. 
We developed a survey following guidelines for survey development.34 Two authors (J.L.B., K.E.H.) reviewed the literature to identify key theories, evidence, and gaps surrounding students’ perceptions of clerkship grading. One school (University of California, San Francisco [UCSF]) held a student town hall on clerkship grading with medical school deans. Based on the literature review and town hall feedback, we developed a model of students’ perceptions of the fairness and accuracy of clerkship assessment, student motivation and effort, perceptions of feedback, clerkship learning environment, and contributors to students’ achievement outcomes (Figure 1). Using this model, we developed and pilot-tested survey items at 2 study schools (UCSF, University of Colorado School of Medicine) with 23 students who provided feedback in writing or in 1 of 4 focus groups. The final survey also included adapted questions from the Manual for the Patterns of Adaptive Learning Scales (PALS) and the Stereotype Vulnerability Scale (SVS).28,35 We modified the PALS Mastery, Performance Approach, and Performance Avoid Classroom Goal Structure scales and SVS stereotype threat items to reference “clerkships.” We eliminated 3 original SVS items because of double-negative wording that confused pilot students.


[최종 106개의 설문조사 항목]은 참가자 인구통계, 자가 보고한 우등상 수상 횟수, 수강한 임상실습 횟수, 의도한 전공, 다양한 영역이 최종 성적에 미치는 영향에 대한 인식(0~10점), 채점에 대한 인식(공정성, 정확성) 및 임상실습 학습 환경(동기 부여, 고정관념 위협)이라는 가설 예측 변수에 관한 것이었습니다. 예측변인 질문은 5점 리커트 척도(매우 동의하지 않음[1] ~ 매우 동의함[5])를 사용했습니다. 개방형 질문 중 하나는 성적 향상을 위한 학생의 추천을 요청하는 질문이었습니다. 
The final 106 survey items addressed participant demographics, self-reported number of honors earned, number of clerkships taken, intended specialty, perceived impact of various domains on their final grade (scored 0–10), and our hypothesized predictors: perceptions of grading (fairness, accuracy) and clerkship learning environment (motivation, stereotype threat). Predictor questions used a 5-point Likert scale (strongly disagree [1] to strongly agree [5]). One open-ended question solicited students’ recommendations to improve grading (see Supplemental Digital Appendix 2 at https://links.lww.com/ACADMED/A720).

요인 분석
Factor analysis

데이터 축소를 위해 주성분 분석을 사용하여 채점 및 임상실습 학습 환경의 공정성과 정확성에 대한 인식에 대해 리커트 척도 질문을 1~5개의 연속 변수로 처리했습니다. 바리맥스 회전을 사용하여 고유값이 1 이상인 요인을 유지하고 수렴하기 전까지 최대 25회 반복했습니다. 누락된 데이터는 쌍별 삭제를 사용했습니다. 카이저-마이어-올킨 검정은 0.80 이상으로 항목 간에 충분한 상관관계가 있음을 나타냅니다. 항목은 가장 큰 로딩을 기준으로 요인에 할당되었습니다. PALS 동기 부여 척도와 SVS는 이전에 검증되었고 약간의 수정에도 여전히 높은 내적 일관성을 보였기 때문에 주성분 분석에 포함되지 않았습니다.28,35 모든 요인에 대해 크론바흐 알파 계수와 비가중 평균 점수를 계산하여 크론바흐 알파가 0.6 이상인 요인만 유지했습니다. 모든 요인 적재량이 양수가 되도록 필요에 따라 항목을 리버스 코딩했습니다. 유지된 각 요인에 대해 해당 요인을 구성하는 항목의 평균과 동일한 연속 변수로 취급하여 척도 점수를 계산했습니다. 척도 점수의 경우 3점 미만은 "동의하지 않음", 3점 이상은 "동의", = 3점은 "중립"으로 분류했습니다. SVS 점수가 3점 이상이면 고정관념 위협에 취약한 것으로 나타났습니다.
We used principal components analysis for data reduction, treating Likert scale questions as continuous 1–5 variables for perceptions of fairness and accuracy of grading and clerkship learning environment. We used varimax rotation, retaining factors with an eigenvalue ≥ 1 and a maximum of 25 iterations before convergence. We used pairwise deletion for missing data. The Kaiser–Meyer–Olkin test was > 0.80, indicating sufficient correlation amongst items. Items were assigned to factors based on their largest loading. Because the PALS motivation scales and SVS were previously validated and still had high internal consistency with our minor modifications, they were not included in the principal component analysis.28,35 For all factors, we calculated the Cronbach alpha coefficient and nonweighted mean score, retaining factors with Cronbach alpha > 0.6. Items were reverse-coded as needed so that all factor loadings were positive. For each retained factor, we calculated a scale score, treated as a continuous variable equal to the mean of the items comprising the factor. For scale scores, we categorized < 3 as “disagree,” > 3 as “agree,” and = 3 as “neutral.” An SVS score > 3 indicated vulnerability to stereotype threat.

통계 분석
Statistical analysis

인구 통계에 대한 기술 통계를 계산했습니다. t 검정으로 연령의 차이를 평가했습니다. 다른 모든 하위 그룹 비교에는 카이제곱 테스트를 사용했습니다. 첫 번째 목표를 조사하기 위해 공정성과 정확성에 대한 학생들의 인식과 임상실습 학습 환경에서의 학생 경험에 대한 기술 통계를 계산했습니다. 성별 및 UIM 상태별 인식의 하위 그룹 비교를 위해 카이제곱 검정을 사용했습니다. 
We calculated descriptive statistics for demographics. t Tests assessed differences in age. For all other subgroup comparisons, we used chi-square tests. To examine our first aim, we calculated descriptive statistics for students’ perceptions of fairness and accuracy and students’ experience in the clerkship learning environment. We used chi-square tests for subgroup comparisons of perceptions by gender and UIM status.

두 번째 목표인 학생 인구통계학적 특성과 인식 및 우등상 수상 간의 관계를 조사하기 위해 [다변량 회귀 분석]을 사용했습니다. 성적 정책의 학교 간 차이를 설명하기 위해 우등상 수상 비율, 해당 학생의 학교 우등상 수상 비율의 평균 및 표준 편차를 사용하여 z 점수를 계산하여 각 학생의 표준화된 우등상을 계산했습니다. 여기서 '우등상 획득'은 각 학생의 표준화된 우등상 값을 의미합니다. 예측 변수는 학생 인구 통계와 학생 인식(PCA 식별 요인, PALS, SVS)의 두 블록으로 입력했습니다. 인구통계학적 변수는 연속형인 나이를 제외하고 이분법으로 처리했습니다. 아프리카계 미국인, 라틴계 미국인, 라틴계 미국인, 히스패닉계 미국인, 아메리카 원주민, 알래스카 원주민, 하와이 원주민 또는 기타 태평양 섬 주민이라고 스스로 밝힌 UIM 학생들.36 2018년 전국 레지던트 매칭 프로그램 데이터를 사용하여 [경쟁이 심한 전문과목]은 매칭 확률 ≤ 90%, 매칭된 지원자의 1단계 점수 중간값 240 이상, 2단계 CK(임상 지식) 중간값 250의 3가지 기준 중 두 가지를 충족하는 것으로 정의했습니다(표 1). 회귀분석에서 16개의 비교를 설명하기 위해 Bonferroni 보정을 수행했으며, P값이 .003 이하인 경우 통계적으로 유의한 것으로 간주했습니다.41 분석에는 Windows용 IBM SPSS 통계 버전 23.0(IBM, Armonk, New York)을 사용했습니다.
We used multivariable regression analysis to explore our second aim, the relationship between student demographics and perceptions and honors earned. To account for interschool differences in grading policies, we computed each student’s standardized honors by calculating a z score using the fraction of clerkships honored, mean and standard deviation of the fraction of clerkships honored for that student’s school. Hereafter, “honors earned” refers to each student’s standardized honors value. We entered predictor variables in 2 blocks: student demographics and student perceptions (PCA-identified factors, PALS, SVS). We treated demographic variables as dichotomous except age, which was continuous. UIM students self-identified as African American, Latino, Latina, Hispanic, Native American, Alaskan Native, Native Hawaiian, or other Pacific Islander.36 Using 2018 National Resident Matching Program data, competitive specialties were defined as meeting 2 of 3 criteria: probability of matching ≤ 90%, median Step 1 score of matched applicants ≥ 240, and median Step 2 CK (Clinical Knowledge) ≥ 25037–40 (Table 1). We performed a Bonferroni correction to account for 16 comparisons in the regression, with a P value ≤ .003 deemed statistically significant.41 We used IBM SPSS Statistics Version 23.0 for Windows (IBM, Armonk, New York) for analyses.

정성적 분석
Qualitative analysis

세 명의 저자(J.L.B., C.J.L., T.L.)가 콘텐츠 분석을 사용하여 댓글을 분석했습니다. 각 저자는 무작위로 추출한 50개의 댓글에서 귀납적으로 [코드북을 개발]했습니다. 토론을 거쳐 [코드를 하나의 코드북으로 통합]한 후 코딩 과정을 통해 [반복적으로 수정]했습니다. 2명의 저자가 Microsoft Excel을 사용하여 각 댓글을 독립적으로 코딩한 후 토론을 통해 불일치하는 부분을 조정했습니다. 코딩에 대한 토론과 코드 간의 관계에 대한 관심을 통해 핵심 주제와 하위 주제가 도출되었습니다. 코딩자 중에는 의대생, 임상실습 책임교수, 평가위원회 책임자가 포함되어 있었기 때문에 코드 조정은 자연스럽게 반성적 사고를 촉진했습니다. 학생의 코멘트 중 어떤 부분이 주어진 코드에 적용되는 코멘트 비율을 계산했습니다.
Three authors (J.L.B., C.J.L., T.L.) analyzed comments using content analysis. Separately, each author inductively developed a codebook from a random sample of 50 comments. After discussion, we combined codes into a single codebook that we iteratively revised throughout the coding process. Using Microsoft Excel, 2 authors coded each comment independently and then reconciled discrepancies through discussion. Discussion of coding and attention to relationships among codes yielded key themes and subthemes. Code reconciliation naturally facilitated reflexivity as the coders included a senior medical student, clerkship director, and assessment committee director. We calculated the percentage of comments for which any portion of a student’s comment applied to a given code.

결과
Results

전체적으로 972명의 학생이 설문조사 초대를 받았고, 757명이 설문조사를 시작했으며, 701명이 설문조사를 완료했습니다. 35명의 학생이 제외 기준을 충족했습니다: 34명은 3개 미만의 임상실습을 이수했으며, 1명은 이수한 임상실습보다 더 많은 우등상을 받았다고 답했습니다. 최종 응답률은 666/937(71.1%)이었습니다. 참가자의 평균 연령(SD)은 27.5세(3.0)였으며, 54.8%가 여성, 16.4%가 UIM이었습니다(표 1). 이 비율은 2018년 전국 AAMC 의과대학 졸업생 설문조사 표본의 비율과 유사하며, 이 중 49.1%가 여성, 15.5%가 UIM이었습니다.42 응답자들은 평균 6.7회(1.1회)의 핵심 서클러십을 이수한 것으로 나타났습니다. 평균 연령, UIM 학생 비율, 경쟁이 치열한 전문 분야에 지원한 비율은 학교별로 통계적으로 유의미한 차이가 있었습니다(표 1).
Overall, 972 students received survey invitations, 757 began the survey, and 701 completed it. Thirty-five students met exclusion criteria: 34 had completed fewer than 3 clerkships, and 1 reported earning more honors than clerkships taken. The final response rate was 666/937 (71.1%). Participants’ mean age (SD) was 27.5 (3.0); 54.8% were women and 16.4% were UIM (Table 1). These percentages are similar to those in the national 2018 AAMC Medical School Graduate Questionnaire sample, among whom 49.1% were women and 15.5% were UIM.42 Respondents had completed a mean (SD) of 6.7 (1.1) core clerkships. There were small, statistically significant differences across schools for mean age, percentage of UIM students, and percentage applying into competitive specialties (Table 1).

최종 성적에 대한 영역의 중요성 인식
Perceived importance of domain on final grade

이 질문에 대한 응답입니다: 한 해 전체를 고려할 때, "귀하의 경험상 최종 임상실습 성적을 결정하는 데 다음 각 영역이 얼마나 중요합니까?"라는 질문에 대한 답변입니다. (보충 디지털 부록 3: https://links.lww.com/ACADMED/A720 참조), 학생들은 "호감도" 8.7/10점(SD = 1.7), "함께 일하는 특정 어텐딩" 8.7점(1.7), "함께 일하는 특정 레지던트" 8.5점(1.9)을 가장 높게 평가했습니다. '개선' 5.7점(2.7), '환자 및 가족과의 관계' 6.0점(2.7)을 가장 중요하지 않은 것으로 평가했습니다.
In response to the question: Considering the year as a whole, “in your experience, how important is each of the following in determining your final clerkship grade?” (see Supplemental Digital Appendix 3 at https://links.lww.com/ACADMED/A720),

students scored

  • “being liked” 8.7/10 (SD = 1.7),
  • “particular attendings you work with” 8.7 (1.7), and
  • “particular residents you work with” 8.5 (1.9) highest.

They rated

  • “improvement” 5.7 (2.7) and
  • “rapport with patients and families” 6.0 (2.7) as least important.

성적 평가에 대한 인식
Perceptions of grading

회전된 PCA 구성 요소 매트릭스는 데이터 세트의 총 분산 중 64.9%를 차지했으며 6개의 예측 요인을 산출했습니다(표 2). 요인들의 내적 일관성은 높았습니다(크론바흐 알파 = 0.73-0.88). 학생들은 채점의 공정성에 대해 낮은 신뢰를 보였으며, 44.4%의 학생만이 평가가 공정하다고 동의했습니다. 임상실습 평가가 정확하거나 받은 피드백이 유용하다고 생각하는 학생은 3분의 2 미만이었습니다(각각 60.8%, 61.7% 동의). 70.0%의 학생이 레지던트 평가 절차가 공정하다는 데 동의한 반면, 주치의 평가 절차가 공정하다는 데 동의한 학생은 41.7%에 불과했습니다.
Our rotated PCA component matrix accounted for 64.9% of the total variance in our dataset and yielded 6 predictor factors (Table 2). Factors had high internal consistency (Cronbach alpha = 0.73–0.88). Students had low confidence in the fairness of grading, with only 44.4% of students agreeing that assessment was fair. Less than two-thirds of students felt that clerkship assessment was accurate or that feedback received was useful (60.8% and 61.7% agreed, respectively). Whereas 70.0% of students agreed that resident evaluation procedures were fair, only 41.7% agreed that attending evaluation procedures were fair.

학생의 1/3(33.6%)이 채점이 편파적이라고 답했습니다. 여성이 남성보다 평가가 편파적이라고 인식하는 비율이 더 높았지만(64.4% vs 25.2%, P < .0005), 여성이 평가가 정확하다고 평가하는 비율도 더 높았습니다(69.2% vs 52.7%, P < .0005). 채점, 피드백의 공정성, 레지던트 및 참석 평가의 공정성에 대한 인식에는 성별 차이가 없었습니다. UIM 학생은 비 UIM 학생보다 평가가 편파적이라고 인식할 가능성이 더 높았습니다(48.1% vs 31.4%, P = .0001). 그 외에는 UIM 학생과 비 UIM 학생의 인식에 차이가 없었습니다. 
One-third of students (33.6%) endorsed grading as biased. While more women perceived bias in evaluations than men (64.4% vs 25.2%, P < .0005), women also more commonly rated evaluations as accurate (69.2% vs 52.7%, P < .0005). There were no gender differences in perceptions of fairness of grading, feedback, or fairness of resident and attending evaluations. UIM students were more likely than non-UIM students to perceive bias in evaluations (48.1% vs 31.4%, P = .0001). Otherwise, UIM and non-UIM students’ perceptions did not differ (see Supplemental Digital Appendix 4 at https://links.lww.com/ACADMED/A720).

임상실습 학습 환경에 대한 인식
Perceptions of the clerkship learning environment

학생들은 임상실습 학습 환경이 숙달 및 성과 지향적이라고 압도적으로 지지했습니다(각각 88.0% 및 85.6%)(표 2). 임상실습 학습 환경이 성과 지향적이라고 답한 학생은 약간 더 적었습니다(68.9%). 성별이나 UIM 여부에 따른 임상실습의 숙련도 또는 성과 지향성에 대한 인식에는 하위 그룹 간 차이가 없었습니다. 
Students overwhelmingly endorsed the clerkship learning environment to be both mastery- and performance-avoid-oriented (88.0% and 85.6%, respectively) (Table 2). Slightly fewer students endorsed clerkships as performance-approach-oriented (68.9%). There were no subgroup differences in perceptions of the mastery or performance orientation of clerkships by gender or UIM status.

전체적으로 학생 응답의 18.3%가 인종에 따른 고정관념 위협에 취약하다고 답했습니다. 여성과 남성은 고정관념의 위협을 비슷하게 인식했습니다. UIM 학생은 그렇지 않은 학생보다 고정관념 위협에 취약하다고 응답한 비율이 훨씬 높았습니다(55.7% 대 10.9%, P < .0005)(보충 디지털 부록 4: https://links.lww.com/ACADMED/A720 참조). 
Overall, 18.3% of student responses indicated vulnerability to stereotype threat based on race. Women and men perceived stereotype threat similarly. UIM students were much more likely than non-UIM students to indicate vulnerability to stereotype threat (55.7% vs 10.9%, P < .0005) (see Supplemental Digital Appendix 4 at https://links.lww.com/ACADMED/A720).

우등상 수상에 따른 다변량 회귀 분석
Honors earned multivariable regression analysis

우등상은 더 [경쟁이 치열한 전문과에 지원]하는 것(베타 = 0.18, P < .0005)과 [평가가 더 정확하다고 인식]하는 것(베타 = 0.29, P < .0005)과 정(+)의 상관관계가 있는 것으로 나타났습니다(표 3). 명예 획득은 [고정관념 위협과 음의 상관관계]가 있었습니다(베타 = -0.162, P < .0005). 획득한 우등과 채점의 공정성에 대한 인식, 참석 또는 레지던트 평가 절차, 임상실습의 숙련도 또는 수행 환경에 대한 인식 간에는 유의미한 연관성이 없었습니다.  
Honors earned was positively associated with applying into a more competitive specialty (beta = 0.18, P < .0005) and perceiving evaluations as more accurate (beta = 0.29, P < .0005) (Table 3). Honors earned was negatively associated with stereotype threat (beta = −0.162, P < .0005). There were no significant associations between honors earned and perception of grading fairness, attending or resident evaluation procedures, or perceptions of mastery or performance environment of clerkships.

정성적 분석
Qualitative analysis

학생들의 의견은 [성적 부여, 평가 과정, 차등 채점의 원인이 되는 편향성, 학습자의 경험] 등 4가지 주제에 대해 다루었습니다(표 4).

  • [성적 부여]의 경우, 많은 응답자가 최종 성적에 기여하는 요소에 [가중치를 부여]하거나, [합격/불합격 채점]을 사용할 것을 권장했습니다(의견의 29.3%). 역량 기반 평가를 도입하거나 위임가능 전문활동 시스템을 사용하자는 의견도 있었습니다.
  • [평가 과정]에서 학생들은 평가자의 평가 지식학생 평가에 사용되는 프레임워크[variability]을 지적했습니다. 이들은 적절한 평가 기법에 대한 평가자 교육(30.6%)을 권장했습니다.
  • [차등 채점의 원인이 되는 편견]을 해결하기 위해 암묵적 편견 교육 또는 평가자를 비교하는 제도적 시스템을 통해 평가자의 개인적 편견(19.2%)을 해결해야 한다는 의견도 있었습니다.
  • [학습자의 경험]을 개선하기 위해 학생들은 보다 정기적이고 실행 가능한 피드백을 통해 학습을 지원하고(14.4%), 시간이 지남에 따라 추적하여 개선 사항을 평가하고 최종 성적에 반영하는 평가(11.6%)를 원했습니다. 

Students’ comments addressed 4 themes: grade assignment, evaluation process, bias causing differential grading, and learners’ experience (Table 4).

  • For grade assignment, many respondents recommended either reweighting components contributing to final grades or using pass/fail grading (29.3% of comments). Some recommended instituting competency-based assessment or using an entrustable professional activities system.
  • In the evaluation process, students noted variability in assessors’ knowledge of assessment and frameworks used to evaluate students. They recommended training evaluators on proper evaluation techniques (30.6%).
  • To address biases causing differential grading, some advocated addressing evaluators’ personal biases (19.2%) with implicit bias training or institutional systems to compare evaluators.
  • To improve learners’ experience, students wanted assessment to support learning through more regular and actionable feedback (14.4%), tracked over time so that improvement was valued and incorporated into final grades (11.6%).

토론
Discussion

여러 기관이 참여한 이 연구에서 핵심 임상실습 평가 및 채점의 공정성에 대한 학생들의 신뢰도가 낮은 것으로 나타났습니다.

  • UIM 학생의 절반 이상이 [고정관념에 의한 위협에 취약]하다고 답했으며, 이는 비 UIM 학생의 5배가 넘는 수치입니다.
  • 당연히 현재 환경에서 [가장 성공한 학생들, 즉 더 많은 우등상을 받은 학생들]이 평가의 정확성을 더 높게 평가하고, 경쟁이 치열한 전문 분야에 지원할 계획이며, 고정관념 위협에 덜 취약하다고 답한 것은 당연한 결과입니다.

학생들의 서술형 의견은 평가 및 채점에 대한 변화에 대한 학생들의 열망을 뒷받침했습니다.
This multi-institutional study reveals low student confidence in the fairness of core clerkship evaluations and grading. More than half of UIM students endorsed stereotype threat vulnerability, a prevalence greater than 5 times that of non-UIM students. Perhaps unsurprisingly, students who were most successful in the current environment, defined by earning more honors, endorsed greater accuracy of evaluations, planned to apply in competitive specialties, and were less vulnerable to stereotype threat. Students’ narrative comments supported their desire for changes to evaluation and grading.


[Grading에 대한 학생들의 인식]은 학습에 중요한 영향을 미치므로 반드시 해결해야 합니다. 연구 결과에 따르면 학생들은 자신의 성적을 결정하는 가장 강력한 요인을 임상 역량과 별개로 인식하고 있습니다. [낮은 성적을 받은 학생]은 [불공정한 시스템]이나 [특정 팀원의 편차] 등 자신의 [외적 요인]으로 성적을 돌릴 수 있습니다.43,44 이러한 시나리오는 자기 효능감을 위협하고 학생의 노력, 행동 및 향후 학습에 부정적인 영향을 미칠 수 있습니다.25,43 이러한 문제를 해결하기 위해 참가자들은 [평가자 교육을 더 많이 받아야 한다]고 주장했습니다. 평가자 교육은 학생의 성과를 공정하고 정확하게 평가하는 데 필요하지만, [특정 환자를 마주하는 상황과 초점, 평가자 자체]에는 [변동성이 내재]되어 있습니다.13,45 평가자 간의 완벽한 신뢰성을 위해 노력하기보다는, 평가 프로그램에서 평가 데이터를 수집하고 종합하는 엄격한 방법을 개발하는 것이 더 적절한 목표입니다.46 그러나 적절한 직접 관찰은 강력한 평가의 필수 구성 요소이기도 합니다. 학생들이 [전공의의 평가]를 [주치의의 평가]보다 더 호의적으로 본다는 연구 결과는 [전공의]가 환자와 함께 일하는 학생들과 더 많이 직접 접촉하기 때문에 설명될 수 있습니다. 감독자, 특히 주치의의 관찰 횟수를 늘리고 주치의 평가자에 대한 학생의 경험을 개선하기 위한 다른 메커니즘을 모색하면 평가의 공정성에 대한 학생의 인식을 개선할 수 있습니다.
Students’ perceptions of grading have important implications for learning that should be addressed. Our results show that students perceive the strongest determinants of their grades as distinct from their clinical competence. Students who receive lower grades may attribute their grades to factors extrinsic to themselves such as an unfair system or variance of particular team members.43,44 This scenario threatens self-efficacy and can negatively affect students’ effort, behaviors, and future learning.25,43 To address these challenges, our participants advocated for more evaluator training. While rater education is necessary for fair and accurate assessment of students’ performance, there is inherent variability in the context and focus of particular patient encounters and evaluators themselves.13,45 Rather than striving for perfect reliability among raters, a more appropriate goal would be to develop rigorous methods of collecting and synthesizing assessment data in a program of assessment.46 However, adequate direct observation is also a necessary constituent of robust assessment. Our finding that students view residents’ evaluations more favorably than attendings’ may be explained by residents’ greater direct contact with students working with patients. Increasing the number of observations from supervisors, in particular attending physicians, and exploring other mechanisms to improve students’ experience with attending evaluators could improve students’ perceptions of the fairness of evaluations.

우리의 데이터는 현재의 평가 시스템이 학습 또는 성과를 촉진하는지에 대한 의문을 제기합니다.47 학생들은 [성과에 높은 가치]를 부여하는 반면, [개선에 대한 가치는 낮게 평가한다]고 느꼈습니다. '우등' 성적이라는 외재적 동기는 성과 중심의 학습 환경을 조장할 수 있습니다. 반면, "학습을 위한 평가"[관찰을 통해 학습 결과를 평가]하고, [시기적절하고 구체적인 피드백을 제공]하여, [평가를 학생의 학습으로 전환]할 때 발생합니다.9 이 시나리오는 장기적인 성과와 학습의 즐거움을 향상시키는 [숙달 지향적 학습자]를 양성합니다.23 참가자들은 [Grading 방식의 성적을 없애거나 역량 기반 접근 방식으로 변경]하여 [임상실습 평가 구조를 재설계]해야 [숙달 사고방식과 평생 학습을 촉진할 수 있다]고 권고했습니다.48,49 현재 [레지던트 배치에 성적이 중요한 것]은 이미 높은 압박감이 있는 임상실습 환경을 심화시키고 있습니다. 의과대학은 레지던트 선발에 활용되기 때문에 단계별 임상실습 성적을 없애는 것을 주저할 수 있습니다. 본 연구의 범위를 벗어나기는 하지만, 계층형 임상실습 성적이 레지던트 기간 동안의 성과를 효과적으로 예측한다는 사실을 뒷받침하는 자료는 별로 없다.50 [레지던트 프로그램의 총체적 검토 접근법]은 학생의 평가 및 채점 부담을 줄이고 레지던트 선발에 유용한 정보를 제공할 수 있는 가능성을 제공합니다.51
Our data raise questions about whether the current assessment system promotes learning or performance.47 Students felt that performance was highly valued, while improvement was minimally valued. The extrinsic motivation of an “honors” grade may promote a performance-oriented learning environment. In contrast, “assessment for learning” occurs when observations are used to both assess learning outcomes and provide timely, specific feedback, thereby transforming assessment into student learning.9 This scenario cultivates mastery-oriented learners with improved long-term performance and enjoyment of learning.23 Our participants’ recommendations to redesign the clerkship assessment structure by eliminating tiered grades or changing to a competency-based approach could better promote a mastery mindset and lifelong learning.48,49 Currently, the importance of grades for residency placement intensifies an already-high-pressure clerkship environment. Medical schools may hesitate to eliminate tiered clerkship grades because of their use during resident selection. While beyond the scope of our study, minimal data support that tiered clerkship grades effectively predict performance during residency.50 Holistic review approaches by residency programs offer promise to reduce evaluation and grading pressures for students and provide residencies useful information for selection.51

[고정관념 위협 취약성]은 성과에 대한 유의미한 부정적 예측 요인으로 나타났으며, 주로 UIM 학생들에게 영향을 미쳤습니다. 고정관념 위협 취약성을 통제한 후에도 UIM 상태는 성과에 대한 유의미한 예측 변수가 아니었습니다. 
Stereotype threat vulnerability emerged as a significant negative predictor of performance, predominately affecting UIM students. UIM status was not a significant predictor of performance after controlling for stereotype threat vulnerability.

UIM 학생들이 직면한 문서화된 성적 편견 외에도, 이번 연구 결과는 고정관념 위협이 UIM 학생들의 학업 성취도를 더욱 저해할 수 있음을 뒷받침합니다.22,27 이 현상은 다른 곳에서 잘 설명되었음에도 불구하고 의대생들 사이에서는 조사되지 않았습니다. 의학교육에서 고정관념 위협의 범위와 의미를 이해하고 이에 대응하기 위한 개입을 설계하기 위해서는 더 많은 연구가 필요합니다. 고정관념 위협의 영향을 완화하기 위한 구체적인 전략으로는

  • (1) 커뮤니티에 고정관념 위협의 개념을 도입하고,
  • (2) 모든 커뮤니티 이해관계자를 참여시켜 정체성 안전을 증진하며,
  • (3) 리더를 고정관념의 영향을 받는 그룹에 노출을 늘리는 것 등이 있습니다.52

In addition to the documented grading biases facing UIM students, our findings support that stereotype threat may further undermine UIM students’ academic achievement.22,27 Despite being well described elsewhere, this phenomenon has not been explored amongst medical students. More work is needed to understand the scope and implications of stereotype threat in medical education and to design interventions to counteract it. Concrete strategies to mitigate the effects of stereotype threat include

  • (1) introducing the concept of stereotype threat to the community,
  • (2) engaging all community stakeholders to promote identity safety, and
  • (3) increasing exposure to leaders of the stereotyped group.52

이 연구에는 한계가 있습니다. 본 조사 결과는 임상실습 성적에 대한 학생의 관점을 포착한 것이므로 교육자의 의견은 다를 수 있습니다. 이 횡단면 설문조사는 인과관계를 보여주지 않습니다. 측정되지 않은 다른 요인들이 학생의 성과에 영향을 미칠 수 있습니다. 연구 대상 학교는 미국 내 한 지역에 위치하고 있어 다른 학교로 일반화할 수 없지만, 연구 집단은 인구통계학적으로 전국 학생과 유사했습니다. 우리는 PALS 교실 목표 구조와 SVS를 약간 수정했으며, 원래 척도가 서로 다른 집단에서 타당성을 보인다는 가정 하에 타당성을 가정했습니다. 설문조사 응답과의 상관관계를 파악하기 위해 성과 데이터를 수집하지 않았으며, 학생들의 전공 선호도는 시간이 지남에 따라 바뀔 수 있습니다. 마지막으로, 질적 결과는 학생들이 더 많은 질문을 통해 드러날 수 있는 임상실습 성적에 대한 추가 권장 사항이 있을 수 있고 모든 학생이 의견을 작성한 것은 아니므로 신중하게 해석해야 합니다.53
This study has limitations. Our results capture students’ perspectives on clerkship grading; educators’ opinions might differ. This cross-sectional survey does not show causation. Other unmeasured factors may contribute to student performance. Study schools are located in 1 U.S. region and may not generalize to other schools, although our study population was similar demographically to students nationally. We made small modifications to the PALS Classroom Goal Structures and SVS and assumed validity based on the original scales’ validity in distinct populations. We did not collect performance data to correlate with survey responses, and students’ specialty preferences may change over time. Finally, our qualitative results must be interpreted cautiously because students may have additional recommendations for clerkship grading that could have emerged with more questions, and not all students wrote comments.53

연구 결과에 따르면 많은 의대생들이 핵심 임상실습 기간 동안의 평가와 채점을 공정하다고 생각하지 않으며, 개선에 대한 보상보다는 성과를 장려하는 환경을 지지하는 것으로 나타났습니다. 평가와 채점에 대한 부정적인 인식은 학업 성취도 저하와 관련이 있습니다. UIM 학생은 임상실습 환경에서 추가적인 불리한 압력에 직면할 수 있습니다. 공정한 평가 시스템에는 평등과 형평성을 증진하는 정책과 절차가 필요합니다.54 본 모델(그림 1)에서 가설로 설정한 많은 기여 요인들이 학생 성과와 연관성을 보이지 않았지만, 이러한 영역에서의 차별적 인식은 학습 행동의 변화 또는 학생 복지와 같은 다른 영향을 미칠 수 있습니다.55,56 이러한 결과는 강력한 평가를 촉진할 뿐만 아니라 모든 학생의 학습을 가능하게 하는 학습 환경을 조성하기 위해 핵심 임상실습의 평가 문화를 재정의할 필요가 있다는 것을 뒷받침합니다.
Our findings demonstrate that many medical students do not view evaluation and grading during core clerkships as fair, and they endorse an environment that encourages performance rather than rewards improvement. Negative perceptions of evaluation and grading are associated with decreased academic achievement. UIM students may face additional adverse pressures in the clerkship environment. A fair assessment system requires policies and procedures that promote equality and equity.54 While many of the contributors hypothesized in our model (Figure 1) did not show associations with student performance, differential perceptions in these domains may have other effects such as changes in learning behaviors or student well-being.55,56 These results support a need to redefine the culture of assessment on core clerkships to create learning environments that not only facilitate robust assessment but also enable learning for all students.


 

Acad Med. 2019 Nov;94(11S Association of American Medical Colleges Learn Serve Lead: Proceedings of the 58th Annual Research in Medical Education Sessions):S48-S56. doi: 10.1097/ACM.0000000000002905.

In Pursuit of Honors: A Multi-Institutional Study of Students' Perceptions of Clerkship Evaluation and Grading

Affiliations collapse

1J.L. Bullock is a first-year resident in internal medicine, Department of Medicine, University of California, San Francisco School of Medicine, San Francisco, California. The author was a fourth-year medical student at the time of writing. C.J. Lai is director of internal medicine clerkships and professor, Department of Medicine, University of California, San Francisco School of Medicine, San Francisco, California. T. Lockspeiser is director of the assessment/competency committee and associate professor, Department of Pediatrics, University of Colorado School of Medicine, Aurora, Colorado. P.S. O'Sullivan is director of research and development in medical education and professor, Department of Medicine and Department of Surgery, University of California, San Francisco School of Medicine, San Francisco, California. P. Aronowitz is clerkship director of internal medicine and professor, Department of Internal Medicine, University of California, Davis School of Medicine, Davis, California. D. Dellmore is director of medical student education and associate professor, Department of Psychiatry and Behavioral Sciences, University of New Mexico School of Medicine, Albuquerque, New Mexico. C.-C. Fung is assistant dean for medical education and associate professor, Keck School of Medicine of USC, Los Angeles, California. C. Knight is associate clerkship director and associate professor, Division of General Internal Medicine, University of Washington School of Medicine, Seattle, Washington. K.E. Hauer is associate dean for competency assessment and professional standards and professor, Department of Medicine, University of California, San Francisco School of Medicine, San Francisco, California.

PMID: 31365406

DOI: 10.1097/ACM.0000000000002905

Abstract

Purpose: To examine medical students' perceptions of the fairness and accuracy of core clerkship assessment, the clerkship learning environment, and contributors to students' achievement.

Method: Fourth-year medical students at 6 institutions completed a survey in 2018 assessing perceptions of the fairness and accuracy of clerkship evaluation and grading, the learning environment including clerkship goal structures (mastery- or performance-oriented), racial/ethnic stereotype threat, and student performance (honors earned). Factor analysis of 5-point Likert items (1 = strongly disagree, 5 = strongly agree) provided scale scores of perceptions. Using multivariable regression, investigators examined predictors of honors earned. Qualitative content analysis of responses to an open-ended question yielded students' recommendations to improve clerkship grading.

Results: Overall response rate was 71.1% (666/937). Students believed that being liked and particular supervisors most influenced final grades. Only 44.4% agreed that grading was fair. Students felt the clerkship learning environment promoted both mastery and performance avoidance behaviors (88.0% and 85.6%, respectively). Students from backgrounds underrepresented in medicine were more likely to experience stereotype threat vulnerability (55.7% vs 10.9%, P < .0005). Honors earned was positively associated with perceived accuracy of grading and interest in competitive specialties while negatively associated with stereotype threat. Students recommended strategies to improve clerkship grading: eliminating honors, training evaluators, and rewarding improvement on clerkships.

Conclusions: Participants had concerns around the fairness and accuracy of clerkship evaluation and grading and potential bias. Students expressed a need to redefine the culture of assessment on core clerkships to create more favorable learning environments for all students.

 

보건의료에서 목소리 높히기의 최신지견 문헌고찰(Adv Health Sci Educ Theory Pract. 2022)
A state‑of‑the‑art review of speaking up in healthcare
Efrem Violato1

서론
Introduction

교육을 포함한 의료 분야에서는 [목소리를 높이는 것]을 지원하고 사람들이 [목소리를 높이는 데] 참여하도록 훈련할 책임과 필요가 있다. [목소리를 높이는 것]은 [다른 사람들이 환자 관리 및 안전과 관련된 우려를 인식하도록 하기 위한 우려 또는 다른 의견을 말하는 것]입니다. 현재 의료 시스템 전반에 걸쳐 [많은 실무자]에게 [환자 안전, 관리 및 위험에 대한 우려를 제기하는 방법]에 대한 인식된 능력(지식과 기술과 함께)은 충분하지 못하다. 목소리를 낼 수 없거나 목소리를 낼 수 없는 것은 환자의 안전, 교육, 전문성 발달 및 전문적 정체성에 부정적인 영향을 미칠 수 있다. 개인적, 대인적, 전문적, 그리고 조직적인 요인을 포함하여, 말을 하는 것을 어렵게 하고 심지어 직업적으로 그리고 개인적으로 위험하게 만드는 수많은 도전들이 존재한다. 일레인 브로밀리는 그러한 요인들의 합체 때문에 죽었다. 여러 자격을 갖춘 전문가들이 방안에 모여 '삽관이 산소를 공급할 수 없는' 상황을 겪은 후 [건강한 사람이 사망한 것]은 [사람들이 목소리를 내지 않거나 내지 못할 때 일어날 수 있는 일]을 상징적으로 보여준다. 대조적으로 팀이 약물과 전달 방법을 선택하도록 유도하는 것과 같이 직접적이고 지속적으로 목소리를 높이는 것은 잠재적으로 치명적인 오류를 피할 수 있다. 
In healthcare, including education, there is a responsibility and need to support speaking up and train people to engage in speaking up (Milligan et al., 2017). Speaking up is voicing concern or a differing opinion to make others aware of a concern related to patient care and safety (Okuyama et al., 2014). Currently, for many practitioners across healthcare systems the perceived ability along with knowledge and skills of how to raise concern about patient safety, care, and risk is suboptimal (Jones et al., 2021; Kim et al., 2020; Milligan et al., 2017; Morrow et al., 2016). The absence of, or inability to, speak up can negatively impact patient safety, education, professional development, and professional identity (Bickhoff et al., 2017). Numerous challenges exist that make speaking up difficult and even professionally and personally risky including individual, interpersonal, professional, and organizational factors (Jones et al., 2021; Kim et al., 2020). Elaine Bromiley died because of the confluence of such factors. The death of a healthy person after a ‘can’t intubate can’t oxygenate’ situation with multiple qualified professionals in the room is emblematic of what can happen when people do not or can not speak up (Harmer, 2005). In contrast, directly and persistently speaking up, such as by questioning a team leads choice of drug and delivery method, can avoid potentially fatal errors (Calhoun et al., 2014).

한동안 존재했지만, [목소리를 높이는 것]에 대한 세계적으로 인정된 도전이 지난 10년 동안 표면화되었다. 문제에 대한 접근 방식은 다양한 관점을 통합했으며 그 결과 지식을 수집하고 문헌을 요약하는 다양한 형식의 리뷰 기사가 있었다. 
While extant for some time, the globally recognized challenge of speaking up has come to the fore in the last ten years (Jones et al., 2021; Peadon et al., 2020). Approaches to the problem have incorporated different perspectives (Holmes et al., 2014; Kohn et al., 1999) and as a result there have been multiple review articles of different formats compiling knowledge and summarizing the literature.

리뷰 기사의 발표로, 말하기에 대한 문헌과 지식의 요약은 분류되고 다른 지식 흐름에 걸쳐 확산될 수 있다. 이미 목소리를 높이는 연구는 기존 증거에 대한 연속성과 연관성이 부족하며 새로운 연구는 이전의 결함을 수정하거나 이전의 발견을 기반으로 하지 않는다. 의료 분야에서 목소리를 높이는 연구는 비교적 초기 단계이기 때문에, 발전을 촉진하고 확산을 방지하기 위해 기존의 지식을 하나의 포괄적인 소스로 통합하는 것이 필수적이다. 이 기사는 목소리를 높이는 연구를 계획하는 연구자, 목소리를 높이는 방법에 대한 정책 입안자, 목소리를 높이는 교육자를 돕기 위해 현재의 "최첨단"을 대표하는 통일된 진술서를 만들고자 한다. 이 목적을 해결하기 위해 다음과 같은 유도 연구 질문이 개발되었다: 
With the promulgation of review articles, summations of the literature and knowledge about speaking up may become fractionated and diffuse across different knowledge streams. Already, research on speaking up lacks continuity and connection to existing evidence and new research does not correct prior flaws or build on previous findings (Jones et al., 2021). Research on speaking up in healthcare is relatively nascent so it is imperative to bring together existing knowledge in a single comprehensive source to promote forward development and prevent diffusion. The present article seeks to create a unified statement representing the current "State of the Art" for speaking up to assist researchers planning studies on speaking up, policymakers deciding on how to improve speaking up, and educators developing curriculum and course material for speaking up. To address this purpose, a guiding research question was developed:

말을 하는 것에 관한 문헌 전반에 걸쳐 공통적인 발견을 식별하고 합성하여, 의료 분야에서 [목소리를 높이는 것]에 대한 연구와 지식의 현재 상태를 통일적으로 이해할 수 있는가?
Can common findings across the literature on speaking up be identified and synthesized to produce a unified understanding of the current state of research and knowledge about speaking up in healthcare?

방법들
Methods

최신 리뷰를 통합하여 향후 조사를 위한 현재 지식과 우선순위에 대한 종합적인 앤솔로지anthology를 효율적으로 구축하는 방식으로 최첨단 검토를 수행하였다. 최첨단 리뷰는 일차 연구의 공식적인 품질 평가를 포함하지 않고 더 높은 수준의 합성을 통해 지식을 병합하기 위한 리뷰를 수집하는 데 초점을 맞춘다. 검토 형식의 다양성은 이차 수준의 공식 평가의 가능성을 배제한다.
A State-of-the-Art review was conducted by amalgamating recent reviews to efficiently create a comprehensive anthology of the current knowledge and priorities for future investigation (Grant & Booth, 2009). A State-of-the-Art review does not include formal quality assessment of primary research but focuses on collating reviews to merge knowledge through higher-level synthesis. The diversity of review formats precludes the possibility of secondary level formal assessment (Fusar-Poli & Radua, 2018; Grant & Booth, 2009).

검색 전략
Search strategy

문헌은 MEDLINE, CINHAL, Scopus, Google Scholar, PsycINFO, ERIC를 사용하여 소스되었다. 수색은 2020년 10월에 처음 실시되었고 2021년 2월과 9월에 다시 실시되었다. 표 1은 검색어와 결과를 보여줍니다. 말하기 및 기타 출판물에 대한 연구를 위한 참조 목록을 손으로 검색했습니다. 
Literature was sourced using MEDLINE, CINHAL, Scopus, Google Scholar, PsycINFO, and ERIC. The search was initially conducted in October 2020 and again in February and September 2021, with no year limits. Table 1 shows search terms and results. Reference lists for studies on speaking up and other publications were hand-searched (e.g., Mannion et al., 2018).

추출.
Extraction

검색 및 검토 프로세스는 체계적 검토 및 메타 분석(PRSMA)을 위한 선호 보고 항목 형식을 따랐다(그림 1). 중복제거 후 주 조사관이 보다 면밀하게 검토할 수 있도록 제목과 요약을 포함하도록 선별했다.

  • 포함 기준: 목소리를 높이고, 안전한 목소리를 내고, 권위에 도전하는 것에 관한 기사를 검토한다; 모든 검토 유형 및 방법.
  • 제외 기준 : 비검토 기사, 예를 들어, 1차 소스 연구, 내부 고발에 대한 문헌, 일반적인 환자 옹호 및 환자 안전과 관련 없는 측면.

추가 기록은 전문가 간 교육 및 협업을 위한 자료와 팀 커뮤니케이션과 같은 다른 출처의 직접 검색을 통해 확인되었다. 

The search and review process followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) format (Fig. 1) (Moher et al., 2009). After de-duplication, titles and abstracts were screened for inclusion for closer review by the principal investigator. 

  • Inclusion criteria: review articles on speaking up, safety voice, and challenging authority; all review types and methods. 
  • Exclusion criteria: non-review articles, e.g., primary source research, literature on whistleblowing, patient advocacy in general, and unrelated aspects of patient safety.

Additional records were identified through a hand search of other sources such as resources for interprofessional education and collaboration and team communication.

분석.
Analysis

연구 질문, 결과, 한계 및 결론과 같은 공통점을 식별하는 데 초점을 맞춘 상수 비교 분석(American Psychological Association, n.d.)의 반복 과정을 사용하였다. PI는 모든 기사를 읽고 요약하여 내용에 대한 개요와 일반적인 이해를 발전시켰다. 그런 다음 기사를 다시 읽고 각 검토의 결과를 추출했다. 연구 결과가 나오자 최종 합성이 가능할 때까지 모든 논문을 검토하여 새로운 연구 결과와의 일치성을 확인했다.  
An iterative process of constant comparative analysis (American Psychological Association, n.d.), focused on identifying commonalities such as research questions, outcomes, limitations, and conclusions was used. All articles were read and summarized by the PI to develop a compendium and general understanding of the content. Articles were then re-read, and the findings of each review were extracted. As findings emerged, all articles were reviewed to determine concordance with the emergent findings until a final synthesis was possible.

결과.
Results

14개의 리뷰 기사를 분석하였다(표 2). 14개의 리뷰는 양적, 질적, 혼합적 방법 연구, 논평, 이론 논문 등 335개의 기사를 다루었다. 335개의 기사 중 51개는 150개의 복제 기사와 236개의 고유 기사에 대한 다중 리뷰를 포함했다(모든 기사에 대해서는 보충 자료 참조). 가장 자주 검토된 논문은 Pian-Smith 외 연구진(2009)(검토 8배), Raemer 외 연구진(2016)(검토 7배), Schwapach and Gerrick(2014)(검토 6배), Johnson and Kimsey(2012)(검토 6배), Sayre 외 연구진(2012)(검토 6배), Barzo Salazar et al(2014)(검토 5배)이다. 이 기사 세트는 반복적으로 인용되는 기사의 29%를 차지했다. 
Fourteen review articles were analyzed (Table 2). The 14 reviews covered 335 articles, including quantitative, qualitative, mixed methods research, commentaries, and theoretical papers. Fifty-one of the 335 articles included multiple reviews, for 150 replications and 236 unique articles (for all articles, see the supplemental material). The articles most frequently reviewed were: Pian-Smith et al (2009) (reviewed 8×), Raemer et al (2016) (reviewed 7×), Schwappach and Gehrig (2014) (reviewed 6×), Johnson and Kimsey (2012) (reviewed 6×), O’Connor et al. (2013) (reviewed 6×), Sayre et al (2012) (reviewed 6×), Barzallo Salazar et al. (2014) (reviewed 5×). This set of articles accounted for 29% of repeatedly cited articles.

주요 소스는 여러 글로벌 지역의 데이터를 나타냅니다: 아시아(홍콩, 일본, 한국, 네팔, 대만), 아프리카(동아프리카, 케냐), 유럽(불가리아, 키프로스, 핀란드, 아일랜드, 이탈리아, 스웨덴, 스위스), 중동(이란, 이스라엘)이다. 가장 빈번하게 제공된 기사는 호주와 뉴질랜드, 캐나다, 미국, 영국에서 왔다. 대부분의 문헌은 학생들을 포함한 의사와 간호사들에 초점을 맞추었다. 소수의 문헌은 연합된 보건 전문가, 행정, 약사, 환자를 조사했다(보조 자료 참조). 
The primary sources represented data from multiple global regions: Asia (Hong Kong, Japan, Korea, Nepal, Taiwan), Africa (East Africa, Kenya), Europe (Bulgaria, Cyprus, Finland, Ireland, Italy, Sweden, Switzerland), and the Middle East (Iran, Israel). The articles most frequently sourced were from Australia and New Zealand, Canada, the United States, and the United Kingdom. Most of the literature focused on physicians and nurses, including students. A minority of the literature examined allied health professionals, administration, pharmacists, and patients (see supplemental material).

현상의 정의
Definition of the phenomena

모든 기사들은 [목소리를 높이는 것]을 정의했다. Okuyama 외 연구진(2014)은 Leonard 외 연구진(2004)과 Lyndon 외 연구진(2012)에서 파생된 간결하고 포괄적인 정의를 제공했다: "목소리 높이기"는 병원 환경에서 [의료 팀 내의 다른 사람들의 위험하거나 부족한 행동]을 인식하거나 인지할 때, 환자 안전과 의료 품질의 이점에 대한 의료 전문가들의 [우려를 제기]하는 것으로 정의된다
All articles defined speaking up similarly. Okuyama et al. (2014) provided a concise and comprehensive definition, derived from Leonard et al. (2004) and Lyndon et al. (2012), that encompassed the definitions used by all the reviews: “Speaking up' is defined as the raising of concerns by healthcare professionals for the benefit of patient safety and care quality upon recognising or becoming aware of the risky or deficient actions of others within healthcare teams in a hospital environment.”

14개 리뷰에 걸쳐 목소리를 높이기 위한 세 가지 맥락이 조사되었다: 

  • (1) 훈련을 포함한 팀 상황; 
  • (2) 의료 계층 내의 일반적인 맥락 
  • (3) 주로 간호학과 학생들에 초점을 맞춘 전문 그룹들 사이의 차이. 

모리슨의 '직원 목소리' 모델을 변형한 위험한 상황에서 목소리를 높이는 행위인 '안전 목소리'는 대중적인 해석 프레임워크였다.

  • [직원 목소리]더 높은 조직 위치에 있는 누군가에게 관심, 제안, 정보 또는 업무 관련 의견을 자발적으로 전달하는 것입니다.
  • [안전 목소리]은 개념적으로 직원 목소리와 유사하지만, 내용과 메시지가 안전에 초점을 맞추고, 맥락이 안전-중시 조직과 관련되어 있으며, 조치가 법적으로 요구될 수 있으며, 조치가 조직(예: 내부 고발)을 넘어 확장될 수 있다는 점에서 다르다.

Across the 14 reviews, three contexts for speaking up were examined:

  • (1) team situations, including training;
  • (2) general contexts within the hierarchies of healthcare; and
  • (3) differences amongst professional groups, focused primarily on nursing and students.

Safety voice’, the act of speaking up in hazardous situations adapted from Morrison's model of ‘employee voice’ (Morrison, 2011; Noort et al., 2019) was a popular interpretive framework.

  • Employee voice is voluntary communication of concerns, suggestions, information, or work-related opinions to someone in a higher organizational position (Morrison, 2014).
  • Safety voice is conceptually similar to employee voice though differs in that the content and message are focused on safety, the context relates to safety–critical organizations, action may be legally required, and the action can extend beyond the organization e.g., whistleblowing (Noort et al., 2019).

용어 및 언어
Terms and language

도전적인 권위를 정의하고 논의하기 위한 공통 언어는 다음을 포함한다: "Speaking Up", "Hierarchies", "Psychological Safety", "Safety Voice", "Challengeing Authority", "Communication"  
Common language to define and discuss challenging authority, included:

  • "Speaking Up,"
  • "Hierarchies,"
  • "Psychological Safety,"
  • "Safety Voice,"
  • "Challenging Authority," and
  • "Communication." 

'엘리트에 의한 불경', '돌봄 전문가의 춤', '순환 위계', '불도덕성', '청렴의 의무', '수평적 폭력', '불건전한 순종' 등 다양한 용어가 등장했다
Variable language and unique terms arose including

  • "Incivility by Elites” (Morrow et al., 2016),
  • "Dance of Caring Professionals (Morrow et al., 2016),"
  • "Circular Hierarchy (Morrow et al., 2016),"
  • "Immorality (Okuyama et al., 2014),"
  • "Duty of Candor (Milligan et al., 2017),"
  • "Lateral [Horizontal] Violence (Milligan et al., 2017)," and
  • "Unwholesome Submissiveness (Peadon et al., 2020)."

 

연구 질문 및 검토 결과
Research questions and findings of the reviews

각각의 리뷰는 독특한 연구 질문으로 문헌에 대한 독특한 조사를 제안했다. 모든 검토에서 조사 결과를 정리할 수 있는 5가지 공통 질문이 존재했다: 

  • (1) 어떤 개입/훈련이 사용되었으며 얼마나 효과적입니까?; 
  • (2) 어떤 메커니즘이 목소리를 높이는 데 영향을 미칩니까? ; 
  • (3) 목소리를 높이는 것에 대한 사람들의 인식과 경험은 무엇인가? ; 
  • (4) 사람들은 어떻게 말합니까?; 
  • (5) 목소리를 높이는 것이 얼마나 효과적인가?

Each review proposed to be a unique investigation of the literature with unique research questions. Across all reviews five common questions existed under which findings can be organized:

  • (1) what interventions/training have been used and how effective are they? (Jones et al., 2021; Kim et al., 2020; Milligan et al., 2017; Nacioglu, 2016; O’Donovan & McAuliffe, 2020; Okuyama et al., 2014; Omura et al., 2017);
  • (2) what mechanisms (inhibitors/enablers) influence speaking up? (Bickhoff et al., 2017; Nacioglu, 2016; Okuyama et al., 2014; Peadon et al., 2020);
  • (3) what are people's perceptions and experiences of speaking up? (Fagan et al., 2016; Morrow et al., 2016; Noort et al., 2019; Pattni et al., 2019);
  • (4) how do people speak up? (Bickhoff et al., 2017; Morrow et al., 2016; Pattni et al., 2019; Rutherford et al., 2012);
  • (5) how effective is speaking up? (Milligan et al., 2017; Nacioglu, 2016; Okuyama et al., 2014).

어떤 개입/훈련이 사용되었으며, 얼마나 효과적입니까?
What interventions/training have been used, and how effective are they?

[목소리 높이기]는 보건 전문가 교육의 필수적인 부분으로 간주되지만 말하기/주장 훈련/갈등 훈련에 대한 커리큘럼과 체계적인 훈련이 부족하여 교육 격차가 존재한다. 5가지 형태가 확인되었다.

  • 교육 및 시뮬레이션,
  • 교육 및 리더십 비디오,
  • 교육 및 사례 연구,
  • 교육 워크샵
  • 비교육적 개입 (포럼 놀이, 전체론적 촉진 및 행동 연구 회의로 구성)


Speaking up is considered an essential part of health professionals' education, yet an educational gap exists with a lack of curriculum and systematic training on speaking up/assertiveness training/conflict training (O’Donovan & McAuliffe, 2020; Omura et al., 2017; Pattni et al., 2019). Five forms of intervention have been identified:

  • education and simulation,
  • education and leadership videos,
  • education and case studies,
  • educational workshops, and
  • non-educational interventions, comprising forum play, holistic facilitation, and action research meetings (Jones et al., 2021; Kim et al., 2020; O’Donovan & McAuliffe, 2020; Okuyama et al., 2014; Omura et al., 2017).

목소리를 높이기 위한 훈련과 개입에 대한 연구는 비교적 새로운 것으로 [효능 증거는 제한적이다]. 일반적으로, 목소리를 높이기 위한 교육적 개입은 [태도와 자신감]에 대한 짧은 지속적인 긍정적 영향을, [행동]에 대한 영향은 미미하거나 결여되어 있다. [팀 훈련]과 [적극적인 의사소통]에 대한 개입은 어느 정도 효과적이지만, [코칭 기반의 훈련]은 효과적이지 않은 것으로 보인다. 항공, 특히 [CRM(Crew/Crisis Resource Management)]에서 적용된 팀 기술 훈련 및 커뮤니케이션 기법이 긍정적으로 평가되었으며, 태도, 자신감 및 행동에 가장 강력한 긍정적 효과를 보였다. 개입의 구체적인 특징은 일관된 긍정적 효과를 보여주지 않는다.
Research on training and interventions for speaking up is relatively new with limited efficacy evidence. Generally, educational interventions for speaking up have weak short lasting positive effects on attitudes and confidence with effects on behaviour being minimal or absent. Interventions for team training and assertive communication are somewhat effective, coaching based training does not appear to be effective. Team skills training and communication techniques adapted from aviation, specifically Crew/Crisis Resource Management (CRM) were positively evaluated, and showed the strongest positive effects on attitudes, confidence, and behaviour (Fagan et al., 2016; Jones et al., 2021; Kim et al., 2020; Milligan et al., 2017; Nacioglu, 2016; O’Donovan & McAuliffe, 2020; Okuyama et al., 2014; Omura et al., 2017; Pattni et al., 2019; Peadon et al., 2020). No specific characteristics of interventions show consistent positive effects.

어떤 메커니즘(활성화/억제제)이 목소리를 높이는 데 영향을 미칩니까?
What mechanisms (enablers/inhibitors) influence speaking up?

많은 선행 사례들이 목소리를 높이는 것을 억제하거나 가능하게 하는 것으로 확인되었지만, [소수의 메커니즘]만이 억제, 가능 또는 둘 다로 일관되게 확인되었다.

Numerous antecedents have been identified as inhibiting or enabling speaking up, though only a few mechanisms were consistently identified as inhibitory, enabling, or both (Bickhoff et al., 2017; Fagan et al., 2016; Jones et al., 2021; Kim et al., 2020; Morrow et al., 2016; Noort et al., 2019; Omura et al., 2017; Pattni et al., 2019; Peadon et al., 2020).

목소리를 높이는 것을 방해하는 것
Inhibitors of speaking up

확인된 [주요 억제 메커니즘]은 [위계 구조]와 [권력 역학]이었지만, 실험 조건에서 위계 구조를 평탄화하는 것이 지속적으로 목소리 높이기를 개선하지는 않았다. [위계 구조]는 간호사와 학생 간호사 사이와 같은 [Intraprofessional]한 것과, 마취사와 호흡기 치료사 사이와 같은 [Interprofessional]한 것이 있다. [Inter 및 Intra professional한 위계]의 장벽과 수행자 직업 문화는 종종 [하위 계층, 특히 학생들에 의해 극복될 수 없는 것]으로 여겨지나, [고위 의사를 포함하여 더 상급 위치에 있는 사람들에게도 극복될 수 없는 것]으로 여겨졌다. 위계와 문화가 [극복할 수 없을 정도로 명백한 것]은 종종 목소리를 높이려고 노력하는 것조차 허무감을 만들어낸다.

  • [위계구조]와 [권력 역학]은 역할, 책임, 직업적 고정관념에 대한 기대감을 조성하고 뒷받침함으로써 목소리를 높이는 것을 제약하는 것으로 확인되었다.
  • [고정관념]은 목소리를 높이는 것에 대한 수용, 목소리를 높이는 인식 능력, 목소리를 높이는 것에 대한 책임감을 감소시키고 계층의 제약에 도전하는 것을 피하게 할 수 있다.

The primary inhibitory mechanism identified was hierarchies and power dynamics, though in experimental conditions flattening hierarchies does not consistently improve speaking up (Bould et al., 2015; Friedman et al., 2015; Sydor et al., 2013). Hierarchies are both

  • intraprofessional, such as between a charge nurse and a student nurse, and
  • interprofessional such as between an anesthetist and a respiratory therapist.

Inter and intra professional hierarchical barriers and attendant professional cultures are often seen to be insurmountable by those lower in hierarchies, in particular students, but also by those in more advanced positions including senior medical doctors (Peadon et al., 2020). The apparent insurmountably of hierarchy and culture often create a sense of futility in even trying to speak up (Morrow et al., 2016).

  • Hierarchies and power dynamics were identified as constraining speaking up by creating and supporting expectations about roles, responsibilities, and professional stereotypes.
  • Stereotypes can inhibit acceptance of speaking up, perceived ability to speak up, diminish responsibility for speaking up, and lead to avoidance of challenging the constraints of hierarchy.

 

  • [심리적 안전성]의 부족은, 인식이든 실제이든, [보복, 갈등, 괴롭힘, 괴롭힘 등 부정적인 결과에 대한 두려움] 및 [다른 사람들의 능력에 의문을 제기하는 것에 대한 우려]를 유발하여 목소리를 높이는 것을 억제한다.
  • [부정적인 결과에 대한 두려움]은 상황에 따라 달라질 수 있다;
    • 학생은 부정적인 평가를 두려워할 수 있는 반면,
    • 상급 전문가는 동료의 능력에 의문을 제기하는 것처럼 보임으로써 동료의 기분을 상하게 하고 싶지 않을 수 있다.
  • [일관성 없는 언어 및 훈련], 그리고 [목소리 높이기의 구성요소에 대한 이해의 불화]도 억제적이다.
  • A lack of psychological safety, perceived or actual, inhibits speaking up through fear of negative consequences, including reprisal, conflict, bullying and harassment, and concerns about questioning others' competency.
  • Fear of negative consequences can be context-dependent;
    • a student may fear a negative evaluation,
    • while a senior professional may not want to offend a colleague by appearing to question the colleague's competency.
  • Inconsistent language, training, and discordant understandings about what constitutes speaking up are inhibitory.
목소리를 높일 수 있는 원동력
Enablers of speaking up

["안전한 목소리"]를 갖는 것은 목소리를 높이는 것을 가능하게 한다; 그러나 안전 음성의 의미는 "분해되고 초기disintegrated and nascent"이다. 가장 일관성 있게 파악된 지지 요인은 [관리적 및 행정적 지원을 통한 조직적 지원]이었습니다.

  • [조직적 지원]에는 리더십 직책에 있는 사람들이 모델, 멘토링, 격려 및 목소리를 높이는 행동을 받아들이는 것이 포함됩니다.
  • [시스템 지원]은 주로 접근 가능한 보고 시스템과 "내부고발"에 대한 지원으로 식별되었다.

[서포트가 필요하다는 인식]은 [조직적 지원을 만들어내는 데] 필요하다. 이 때 [하향식 지원]을 동반함으로써, [목소리를 낼 수 있다고 느끼게 임파워링empowerng]하면서 [심리적으로 안전한 조직이나 환경]을 만들어낼 수 있다. 
Having "safety voice" is an enabler of speaking up; however, the meaning of safety voice is "disintegrated and nascent" (Noort et al., 2019). The most consistently identified enabler was organizational support through managerial and administrative support.

  • Organizational support includes those in leadership positions modeling, mentoring, encouraging, and accepting speaking up behaviour.
  • System supports were mainly identified as accessible reporting systems and support for “whistleblowing” (Fagan et al., 2016).

Awareness of the need to provide support is required to create organizational support, with top-down support necessary to create an empowering and psychologically safe organization or environment where a person feels it is possible to have a voice (Morrow et al., 2016; O’Donovan & McAuliffe, 2020).

  • [목소리를 높이는 것에 대한 지원]은 또한 환자 중심성, 환자 안전 및 전문적인 원칙 보호와 같은 [전문직업적 의무(예: 환자 옹호자가 되는 것)]로부터 온다.
  • [임상 및 안전 관련 경험과 지식]은 모두 도움이 됩니다.
  • 목소리를 높이는 부정적인 방향의 enabler는 "감정적 빌드업"이다; 사람들은 목소리를 내지 않는 것이 불가능하다는 것을 발견하는 "감정적 빌드업"의 수준에 도달하곤 한다.
  • Support for speaking up also comes from professional mandates such as patient-centeredness, patient safety, and safeguarding professional principles, e.g., being a patient advocate (Bickhoff et al., 2017).
  • Experience and knowledge, both clinical and safety-related, are supportive (Fagan et al., 2016; Okuyama et al., 2014).
  • A negatively oriented enabler of speaking up is “emotional buildup”; people reach a level of “emotional buildup” where they find it impossible not to speak up (Morrow et al., 2016).
목소리를 높이는 것을 억제하는 것/가능하게 하는 것
Inhibitor/enabler of speaking up
  • [위계 구조]는 주요 억제 메커니즘이지만, [상급 위치에 있는 사람들이 목소리를 높이고, 조언하고, 가능하게 하고, 받아들일 때], 목소리를 높이는 것을 긍정적으로 지원할 수 있는 의료의 필수적인 측면으로 간주된다;
    • 예를 들어, 개별 [간호사 관리자]들은 학생들이나 그들 이하의 다른 사람들이 목소리를 높이는지 여부에 긍정적 또는 부정적인 영향을 미칠 수 있다.
  • [상호 존중의 긍정적인 멘토-멘티 관계]의 경우, 너무 많은 존중이 의사소통을 방해하는 반면, 부정적인 관계는 의사소통을 방해할 수 있지만, 목소리를 높이는 것을 지지할 수 있다.
  • [개인적 요소](문화적, 세대적 배경, 태도, 성격, 성별, 경험, 자신감 등)도 억제하거나 가능하게 할 수 있다. 이러한 개별 특성이 어떻게 기능하는지는 상황에 따라 달라질 수 있습니다.
    • [학생]은 조직적인 지원, 임상 지식, 안전에 대한 헌신, 그리고 목소리를 낼 수 있는 자신감을 갖기 위한 [기술과 도구가] 필요할 수 있다.
    • 반면 [경험이 많은 동료]는 조직 지원이나 의견 표명을 위한 [도구의 필요성을 느끼지 못할 수도] 있습니다.
  • While being the primary inhibitory mechanism, hierarchy is also seen as a necessary aspect of healthcare that can positively support speaking up when those in advanced positions model, mentor, enable and accept speaking up; e.g., individual nurse managers can have a positive or negative effect on whether students or others below them speak up (Morrow et al., 2016).
  • A positive mentor–mentee relationship of mutual respect can support speaking up, though too much respect may prevent speaking up, while a negative relationship impedes communication.
  • Individual factors like cultural and generational background, attitudes, personality, sex, experience, and confidence could be inhibitory or enabling. How these individual characteristics function is developing and can vary based on context.
    • A student may need organizational support, clinical knowledge, a commitment to safety, and the skills and tools for speaking up to have the confidence to speak up.
    • Simultaneously, a more experienced colleague may not feel a need for organizational support or tools for speaking up.

관련성은 불분명하지만 [문화적 고려]가 중요하다.

  • [위계적 구배]가 의료 문화에 너무 깊이 뿌리박혀 있어서, 그 효과가 국가 문화적 영향을 대체할 수도 있지만,
  • 반대로 [국가 문화]가 너무 강해서 직업적 변화와 개입을 무시할 수도 있다.

Cultural considerations are important though the relevance is unclear (Jones et al., 2021).

  • Hierarchy gradients may be so ingrained in medical culture that the effects supersede national cultural influences (Kobayashi et al., 2006),
  • or alternatively, national culture is so strong it overrides professional changes and intervention (Roh et al., 2015).

목소리를 높이는 것에 대한 사람들의 인식과 경험은 무엇인가?
What are people's perceptions and experiences of speaking up?

전반적으로, [목소리 높이기]에 대한 인식과 경험은 부정적이며, 이것이 [목소리 높이기]를 억제하는 메커니즘이 될 수 있다. 개인과 조직은 종종 [목소리를 높이는 것]이 가치가 있지만, [적극적으로 장려되지는 않는다]고 말한다. 목소리를 높이고 공개적인 의사소통은 효과적이지 않고, 직업적으로 안전하지 않은 [고-위험, 저-보상 행동]으로 간주되며, 학대와 같은 다른 [부정적인 결과에 대한 두려움]을 수반한다. 게다가, 말하는 행동은 종종 [무시]되고, [언제 목소리를 높여야 하는지, 무엇에 대해 높여야 하는지가 불분명]한 경우가 많다. 사람들은 목소리를 높이는 것이 동료들과 [동료들에게 굴욕감을 줄 것]이라고 생각할지도 모른다. 부정적인 결과에 대한 학생들의 우려는 주로 [(자신에 대한) 부정적인 평가]와 [미래의 연습practice에 대한 영향]과 관련이 있다. 안전하지 않은 상황과 환자에 대한 우려를 개선하는 것이 목소리를 높이는 주된 이유입니다. 목소리를 높이는 것은 [감정적인 부담]을 수반할 수 있으며, 조직, 고위직 동료, 동료들의 지원과 함께 [용기가 필요]하다.
Overall, perceptions and experiences of speaking up are negative and can be inhibitory mechanisms on speaking up. Individuals and organizations frequently indicate speaking up is valued but is not actively encouraged. Speaking up and open communication is viewed as a high-risk, low reward behaviour that is ineffective and professionally unsafe, carrying a fear of abuse and other negative consequences. Further, speaking up behaviour is often ignored, and it is often unclear when to speak up and what speaking up entails. People may think speaking up will humiliate peers and colleagues. Student concerns over negative consequences primarily relate to negative assessment and impacts on future practice. Improving unsafe situations and concern for the patient is the primary reason for speaking up. Speaking up may carry an emotional burden, and courage along with support from organizations, colleagues of senior status, and peers is required (Bickhoff et al., 2017; Fagan et al., 2016; Milligan et al., 2017; Morrow et al., 2016; Nacioglu, 2016; Noort et al., 2019; Okuyama et al., 2014; Peadon et al., 2020).

사람들은 어떻게 목소리를 높일까요?
How do people speak up?

사람들이 실제로 어떻게 목소리를 높이는지에 대한 연구가 부족하다. 종종 사람들은

  • 질문이나 제안을 위한 [매니퓰레이팅 발화]같은 [간접적 방법]을 사용하거나
  • [행동을 피하고], [오더를 넣치 않는 것]과 같은 [하극상subversion]을 사용하는 [부정적이거나 침묵적인 방법]을 사용한다.
  • [긍정적인 방법]은 사실을 수집하고, 목소리를 높이려는 의도를 보여주고, 언제 어디서 목소리를 내야 하는지에 대한 구체적인 대상을 파악하는 것이다.

There is a paucity of research on how people speak up in practice (Bickhoff et al., 2017; Milligan et al., 2017; Morrow et al., 2016; Noort et al., 2019; Okuyama et al., 2014; Rutherford et al., 2012). Often people 

  • engage in indirect methods by manipulating speech to use questions or suggestions or
  • use negative or silent methods by avoiding action and using subversion, such as not filling an order.
  • Positive methods include collecting facts, showing an intent to speak up, and identifying specific targets for when and where to speak up.

[목소리 높이기 행동]에 대한 많은 증거는 학생에 관한 것이다. 학생들은 언제 누구에게 걱정을 보고해야 할지 모른다. 목소리를 높이는 것을 생각할 때, 사람들은 [결과에 대한 효용과 기대를 저울질하고 평가]한다. 전문가 간의 연구와 행동 등급 척도와 같은 객관적인 척도의 부재로 인해 목소리를 높이는 데 있어 전문적인 차이에 대한 경험적 증거가 부족하다.

Much of the evidence on speaking-up behaviour is for students, who often do not know when and to whom to report concerns. When considering speaking up people weigh and evaluate the utility and expectations about outcomes. Empirical evidence for professional differences in speaking up is lacking due to an absence of cross-professional research and objective measures like behavioural rating scales.

목소리를 높이는 것은 얼마나 효과적인가?
How effective is speaking up?

연구 부족으로 인해 실제로 목소리를 높이는 것의 효과는 알려져 있지 않다. 연구와 증거의 우세는 교육 개입과 훈련의 효과에 초점을 맞추고, 목소리를 높이면서 스스로 보고한 경험에 초점을 맞춘다. 환자의 목소리를 높이는 것이 환자의 목소리를 높이는 것이 잘 받아들여지지 않는다는 것을 발견한 연구는 거의 없다. 
In practice effectiveness of speaking up is unknown due to a lack of research. The preponderance of research and evidence focuses on the effectiveness of educational interventions and training and self-reported experiences with speaking up. Very few studies have investigated patient speaking up those that have found patient speaking up is not well received (Nacioglu, 2016).

시뮬레이션의 중요성
The importance of simulation

시뮬레이션은 교육 및 훈련을 조사하는 데 중요한 것으로 간주된다; 그러나 시뮬레이션 기반 연구는 부족하다. 14개 리뷰에 걸쳐 17개의 시뮬레이션 기반 연구가 명시적으로 식별되었다(추가 자료 참조). Patni 외 연구진(2019)은 시뮬레이션 연구와 대조적으로 자기 보고서 연구가 널리 퍼져 있다고 확인했다. 시뮬레이션 기반 연구의 [세 가지 유형]이 확인되었다: 

  • 교육 검증
  • 장벽의 효과(예: 엄격한 계층) 검증
  • 목소리를 높이기 위한 방법 검증(예: 2단계 규칙).

시뮬레이션 연구의 논의는 [효용, 과제, 개선]의 세 가지 영역에 초점을 맞추었다.
Simulation is considered important for investigating education and training; however, there is a dearth of simulation-based research. Across the fourteen reviews, seventeen simulation-based studies were explicitly identified (see supplemental material). Pattni et al. (2019) identified self-report studies were pervasive contra to simulation studies. Three types of simulation-based studies were identified:

  • testing education,
  • testing effects of barriers (e.g., strict hierarchy), and
  • testing methods to speak up (e.g., two-challenge rule).

The discussion of simulation research focused on three areas: utility, challenges, and improvement (Fagan et al., 2016; Kim et al., 2020; Morrow et al., 2016; Noort et al., 2019; O’Donovan & McAuliffe, 2020; Okuyama et al., 2014; Omura et al., 2017; Pattni et al., 2019; Peadon et al., 2020).

시뮬레이션의 효용
Utility of simulation

[시뮬레이션]은 [목소리 높이기와 관련된 교육, 훈련 및 개념을 검증]하기 위한 유망한 방법으로 간주된다. 주요 효용은 [성과와 행동의 객관적인 측정을 얻는 것]으로 간주된다. 또한 시뮬레이션은 개별 요인(예: 태도, 성격, 의사소통 능력) 및 수식자와 장벽(예: 평평한 팀 대 계층 구조 팀)의 영향을 측정하는 데 유용하다. 시뮬레이션은 또한 사람들이 기술을 연습하고 교훈적으로 전달된 지식을 통합할 수 있게 하며 시나리오 기반 학습에 대한 학생들의 욕구를 충족시키며 새로운 말하기 방법을 테스트하고 개발하는 환경으로 사용될 수 있다.
Simulation is viewed as a promising modality for education, training, and testing concepts related to speaking up (Fagan et al., 2016; Noort et al., 2019; O’Donovan & McAuliffe, 2020; Okuyama et al., 2014; Omura et al., 2017; Pattni et al., 2019; Peadon et al., 2020). The primary utility is seen as obtaining objective measurements of performance and behaviour. Additionally, simulation is useful in measuring the effects of individual factors (e.g., attitudes, personality, communication skills) and the impact of modifiers and barriers (e.g., flat vs. hierarchically structured teams). Simulation can also allow people to practice skills and incorporate knowledge delivered didactically and fulfills students' desire for scenario-based learning and  can be used as an environment to test and develop new speaking up methods.

시뮬레이션 과제
Challenges of simulation

[거의 모든 리뷰는 시뮬레이션을 비판]했으며, [시뮬레이션 기반 연구는 한계로 간주]되었다. [일차적인 비판]은 시뮬레이션이 임상 환경과 일치하지 않으며, 시뮬레이션에 의존하는 개입의 효과가 부족한 원인이 된다는 것이었다. 시뮬레이션 중에 참가자로부터 인식된 [책임과 긴급성이 결여]되어 있다고 가정한다. 흥미롭게도 [저 위험 설정]으로 간주되는 시뮬레이션 연구와 참가자들이 훈련을 받은 후에, 목소리를 높이는 [효과가 모든 곳에서 발생하지는 않았다]. 참가자에게 생소한 시나리오와 맥락, 시간 압축, 종단적 연구 부족 등 [시뮬레이션 방법에 초점을 맞춘 추가 비평]이 이어졌다. 시뮬레이션 동안 [사용된 평가의 신뢰성]은 부적절한 척도와 일관성 없는 관찰자/평가자 훈련으로 인해 제한된 것으로 간주되었다. 확인된 유일한 무작위 대조군 시험은 시뮬레이션 연구였지만, 전반적인 품질은 낮았다.
Nearly every review critiqued simulation, with simulation-based research being considered a limitation. The primary critique was that simulation does not match the clinical environment and the lack of verisimilitude is a cause of the lack of effectiveness of interventions that rely on simulation. It is assumed there is a lack of perceived responsibility and urgency from participants during simulation. Interestingly across simulation studies, which are considered low risk settings, and after participants received training, speaking up was not ubiquitous (O’Donovan & McAuliffe, 2020; Omura et al., 2017; Pattni et al., 2019; Peadon et al., 2020). Additional critique focused on simulation methods, including scenarios and contexts unfamiliar to participants, time compression, and lack of longitudinal study (Omura et al., 2017). The reliability of assessments used during simulation was considered limited due to inadequate measures and inconsistent observer/rater training. While the only Randomized Control Trials identified were simulation studies, the overall quality was low (Omura et al., 2017).

시뮬레이션 개선
Improvement to simulation

오무라 외 연구진(2017)에 따르면 한계에도 불구하고 시뮬레이션은 유리하고 중요하며 계속 사용되어야 한다. "…시뮬레이션을 사용한 관찰자 기반 결과 측정은 반응, 지식 및 태도뿐만 아니라 행동 변화도 평가할 수 있음이 분명하다." 
리뷰 전반에 걸쳐 [시뮬레이션 개선]은 다음과 같은 6가지 제안에 초점을 맞추고 있다: 

  • (1) 시뮬레이션에서 행동 결과를 자주 측정하고, 행동을 평가하는 더 나은 방법을 개발한다; 
  • (2) 실제 시나리오, 텍스트, 언어 및 대화를 사용하여 실제 관측에서 역방향으로 시뮬레이션을 설계함으로써 생태학적 대응성이 향상된 시뮬레이션을 설계; 
  • (3) 실험 방법과 사전 테스트 방법을 사용하여 [목소리 높이기의 원인 메커니즘과 장벽을 이해]한다; 
  • (4) 학습을 촉진하고 신뢰를 유지하기 위해 좋은 디브리핑 전략을 사용한다(특히 속임수를 사용할 때에
  • (5) 시뮬레이션에서 교육 및 훈련의 효과를 실무 결과와 연결하기 위해 종단 연구를 사용한다; 
  • (6) 시뮬레이션 교육을 위해 멀티미디어 및 확장 현실을 사용하여 탐색합니다.

Despite the limitations, simulation is viewed as advantageous, important and should continue to be used, according to Omura et al. (2017), "…it is evident that observer-based outcome measures with simulation can evaluate not only, reaction, knowledge and attitudes, but also behaviour change". Across the reviews, improvements to simulation focus on six suggestions (Fagan et al., 2016; Kim et al., 2020; Morrow et al., 2016; Noort et al., 2019; O’Donovan & McAuliffe, 2020; Okuyama et al., 2014; Omura et al., 2017; Pattni et al., 2019; Peadon et al., 2020):

  • (1) Frequently measure behavioural outcomes in simulation and develop better methods of assessing behaviour;
  • (2) Design simulations with better ecological correspondence by designing simulations backward from real-world observations using real scenarios, text, language, and conversation;
  • (3) Use experimental methods and pre-test measures to understand causative mechanisms and barriers of speaking up;
  • (4) Use good debriefing strategies to facilitate learning and maintain trust, especially when deception is used;
  • (5) Use longitudinal studies to connect the effects of education and training in simulation to practice outcomes;
  • (6) Explore using multi-media and extended reality for simulation training.

검토한 리뷰에서 의견을 제시하는 능력을 향상시키기 위한 권장 사항
Recommendations in the reviewed reviews to improve speaking up

다음과 같은 다섯 가지 공통 권고안이 제시되었다:

  • (1) 조직 지원 및 문화 변화; 
  • (2) 목소리를 높이는 연구를 개선 
  • (3) 원인 메커니즘에 대한 더 나은 이해 개발
  • (4) 교육 및 훈련 개선; 
  • (5) 외부 지식을 통합하는 것.

Five common recommendations arose:

  • (1) organizational support and culture change;
  • (2) improving the research on speaking up;
  • (3) developing a better understanding of causative mechanisms;
  • (4) improving education and training;
  • (5) incorporating outside knowledge.

조직 지원 및 문화 변화
Organizational support and culture change

[조직의 지원과 문화적 변화]는 심리적 안전, 시민의식, 그리고 모든 수준의 의료 전문가들에 걸친 파트너십을 통해 정당한 문화를 만들어낼 수 있다. 하향식 조직 및 문화 변화는 목소리를 낼 수 있는 [제도적 장벽과 요인]에 초점을 맞춰야 하고 이것은 대체로 바뀔 수 있다. 목소리를 높이는 것은 [경영진과 행정부를 포함]한 [모든 직원들의 지지]를 받아야 하며, 조직들은 [위계질서와 직업적 긴장]을 포함하여 기존의 불리한 직장 문화를 해결해야 한다. [지도적 위치]에 있는 사람들과 직업과 팀의 고위 구성원들은 모델링, 멘토링, 격려 및 말하기 행동을 수용함으로써 변화를 창출할 수 있다. 이러한 기술과 행동에 대한 [교육, 훈련, 격려 및 지원이 지도직에 제공]되어야 한다; 행동을 강요하는 것이 필요할지도 모른다. 또한 조직변화를 구현하고자 할 때는 구현장소에 고유한 국가문화, 사회경제적 맥락, 정치적 맥락 등을 고려할 필요가 있다.
Organizational support and cultural change can create just cultures through psychological safety, civility, and partnership across all levels of healthcare professionals. Top-down organizational and cultural change should focus on institutional barriers and enablers to speaking up that are mostly modifiable (Bickhoff et al., 2017; Fagan et al., 2016; Kim et al., 2020; Morrow et al., 2016; Noort et al., 2019; Omura et al., 2017; Pattni et al., 2019). Speaking up must be supported by all staff members, including management and administration, and organizations must address existing adverse workplace cultures, including hierarchy and professional tension. Those in leadership positions and senior members of professions and teams can create change through modeling, mentoring, encouraging, and accepting speaking-up behaviour. Education, training, encouragement, and support for these skills and behaviours should be provided to leadership positions; mandating the behaviours may be necessary. Additionally, when attempting to implement organizational change, it is necessary to consider national culture, socio-economic contexts, and political contexts endemic to the location of implementation.

목소리를 높이는 연구 개선
Improve research on speaking up

말하기가 어떻게 발생하는지 더 잘 이해하고 교육과 훈련을 개선하기 위해서는 더 엄격한 연구가 필요하다. 7가지 개선점이 확인되었다: 

  • (1) 교육 및 훈련 중 평가를 포함하여 보다 객관적인 측정 및 평가를 사용하여 말하기 평가 방법을 개선합니다; 
  • (2) 사람들이 목소리를 높이는 방법, 목소리를 높이는 결과 및 목소리를 높이는 방법에 대한 이해를 향상시키기 위해 생태학적 관찰이 필요하다; 
  • (3) 연구는 병원 밖, 의학과 간호 외의 직업, 다른 직장과 국가 문화, 그리고 문화적으로 다양한 팀에서 필요하다. 연구는 초국가적 및 국가적 정책의 효과를 이해하기 위해 지역적 이행을 넘어서야 한다; 
  • (4) 실험 방법 및 시뮬레이션을 사용하여 목소리 높이기의 원인 메커니즘, 활성화 요인 및 억제 요인을 테스트한다; 
  • (5) 교육에 보다 현실적인 시뮬레이션을 사용한다; 
  • (6) 시행 격차를 해소하고 교육 및 훈련의 장기적 효과와 개인 및 조직의 변화가 어떻게 발생하는지 이해하기 위해 다양한 맥락에서 종단적 연구를 수행한다; 
  • (7) 연구에 대한 이해관계자의 참여를 높이고 더 많은 자금 및 리소스를 포함하여 목소리를 높이는 데 연구의 우선순위를 지정합니다.

More rigorous research is required to better understand how speaking up occurs and to improve education and training (Bickhoff et al., 2017; Jones et al., 2021; Nacioglu, 2016; Omura et al., 2017; Rutherford et al., 2012). Seven improvements were identified:

  • (1) Improve how speaking up is measured by using more objective measures and assessments, including assessment during education and training;
  • (2) Ecological observation is required to improve understanding about how people speak up, outcomes of speaking up, and how speaking up is received;
  • (3) Research is needed in contexts outside of hospitals, with professions outside of medicine and nursing, in different workplace and national cultures, and with culturally diverse teams. Research should move beyond local implementation to understand the effects of transnational and national policies;
  • (4) Using experimental methods and simulation, to test causative mechanisms, enablers, and inhibitors of speaking up;
  • (5) Use more situationally realistic simulation for training;
  • (6) Conduct longitudinal research in multiple contexts to bridge the implementation gap and understand the long-term effects of education and training and how change occurs in individuals and organizations;
  • (7) Increase stakeholder involvement in research and prioritize research on speaking up, including more funding and resources.

원인 메커니즘에 대한 더 나은 이해 개발
Developing a better understanding of causative mechanisms

[원인 메커니즘에 대한 더 나은 이해]는 연구를 개선하는 것과 동시에 이루어진다. 특정 선행사례들은 업무량, 말하기 수용성, 말하기에 대한 인식된 가치, 환자에게 피해를 주지 않으려는 욕구, 협박, 괴롭힘, 괴롭힘과 같은 부정적 결과의 영향을 통해 말하기를 억제하는 것을 포함하여 적절한 양의 연구를 하는 것으로 확인되었다. 개인의 차이와 성격, 그리고 직장, 직업, 문화적 맥락의 영향을 포함하지 않고 사람들이 목소리를 높이도록 이끄는 사회적, 심리적 과정에 대한 더 많은 지식이 필요하다. 다양한 교육 모드의 효과와 조직 변화에 대한 적용된 시도가 목소리를 높이는 데 어떻게 영향을 미치는지에 대한 더 나은 이해가 필요하다. 
A better understanding of causative mechanism is concurrent with improving research (Jones et al., 2021; Kim et al., 2020; Milligan et al., 2017; Noort et al., 2019; Omura et al., 2017; Peadon et al., 2020). Certain antecedents were identified as having an adequate amount of research including workload, receptiveness to speaking up, perceived value in speaking up, the desire to avoid patient harm, and the inhibition of speaking up through the effect of negative consequences such as intimidation, bullying, and harassment. More knowledge is required about the social and psychological processes that lead people to speak up or not including individual differences and personality, and effects of workplace, professional, and cultural contexts. Better understanding of the effectiveness of different modes of training and how applied attempts at organizational change affect speaking up is necessary.

[목소리 높이기]는 일반적으로 [사회적으로 구성된 현상]으로 간주되어 왔으며, [목소리 높이기]가 왜 이뤄지지 않는지에 대한 가정, 추론, 의미의 속성이 만들어졌다. [일어나지 않은 일]을 평가하고 부재를 추리에 의해서 해석하는 것은 동일한 결과에 대해 반대되거나 상충되는 해석을 만들 수 있다. [목소리를 높이는 것]을 [본질적으로 의미 있고 사회적으로 내재된 현상]으로 보고 [경험적 접근법]을 사용하면 목소리를 높이는 것에 대한 보다 구체적인 이해로 이어질 수 있다.
Speaking up has typically been viewed as a socially constructed phenomenon, and assumptions, inferences, and attributions of meaning are made about the absence of speaking up. Assessing things that have not happened and interpreting absence inferentially can create opposing or conflicting interpretations of the same outcomes. Viewing speaking up as an inherently meaningful and socially embedded phenomenon and using an empirical approach can lead to a more concrete understanding of speaking up (Noort et al., 2019).

교육 및 훈련 개선
Improvement to education and training

모든 의료 수준에서 목소리를 높이는 것에 대한 [더 많은 교육과 훈련]이 필요하다. 학습자는 더 많은 교육과 훈련을 통해 목소리를 높일 수 있는 더 많은 지원을 원합니다. 교육과 훈련은 학부 수준에서 시작되어야 하며, 대학원 이후의 훈련과 지속적인 교육에서 계속되어야 한다. [목소리를 높이는 것]에 대한 교육과 훈련의 효과에 대한 의구심이 있으며, [개입의 맞춤형 및 이질적 특성]은 일관된 훈련 효과의 부족에 기여할 가능성이 있다. 목소리를 높이는 [도구와 기법에 대한 표준화된 훈련과 표준화된 평가 조치]는 개인의 경력 전반에 걸쳐 그리고 교육 및 훈련 자체에 대한 지속적인 개선 프로세스를 창출할 수 있다. 교육 및 훈련은 대면 및 기타 전달 방법을 사용하여 주로 [CRM과 같은 높은 효율성을 가진 방법]에 초점을 맞춰야 한다.
More and better education and training on speaking up is needed at all levels of healthcare (Fagan et al., 2016; Kim et al., 2020; Morrow et al., 2016; Noort et al., 2019; O’Donovan & McAuliffe, 2020; Okuyama et al., 2014; Omura et al., 2017; Pattni et al., 2019). Learners want more support for speaking up through more education and training. Education and training should begin at the undergraduate level and continue in post-graduate training and continuing education. Education and training on speaking up have questionable efficacy, with the bespoke and heterogenous nature of interventions likely contributing to the lack of consistent training effects. Standardized training for the tools and techniques of speaking up and standardized assessment measures can create continuous improvement processes for individuals throughout their career and for education and training itself. Education and training should focus on methods with high efficacy, primarily CRM, using face-to-face and other delivery methods.

다양한 직업의 특정 상황과 작업 환경에 맞춘 [시뮬레이션과 개입을 통한 경험적 시나리오 기반 학습]은 교육과 훈련을 개선할 수 있다. 학습자부터 고위 경영진까지 [모두가 참여하는 교육훈련] 설계 단계 중 [다층적 조직적 투입]이 필요하다. 교육 및 훈련 개선은 복잡하고 연구, 전략 계획 및 조직 변화를 통한 [하향식 노력의 얽힘]이 필요하다. 교육의 개선과 목소리를 높이기 위한 개입이 발생함에 따라 [새로운 세대]에게 더 나은 훈련이 제공될 수 있다. [목소리 높이기를 격려 및 수용하고, 행동을 모델링하고, 신입생들을 멘토링]하는 새로운 세대의 전문가들을 만들어 냄으로써 [광범위한 문화 변화]가 일어날 수 있다. [다양한 사고 방식과 실천 방식을 가진 새로운 전문가들]이 의료 시스템으로 이동하기 시작함에 따라, 기존 시스템과 실무자를 개선하기 위한 노력을 통해 발생하는 [하향식 조직 변화를 반영하는 상향식 변화가 발생]할 것이다.

Experiential scenario-based learning through simulation and interventions tailored to different professions' specific contexts and work environments may improve education and training. Multilevel organizational input during the design phase of education and training that includes everyone from learners to senior management is needed. Improving education and training is complex and requires the intertwining of top-down efforts through research, strategic planning, and organizational change. As improvements to education and interventions for speaking up occur better training can be provided to new generations. By creating a new generation of professionals who encourage and accept speaking up, model the behaviour, and mentor new students widespread culture change can occur. As new professionals with different ways of thinking and practicing begin moving into the healthcare system bottom-up change will occur to mirror the top-down organizational changes occurring through efforts to improve existing systems and practitioners.

외부 지식 통합
Incorporating outside knowledge

[의료 외 분야, 특히 의학 외 분야의 지식]은 교육과 훈련을 개선하고 말하기 이해에 대한 연구를 가속화함으로써 말하기 문제를 해결하는 데 도움이 될 수 있다. 가장 가치 있는 것으로 확인된 두 가지 주요 지식 분야는 [조직 행동 연구]를 포함하여 [심리학]과 [비즈니스]이다. [복잡성]과 [시스템 이론]도 도움이 될 수 있다. 의료 외의 관련 연구는 일반적으로 통합되지 않고 활용되지 않는다.
Knowledge from areas outside of healthcare, particularly outside of medicine, can help address the problem of speaking by improving education and training and accelerating research on understanding speaking up (Jones et al., 2021; Nacioglu, 2016; Noort et al., 2019; Peadon et al., 2020). The two main knowledge areas identified as being most valuable are psychology and business, including organizational behavioural research. Complexity and systems theories could also prove helpful. The large amounts of relevant research outside of healthcare are generally not incorporated and utilized.

목소리를 높이는 것에 대한 기존 지식의 종합
Synthesis of the existing knowledge on speaking up

[목소리를 높이는 것]은 주로 [위계 구조, 영향에 대한 두려움, 조직 지원의 부족]을 포함하는 [깊이 뿌리박힌 장벽] 때문에 효과적이지 못한 것으로 인식된다. [리더십과 선배 동료들의 지원, 환자에 대한 관심, 개인의 특성]이 목소리를 높이는 것을 지원한다. 목소리를 높이는 원인 메커니즘, 촉진자 및 억제자의 영향은 복잡한 상호작용에 존재한다. [원인 메커니즘]의 진정한 효과는 잘 이해되지 않으며, 그 효과를 명확하게 말할 수 없다. [목소리 높이기]를 더 잘 이해하기 위해서는 더 나은 평가, 실험, 종적 연구, 생태 관찰을 포함한 개선된 연구 방법이 필요하다. [상향식 조직 변화와 상향식 교육 및 훈련]은 원활한 목소리 내기에 필수적이다.  현재, 교육과 훈련은 충분히 자주 제공되지 않으며 일관적이거나 적절하게 효과적이지 않다; 개선이 필수적이다. 시뮬레이션은 교육, 훈련 및 연구에서 중요한 역할을 할 수 있지만 또한 개선이 필요하다. 현재 목소리를 높이는 것과 관련된 지식과 연구가 대두되고 있으며, 단점을 해결하고 질문을 명확히 하고 개선할 수 있는 방법을 찾을 수 있는 기회가 많다. 
Speaking up is perceived to be ineffective due to deeply ingrained barriers that primarily include hierarchy, fear of repercussions, and a lack of organizational support. Support from leadership and senior colleagues, concern for the patient, and individual characteristics support speaking up. The influence of causative mechanisms, facilitators and inhibitors of speaking up exist in complex interactions. Causative mechanisms' true effects are not well understood, and effects cannot be stated definitively. Improved research methods including better assessment, experiment, longitudinal study, and ecological observation are needed to better understand speaking up. Top-down organizational change and bottom-up education and training are essential to facilitate speaking up. Currently, education and training are not delivered frequently enough and are not consistently or adequately effective; improvement is essential. Simulation can have an important role in education, training, and research, but also requires improvement. Currently, knowledge and research related to speaking up is emergent, with many opportunities to address shortcomings, clarify questions, and identify avenues for improvement.

논의
Discussion

의료 분야에서 목소리를 높이는 것에 대한 지식의 현재 상태에 대한 통합된 이해를 개발하기 위해 14개의 리뷰를 합성했다. 목소리를 높이는 것은 전 세계적으로 관련된 의료 문제입니다. 국가와 의료 시스템 전반에 걸쳐 목소리를 높이는 것이 문제라는 공통된 이해가 있지만, 문제를 효과적으로 해결하는 방법에 대한 지식은 제한적이다. 
Fourteen reviews were synthesized to develop a unified understanding of the current state of knowledge on speaking up in healthcare. Improving speaking up is a globally relevant healthcare issue. Across countries and healthcare systems there is a shared understanding that difficulty in speaking up is a problem, however, knowledge about how to effectively address the problem is limited.

최신 검토를 기반으로 한 권장 사항
Recommendations based on the state-of-the-art review

언어의 일관된 사용
Consistent use of language

[목소리 높이기]를 조사하기 위해서는 일관된 언어가 필요하다. Okuyama 외 연구진(2014)은 모든 검토에 반영된 간결하고 구체적인 정의를 제공하며 적용 가능성을 확대하기 위해 "병원 환경"을 제거하는 경우 합의된 정의를 위해 충분히 일반적이다:

  • "말하기"는 의료 전문가들이 의료 팀 내 다른 사람들의 [위험하거나 부족한 행동을 인식하거나 인지]할 때, 환자의 안전과 의료 품질의 이점에 대한 우려를 제기하는 것으로 정의된다 

Consistent language is necessary to investigate speaking up. Okuyama et al (2014) provide a concise, concrete definition, reflected in all the reviews and sufficiently general for a consensus definition if "hospital environment" is removed to broaden applicability:

  • "Speaking up' is defined as the raising of concerns by healthcare professionals for the benefit of patient safety and care quality upon recognizing or becoming aware of the risky or deficient actions of others within healthcare teams."

'스피킹 업'은 '안전 보이스'보다 장점이 있는데, 이는 [안전 보이스]가 수동적이고 정의적으로 '분해되고 초기'이기 때문이다. 모리슨의 프레임워크 내에서 [안전 목소리]는 미래의 이해를 지원할 수 있지만, 지식은 창발적이고, 더 간단한 개념화가 권장된다.
'Speaking up'
is advantageous over 'safety voice' as speaking up is actionable and precise while safety voice is passive and definitionally "disintegrated and nascent" (Noort et al., 2019). Safety voice within Morrison's framework may support future understandings; however, knowledge is emergent, and a simpler conceptualization is recommended.

명확하고 객관적인 언어는 목소리를 높이고 토론하고 연구하는 데 사용되어야 한다. "엘리트의 무례함"과 "불건전한 복종" 등의 용어는 모호하고 비난적이다. [Value-laden terms]를 통해 [책임, 죄책감 및 보복의 필요성]을 명시적으로 할당하는 것은 [개방적 의사소통과 책임 없는 문화의 기본적인 환자 안전 원칙을 훼손]한다. 최적의 환자 치료를 제공하기 위해 목소리를 높이도록 장려하고 지원하는 대학 팀과 조직을 만드는 것은 개인이나 그룹을 권력 투쟁과 자연스러운 갈등 상태에 있다고 보는 데서 오는 것이 아니다.
Clear, objective language should be used for discussing and researching speaking up. Terms such as "incivility of elites” and "unwholesome submissiveness” etc., are ambiguous and accusatory. Explicitly assigning blame, guilt, and the need for reprisal through value-laden terms undermines fundamental patient safety principles of open communication and blame-free cultures (Helmreich et al., 1999; Kohn et al., 1999). Creating collegial teams and organizations that encourage and support speaking up to provide optimal patient care does not come from viewing individuals or groups as being in a power struggle and natural state of conflict.

연구 방향
Research directions

말하기를 이해하기 위한 [기본 질문]을 설정할 수 있습니다. 말하기에서 가장 큰 차이를 설명하는 변수와 공분산은 무엇입니까? 여기서 변수는 발화의 단일 행동 결과를 억제하거나 가능하게 하는 원인 메커니즘 또는 선행 요소로 정의될 수 있다. 단순화된 접근 방식은 목소리를 높이기 위한 변수의 복잡한 상호 작용에 대한 인과 관계 조사를 지원한다. 1차 활성화 또는 억제 변수에 대한 지식이 증가함에 따라, 데이터 수집은 구조 방정식 모델링(SEM)을 사용하여 경험에 기반한 말하기 모델을 개발하기 위해 더 효율적으로 초점을 맞출 수 있다. SEM을 활용한 초기 시도는 부분적으로 engagement에 의해 매개된 "음성"을 촉진하는 지원 및 제어의 이중 요인 모델을 확인했다. 단순화된 개념화와 경험적으로 파생된 모델은 연구, 행동 관찰 및 평가를 지원할 것이다.
A foundational question for understanding speaking up can be set: which variables account for the most variance in speaking up, and what are their covariance? Here variables can be defined as the causative mechanisms or antecedents that inhibit or enable the single behavioral outcome of speaking up. A simplified approach supports causal investigation of the complex interactions of the variables for speaking up. As knowledge about the primary enabling or inhibitory variables increases, data collection can be more efficiently focused to develop an empirically based model of speaking up using Structural Equation Modelling (SEM) (Kenny, 2019). An early attempt utilizing SEM identified a bifactorial model of support and control promoting “voice” that was partially mediated by engagement (Voogt et al., 2019). Simplified conceptualization and an empirically derived model would support research, behavioural observation, and assessment.

[심리학의 지식]을 연구와 교육, 특히 사회심리학의 준법, 권위에 대한 복종, 순응에 대한 방대한 연구와 함께 통합하면 분야를 가속화할 수 있다. 심리학은 치료의 대인 관계를 포함한 의료 제공의 많은 영역과 교차하는 과학이다. 맥락은 중요하지만, 인간 행동과 인지의 원칙은 일반적으로 보편적이며, 이미 알려진 것을 재발견할 필요는 없다.
Incorporating knowledge from psychology in research and education, specifically social psychology with its vast research on compliance, obedience to authority, and conformity (Cialdini & Griskevicius, 2010; Cialdini & Trost, 1998) can accelerate the field. Psychology is a science that intersects with many areas of healthcare delivery, including the interpersonal aspects of care. Context is important; however, the principles of human behaviour and cognition are generally universal (Ariely, 2008; Olsson-Collentine et al., 2020), and it is not necessary to rediscover what is already known (Avorn, 2018; Croskerry et al., 2017).

연구에서 도출된 추론을 개선하기 위해서는 수행되는 연구의 양과 질을 개선하는 것이 필요하다. 두 측면 모두에서 지속적인 개선은 말하기를 촉진하는 교육과 방법을 만들고 신뢰성과 타당성이 좋은 객관적 관찰 평가에 의해 안내되는 효과적인 표준화된 개입을 개발하는 데 도움이 될 수 있다. 
Improvements to the quantity as well as quality of research being conducted is necessary to improve the inferences drawn from the research (Vazire et al., 2022). Continuous improvement in both aspects can help create education and methods to promote speaking up and develop effective standardized interventions with long-lasting effects guided by objective observational assessments with good reliability and validity.

시뮬레이션 향상
Enhancement of simulation

연구와 교육을 위한 시뮬레이션에 대한 인식 전환이 필요하다. 시뮬레이션의 한계는 모달리티 자체에 내포되어 있습니다. 제한으로 간주되기보다는, 시뮬레이션을 목소리를 높이기 위한 훈련과 원인 메커니즘을 실험적으로 조사할 수 있는 풍부한 가능성을 제공하는 [생태학적/외부적 타당성이 높은 시험장]으로 보아야 한다. 임상 연구 및 관찰 시뮬레이션과 함께 이론 개발, 실험, 교육 및 평가를 위한 주요 이점이다. 예를 들어, [목소리 높이기] 예측 모델은 임상 및 조직 수준에서 조사로 이동하기 전에 시뮬레이션에서 평가할 수 있다. 임상 환경으로 발전하는 과정에서 시뮬레이션은 목소리를 높이는 기술을 연습하는 매개체가 될 수 있다. 구현 과학의 통합은 시뮬레이션에서 임상 공간에 이르는 학습을 전파하고 임상의, 교육자 및 행정부의 새로운 지식과 관행을 활용하는 데 도움이 될 수 있다.
A shift in the perception of simulation for research and education is necessary. Simulation’s limitations are implied in the modality itself. Rather than being seen as a limitation simulation should be viewed as a testing ground with high ecological/external validity that provides abundant possibilities to experimentally examine training and causative mechanisms for speaking up (Friedman et al., 2015). In conjunction with clinical research and observation simulation is a major boon for theory development, experimentation, education, and assessment. For example, predictive models of speaking up can be evaluated in simulation before moving to investigation at the clinical and organizational level. In the progression to the clinical environment, simulation can be an intermediary to practice the skill of speaking up. The incorporation of implementation science can help to disseminate the learnings from simulation to the clinical space and uptake of new knowledge and practices by clinicians, educators, and administration.

현재 검토의 한계
Limitations of the present review

  • (1) 관련 문헌이 누락되었을 가능성이 높다. 최근의 특정 기사와 관련 기사들은 어떤 리뷰에도 포함되지 않았다. 
  • (2) 현재 검토는 사용된 검토의 검색, 분석 및 보고에 따라 결정된다. 검토 방법론적 품질의 평가는 현재 방법론적 검토가 반드시 관련이 있는 것은 아니기 때문에 수행되지 않았다. 검색에서 메타 분석이 확인되지 않았으며, 이는 해당 분야가 개발 단계에 있음을 나타낸다. 많은 변수와 여러 검토 유형에 걸쳐 다양한 효과 크기가 분산되어 있기 때문에 현재 체계적인 평가가 불가능하다. 
  • (3) 본 검토는 환자 치료의 맥락에서 목소리를 높이는 데 초점을 맞췄다. 개인의 존엄성을 지지하기 위해 목소리를 높이는 것이 필요한 다른 교육적이고 전문적인 맥락이 존재하며 고려되어야 한다.
  • (1) Relevant literature was likely missed. Certain recent and relevant articles were not included in any of the reviews (Calhoun et al., 2014; Daly Guris et al., 2019; Fisher & Kiernan, 2019; Kaba et al., 2016; Kuo et al., 2020; Oner et al., 2018; Shanks et al., 2020).
  • (2) The present review is contingent on the search, analysis, and reporting of the reviews used. An assessment of the reviews methodological quality was not conducted as, at present, a methodological review is not necessarily relevant. No meta-analyses were identified in the search, indicating the field is in a developmental stage. With numerous effect sizes spread over many variables and multiple review types at this time systematic evaluation is not possible (Noort et al., 2019).
  • (3) The present review focused on speaking up in the context of patient care. Other educational and professional contexts exist, and should be considered, where speaking up is necessary to support individual dignity (Gianakos et al., 2022).

결론
Conclusion

환자 위해 문제가 지속되고 환자 안전을 발전시키고 지속적으로 개선할 필요가 있는 상황에서, 그 임무를 추진하기 위한 공동의 노력이 필요하다. 행동의 명백한 단순성에도 불구하고, 목소리를 높이는 것은 잘 이해되지 않고, 연구는 초기 단계에 있으며, 교육과 훈련에 대한 개선이 필요하며, 외부의 견해가 필요하다. 초기 분야의 리뷰는 중복성을 줄이고 현재 지식의 포괄적인 통합을 제공하기 위해 단일의 포괄적인 문서를 작성할 필요성을 뒷받침한다. 현재의 합성은 통일된 이해를 촉진하고, 기존의 지식을 명확하게 하며, 향후 연구의 공통된 출발점으로 기능하고자 한다.
With the persistent issue of patient harm and the need to advance and continuously improve patient safety, a concerted effort is needed to carry the mission forward (Cohen & Patel, 2020). Despite the apparent simplicity of the behaviour, speaking up is not well understood, research is in early stages, improvements to education and training are required, and outside views are needed. The plethora of reviews in an incipient field supports the need to create a single comprehensive document to reduce redundancy and provide a comprehensive synthesis of the current knowledge. The current synthesis intends to foster a unified understanding, clarify existing knowledge, and function as a common starting point for future research.

 


Adv Health Sci Educ Theory Pract. 2022 Oct;27(4):1177-1194. doi: 10.1007/s10459-022-10124-8. Epub 2022 Jun 6.

A state-of-the-art review of speaking up in healthcare

Affiliations collapse

1Center for Advanced Medical Simulation, Northern Alberta Institute of Technology, CAT 126, 11762 106 St NW, Edmonton, AB, T5G 2R1, Canada. efremv@nait.ca.

PMID: 35666354

DOI: 10.1007/s10459-022-10124-8

Abstract

Throughout healthcare, including education, the need for voicing of concern by speaking up is a globally recognized issue that has come to the fore in the last ten years. There has been a rapid growth in the number of review articles on the topic. To prevent diffusion of knowledge and support future research it is necessary to gather the existing knowledge in a single place. The purpose of the present article is to bring together the existing reviews on speaking up to create a source of unified knowledge representing the current "State of the Art" to advance future research and practice. A State-of-the-Art review was conducted to synthesize the existing knowledge on speaking up. Six databases were searched. Fourteen review articles spanning 2012 to 2021 were identified. Five main research questions have been investigated in the literature and five common recommendations for improvement are made, the knowledge across all reviews related to the research questions and recommendations was synthesized. Additionally, simulation-based research was frequently identified as an important though limited method. Further issues in the literature are identified and recommendations for improvement are made. A synthesis was successfully developed: knowledge about speaking up and research related to speaking up is in an emergent state with more shortcomings, questions, and avenues for improvement than certitude. The whys and how of speaking up remain open questions.

Keywords: Challenging authority; Healthcare professional; Healthcare student; Interprofessional; Patient safety; Safety voice; Speaking up.

구색만 갖추기: 어거지로 하는 설문이 자료 퀄리티에 미치는 영향(EDUCATIONAL RESEARCHER, 2021)
Assessing Survey Satisficing: The Impact of Unmotivated Questionnaire Responding on Data Quality
Christine Calderon Vriesema1 and Hunter Gehlbach2

 

교육 연구자들을 포함한 사회과학자들은 오랫동안 설문조사와 애증 관계를 유지해왔다. 한편, 설문조사는 응답자의 가치, 인식 및 태도를 효율적이고 규모에 맞게 파악한다(Gehlbach, 2015; Gilbert, 2006; West et al., 2017). 설문조사의 유연성을 통해 응답자는 자신, 다른 개인 또는 전체 계층 또는 공동체에 대한 인식을 보고할 수 있습니다. 
Social scientists, including educational researchers, have long maintained a love–hate relationship with surveys. On the one hand, surveys uncover respondents’ values, perceptions, and attitudes efficiently and at scale (Gehlbach, 2015; Gilbert, 2006; West et al., 2017). Surveys’ flexibility allows respondents to report on themselves (i.e., self-report measures), other individuals, or their perceptions of a whole class or community.
반면, [회의론자]들은 설문조사 데이터의 가치를 비판해 왔으며, 종종 세 가지 우려에 초점을 맞추고 있다.
  • 첫째, 일부는 참가자들이 양질의 답변을 제공하는 데 필요한 자기성찰 능력에 대해 의문을 제기한다. 예를 들어, Nisbett와 Wilson(1977a, 1977b)은 사람들이 자신의 선택을 이해하려는 잘못된 시도의 여러 예를 제공했다. 다른 것들은 응답자들이 존재하지 않는 정책에 대해 어떻게 쉽게 보고하는지, 따라서 사람들이 어떻게 그들이 가질 수 없는 의견을 보고하는지를 보여준다(Bishop et al., 1980).
On the other hand, skeptics have critiqued the value of survey data, often focusing on three concerns. First, some raise questions about the introspective abilities participants need to provide high-quality answers. For example, Nisbett and Wilson (1977a, 1977b) provided multiple examples of people’s erroneous attempts to understand their own choices. Others show how respondents readily report on policies that do not exist, thus, showing how people report opinions that they could not possibly have (Bishop et al., 1980).
  • 두 번째 도전은 사람들이 자신의 태도를 알 수 있다는 것을 인정하면서도, 미묘한 힘이 응답자의 정확한 보고를 방해할 수 있다고 우려하는 비평가들로부터 발생한다. 이러한 힘에는 묵인 편향, 사회적 만족도, 바닥/천장 효과, 편향된 질문 표현, 대응 순서 효과 등과 같은 현상이 포함된다(예: Krosnick, 1999).
A second challenge arises from critics who acknowledge that people might know their own attitudes but worry that subtle forces may inhibit respondents’ accurate reporting. These forces include phenomena such as acquiescence bias, social desirability, floor/ceiling effects, biased question wording, response order effects, and so forth (e.g., Krosnick, 1999).
그러나 설문조사 설계자는 설문조사를 [응답자가 합리적으로 의견을 가질 수 있는 주제]로 제한할 수 있다. 또한 설문조사 연구자들이 개발한 많은 모범 사례에 부합하도록 설문조사를 설계할 수 있다(Gehlbach & Artino, 2018). 따라서 조사 연구의 방법론으로서의 이 [두 가지 잠재적인 문제는 현실적이고 진지하게 받아들여야 하지만 극복할 수 없는 경우는 거의 없다]. 
Yet, survey designers can delimit surveys to topics that respondents might reasonably have opinions on. Furthermore, they can design surveys to accord with many of the best practices that survey researchers have developed (Gehlbach & Artino, 2018). So, although these two potential problems with survey research as a methodology are real and need to be taken seriously, they are rarely insurmountable.
  • 셋째, 잠재적으로 더 도전적인 것은 참가자들이 설문조사를 진지하게 받아들이려는 동기에 대한 우려이다. 가장 극단적인 형태로, 어떤 사람들은 아마도 지루하거나 웃기려는 시도로 거짓 대답을 하기 위해 적극적으로 노력하는 "장난꾸러기 응답자"가 될 수도 있다. Krosnick(1991)은 응답자들이 응답에 최선의 노력을 기울이지 못하는 더 온화하고 잠재적으로 더 보편적인 형태의 "구색만갖추기satisficing"을 설명한다. 이러한 동기 부여 문제는 충분히 일반적이어서 일부 연구자들은 성실성이라는 성격 특성을 측정하기 위한 수행 과제로 설문지에 대한 응답자의 노력(또는 그 부족)을 사용했다(Hitt et al., 2016; Zamarro et al., 2018).
Third, and potentially more challenging, are concerns about participants’ motivation to take the survey seriously. In its most extreme form, some may become “mischievous responders” (Robinson-Cimpian, 2014) who actively strive to give false answers, perhaps out of boredom or an attempt to be funny. Krosnick (1991) describes milder, and potentially more prevalent forms of “satisficing,” where respondents fail to put forth their best efforts in responding. This motivation problem is sufficiently common that some researchers have even used respondents’ effort (or lack thereof) on questionnaires as a performance task to measure the personality trait of conscientiousness (Hitt et al., 2016; Zamarro et al., 2018).
학교들이 설문조사 자료로 정책을 알리는 것을 점점 더 목표로 함에 따라, 이러한 동기 부여 문제는 독특한 도전 과제를 제시한다. 만약 응답자들이 항목을 건너뛰거나, 일찍 그만두거나, 매번 같은 대답을 해서 설문조사를 빠르게 통과하고 싶다면, 연구원들은 그것을 막기 위해 할 수 있는 일이 거의 없다. 실무자와 정책 입안자는 보완적인 문제에 직면해 있습니다: 고객은 만족도가 데이터 품질에 어느 정도 영향을 미치는지 파악하기 위해 [satisficing의 만연성]을 이해해야 합니다. 우리는 캘리포니아에서 초등학생과 중등학생의 사회적 정서적 학습에 대한 지속적인 대규모 조사에서 만족도를 조사함으로써 두 가지 과제를 모두 해결한다. 이 기사는 설문 조사 데이터의 만족도를 감지, 평가 및 회계 처리하기 위한 간단한 전략을 개략적으로 설명합니다. 참가자 만족도에 관한 더 큰 문헌(예: Burge & Gehlbach, 2012; Hitt et al., 2016; Krosnick, 1991; Soland, 2018) 내에서, 우리는 이 연구가 잠재적으로 문제가 있는 대응 패턴을 식별하기 위한 접근 가능한 도구를 교육 의사결정자에게 제공하기를 바란다. 
As schools increasingly aim to inform their policies with survey data, this motivation problem presents a unique challenge. If respondents want to skip items, quit early, or speed through the survey by giving the same answer each time, researchers can do little to prevent it. Practitioners and policymakers face a complementary problem: They need to understand the pervasiveness of satisficing to determine to what extent satisficing affects data quality. We address both challenges by investigating satisficing in an ongoing, large-scale survey of elementary and secondary students’ social–emotional learning in California. This article outlines straightforward strategies for detecting, assessing, and accounting for satisficing in survey data. Within the larger literature around participant satisficing (e.g., Barge & Gehlbach, 2012; Hitt et al., 2016; Krosnick, 1991; Soland, 2018), we hope this study provides educational decision makers with accessible tools for identifying potentially problematic response patterns.

구색만 갖추기
Satisficing

전통적으로 "satisficing"은 사람들이 [정신 에너지를 보존하기 위해 차선의 의사 결정 전략을 사용하는 인지 휴리스틱]을 의미한다(Simon, 1957; Simon & Stedry, 1968). 예를 들어, 일부 의사결정자들은 "최적" 해결책을 찾기보다는 "충분히 가까워 보인다"는 만족스러운 첫 번째 대안을 선택하여 시간과 노력을 절약한다. 설문조사 학자들은 응답자들의 차선의 행동을 설명하기 위해 이 개념을 적용했다(Krosnick, 1991; Tourangeau, 1984). 설문조사에서 만족하는 것은
  • 첫 번째 합리적인 응답 옵션 선택,
  • 제시된 모든 진술에 동의하는 것,
  • 여러 항목에 걸쳐 동일한 옵션을 직선으로 반복적으로 선택하는 것,
  • "모름" 또는 "해당되지 않는" 응답을 일관되게 선택하는 것을 포함할 수 있다(Barge & Gehlbach, 2012; Krosnick, 1991).
Traditionally, “satisficing” refers to a cognitive heuristic in which people employ suboptimal decision-making strategies to conserve mental energy (Simon, 1957; Simon & Stedry, 1968). For example, rather than searching for an “optimal” solution, some decision makers select the first satisfactory alternative that seems “close enough,” thereby saving time and effort. Survey scholars have adapted this concept to explain respondents’ suboptimal behaviors (Krosnick, 1991; Tourangeau, 1984). On surveys, satisficing can include
  • selecting the first reasonable response option,
  • agreeing with all the statements presented,
  • selecting the same option repeatedly in a straight line across multiple items, and
  • consistently selecting the “don’t know” or “not applicable” responses (Barge & Gehlbach, 2012Krosnick, 1991).
일부 조사 연구자들이 참가자들의 satisficing에 대해 보고했지만, 이러한 세부사항을 체계적으로 포함하는 사람은 거의 없다. Barge와 Gehlbach(2012)는 대학생을 대상으로 실시한 두 가지 설문조사를 대상으로 척도 간 신뢰도와 연관성에 대한 만족도의 영향을 조사했다. 저자들은 대부분의 학생들이 적어도 한 가지 형태의 satisficing에 관여한다는 것을 발견했다(두 조사에서 61%와 81%의 학생들). 이러한 satisficing은 인위적으로 부풀린 내부 일관성 추정치와 척도 간 상관관계를 초래했다. 이러한 관행의 만연성과 데이터 해석에 대한 시사점은 특히 특정 항목이 작성되는 방법에 어려움을 겪을 수 있는 어린 학생들에게 만족스러운 설문조사와 잠재적 결과를 추가로 탐구할 필요성을 강조한다(예: 부정적으로 단어화된 항목; Benson & Hocevar, 1985). 대규모 데이터가 정책 및 실천을 위한 의사 결정을 안내하는 데 점점 더 많이 사용되고 있기 때문에 이러한 지식은 특히 중요하다(Marsh et al., 2018).

Although some survey researchers have reported on participant satisficing, few systematically include these details. Barge and Gehlbach (2012) examined the effects of satisficing on the reliability of and associations between scales for two surveys administered to college students. The authors found that most students engaged in at least one form of satisficing (61% and 81% of students across the two surveys). This satisficing resulted in artificially inflated internal consistency estimates and correlations between scales. The pervasiveness of these practices and implications for data interpretation underscore the need to further explore survey satisficing and its potential consequences, especially for younger students who may struggle with how certain items are written (e.g., negatively worded items; Benson & Hocevar, 1985). This knowledge is particularly important now as large-scale data are increasingly used to guide decisions for policy and practice (Marsh et al., 2018).

satisficing 검출 전략에는 복잡도가 다양한 방법이 포함된다(Barge & Gehlbach, 2012; Steedle et al., 2019). 이상적으로, 만족도를 해결하기 위한 절차의 집합은 가능한 한 광범위하게 접근할 수 있어야 한다. 이를 위해, 우리는 연구자, 실무자 및 정책 입안자가 거의 모든 설문 조사 기반 데이터 세트 내에서 평가할 수 있는 세 가지 응답자 행동에 초점을 맞춘다
  • 조기종결
  • 미응답
  • 한줄긋기

Strategies for detecting satisficing include a range of methods that vary in complexity (Barge & Gehlbach, 2012; Steedle et al., 2019). Ideally, any set of procedures to address satisficing should be as broadly accessible as possible. Toward this end, we focus on three respondent behaviors that researchers, practitioners, and policymakers can assess within almost all survey-based data sets: 
  • early termination—when respondents fail to complete the full survey; 
  • nonresponse, or omitted items; and 
  • straight-line responding—when respondents select the same response option repeatedly (e.g., for at least 10 consecutive items).
본 연구에서는 선행연구(Barge & Gehlbach, 2012)를 바탕으로 직선응답을 10개의 연속항목으로 조작화하였으며, 이는 역점수 항목의 배치를 고려할 때 이 특정 설문조사의 맥락에 부합하기 때문이다. 학생들이 동일한 구조 내의 여러 항목에 대해 유사하게 반응하는 것이 타당해 보였지만, 여러 구조와 역 점수 항목에 걸쳐 연속적으로 10개의 동일한 반응이 나타날 가능성은 거의 없어 보였다. 이러한 조작화는 또한 carryover effect와 같이 표면적으로 유사한 인지적 편향과 직선화를 구별하는 데 도움이 될 것이다. 직선적인 반응은 응답자들이 인지적 노력을 보존하는 데 도움이 된다. 대조적으로, 참가자들이 한 조사 항목과 후속 항목의 유사성을 인식할 때 이월 효과(Dillman et al., 2014)가 발생할 수 있으며, 따라서 유사한 응답을 장려한다(과다하게). 설문조사의 모든 10개 항목 집합에는 여러 개의 구성요소가 포함되므로, 참가자는 항목 간의 개념적 차이를 확인해야 합니다. 
In this study, we operationalized straight-line responding as 10 consecutive items based on prior research (Barge & Gehlbach, 2012), and because it fit the context of this particular survey given the placement of reverse-scored items. Although it seemed plausible for students to respond similarly across multiple items within the same construct, the likelihood of 10 identical responses in a row spanning multiple constructs and reverse-scored items seemed vanishingly small. This operationalization also should help distinguish straight-lining from ostensibly similar cognitive biases, such as carryover effects. Straight-line responding helps respondents conserve cognitive effort. By contrast, carryover effects (Dillman et al., 2014) can occur when participants perceive similarities from one survey item to a subsequent item, thereby encouraging (overly) similar responses. Because multiple constructs are included in all 10-item sets within the survey, participants should see conceptual differences between items.
요약하면, 우리는 조기 종료, 무응답 또는 직선 응답이라는 세 가지 차선의 응답 패턴 중 하나 이상에 관여함으로써 satisficing을 조작화했다. 다른 접근법이 존재하지만(예: Hit 등, 2016; Robinson-Cimpian, 2014; Steedle 등, 2019), 우리는 대규모 학생 설문 조사 데이터에서 만족스러운 것을 체계적으로 정의, 계산 및 보고하기 위한 세 가지 간단하고 접근 가능한 전략에 중점을 두었다. 그렇게 함으로써, 우리는 이러한 간단한 단계들이 그들의 특정 교육적 맥락 내에서 가능한 한 많은 조사 데이터 사용자들에 의해 널리 채택되기를 희망했다. 
In sum, we operationalized satisficing as engaging in one or more of these three suboptimal response patterns: early termination, nonresponse, or straight-line responding. Although other approaches exist (e.g., Hitt et al., 2016; Robinson-Cimpian, 2014; Steedle et al., 2019), we focused on three straightforward, accessible strategies for systematically defining, calculating, and reporting satisficing in large-scale student survey data. By doing so, we hoped that these simple steps might be widely adopted by as many users of survey data as possible within their specific educational contexts.

연구 질문과 가설
Research Questions and Hypotheses

연구 결과의 투명성과 신뢰성을 높이기 위해 권장 사례에 따라 일련의 가설을 사전 등록했다(Gehlbach & Robinson, 2018). 구체적으로, 우리는
  • (a) 학생들이 어느 정도까지 조사 satisficing에 참여했는지,
  • (b) 어떤 형태의 satisficing이 조사 데이터에 가장 큰 위협이 되는지,
  • (c) 이 전략이 조사 척도에서 학생들의 평균 점수에 어떻게 영향을 미칠 수 있는지 더 잘 식별하기 위해, 학생들이 한줄긋기를 할 때 가능성이 높은 응답 옵션이 무엇인지
  • (d) 어떤 학생들이 가장 satisficing을 할 것 같은지
To boost the transparency and credibility of our findings, we preregistered a set of hypotheses (https://osf.io/36zqk/) according to recommended practices (Gehlbach & Robinson, 2018). Specifically, we wanted to know
  • (a) to what extent students engaged in survey satisficing,
  • (b) which form of satisficing posed the largest threat to survey data,
  • (c) which response option students were most likely to select when straight-lining in order to better discern how this strategy might affect students’ mean scores on the survey scales, and
  • (d) which students were most likely to satisfice.
탐사 파일럿 데이터와 이전 연구를 통해 다음과 같은 사전 지정된 가설을 테스트했다:
Informed by our exploratory pilot data and prior research, we tested the following prespecified hypotheses:
  • 가설 1: 전체 표본의 최소 10%가 만족스러운 형태로 사용됩니다.
  • 가설 2: 세 가지 유형의 만족도 조사 중 직선이 가장 많은 총 조사 항목에 영향을 미칩니다.
  • 가설 3: 직선 라이닝은 데이터의 품질에 영향을 미칩니다. 구체적으로:
    • 가설 3a. 직선을 이루는 참가자는 대부분 척도의 오른쪽에서 가장 극단적인 반응 옵션을 선택합니다.
    • 가설 3b: 역점수 항목을 고려한 후 직선 응답은 네 가지 조사 척도의 평균 점수에 유의한 영향을 미친다.
  • 가설 4: 남학생이 여학생보다 만족도가 높을 것이다.
Hypothesis 1: At least 10% of the total sample will engage in some form of satisficing.
Hypothesis 2: Of the three types of satisficing examined, straight-lining will affect the greatest number of total survey items.
Hypothesis 3: Straight lining will affect the quality of the data. Specifically:
Hypothesis 3a. Participants who straight-line will select the most extreme response option on the right-hand side of the scale the majority of the time.
Hypothesis 3b: After accounting for reverse-scored items, straight-line responding will significantly affect the mean scores across the four survey scales.
Hypothesis 4: Male students will be more likely to satisfice than female students.

방법
Method

샘플
Sample

본 연구는 캘리포니아 교육의 Core-PACE 연구 파트너십을 위한 정책 분석을 통해 수집된 2차 자료를 분석하였다. 우리는 2014-2015년과 2015-2016년 학기 동안 여러 캘리포니아 학군을 대상으로 대규모 설문 조사의 일환으로 관리된 일련의 사회 정서 학습(SEL) 항목에 대한 학생들의 반응을 분석했다. 전체 설문조사에는 SEL 항목이 포함되었고 그 다음으로 학교 문화와 기후 항목이 포함되었다. 그러나 문화와 기후 항목의 수는 학군과 학년에 따라 달랐다. 따라서, 우리는 분석을 SEL 항목으로 제한했다.
This study examined secondary data collected through the Policy Analysis for California Education’s CORE-PACE Research Partnership. We analyzed student responses to a set of social–emotional learning (SEL) items administered as part of a larger survey to several California school districts during the 2014–2015 and 2015–2016 school years. The full survey included SEL items followed by a set of school culture and climate items; however, the number of culture and climate items varied across districts and school years. Thus, we restricted our analyses to the SEL items.

파일럿의 경우 2014-2015학년도 데이터에 대한 탐색적 분석을 수행했습니다. 이러한 분석은 2015-2016년 데이터에 대해 테스트한 사전 등록된 가설을 생성했다. 확인 연구를 위한 표본(N = 409,721)은 CORE 학군의 하위 집합에서 온 학생들을 포함했다. 2학년 2명을 제외한 학생들은 3학년부터 12학년까지 다양했다. 표본은 146,126명의 초등학생들, 125,747명의 중학생들, 그리고 137,838명의 고등학생들을 포함했다. 표 1은 학생 인구 통계의 완전한 설명을 제공한다. 

For our pilot, we conducted exploratory analyses on the 2014–2015 school year data. These analyses generated the preregistered hypotheses that we tested on the 2015–2016 data. The sample (N = 409,721) for our confirmatory study included students from a subset of the CORE school districts (see the full list here: https://edpolicyinca.org/initiatives/core-pace-research-partnership). Except for two second graders, students ranged from Grades 3 through 12. The sample included 146,126 elementary school students; 125,747 middle school students; and 137,838 high school students. Table 1 provides a complete description of student demographics.

 

방안
Measures

25개 항목으로 구성된 SEL 조사 성장 마인드를 측정하는 척도 4개(n = 4개 항목), 규제(n = 9개 항목), 자기 인식(n = 4개 항목), 사회 인식(n = 8개 항목)을 포함하였다. 각 척도에 대한 예제 항목은 다음과 같습니다: "내 지능은 내가 별로 바꿀 수 없는 것", "마지막까지 기다리지 않고 바로 일을 끝냈다", "어려울 때도 모든 시험을 잘 볼 수 있다", "다른 사람들의 의견을 얼마나 주의 깊게 들었느냐"는 질문이 각각 나왔다. 우리는 전체 척도에 대한 모든 항목과 5가지 대응 옵션을 온라인 보충 자료의 부록 A(저널 웹사이트에서 이용 가능)에 제시한다. 평균, 표준 편차 및 내부 일관성 신뢰성 추정치는 표 2에 나와 있습니다. 전반적으로 이러한 구조를 측정하는 목적은 학생들의 SEL을 더 잘 이해하고 학교 수준의 데이터 중심 의사결정을 지원하는 것이었다. 이 설문조사의 개발에 대한 자세한 내용(예: SEL 역량이 결정된 방법, 조치를 선택하고 테스트한 방법 등)은 West 등(2018)을 참조하십시오. 
The 25-item SEL survey included four scales measuring growth mindset (n = 4 items), regulation (n = 9 items), self-efficacy (n = 4 items), and social awareness (n = 8 items). Example items for each scale included: “My intelligence is something that I can’t change very much,” “I got my work done right away instead of waiting until the last minute,” “I can do well on all my tests, even when they’re difficult,” and “How carefully did you listen to other people’s points of view?” respectively. We present all items for the complete scales and their five response options in Appendix A of the online supplementary materials (available on the journal website). Means, standard deviations, and internal consistency reliability estimates are in Table 2. Overall, the goal of measuring these constructs was to better understand students’ SEL and to support school-level data-driven decision making. For more information about the development of this survey (e.g., how SEL competencies were determined, how measures were selected and tested, etc.), please see West et al. (2018).

절차들
Procedures

각 만족스러운 행동에 대해 응답자들이 특정 대응 전략(각각 1 또는 0으로 코딩됨)에 참여하는지 여부를 결정했다. 우리는 최종 조사 항목(즉, 항목 25)을 완료하기 전에 조사를 종료함으로써 조기 종료를 조작화했다. 응답자가 마지막으로 완료한 항목 이전에 설문조사에서 하나 이상의 항목이 누락하는 것으로 무응답을 조작화하였습니다. 이 접근법을 통해 비응답자와 초기 종료자를 이중으로 계산하는 것을 피할 수 있었다. 
For each satisficing behavior, we determined whether respondents engaged in the specific response strategy or not (coded as 1 or 0, respectively). We operationalized early termination as ending the survey prior to completing the final survey item (i.e., Item 25). Nonresponse was operationalized as omitting at least one item in the survey prior to a respondent’s last completed item. This approach allowed us to avoid double-counting nonresponders and early terminators.
직선 응답을 식별하기 위해 설문 조사 전체에 걸쳐 10개 항목의 각 순차적 집합에 대한 표준 편차를 분석했다(예: 항목 1-10, 항목 2-11, 항목 3-12 등). 주어진 집합에 대한 표준 편차 0은 학생이 10개 항목 각각에 대해 동일한 반응 옵션을 선택했음을 나타냅니다. 따라서 16개의 가능한 구간(즉, 10개의 순차 항목의 16세트)에서 학생들은 전략을 적어도 한 번 이상 사용할 경우 직선자로서 자격을 얻었다. 마지막으로, 우리는 이 세 가지 이진수 값을 합산하여 전반적인 satisficing(조사 중 어느 시점에서 학생이 satisficing했는지 여부)를 결정했다. 0보다 큰 값은 학생이 조사 중 어느 시점에서 만족했음을 나타낸다. 이러한 계산에 대한 자세한 설명은 온라인 보충 자료(저널 웹사이트에서 제공)의 부록 B를 참조하십시오.
To identify straight-line responding, we analyzed the standard deviation for each sequential set of 10 items across the survey (e.g., Items 1–10, 2–11, 3–12, etc.). Standard deviations of zero for a given set indicated that the student selected the same response option for each of the 10 items. Thus, across the 16 possible intervals (i.e., the 16 sets of 10 sequential items), students qualified as straight-liners if they used the strategy at least once. Finally, we determined overall satisficing—whether a student satisficed at any point during the survey—by summing these three binary values; values greater than zero indicated that a student satisficed at some point during the survey. Please see Appendix B in the online supplementary materials (available on the journal website) for detailed descriptions of these calculations.

사전 등록된 결과
Preregistered Results

가설 1: 전체 만족도
Hypothesis 1: Overall Rate of Satisficing

만족 학생 수를 전체 참가자 수로 나누어 표본의 최소 10%가 만족도 조사에 참여할 것이라는 첫 번째 가설을 실험하였다. 우리의 데이터는 30.36%의 학생들이 적어도 하나의 satisficing 형태에 참여하면서 그 가설을 지지했다. 만족도는 조기종료 3.73%, 무응답 24.99%, 직선응답 5.38% 등이었다. 일부 학생들은 다양한 형태의 satisficing을 보였다(3.26%는 두 가지 형태, 0.14%는 세 가지 형태 모두에 참여했다).
We tested our first hypothesis that at least 10% of the sample would engage in survey satisficing by dividing the number of students who satisficed by the total number of participants. Our data supported the hypothesis with 30.36% of students engaging in at least one form of satisficing. The satisficing included 3.73% early termination, 24.99% nonresponse, and 5.38% straight-line responding. Some students engaged in multiple forms of satisficing (3.26% engaged in two forms, and 0.14% engaged in all three).

가설 2: 측량 영향
Hypothesis 2: Survey Impact

우리는 세 가지 응답 패턴 중에서 직선 응답이 가장 많은 총 조사 항목에 영향을 미칠 것이라는 가설을 세웠다. 단일 항목에만 영향을 미칠 수 있는 무응답 및 조기 종료와 대조적으로, 정의상 한 번만 응답해도 최소 10개의 항목이 포함된다. 결과는 직선으로 선을 그은 학생들이 평균 3.90개의 간격(각 간격은 10개의 잠재적으로 중복되는 항목 집합을 나타낸다; SD = 4.04) 동안 이 행동에 참여했다는 우리의 가설을 뒷받침했다. 이 평균은 거의 13개 항목을 연속으로 동일한 반응 옵션을 선택하는 것에 해당합니다. 이에 비해 평균 무응답은 건너뛴 항목 1.77개에 해당했고, 평균 조기 종료로 3.52개 항목이 조기 종료됐다. 따라서 직설적 응답(각각 5.38% 대비 24.99%)에 비해 무응답 학생이 더 많았음에도, 무응답 관련 항목은 더 적었다.
We hypothesized that out of the three response patterns, straight-line responding would affect the greatest number of total survey items. In contrast to nonresponse and early termination, which might affect as little as a single item, straight-line responding even once implicates a minimum of 10 items, by definition. The results supported our hypothesis in that students who straight-lined engaged in this behavior for a mean of 3.90 intervals (each interval represents a set of 10, potentially overlapping items; SD = 4.04). This average corresponds to selecting the same response option almost 13 items in a row. In comparison, average nonresponse corresponded to 1.77 skipped items, and average early termination resulted in ending 3.52 items early. Thus, even though more students engaged in nonresponse compared with straight-line responding (24.99% compared with 5.38%, respectively), fewer items were implicated by nonresponse.

가설 3a 및 3b: 직선 반응
Hypotheses 3a and 3b: Straight-Line Responding

우리는 첫 번째, 두 번째, 세 번째, 네 번째, 다섯 번째 반응 옵션에 대해 직선 응답이 얼마나 자주 발생했는지에 대한 빈도를 조사함으로써 직선 응답을 한 참가자가 절반에 걸쳐 척도의 오른쪽에서 가장 극단적인 반응 옵션을 선택한다는 가설 3a를 테스트했다. 그런 다음 다섯 번째 반응 옵션(즉, 오른쪽으로 가장 먼 선택)을 사용하여 발생한 직선 반응의 백분율을 계산했습니다. 참가자들은 16개 구간에 걸쳐 46.02%의 경우에 가장 오른쪽 응답 옵션을 선택했는데, 이는 우리가 예측한 대다수에 미치지 못하는 것입니다. 두 번째로 많이 선택된 옵션은 중간 옵션(M = 29.97%)이었다.
We tested Hypothesis 3a, that participants who straight-lined would select the most extreme response option on the right-hand side of the scale over half the time, by examining the frequencies for how often straight-line responding occurred for the first, second, third, fourth, and fifth response options. We then calculated the percentage of straight-line responding that occurred using the fifth response option (i.e., the choice farthest to the right). Participants selected this response option 46.02% of the time across the 16 intervals–short of the majority we had predicted. The second most frequently selected option was the middle option (M = 29.97%).
직선 응답이 4개 척도(가설 3b)에 대한 학생들의 평균 점수에 영향을 미치는지 여부를 조사하기 위해 척도별로 일련의 2-표본 t 검정을 수행했습니다. 우리는 전체 샘플을 충실도가 높은 샘플(즉, 직선을 긋는 응답자를 제외한 후 샘플)과 비교했다.1 
To examine whether straight-line responding affected students’ mean scores for the four scales (Hypothesis 3b), we conducted a series of two-sample t tests for each scale. We compared the complete sample with the high-fidelity sample (i.e., the sample after excluding respondents who straight-lined).1
효과 크기는 작았지만, 전체 표본은 고출실도 표본보다 높은 평균 점수를 보였다: 규정, t(796909) = 9.68, p <.001, 99% CI 0.01, 0.02, Cohen's d = 0.02, 자가 진단, t(794575) = 16.19, p <.001, 99% CI [0.03, 0.04], 사회적 인식, C.03, C.03, C.03, 14.03, C.03, C.03, C.03, C.93).03, C.03, C.03, C.03oen's d = 0.03. 성장 마인드 척도에서도 동일한 패턴이 나타났지만 항목은 역점을 받았다. 척도의 맨 오른쪽에 일직선으로 그은 학생들(즉, 반응 옵션 5를 선택)은 성장 사고방식의 개념적 반대를 지지했다. 따라서 역점수 성장 마인드 항목을 고려한 결과 성장 마인드 점수는 다른 척도의 패턴을 반영하는 것으로 나타났다. 구체적으로, 완전표본은 고순도표본 t(794700) = -6.51, p < .001, 99% CI [-0.02, -0.01], Cohen's d = 0.01 (표 2 참조)보다 낮은 점수를 받았다. 요약하면, 학생들이 직선 반응에 참여하는 방식은 각 구성에 대한 전체 평균 점수에 영향을 미쳤다. 
Although the effect sizes were small, the complete sample had higher mean scores than the high-fidelity sample for: regulation, t(796909) = 9.68, p < .001, 99% CI 0.01, 0.02], Cohen’s d = 0.02; self-efficacy, t(794575) = 16.19, p < .001, 99% CI [0.03, 0.04], Cohen’s d = 0.04; and social awareness, t(795008) = 14.93, p < .001, 99% CI [0.02, 0.03], Cohen’s d = 0.03. The same pattern emerged for the growth mindset scale; however, the items were reverse scored. Students who straight-lined on the far right-hand side of the scale (i.e., selecting Response Option 5) endorsed the conceptual opposite of growth mindset. Thus, after accounting for the reverse-scored growth mindset items, we found that the growth mindset scores mirrored the pattern of the other scales. Specifically, the complete sample had lower scores than the high-fidelity sample, t(794700) = −6.51, p < .001, 99% CI [−0.02, −0.01], Cohen’s d = 0.01 (see Table 2). In sum, the pattern of how students engage in straight-line responding affected the overall mean scores for each construct.

가설 4: 만족자 식별
Hypothesis 4: Identifying Satisficers

로지스틱 회귀 분석을 사용하여 남학생이 여학생보다 만족할 가능성이 높다는 가설을 검증했습니다. 결과에 따르면 satisficing을 할 확률은 남성이 여성보다 16% 높았다(B = 0.15, SE = 0.01, 승산비 = 1.16, 99% CI [1.14, 1.18]).
We used a logistic regression to test our hypothesis that male students would be more likely to satisfice than female students. Results showed that the odds of satisficing were 16% higher for males than females (B = 0.15, SE = 0.01, odds ratio = 1.16, 99% CI [1.14, 1.18]).

탐색 결과
Exploratory Results

전체적으로, 우리의 결과는 우리의 샘플에서 satisficing하는 학생의 만연성을 보여주었고, 30% 이상의 학생들이 이런 행동에 관여하고, 직선적인 응답을 하는 것은 가장 많은 항목을 의미한다. 이러한 사전 등록된 가설은 중요한 확인 데이터를 제공하는 동시에 일련의 탐색적 분석을 통해 추구한 추가적인 질문을 제기하기도 했다. 
  • 먼저 성별 외에 다른 학생 특성이 전반적인 만족도를 예측하는지 살펴보았다.
  • 둘째, 직선 응답이 학생 하위 그룹 비교와 설문 조사의 심리학적 속성(예: 크론바흐의 알파 계수)에 미치는 영향을 추가로 조사했다
  • 마지막으로, 우리는 학생들이 이러한 형태의 만족에 가장 자주 참여한다는 점을 고려하여 무응답을 더 자세히 탐구했다.2
우리는 향후 연구를 위한 보다 정교한 권장 사항을 개발하기 위해 이러한 탐색적 분석을 추구했다.
Overall, our results showed the pervasiveness of student satisficing in our sample, with over 30% of students engaging in some form of satisficing and straight-line responding implicating the greatest number of items. While providing important confirmatory data, these preregistered hypotheses also raised additional questions that we pursued through a series of exploratory analyses. Specifically,
  • we first explored whether other student characteristics in addition to gender predicted overall satisficing.
  • Second, we further examined the effects of straight-line responding on student subgroup comparisons and the psychometric attributes of the survey (e.g., Cronbach’s alpha coefficients.)
  • Last, we explored nonresponse in further detail, given that students engaged in this form of satisficing most frequently.2 
We pursued these exploratory analyses to develop more refined recommendations for future research.

탐색적 분석: 전반적인 만족도
Exploratory Analyses: Overall Satisficing

로지스틱 회귀 모형을 적합하여 다른 학생 특성도 만족스러운 설문 조사를 예측했는지 여부를 조사합니다. 성별 외에 인종/민족, 학년, 영어 학습자 상태, 장애 상태 학생, 무료 또는 할인된 가격의 점심 자격, 정학을 포함했습니다. 결과에 따르면 유색인종 학생들은 6%, 무료 또는 할인된 점심식사를 할 수 있는 학생들은 8%, 영어 학습자로 분류된 학생들은 22%, 그리고 장애가 있는 학생들은 45%의 만족 가능성이 증가했다저학년 학생들은 만족할 확률이 13% 감소했다. 정학 횟수는 학생들의 만족도를 예측하지 못했다(표 3 참조). 
We fit a logistic regression model to examine whether other student characteristics also predicted survey satisficing. In addition to gender, we included race/ethnicity, grade, English Language Learner status, student with a disability status, free or reduced-price lunch qualification, and suspensions. Results indicated that odds of satisficing increased 6% for students of color, 8% for students qualifying for free or reduce price lunch, 22% for students classified as English language learners, and 45% for students with disabilities. Odds of satisficing decreased by 13% for students in younger grades. The number of suspensions did not predict student satisficing (see Table 3).

탐색적 분석: 직선 응답
Exploratory Analyses: Straight-Line Responding

직선응답은 다른 어떤 형태의 만족도보다 총조사 항목에 더 많은 영향을 미쳤고 4개 척도에 대한 학생들의 평균 점수에 영향을 미쳤다는 점을 고려하여, 우리는 이러한 구체적인 만족도 형태에 대한 몇 가지 후속 질문을 추구했다. 우리는 잠재적인 성별 차이, 크론바흐의 알파 및 상관 계수의 차이, 직선 응답 패턴에 초점을 맞추었다. 
Given that straight-line responding affected more total survey items than any other form of satisficing and affected students’ mean scores for the four scales, we pursued several follow-up questions for this specific form of satisficing. We focused on potential gender differences, differences in Cronbach’s alpha and correlation coefficients, and the pattern of straight-line responding.

성별 차이
Gender differences

남학생들이 여학생들보다 더 자주 satisficing한다는 사전 등록된 연구 결과를 추적하기 위해, 우리는 연구자들이 만족을 설명하지 못한다면 이러한 경향이 성별 차이를 강조하거나 약화시킬 수 있을지 궁금했다. 구체적으로, 우리는 전체 샘플과 충실도가 높은 샘플(즉, 직선으로 선 학생들을 제거하기 전과 후)을 비교했다. 결과는 직선제를 실시한 학생들을 제거한 후 평균 점수가 0.01점에서 0.02점 사이에서 변화한 것으로 나타났다(예: 여학생들의 평균 자기 조절 점수는 4.16점에서 4.14점으로 떨어졌다). 그러나, 평균 점수가 변경되었음에도 불구하고, 여학생과 남학생의 차이의 크기는 완전 표본에 기초한 분석인지 아니면 충실도가 높은 표본에 기초한 분석인지에 관계없이 일관성을 유지했다. 여학생들은 남학생들보다 자기조절(Cohen's d = 완전체 0.28, 고충실체 0.27), 항목을 역선택한 후 성장 마인드(Cohen's d = 완전체 0.04, 고충실체 0.03), 사회적 인식(Cohen's d = 완전체 0.22, 고충실체 0.22)이 더 높게 나타났다. 대조적으로, 남학생들은 여학생들보다 더 높은 자기 효능감을 보였다(완전한 경우에는 0.08, 고충실도 응답 경우에는 0.10). 
To follow up on our preregistered finding that male students satisficed more often than their female counterparts, we wondered if this tendency might accentuate or mute gender differences if researchers failed to account for satisficing. Specifically, we compared the complete sample and the high-fidelity sample (i.e., before and after removing students who straight-lined). Results indicated that mean scores changed between 0.01 and 0.02 points after removing the students who straight-lined (e.g., mean self-regulation scores for female students fell from 4.16 to 4.14). However, even though the mean scores changed, the magnitude of differences between female and male students remained consistent regardless of whether analyses were based on the complete sample or the high-fidelity sample.
  • Female students reported higher
    • self-regulation (Cohen’s d = 0.28 for complete, 0.27 for high fidelity),
    • growth mindset after reverse-scoring the items (Cohen’s d = 0.04 for complete, 0.03 for high fidelity), and
    • social awareness (Cohen’s d = 0.22 for complete, 0.22 for high fidelity) than male students.
  • In contrast, male students reported higher
    • self-efficacy than female students (Cohen’s d = 0.08 for complete, 0.10 for high fidelity).

Cronbach의 알파 및 상관 계수
Cronbach’s alpha and correlation coefficients

둘째, 우리는 Feldt의 (1969) W 통계량을 사용하여 Cronbach의 알파 계수를 비교했다. 표 2에서 알 수 있듯이 성장 마인드, 조절, 자기효능감, 사회적 인식에 대한 알파 계수는 충실도가 높은 표본에 비해 전체 표본이 .01에서 .02 사이 더 높았다. 이러한 결과는 0.001 미만의 p 값에 해당한다(표 2 참조). 
Second, we compared Cronbach’s alpha coefficients by using Feldt’s (1969) W statistic. As Table 2 shows, the alpha coefficients for growth mindset, regulation, self-efficacy, and social awareness were between .01 and .02 higher for the complete sample as compared with the high-fidelity sample; these findings correspond to a p value of less than .001 (see Table 2).
셋째, Fisher's z를 사용하여 전체 표본과 충실도가 높은 표본 간의 상관 계수를 비교했습니다. 성장 마인드와 조절(z = -12.65), 자기 인식(z = -13.23), 사회 인식(z = -5.12)의 상관관계는 고순도 표본보다 완전 표본에서 더 높았다. 자기인식(z = 13.20)과 사회인식(z = 13.16)의 규제 상관관계와 자기인식과 사회인식의 상관관계(z = 21.80)를 분석할 때도 동일한 양상이 나타났다. 모든 상관관계는 p < .001에서 유의했다(표 4 참조). 결론적으로, 내부 일관성과 척도 간 상관관계에 대한 완전 표본과 고충실도 표본 간의 차이는 작았다. 
Third, we used Fisher’s z to compare the correlation coefficients between the complete sample and the high-fidelity sample. Correlations for growth mindset with regulation (z = −12.65), self-efficacy (z = −13.23), and social awareness (z = −5.12) were higher for the complete sample than the high-fidelity sample. The same pattern emerged when examining the correlations for regulation with self-efficacy (z = 13.20) and social awareness (z = 13.16), as well as the correlation between self-efficacy and social awareness (z = 21.80). All correlations were significant at p < .001 (see Table 4). In sum, the differences between the complete and high-fidelity samples for internal consistency and correlations between scales were small.

직선 응답 패턴
Pattern of straight-line responding

지금까지, 우리의 데이터는 직선 응답이 평균 점수, 척도 신뢰성 및 상관 계수에 영향을 미친다는 것을 시사했다. 이것은 설문 조사에서 학생들이 가장 자주 직선을 긋는 부분(예: 학생들이 피곤해하고 설문 조사가 끝날 때쯤 직선으로 응답하기 시작할 수 있음)에 대한 질문을 제기한다. 그렇지 않으면 전체 설문조사를 완료한 모든 학생(즉, 항목을 누락하거나 설문조사를 일찍 종료하지 않은 학생을 포함)을 대상으로 응답 패턴을 조사했다. 우리는  
  • 첫 번째 간격 이후 학생 직선(a)이 감소했고,
  • (b) 다음 13 간격 동안 상당히 일정하게 유지되었지만
  • (c) 조사의 마지막 세 간격 동안 증가했다는 것을 발견했다(그림 1 참조).
Thus far, our data suggested that straight-line responding affected mean scores, scale reliability, and correlation coefficients. This raises the question of where, within the survey, students straight-lined most frequently (e.g., students may get fatigued and start straight-line responding toward the end of the survey). We examined the response pattern for all the students who otherwise completed the full survey (i.e., we included the students who did not omit items or end the survey early). We found that student straight-lining
  • (a) decreased after the first interval,
  • (b) remained fairly consistent for the next 13 intervals
  • but (c) increased during the last three intervals of the survey (see Figure 1).

탐색적 분석: 무응답
Exploratory Analyses: Nonresponse

사전 등록된 분석에 따르면 직선 응답은 무응답보다 설문 항목 수가 더 많은 것으로 나타났다. 그러나, 무응답이 대부분의 학생들이 참여하는 만족스러운 행동이라는 점을 고려하여, 우리는 (a) 누락된 데이터가 4개 척도의 평균 점수에도 영향을 주는지 여부와 (b) 누락된 데이터의 패턴을 조사하기 위해 두 가지 탐색적 분석을 추구했다. 
Our preregistered analyses indicated that straight-line responding implicated a greater number of survey items than nonresponse. However, given that nonresponse was the satisficing behavior most students engaged in, we pursued two exploratory analyses to examine (a) whether missing data also affected mean scores for the four scales and (b) the pattern of missing data.

평균 차이
Mean differences

네 가지 척도에 걸쳐 완전 표본과 충실도가 높은 표본 간에 차이가 있는지 조사했습니다. 결과는 전체 표본이 높은 충실도 표본에 비해 규제 및 성장 사고방식에 대한 평균 점수가 유의하게 낮다는 것을 보여주었다. 그러나 차이의 크기는 규제(Cohen's d = 0.06)와 성장 사고방식(Cohen's d = 0.03) 모두에서 미미했다. 평균 차이는 자기효능감과 사회적 인식에서 훨씬 더 작았지만, 표본 크기를 고려할 때 이러한 차이도 상당했다. 전체 표본은 고순도 표본(Cohen's d = 0.01)보다 낮은 자기 만족도를 보였습니다. 사회적 인식은 거의 차이가 없었다(Cohen's d = 0.001). 표 5를 참조하십시오. 
Across the four scales, we examined whether there were differences between the complete and high-fidelity3 samples. Results indicated that the overall sample had significantly lower mean scores for regulation and growth mindset than the high-fidelity sample. However, the magnitude of the differences was modest for both regulation (Cohen’s d = 0.06) and growth mindset (Cohen’s d = 0.03). The mean differences were even smaller for self-efficacy and social awareness, although given our sample size, these differences also were significant. The overall sample had lower self-efficacy than the high-fidelity sample (Cohen’s d = 0.01). There was almost no difference for social awareness (Cohen’s d = 0.001). See Table 5.

무응답 패턴
Pattern of nonresponse

직선 응답 패턴을 탐구하는 분석과 유사하게, 우리는 이 분석에서 샘플을 전체 설문 조사를 완료한 학생(즉, 항목 25에 응답한 학생)으로 제한했다. 따라서 그림 2는 조사의 처음 24개 항목에 걸쳐 누락된 데이터의 패턴을 보여줍니다. 값은 각 항목에 대한 결측 반응의 총합을 나타냅니다. 이 패턴은 학생들이 일반적으로 하반기에 비해 상반기에 더 많은 항목을 누락했다는 것을 시사한다. 이번 조사에서 가장 많이 누락된 항목은 10번과 11번으로 각각 13,627건과 10,104건의 누락 응답이 있었다. 두 항목 모두 역채점 성장 마인드 항목이었다. 
Similar to our analysis exploring the pattern of straight-line responding, we restricted our sample in this analysis to the students who completed the full survey (i.e., those who responded through to Item 25). Thus, Figure 2 shows the pattern of missing data across the first 24 items in the survey. The values represent the total missing responses for each item. The pattern suggests that students generally missed more items in the first half of the survey compared with the second half. The two most-missed items in the survey were Items 10 and 11, where there were 13,627 and 10,104 missing responses, respectively. Both items were reverse-scored growth mindset items.

논의
Discussion


이 기사에서, 우리는 고도로 훈련된 통계학자의 도움 없이도 다른 사람들이 이러한 단계를 쉽게 복제할 수 있도록 만족도를 정의하고 계산하기 위해 의도적으로 간단한 접근법을 취했다. 광범위한 satisficing에도 불구하고 데이터는 조기 종료, 무응답 및 직선 응답에 놀라울 정도로 robust한 것으로 나타났다. 우리는 우리의 연구 결과와 연구자, 실무자 및 정책 입안자가 응답자의 만족에 대응하여 무엇을 할 수 있는지 논의한다. 

In this article, we took a deliberately simple approach to defining and calculating satisficing so that others might easily replicate these steps, even without the help of highly trained statisticians. Despite extensive satisficing, the data appeared to be surprisingly robust to early termination, nonresponse, and straight-line responding. We discuss our findings and what researchers, practitioners, and policymakers might do in response to respondent satisficing.

전체 만족도
Total Satisficing

전체적으로 30%가 조금 넘는 표본이 적어도 하나의 satisficing 형태에 관여했다. 그러나 만족도가 응답자의 동기 부여 부족을 반영한다는 점을 고려할 때, 연구자들은 satisficing 행동을 어떻게 조작화 할 것인지를 고려하는 것이 중요하다. 우리는 우리의 정의에 대해 포괄적인 접근법을 취했다. 특히, 응답자들은 한 항목을 생략하는 한 비응답자로서의 자격이 있었다. 이렇게 큰 데이터 세트를 사용하면 일부 의욕적인 학생(예: 실수로 항목을 놓친 학생)을 실수로 잘못 분류하여 잠재적으로 학생 메타데이터를 잘못 표현하는 것이 불가피해 보인다(Soland 등, 2019)., 2019). 반면에, 연속적으로 10개 항목의 직선화 임계값은 차선의 노력을 기울이고 있는 일부 응답자(예: 9개 항목에 대해 직선화한 응답자)를 제외했을 가능성도 있다. 자체 데이터 만족도를 조사하는 연구자, 실무자 및 정책 입안자는 특정 컨텍스트에 대한 만족도 및 운영화에 대한 의미 있는 정의를 결정해야 합니다. 
Overall, a little more than 30% of the sample engaged in at least one form of satisficing. Given that satisficing reflects a lack of respondent motivation, however, it is important for researchers to consider how they operationalize satisficing behaviors. We took an inclusive approach to our definitions. In particular, respondents qualified as nonresponders as long as they omitted one item. With a data set this large, it seems inevitable that we inadvertently miscategorized some motivated students (e.g., those who missed an item by accident), thereby potentially misrepresenting student metadata (Soland et al., 2019). On the other hand, it is also possible that our straight-lining threshold of 10 items in a row excluded some respondents (e.g., those who straight-lined for nine items) who were putting forth suboptimal effort. Researchers, practitioners, and policymakers examining satisficing in their own data will need to determine meaningful definitions of satisficing and operationalizations for their specific contexts.

설문조사 데이터에 미치는 영향
Impact on Survey Data

세 가지 만족행동 중 학생 직선응답이 전체 설문항목(평균 13개 항목)에 가장 많은 영향을 미쳤고, 무응답 및 조기종료(각각 1.77개, 3.52개 항목)에 비해 영향을 미쳤다. 우리는 이번 조사에 성장 마인드를 측정하는 역점수 항목들이 포함되어 있었기 때문에 직선으로 선 학생들이 자신의 태도를 정확하게 보고하지 않았다고 합리적으로 확신한다. 따라서 오른쪽 대응 옵션은 고정된 사고방식, 즉 성장 사고방식의 개념적 반대를 의미했습니다. 직선 응답자 중 성장 마인드 점수가 가장 낮은 학생은 자기효능감과 규제가 가장 높은 것으로 조사됐다. 이러한 결과는 더 강한 성장 마인드셋과 더 높은 자기효능감을 연결하는 동기부여 연구와 일치하지 않을 것이다(Deck & Master, 2009). 
Of all three satisficing behaviors, student straight-line responding affected the greatest number of total survey items (almost 13 items on average) compared with nonresponse and early termination (1.77 and 3.52 items, respectively). We are reasonably confident that the students who straight-lined were not accurately reporting their attitudes because the survey included a set of reverse-scored items measuring growth mindset. The right-hand response option therefore signaled a fixed mindset—the conceptual opposite of growth mindset. Among straight-line responders, our survey results show that the students with the lowest growth mindset scores have the highest self-efficacy and regulation. These findings would be incongruous with the motivation research linking stronger growth mindsets with higher self-efficacy (Dweck & Master, 2009).
또한, 우리 표본에서 한줄긋기를 한 학생들은 거의 절반의 경우(M = 46.02%)을 맨 오른쪽에서 반응 옵션을 선택했기 때문에, 이러한 만족스러운 행동은 네 가지 척도에 걸쳐 학생들의 점수에 영향을 미쳤다. 그러나, 상대적으로 작은 효과 크기는, 통계적으로 유의미하긴 하나, 샘플 간의 차이가 반드시 우리의 연구 결과에 대한 해석에 상당한 위협을 나타내는 것은 아니라는 것을 시사했다. 우리가 미리 지정한 가설에서, 코헨의 d 계수는 0.01에서 0.04 사이로, 일반적으로 "작은" 효과 크기에 대해 예약된 0.20 컷오프 아래로 떨어졌다(코헨, 1988). 우리는 탐색적 분석(Cohen의 d 계수 0.001 ~ 0.06)에 대해 유사한 결과를 얻었다. 이는 일반적으로 우리의 데이터에 대한 실질적으로 다른 해석을 정당화할 정도로 평균이 충분히 다르지 않았음을 나타낸다. 물론 효과 크기의 크기는 연구 맥락에 따라 다양하다. 한 영역에서 작을 수 있는 것은 다른 영역에서 의미 있는 차이를 나타낼 수 있다(Kraft, 2020). 또한, 일부 연구자들은 효과 크기 컷오프가 상대적으로 임의적이며, 대신 효과가 야기할 수 있는 결과의 관점에서 해석되어야 한다고 주장한다(Funder & Ozer, 2019). 따라서 로컬 컨텍스트는 차이가 의미 있는 경우 안내하는 데 도움이 될 수 있습니다. 
Moreover, because the students who straight-lined in our sample selected the response option on the far right-hand side almost half the time (M = 46.02%), this satisficing behavior affected students’ scores across the four scales. Yet, the relatively modest effect sizes suggested that, while significant, the differences between samples did not necessarily represent a substantial threat to interpretations of our findings. In our prespecified hypotheses, the Cohen’s d coefficients ranged from 0.01 to 0.04, falling below the 0.20 cutoff typically reserved for “small” effect sizes (Cohen, 1988). We obtained similar findings for our exploratory analyses (Cohen’s d coefficients from 0.001 to 0.06)—indicating that, in general, the means were not sufficiently different to warrant substantially different interpretations of our data. Of course, the magnitude of effect sizes ranges across research contexts—what may be small in one domain may represent a meaningful difference in others (Kraft, 2020). Moreover, some researchers argue that effect size cutoffs are relatively arbitrary and should instead be interpreted in terms of the consequences that the effects could cause (Funder & Ozer, 2019). Local contexts can therefore help guide when the differences are meaningful.
우리는 또한 직선 응답과 무응답의 패턴을 조사했다. 설문조사가 [끝날 무렵에는 한줄긋기 응답]이 더 많이 발생한 반면, [전반기에는 무응답]이 더 빈번하게 발생했다. 주목할 점은 가장 많이 놓친 항목이 성장 마인드 항목이었다는 점이다. 이는 본 SEL 설문조사에 사용된 성장 사고방식 척도와 역점수 항목의 포함에 대한 우려와 일치한다(Meyer et al., 2018). 역스코어링 항목 이외의 조사 설계 전략을 사용하면(예: 한 번에 한 척도에서 모든 항목을 제시하는 대신 조사 전체에 걸쳐 서로 다른 구조의 항목을 삽입하는 것) 역스코어링 항목에 필요한 인지 복잡성을 추가하지 않고도 응답자 satisficing를 최소화하는 데 도움이 될 수 있다(Gehlbach & Barge, 2012) 항목(예: Benson & Hocevar, 1985). 
We also examined patterns in straight-line responding and nonresponse. Straight-line responding occurred more toward the end of the survey, whereas nonresponse happened more frequently in the first half. Of note is that the most frequently missed items were growth mindset items. This aligns with concerns about the growth mindset scale used in this SEL survey and its inclusion of reverse-scored items (Meyer et al., 2018). Using survey design strategies other than reverse-scoring items (e.g., interspersing items from different constructs throughout the survey rather than presenting all items from one scale at a time) may help to minimize respondent satisficing (Gehlbach & Barge, 2012) without adding the cognitive complexity required by the wording of reverse-scored items (e.g., Benson & Hocevar, 1985).
요약하면, 본 연구의 결과는 설문조사 데이터 사용자가 만족도가 각 표본의 데이터 품질에 어떤 영향을 미치는지 인지할 필요가 있지만, 이러한 행동이 satisficing이 높은 경우에도 전체 결과의 무결성을 항상 위협하지는 않을 수 있음을 시사한다. 이 기사에서 평가한 세 가지 전략(예: Robinson-Cimpian, 2014; Steedle et al., 2019)을 초과하는 만족도 감지를 위한 추가 또는 다른 전략을 원하는 설문조사 데이터 사용자는 응답 행동이 특정 교육 환경에서 데이터에 어느 정도 영향을 미치는지 유사하게 결정해야 한다. 
In sum, the findings of this study suggest that although survey data users need to be aware of how satisficing affects data quality in their respective samples, these behaviors may not always threaten the integrity of the overall results even when rates of satisficing are high. Users of survey data who want additional or different strategies for detecting satisficing that extend beyond the three assessed in this article (e.g., strategies described in Robinson-Cimpian, 2014; Steedle et al., 2019) will need to similarly determine to what extent the response behaviors affect the data in their specific educational context.

응답자 특성
Respondent Characteristics

만족이 데이터 품질에 미치는 영향을 조사하는 것 외에도 satisficing이 샘플의 특성을 체계적으로 변화시킬 수 있는지도 조사했다. 표본에서 만족자를 제거하여 응답자 동기 부여 문제를 해결하는 것은 특정 집단이 다른 집단보다 더 높은 비율로 satisficing을 하는 경우 대표성이 없는 표본으로 이어질 수 있다. 현재 표본에서는 남학생이 여학생보다 만족도가 높았다. 탐색적 분석은 또한 인종/인종, 언어 상태 및 장애 상태를 만족과 관련된 다른 요인으로 식별했다. 이러한 결과는 Soland(2018)가 학생 하위 그룹 간 만족도 차이를 발견한 결과를 반영한다. 구체적으로 학업진보도 평가에서 남학생과 유색인종 학생들의 빠른 추측 비율이 높았다(Soland, 2018). 그러나 현재 연구와 다른 연구(예: Soland, 2018)에서 하위 그룹 차이의 작은 특성 때문에 정책 결정은 학생의 하위 그룹을 다르게 다룰 필요가 없을 수 있다. 그러나 언급한 바와 같이, 연구자들은 연구 결과를 발표하기 전에 만족자를 제거할 필요가 있는지 여부를 확인하기 위해 만족도가 자신의 데이터 품질에 어느 정도 영향을 미치는지 결정해야 한다.
In addition to investigating the impact of satisficing on data quality, we also examined whether satisficing might change the nature of the sample in systematic ways. Addressing the problem of respondent motivation by removing satisficers from the sample could lead to unrepresentative samples if certain groups satisfice at higher rates than others. In the present sample, male students were more likely to satisfice than their female counterparts. Exploratory analyses also identified race/ethnicity, language status, and disability status as other factors associated with satisficing. These findings reflect those obtained by Soland (2018), who found differences in satisficing across student subgroups. Specifically, on the Measures of Academic Progress assessment, male students and students of color engaged in higher rates of rapid guessing (Soland, 2018). However, because of the modest nature of the subgroup differences in the current study and elsewhere (e.g., Soland, 2018), policy decisions may not need to treat different subgroups of students differently. As noted, however, researchers will need to determine to what extent satisficing affects their own data quality in order to ascertain whether it is necessary to remove satisficers prior to presenting their findings.

연구자, 실무자 및 정책 입안자를 위한 권장 사항
Recommendations for Researchers, Practitioners, and Policymakers

본 연구를 바탕으로 학생 설문조사 데이터를 처리하기 위한 5가지 지침을 제안한다. 
  • 첫째, 연구자, 실무자 및 정책 입안자는 설문 조사의 맥락 내에서 의미 있는 satisficing 의 정의를 결정해야 한다. 만족스럽게 정의하고 운영하기 위해 비교적 포괄적인 접근 방식을 취했지만, 일부 구역에서는 더 보수적인 접근 방식(예: 무응답을 하나가 아닌 네 개의 누락된 항목으로 정의)이 필요할 수 있다. 다행히도 다양한 정의를 만족하고 영향을 검토하는 테스트는 비교적 저렴한 비용으로 추가 분석을 수행하는 데 걸리는 시간에 불과합니다. 데이터 분석가들이 만족의 영향을 더욱 탐구함에 따라, 우리는 주어진 맥락에서 가장 합리적인 것이 무엇인지 보기 위해 다양한 정의를 테스트하는 것을 추천한다.
Based on this study, we recommend the following five guidelines for handling student survey data. First, researchers, practitioners, and policymakers will need to determine meaningful definitions of satisficing that make sense within the context of their surveys. While we took a relatively inclusive approach to defining and operationalizing satisficing, some districts might require more conservative approaches (e.g., defining nonresponse as four missed items rather than one). Fortunately, testing different definitions of satisficing and examining the repercussions is relatively low cost—merely the time taken to conduct additional analyses. As data analysts further explore the impact of satisficing, we recommend testing various definitions to see what is most sensible for a given context.
  • 둘째, satisficing 행동이 결과 해석에 얼마나 영향을 미치는지 평가하기 위해 연구자, 실무자 및 정책 입안자가 satisficing 유무에 관계없이 데이터를 검토할 것을 권장한다. CORE 맥락 내에서, 직선 응답과 무응답은 결과의 주요 해석을 바꾸지 않았다. 예를 들어, 직선 응답 표본과 높은 충실도 표본 간의 차이의 크기는 매우 작았습니다. 그러나, 교육의 맥락 의존적 특성을 고려할 때, 이러한 결과는 교육 환경에 따라 달라질 수 있다. 또한 다른 유형의 분석은 다른 방식으로 영향을 받을 수 있다. 특정 부분군 비교(예: 학년 수준, 학교, 성장률 등) 또는 항목 구조와 관련된 분석(예: 요인 분석 기법)은 만족자의 포함 또는 제외에 더 민감할 수 있다. 따라서, 우리는 설문 조사 데이터 사용자가 다양한 분석 및 설정에 걸쳐 자체 설문 조사에서 만족도가 척도에 얼마나 영향을 미칠 수 있는지 조사할 것을 권장한다. 만족으로 인한 결과적 차이가 언제, 어디서, 왜 더 많이 나타나는지를 배우는 것이 앞으로 나아가는 중요한 지식이 될 것이다.
Second, we recommend that researchers, practitioners, and policymakers examine their data with and without satisficers in order to evaluate how much the satisficing behaviors affect interpretation of the findings. Within the CORE context, straight-line responding and nonresponse did not change the main interpretation of findings. For example, the magnitude of differences between the straight-line responding and high-fidelity samples was quite small. However, given the context-dependent nature of education, results like these might vary across educational settings. Furthermore, other types of analyses might be affected in different ways. Perhaps certain subgroup comparisons (e.g., grade levels, schools, growth rates, etc.) or analyses involving item structure (e.g., factor analytic techniques) are more sensitive to the inclusion or exclusion of satisficers. Thus, we encourage users of survey data to examine how satisficing may affect scales in their own surveys across a range of analyses and settings. Learning when, where, and why there are more consequential differences in findings due to satisficing will be important knowledge moving forward.
  • 셋째, satisficing하는 모든 학생으로부터 모든 데이터를 제외하지 않는 것이 좋습니다. 대신, 연구자, 실무자 및 정책 입안자는 결함이 있는 데이터만 제거함으로써 더 많은 이익을 얻을 수 있다(즉, 목록별 삭제보다는 사례별 삭제). 특히, 직선 응답과 무응답은 학생들의 평균 점수에 영향을 미칠 수 있으므로, 데이터 분석가들이 이러한 두 가지 반응 패턴에 초점을 맞출 것을 제안한다. 결함이 있는 데이터를 제거하면 분석가들이 데이터 중심의 의사 결정을 지원할 때 잠재적으로 손상된 데이터와 함께 양질의 데이터를 낭비하지 않도록 하는 데도 도움이 될 것입니다. 그러나 이 과정의 일부로, 우리는 또한 데이터 분석가들이 먼저 데이터를 제외하는 것이 표본 모집단의 성격을 크게 바꾸지 않는다는 것을 확인하도록 권장한다(예: 불균형한 수의 특정 인구 통계 그룹을 제거함으로써).
Third, we recommend against excluding all data from every student who satisfices. Instead, researchers, practitioners, and policymakers may benefit more from removing only the flawed data (i.e., case-wise rather than listwise deletion). Specifically, because straight-line responding and nonresponse may affect students’ mean scores, we suggest that data analysts focus on these two response patterns. Removing flawed data will also help ensure that analysts are not throwing quality data away along with the potentially compromised data when working to support data-driven decision making. As part of this process, however, we also encourage data analysts first to confirm that excluding data does not markedly change the nature of the sample population (e.g., by removing a disproportionate number of certain demographic groups).
  • 넷째, 설문조사에 [역스코어 항목을 포함하는 것]은 직선 응답자를 탐지하기 위한 효과적인 전략으로 보일 수 있다. 그러나 우리는 이 전술을 사용하지 않아야 한다고 경고한다. 역스코어 항목은 척도의 신뢰성을 떨어뜨리고 참가자들이 답변하기 어렵다(Benson & Hocevar, 1985; Gehlbach & Brinkworth, 2011; Swain et al., 2008). 대신에 조사 설계자는 [서로 다른 구인의 항목을 분산시켜 삽입]하고(Gehlbach & Barge, 2012) [응답 옵션이 구인 특이적이도록 보장]하여 직선 응답을 완화할 수 있다(Gehlbach & Brinkworth, 2011).
    • 항목을 [interspersing]하면 참가자가 동일하거나 유사한 구조의 항목을 서로 옆에 배치할 때 발생할 수 있는 고정 및 조정과 같은 인지 편향에 참여할 가능성이 줄어든다(Gehlbach & Barge, 2012).
    • 또한 각 항목에 대해 [완전한 레이블이 지정된 응답 옵션]을 포함하고 항목과 응답 옵션 모두에서 동일한 구인별 언어를 사용하면 응답자들이 유사한 질문을 반복적으로 하는 것이 아니라 별개의 현상에 대해 설문조사가 질문한다는 점을 강화할 수 있다(Gehlbach, 2015).

Fourth, including reverse-scored items in a survey may seem like an effective strategy for detecting straight-line responders. However, we caution against using this tactic. Reverse-scored items reduce scale reliability and are difficult for participants to answer (Benson & Hocevar, 1985; Gehlbach & Brinkworth, 2011; Swain et al., 2008). Instead, survey designers can attempt to mitigate straight-line responding by interspersing items from different constructs (Gehlbach & Barge, 2012) and ensuring that response options are construct specific (Gehlbach & Brinkworth, 2011).
  • Interspersing items reduces the chances that participants will engage in cognitive biases, like anchoring and adjusting, which can occur when items from the same or similar constructs are placed next to each other (Gehlbach & Barge, 2012).
  • Furthermore, including fully labeled response options for each item and using the same construct-specific language in both the items and response options can help to reinforce to respondents that the survey is asking about distinct phenomena as opposed to asking similar questions over and over (Gehlbach, 2015).
마지막으로, 설문조사에 인지적 노력을 기울이는 학생들의 동기는 가변적이기 때문에 설문조사를 이용하는 학생들은 설문조사 도구를 사용하기 전에 학생들의 참여를 유도할 것을 권장한다. 미래 연구는 또한 학생들이 만족스러운 행동에 참여할 가능성이 가장 높은 시기와 장소를 더 잘 이해하기 위해 설문 조사(예: 인지 면접을 통해)에서 학생들의 만족 동기의 질적 차이를 탐구함으로써 이익을 얻을 수 있다. 설문조사 과정 초기에 응답자의 동기를 향상시키기 위해 증거 기반 전략(예: Dillman 등, 2014)을 사용하면 학생들이 사용하는 만족스러운 행동의 일부를 줄일 수 있다.
Last, because students’ motivation to put cognitive effort into a survey is malleable, we urge those using survey research to cultivate buy-in from students prior to administering the survey instruments. Future research may also benefit from exploring qualitative differences in students’ motivation to satisfice on surveys (e.g., through cognitive interviewing) to better understand when and where students are most likely to engage in satisficing behaviors. Using evidence-based strategies (e.g., Dillman et al., 2014) to enhance respondent motivation early in the survey process may reduce some of the satisficing behaviors utilized by students.

결론
Conclusion

조사 데이터에 대한 비판이 많다. 이러한 비판은 응답자들이 자신의 태도를 이해하고, 태도를 정확하게 보고하고, 충분한 동기를 가지고 설문에 참여하는 능력에 의문을 제기한다. 낮은 응답자 동기 부여는 이러한 유형의 응답자 행동에 대한 연구자 통제가 부족하다는 점을 고려할 때 데이터 해석에 가장 도전적인 위협이 될 수 있다. 그러나, 우리의 연구 결과는 우리의 샘플에서 satisficing하는 학생들이 널리 퍼져 있음에도 불구하고, 이 관행이 데이터 품질에 미치는 영향은 놀라울 정도로 작아 보였다는 것을 보여준다. 교육의 맥락별 특성 때문에, 우리는 다른 사람들이 자신의 데이터 세트에서 만족하는 설문 조사의 유병률과 영향을 유사하게 결정할 것을 촉구한다. 집단적인 노력을 통해, 우리는 조사 결과가 만족스러운 행동에 얼마나 강력한지 알 수 있다. 이 중요한 작업을 추구하는 연구자, 실무자 및 정책 입안자를 지원하기 위해, 우리는 대규모 데이터 세트에서 만족하는 학생을 정의하고 계산하기 위한 접근 가능한 기반을 제공했다. 우리는 이러한 전략이 궁극적으로 학교가 더 나은 데이터 중심 결정을 내리도록 도우려는 개인들을 촉진하기를 바란다.
Critiques of survey data abound. These criticisms question respondents’ ability to understand their own attitudes, accurately report their attitudes, and engage in surveys with sufficient motivation. Low respondent motivation may present the most uniquely challenging threat to data interpretation given the lack of researcher control over this type of respondent behavior. However, our findings indicate that despite the prevalence of student satisficing in our sample, the impact of this practice on data quality appeared surprisingly small. Because of the context-specific nature of education, we urge others to similarly determine the prevalence and impact of survey satisficing in their own data sets. Through a collective effort, we can learn how robust survey findings are to satisficing behaviors. To support researchers, practitioners, and policymakers pursuing this important task, we have provided an accessible foundation for defining and calculating student satisficing in large-scale data sets. We hope these strategies ultimately facilitate those individuals who are trying to help schools to make better data-driven decisions.

Abstract

Education researchers use surveys widely. Yet, critics question respondents’ ability to provide high-quality responses. As schools increasingly use student surveys to drive local policy making, respondents’ (lack of) motivation to provide quality responses may threaten the wisdom of using questionnaires for data-based decision making. To better understand student satisficing—the practice of suboptimal responding on surveys—and its impact on data quality, we examined its pervasiveness and impact on a large-scale social–emotional learning survey administered to 409,721 elementary and secondary students. Findings indicated that despite the prevalence of satisficing, its impact on data quality appeared more modest than anticipated. We conclude by outlining an accessible approach for defining and calculating satisficing for researchers, practitioners, and policymakers.

 

 

 

"요즘 애들": 밀레니얼 학습자에 대한 우리의 대화 다시 생각하기 (Med Educ, 2019)
“Kids these days”: Reconsidering our conversations about Millennial learners
Glenn Regehr1,2

Juaregui 등이 밀레니얼 학습자의 신화에 대한 그들의 논문에서 지적했듯이, 현재 우리 보건의료계열 학습자들의 "독특한" 교육적 요구를 해결하려는 노력은 "세대적 타자화" 과정이 공식화된 것에 불과할 수 있다. 실제로 우리 보건의료기관 복도와 건강관리일지 페이지에는 현 세대의 학습자들이 '예전 같지 않다'는 다양한 설명이 울려 퍼지는 것 같다. 그리고 적어도 우리의 저널에서는, 이것이 학습자로서 그들이 있는 곳에서 그들을 만나는 것에 대한 우려로 여겨지지만, 이러한 논의의 밑바탕에는 특정한 시선의 굴림과 "요즘 아이들"의 한숨이 깔려 있다는 느낌을 갖지 않을 수 없다. 
As Juaregui et al1 point out in their article on the myth of the Millennial learner, efforts to address the “unique” educational needs of our current health professional learners may be little more than a formalisation of the process of “generational othering”. Indeed, the hallways of our health care institutions and the pages of our health care journals seem to echo with various descriptions of how the current generation of learners is “not like we used to be”. And although, at least in our journals, this is couched as concern for meeting them where they are as learners, one cannot help but have the sense that underlying these discussions is a certain rolling of the eyes and a sigh of “Kids these days”.

물론, 세대 간의 차이는 새로운 것이 아니다. 모든 세대는 다음 세대에 대해 불평한다. 아마도 그의 제자들에 대한 우려가 후손들이 따를 것을 맹세하기로 한 히포크라테스의 결정의 근저에 놓여있을 것이다. 따라서 후아레귀 등의 부름에 귀를 기울여 다음 세대에 대한 이 영원한 불만의 근원을 탐구하고 다음과 같은 질문을 하는 것이 가치 있을 수 있다: 교육자인 우리가 어떻게 대화를 더 생산적인 방향으로 바꾸기 위해 "세대적 겸손"을 호출할 수 있을까요?
Of course, generational othering is not new. Every generation complains about the next. It may be that concerns regarding his own students lay at the root of Hippocrates’ decision to put in writing an oath for future generations to follow. Thus, heeding the call of Juaregui et al,1 it might be valuable to explore the source of this perennial complaint about the next generation and to ask: How might we, as educators, invoke “generational humility” to shift the conversation in a more productive direction?

아마도 (이번 호의 주제와 일치하여) 다가오는 각 세대에 대한 반복적인 불만의 원천 중 하나는 [개인적 안정stability의 신화]일 것이다. 우리는 모두 우리 [자신이 시간이 지남에 따라 연속성을 갖는 일관된 정체성을 가지고 있다]고 생각하기를 좋아한다. 그래서 우리는 우리의 현재 지혜와 행동 상태가 [경험과 시간의 어렵게 얻은 결과]라기보다는, (지금 그러하고, 과거에도 그랬던) 오래 지속된 것이라고 상상한다. 우리는 일반적으로 우리가 현재 학생들을 인식하는 것처럼 미성숙하고 순진하며 자기중심적이었다고 생각하는 것을 어려워한다. 그래서 우리는 그들이 우리가 그들 나이였을 때보다 더 나쁘다고 가정합니다.
Perhaps (in keeping with the theme of this issue of the journal) one source of the recurring complaints about each upcoming generation is the myth of personal stability. We all like to think of ourselves as having a coherent identity that has continuity over time. So we imagine that our current state of wisdom and demeanour is longstanding (as it is, so it always was) rather than being the hard-won result of experience and time. We generally struggle to conceive of ourselves as ever having been as immature, naïve and self-absorbed as we perceive our current students to be, and so we assume that they must be worse than we were at their age.

하지만, 이 후렴구의 또 다른 가능한 출처가 있는데, 그것은 우리의 학습자들뿐만 아니라 전문직 전체를 위해 중요할 수도 있다. 더글러스 애덤스는 그의 저서 "The Salmon of Double"에서 다음과 같이 주장한다:
However, there is another possible source of this refrain, one that might be important not only for our learners, but for the profession as a whole. In his book The Salmon of Doubt, Douglas Adams asserts:

여러분이 태어날 때 세상에 있는 모든 것은 평범하고 평범하며 세상이 돌아가는 방식의 자연스러운 부분일 뿐입니다. 당신이 15살에서 35살 사이에 발명된 것은 무엇이든 새롭고 흥미롭고 혁신적이며, 아마도 그 분야에서 경력을 쌓을 수 있을 것이다. 35세 이후에 발명된 것은 자연스러운 질서에 어긋나는 것이다.
Anything that is in the world when you're born is normal and ordinary and is just a natural part of the way the world works. Anything that's invented between when you're 15 and 35 is new and exciting and revolutionary and you can probably get a career in it. Anything invented after you're 35 is against the natural order of things.

사실, 사회는 끊임없이 변화하고 있다. 우리가 우리 자신과 학생들 사이에서 보는 차이점들은 부분적으로 그러한 [사회적 변화의 반영]이다. 35세 이상의 사람들에게, 그러한 변화는 어렵고 심지어 무서울 수도 있다. 나는 문자 메시지가 어떻게 우리 아이들의 철자와 문법 사용 능력을 파괴하는지에 대한 토론을 들은 적이 있는데, 그 동안 한 연사는 다음과 같은 취지의 논평을 했다:
Society is, in fact, constantly changing. The differences we see between ourselves and our students are, in part, reflections of that societal change. For those of us over the age of 35, such change can be difficult and even frightening. I was once listening to a debate about how texting was destroying our children's ability to spell and use grammar, during which one speaker offered a comment to the effect that:

우리는 항상 변화 속에서 잃어버리고 있는 것에 몰두한다. 왜냐하면 그것이 우리가 편안하게 느끼는 것이기 때문이다. 나는 첫 번째 동물이 바다를 벗어나 육지에 오르려고 발버둥쳤을 때, 다른 물고기가 아가미를 잃을 것이라고 초조해 했을 것이라고 추측한다. 하지만 결과적으로 우리가 얻은 것을 보세요. 
We are always preoccupied with what is being lost in a change because that is what we are comfortable with. I suspect that when the first animal struggled out of the ocean and on to land, the other fish fretted that it would lose its gills. But look what we gained as a result. 

수십 년 동안, 의학은 상당히 발전해 왔다. 이것은 기술적인 관점에서 명백하지만, 전문성의 영역에서도 사실이다. 더 이상 의사가 환자의 남편과 그녀의 참여(또는 최소한의 동의) 없이 최선이라고 생각하는 수술 절차를 논의하는 것은 적절하지 않다고 여겨진다. 그러나 [전문적인 관행의 표준]이 이렇게 바뀌게 된 것이 [모든 전문가들이 어느 날 갑자기 자신들이 하고 있는 일이 비전문적이고, "옳은" 일에 대한 새로운 인식을 수용하려면 변화가 필요하다는 것을 깨달았기 때문에] 발생하지는 않았을 것이다. [변화하는 사회적 기대에 처음으로 반응]하고, 여성에게 자신의 건강 관리에 대해 직접적으로 이야기하기 시작한 의사들은 [전통적인 접근 방식에 익숙해진 사람들]에게는 전문적이지 못한 것으로 보였을 가능성이 더 높다.
Over the decades, medicine has evolved substantially. This is obvious from a technological perspective, but it is also true in the sphere of professionalism. No longer is it considered appropriate for a physician to discuss with a patient's husband the operative procedure they think is best for her without her participation (or at least consent). However, this change in the standards of professional practice is unlikely to have occurred because every member of the profession realised one day that what they were doing was unprofessional and required to be altered to accommodate their new awareness of the “right” thing to do. It is more likely that those physicians who first began to respond to shifting societal expectations and to talk to women directly about their own health care were seen as unprofessional by those who were accustomed to the traditional approaches.

우리가 하는 모든 것이 이제는 "옳기" 때문에, 의료 전문성이 진화를 멈췄다(또는 멈춰야 한다)고 가정하는 것은 어리석은 일이다. 의학이 사회적 요구에 계속 반응하는 상태에 있으려면, 의학이 놓여있는 더 넓은 문화와 함께 진화해야 한다. 그러나 얼마 동안 존재해온 우리들 중 누군가는 자연의 질서에 반하는 것과 같은 변화를 경험할 것이다. 그러므로, 우리는 우리가 사회에 반응할 수 있도록 새로운 세대의 학습자들에게 의존해야 한다. 우리의 불편함에도 불구하고 그 직업을 계속 발전시키는 것은 그들의 책임이다.
It would be foolish to assume that medical professionalism has (or should have) stopped evolving, that we now have everything “right”. If medicine is to stay responsive to societal needs, it must evolve with the broader culture in which it is embedded. Those of us who have been around for some time, however, may well experience such changes as being against the natural order. Hence, we must count on our new generations of learners to keep us responsive to society. It is their responsibility to keep the profession evolving despite our discomfort.

그렇긴 하지만, 전통은 목적을 위해 존재한다. 보존되어야 할 핵심 가치들이 있다. 사실, 건강 전문 교육의 중요한 측면은 우리를 직업으로 만드는 [핵심 가치, 사고 방식과 지식의 전달]입니다. 그래서 교육자로서 우리가 씨름해야 하는 진짜 문제는 [어떻게 하면 변화와 전통 사이의 긴장을 효과적으로 균형을 맞출 수 있는가] 하는 것입니다. 무엇이 진화해야 하고 무엇이 진화해서는 안 되는가? 콜린스와 포라스의 작품을 그들의 책 "Build to Last: 비전을 가진 기업의 성공적인 습관"에서 다음과 같이 말한다:
That said, tradition exists for a purpose. There are core values that ought to be preserved. In fact, a critical aspect of health professional education is the transmission of those core values, the ways of thinking and knowing that make us who we are as a profession. So the real question with which we must grapple as educators is about how we can effectively balance this tension between change and tradition. What should evolve and what should not? Here, we might draw on the work of Collins and Porras in their book
 Built to Last: Successful Habits of Visionary Companies, in which they state:

비전이 있는 기업은 시대를 초월한 핵심 가치와 지속적인 목적(절대 변경되어서는 안 됨)을 운영 관행 및 비즈니스 전략(변화하는 세상에 대응하여 끊임없이 변화해야 함)과 구별합니다.
Visionary companies distinguish their timeless core values and enduring purpose (which should never change) from their operating practices and business strategies (which should be changing constantly in response to a changing world).3

그러나 Collins와 Porras3가 지적했듯이, 우리의 [실천은 우리의 가치의 구현]이며, 때로는 [가치 자체로 오인]되기 때문에 [무엇이 가치이고 무엇이 실천인지를 인식하는 것이 종종 어렵다]. 이런 점에서, 학습자들과 그들이 때때로 시행하는 [불편한 행동]들이 우리의 가장 큰 자산이 될 수 있다. 예를 들어, [복장]은 종종 전문직업성의 문제로 간주된다. 그러나 "전문적으로 옷을 입는" 것이 무엇을 의미하는지는 시간이 지남에 따라 변한다. 한때 전문적인 복장의 특징이었던 넥타이는 현재 많은 상황에서 부적절한 형식적인 것으로 간주된다. 따라서 학생이 반사적으로 행동을 꾸짖기보다는 부적절하게 평상복 차림으로 진료소에 도착했을 때, [어떤 가치가 침해당했다고 느끼는지에 대해 성찰하는 시간]을 갖는 것이 가치 있을 수 있다.

  • 복장 규정은 [전문직 정체성과 개인적 정체성 사이의 경계를 존중하는 것]에 관한 것인가? 
  • 전문가가 어떻게 옷을 입는지에 대한 [환자의 기대를 존중하는 것]에 관한 것인가? 
  • 아니면 [그냥 "맞추기"에 관한 것]인가요? (우리처럼 되고 분명히 "다른" 사람이 되지 않는 것인가요?

As Collins and Porras3 point out, however, it is often difficult to recognise what is a value and what is a practice because our practices are the embodiment of our values, and sometimes are mistaken for the values themselves. In this regard, our learners and the discomfiting behaviours they sometimes enact may be our greatest asset. For example, modes of dress are often seen as an issue of professionalism. Yet what it means to “dress professionally” changes over time. Ties, once a hallmark of professional dress, are in many contexts now considered inappropriately formal (and more than a little questionable from the perspective of infection control). So when a student arrives at one's clinic in inappropriately casual dress, rather than reflexively chastising the behaviour, it might be worthwhile taking the time to reflect on what value one feels has been violated.

  • Is the dress code about respecting the boundaries between professional and personal identity?
  • Is it about respecting patient expectations of how a professional dresses?
  • Or is it just about “fitting in” (being like us and not being obviously “other”)?

일단 대화가 [가치의 수준]으로, 그리고 그것들이 [실제로 어떻게 구체화될 수 있는지에 대한 수준]으로 올라가면, 학생과 지도자 모두에게 흥미로운 학습이 일어날 수 있다. 예를 들어,

  • [학생들]그들의 "불쾌한" 행동에도 불구하고, 또는 심지어는 그런 행동을 통해서, 특정한 가치가 어떻게 보존되고 제정될 수 있는지를 고려하기 시작할 수 있다.
  • [교사들]왜 그들이 그 행동을 그렇게 소중하게 여기는지에 대한 재조사를 통해 그들 자신의 행동을 재고할 수 있다.
  • [모든 사람들]은 그 가치 자체를 더 깊이 이해할 수 있다. 

예를 들어, 복장에 대한 대화는 환자마다 다른 것을 기대하기 때문에 "환자에 대한 존중을 보여주는 것"이 가치에 대한 불충분한 구성이라는 이해로 이어질 수 있다. 따라서 가치의 보다 정교하고 환자 중심적인 버전을 진화시킬 수 있다. 즉, 존중을 보여주는 것(가치의 행위자 중심적인 표현)이 아니라 각 환자가 존중받는 느낌을 갖도록 하는 것이다. 이는 결국 행동 자체에 관한 것이 아니라 개별 환자의 기대에 대한 탐색에 기초한 행동의 효과적인 적응에 관한 것이라는 인식으로 이어질 수 있다.
Once the conversation is raised to the level of values and how they might be embodied in practice, interesting learning can occur for both student and preceptor.

  • Students, for example, can begin to consider how a particular value might be preserved and enacted despite, or even through, their “offending” behaviour.
  • Preceptors can reconsider their own behaviour through a re-examination of why they value that behaviour so dearly.
  • And everyone can come to a deeper appreciation of the value itself.

For example, conversations about dress might lead to an understanding that “showing respect for patients” is an insufficient construction of the value, because each patient expects something different. Thus it is possible to evolve a more sophisticated, patient-centred version of the value: it is not about showing respect (an actor-centred representation of the value), but about ensuring that each patient feels respected. This, in turn, can lead to the realisation that it is not about the behaviour itself, but about effective adaptation of the behaviour based on an exploration of the individual patient's expectations.

아마도 우리가 "요즘 아이들" 순간을 보낼 때 학습자들과 관련된 접근법을 취하는 것이 "세대적 겸손"에 대한 후아레기 등의 권고를 제정할 수 있는 하나의 메커니즘일 것이다. 물론 그러한 접근법을 취하게 되면, 우리는 [내가 옳다는 속편한 마음]은 갖지 못할 수도 있지만, 결과적으로 우리가 얻을 수 있는 것은 무엇인지 살펴봐야 한다. 
Perhaps taking such an approach to engaging with learners when we are having a “kids these days” moment is one mechanism by which we can enact Juaregui et al's recommendation for “generational humility”.
1 Of course, in taking such an approach, we may lose our comfortable sense of self-righteousness, but look what we might gain as a result.

 


Med Educ. 2020 Jan;54(1):10-12. doi: 10.1111/medu.14002.

"Kids these days": Reconsidering our conversations about Millennial learners

Affiliations collapse

1Centre for Health Education Scholarship, Faculty of Medicine, University of British Columbia, Vancouver, British Columbia, Canada.

2Department of Surgery, Faculty of Medicine, University of British Columbia, Vancouver, British Columbia, Canada.

PMID: 31849098

DOI: 10.1111/medu.14002

뒤를 돌아보고 앞으로 나아가기: 의과대학 1학년의 포트폴리오 글쓰기에 대한 메타-성찰(Acad Med, 2018)
Looking Back to Move Forward: First-Year Medical Students’ Meta-Reflections on Their Narrative Portfolio Writings
Hetty Cunningham, MD, Delphine Taylor, MD, Urmi A. Desai, MD, MS, Samuel C. Quiah, MSW, Benjamin Kaplan, Lorraine Fei, MPH, Marina Catallozzi, MD, MSCE, Boyd Richards, PhD, Dorene F. Balmer, PhD, and Rita Charon, MD, PhD 

의학 교육자들은 종종 [성찰적인 글쓰기]를 성찰적 능력을 배양하고 유능하고 동정심 있는 의사들의 전문적 정체성 구축을 촉진하는 귀중한 도구로 언급한다. 일상적인 의학 교육의 엄격함은 종종 학습자들이 그들이 어떤 사람이 되고 있는지에 대한 종적인 관점을 얻는 것을 방해한다. 훈련 중에 번아웃의 상당한 위험과 공감의 감소를 고려할 때, 학생들이 성찰 능력을 육성하는 데 도움을 줄 필요성은 특히 두드러진다. 게다가, 현재 [표준화와 역량의 문서화]에 초점을 맞추고 있는 것은 개인의 재능, 가치, 목표의 개인적 평가를 떨어뜨림으로써 [전인적 직업적 정체성]의 구축에 도전하거나 심지어 대항할 수 있다 —성찰을 지원하기 위한 의도적인 노력이 더욱 필요합니다.
Medical educators frequently cite reflective writing as a valuable tool to foster reflective capacity and to promote construction of a professional identity in competent and compassionate doctors.1–6 The day-to-day rigors of medical education often preclude learners from gaining a longitudinal perspective on who they are becoming. The need to aid students in fostering reflective capacity is particularly salient given the significant risks of burnout and a decrease in empathy during training.7–9 Furthermore, the current focus on standardization and documenting competency may challenge or even counteract the construction of a holistic professional identity by devaluing individuation of personal talents, values, and goals—which, in turn, renders intentional efforts to support reflection all the more necessary.10–13

비록 전세계의 의대들이 이러한 많은 문제들을 해결하는 수단으로 성찰적인 글쓰기를 받아들였지만, 일부 교육자들은 [글쓰기 커리큘럼]이 [역량을 문서화하도록 구조화]되거나, 학생들의 노력이 [환원주의적 관점으로 평가될 때] 성찰 능력을 육성하려는 노력이 역효과를 낼 수 있다는 우려의 목소리를 냈다. 실제로, Ng와 동료들은 전통적인 의료 기관의 "지배적 인식론적 렌즈"와 일치하는 공리적 결과를 제공하기 위해 반사적 글쓰기를 사용할 때의 잠재적인 함정을 웅변했다. 이러한 위험으로부터 보호하기 위해 일부 의학 교육자들은 [서사 의학적 전통에 앵커링된 성찰 도구]를 학생들이 효과적으로 성찰 능력을 개발하고 전체론적인 전문적 정체성을 구축할 수 있는 "개방된 교육 공간"을 육성하는 데 이상적으로 적합한 것으로 보았다.
Although medical schools around the globe have embraced reflective writing as a means of addressing many of these issues,14 some educators have voiced concerns that efforts to foster reflective capacity may be counterproductive when writing curricula are structured to document competencies or when student efforts are evaluated with a reductionist lens.14,15 Indeed, Ng and colleagues15 eloquently described potential pitfalls of using reflective writing to serve a utilitarian outcome consistent with the “dominant epistemological lens” of traditional medical institutions. To protect against these risks, some medical educators have looked to reflective tools anchored in the narrative medicine tradition as ideally suited to foster “open educational spaces,” sometimes called “reflective spaces,” wherein students may effectively develop their reflective capacity and construct a holistic professional identity.16,17

서사 의학
Narrative Medicine

[서사 의학 프레임워크] 내에서, [성찰적 공간]은 [엄격한 교육학적 행동 순서]를 통해 만들어진다. 참가자들은 [창의성과 세부사항]에 대한 관심을 증진시키는 [텍스트 작품을 공유하는 것]으로 시작한다. [공유된 읽기와 해석]은 그룹 내 신뢰를 강화합니다. 다음으로, 읽기를 반향echo하는 [개방형 프롬프트에 대한 응답]으로 글쓰기는 의대생들이 [결정적이고 증거에 기초한 답을 제공하는 것]에서, [실존적이고 모호한 질문을 숙고하는 것]으로 [환원주의]에서 보다 [확장적인 사고 방식]으로 전환하도록 장려한다. 그리고 나서, 참가자들이 방금 [쓴 것을 큰 소리로 읽고 듣는 것]은 서로의 글에 반응하는 것과 함께, 안전과 상호 존중의 분위기에서 창의적인 글의 발견 가능성에 대한 깊은 존경심을 불러일으킨다.  
Within a narrative medicine framework, reflective space is created through a rigorous sequence of pedagogical actions.18 Participants start with a shared close reading of textual works that promote creativity and attention to detail. The shared reading and interpreting enhance trust within the group. Next, writing in response to open-ended prompts that echo the reading encourages medical students to shift from a reductionist to a more expansive way of thinking, from providing definitive, evidence-based answers to pondering existential and ambiguous questions. Then, reading aloud and listening to what participants have just written, along with responding to one another’s writing, inspires deep respect for the discovery potential of creative writing within a climate of safety and mutual respect.18

내러티브 포트폴리오 개발
Development of a Narrative Portfolio

컬럼비아 대학교 Vagelos 의과대학(P&S)의 교육자들은 학생들이 성장하고 공유할 수 있는 안전하고 목적이 있는 탐구적인 성찰 공간을 제공하기 위해 이러한 서사 의학 개념을 통합한 종단적 성찰적 글쓰기 포트폴리오(아래 설명)를 개발했다. 종단적 관점를 향상시키고, 이 포트폴리오 내에서 성찰적 쓰기의 이점을 높이기 위해 Signature Reflection이라는 학기말 메타 성찰 과제를 설계했다. 매 학기마다, [시그니처 리플렉션 과제]는 학생들이 [포트폴리오 항목의 아카이브를 검토]하고, 이전 글에서 [자신에 대해 알게 된 것]에 대해 쓰고, 이 연습을 통해 [자신의 과거와 미래의 자신이 연결]되도록 촉구한다. 이 글은 우리 학교의 종단적 글쓰기 포트폴리오가 만들어낸 '반사적 공간'에 대한 이론적 근거와 교육과정 내용, 그리고 초기 분석에 초점을 맞추고 있다.
Educators at the Columbia University Vagelos College of Physicians and Surgeons (P&S) have developed a longitudinal reflective writing portfolio (described below), incorporating many of these narrative medicine concepts, to provide a safe, purposeful, exploratory reflective space in which students may grow and share. To enhance the longitudinal scope and to augment the benefits of reflective writing within this portfolio, we designed an end-of-semester meta-reflection assignment called the Signature Reflection. Each semester, the Signature Reflection assignment prompts students to review their archive of portfolio entries, to write about what they notice about themselves in their previous writing, and through this exercise, to connect with their former and future selves. This article focuses on the rationale, curriculum content, and early analysis of the “reflective space” created by our school’s longitudinal writing portfolio.

의대 1학기 후 학생들이 작성한 [시그니처 리플렉션]에 대한 우리의 분석은 포트폴리오에 표현된 대로 의대 여정에 대한 학생들의 관점을 보여주는 독특한 창을 제공했다. 우리는 학생들이 자신의 글을 돌아볼 수 있는 기회가 주어졌을 때 그들 스스로 무엇을 볼 수 있는지 배우고 싶었다. 우리는 또한 그들의 말이 포트폴리오의 학습 결과를 밝힐 수 있는지, 그리고 만약 그렇다면, 어떻게 밝혀질지 알고 싶었다. 마지막으로, 우리는 학생들이 전문적인 정체성 구축을 탐구하기 위해 포트폴리오를 사용하는지, 그리고 만약 그렇다면, 어떻게 사용하는지를 이해하고 싶었다.
Our analysis of Signature Reflections written by students after their first semester of medical school offered a unique window into the students’ perspectives on their own medical school journeys as expressed in their portfolios. We wanted to learn what students themselves would see when given the opportunity to look back on their own writings. We also wanted to learn whether their words might shed light on the learning outcomes of the portfolio, and if so, how. Finally, we wanted to understand if students used the portfolio to explore professional identity construction, and if so, how.

맥락
Context

20년 이상 P&S 의대생들은 서사의학 프로그램을 통해, 인문학 내 창의적이고 탐구적인 세미나를 선택하고 참여하도록 했다. 시각적 작업이든 서면 작업이든, 학생들은 부분적으로 이러한 경험에 대해 글을 썼는데, 이는 그들이 면밀한 관찰과 정확하고 상세한 표현의 기술을 강화할 수 있고, 이는 결과적으로 환자 및 동료들에게 참여하고 심지어 연결하는 능력을 강화할 수 있다. P&S 서사의학 선택지에 대한 질적 분석은 학생들이 미래의 의사로서 훈련과 성장에 있어 서사의학의 역할을 인식하고 높이 평가하는 것으로 나타났다.  
For more than 20 years, through the narrative medicine program, medical students at P&S have been required to select and participate in creative and exploratory seminars within the humanities. Whether responding to visual or written works, students have written about these experiences, in part so that they might strengthen their skills of close observation and accurate and detailed representation, which can, in turn, deepen their capacity to attend to and even connect with patients and peers. Qualitative analysis of the P&S Narrative Medicine selectives has indicated that students recognize and appreciate the role of narrative medicine in their training and growth as future physicians.4,19,20

의과대학의 사회적, 행동적, 인문학적, 예술적 교육과정을 향상시키기 위한 [국립보건원 지원 프로젝트]의 맥락에서, 2011년 교수진과 학생들로 구성된 태스크포스가 소집되어, [서사 의학적 전통의 방법]을 사용하여 교육과정 전반에 걸쳐 [성찰적 글쓰기를 증가시키는 방법]을 모색했다. 그 목적은 학생들에게 시간이 지남에 따라 [그들이 되어고 있는 전문직]에 대한 미묘한 시각을 제공하는 것이었다. 그 특별 위원회는 반성문에 관한 문헌을 검토하고 동료 기관들의 경험을 조사하기 위해 1년 동안 매월 만났다. 태스크 포스는 적절한 텍스트와 프롬프트의 선택, 학생 참여 및 신뢰 유지(특히 개인 정보 보호 문제를 고려할 때), 교수진 훈련 및 평가 메커니즘을 고려했다. 결과적으로, 2013년에, P&S는 1년간의 자발적인 파일럿 프로그램에서 배운 후, 들어오는 모든 의대생들을 위한 성찰적인 글쓰기 포트폴리오 요건을 시작했다. 태스크포스는 포트폴리오가 다음을 할 것을 기대했다.

  • 기관이 과거에 서사의학으로 성공했던 경험을 활용하고,
  • 학생들이 종단적 글쓰기 커리큘럼의 혜택을 받을 수 있도록 하고,
  • 학생들에게 시간이 지남에 따라 그들이 어떤 사람이 되고 있는지에 대한 개인화된 관점을 제공한다.

In the context of a National Institutes of Health–supported project to enhance the social, behavioral, humanities, and arts curricula in the medical school, a task force of faculty and students convened in 2011 to look for ways to increase reflective writing throughout the curriculum using methods from the narrative medicine tradition. The aim was to provide students with a nuanced view, over time, of the professionals they are becoming. The task force met monthly for a year to review the literature on reflective writing and to examine peer institutions’ experiences. The task force considered the selection of appropriate texts and prompts, the maintenance of student engagement and trust (especially given privacy concerns), faculty training, and a mechanism for evaluation. As a result, in 2013, after learning from a one-year voluntary pilot, P&S launched a reflective writing portfolio requirement for all incoming medical students. The task force hoped the portfolio would

  • capitalize on the institution’s prior success with narrative medicine;
  • allow students to benefit from a longitudinal writing curriculum; and
  • offer students an individualized view, over time, of who they are becoming.

P&S 포트폴리오
The P&S portfolio

전임상에서 [임상 사회 및 행동 과학]과 [핵심 임상 기술]에 초점을 맞춘 P&S 과정인 Foundations of Clinical Medicine(FCM)에 뿌리를 둔 포트폴리오 목표는 [학생들의 성찰 능력을 강화]하고, [전인적 전문적 정체성의 구축을 촉진]하는 것이다. 이를 위해서는 다양한 기회를 제공하는 종단적 프로그램을 통해, 학생들은 동료와 교수진-학생들의 의견을 반영하여 [유도된 성찰]에 참여하도록 합니다. 우리는 표 1에 포트폴리오의 주요 기능을 설명합니다.
Rooted within Foundations of Clinical Medicine (FCM), a P&S course focused on preclinical social and behavioral sciences and on core clinical skills, the portfolio goals are to reinforce students’ reflective capacity and to foster the construction of a holistic professional identity through a longitudinal program that offers multiple occasions for students to engage in guided reflections with input from peers and faculty–mentors. We describe key features of the portfolio in Table 1.

포트폴리오의 목적이 [글쓰기에서 성찰을 보여주는 것]이 아니라, [글쓰기를 통해 성찰을 배우는 것]이기 때문에, 출품작은 대부분 편집되지 않은 [강의식 서사 의학 연습]이다. 추가적인 글쓰기 프롬프트는 학생들이 시체를 처음 만난 후에 작성된 성찰부터 초기 임상 경험에 대한 고려사항까지 다양하다. 학생들은 포트폴리오 작성 항목을 대학 포트폴리오 웹사이트에 업로드합니다. 학생들은 3학기마다 최소 6개의 글쓰기 항목과 1개의 학기말 메타성찰문을 제출한다; 이후 임상실습생 자격으로 계속 입력해야 합니다. 교수진은 학생들이 P&S에서 학부 의학 교육을 받는 동안 추가적인 자발적인 참가를 제출하도록 장려한다.
Because the purpose of the portfolio is to learn reflection through writing, as opposed to the goal of demonstrating reflection in writing, entries are mostly in-class, unedited narrative medicine exercises. Additional writing prompts are diverse, ranging from reflections written after the students first meet their cadavers to considerations of their early clinical experiences. Students upload their portfolio writing entries to a university portfolio website. Students submit at least six writing entries and one end-of-semester meta-reflection during each of three preclinical semesters; entries continue during subsequent clinical clerkships. Faculty encourage students to submit additional voluntary entries throughout their undergraduate medical education at P&S. 

[교실에서 연습할 때 사용되는 서사적 의학 방법]은 다른 곳에서 잘 설명되었다. 간단히 말해, [서사 의학 전통을 훈련]받은 FCM [교수진-멘토]들이 이끄는 [12~14명의 학생들]로 구성된 [소규모 그룹]이 복잡한 문학 텍스트에 대한 [긴밀한 독서와 토론]에 함께 참여한다. 교수진은 진행 중인 과정 주제(예: 임상적 만남에서 여러 관점에 대한 인식)와 암묵적으로 일치하기 위해 학생들의 의대 경험의 아크arc와 유사한 서사적 텍스트를 선택한다. 단, 교직원은 학생들에게 [커리큘럼 및 주제 관련성을 설명하지 않는다];  오히려, 학생들은 [텍스트에 대한 그들만의 관찰과 해석]을 하고, 그들이 [알아차린 것을 명확하게 하고 해석하는 연습]을 하도록 장려된다.

The narrative medicine method used during in-class exercises has been well described elsewhere.18,21 Briefly, small groups of 12 to 14 students, led by FCM faculty–mentors who are trained in the narrative medicine tradition, engage together in close reading and discussion of complex literary texts.3 Faculty select the narrative texts, which parallel the arc of students’ medical school experiences, for their implicit congruence with the ongoing course themes (e.g., awareness of multiple perspectives in a clinical encounter). However, faculty do not explicate the curricular and thematic connections for students; rather, students are encouraged to make their own observations and interpretations of the text, and to practice articulating and interpreting what they notice.

학생들의 관찰력과 서술력을 높이기 위해, 소그룹 토론은 [형태, 시간, 은유, 분위기]와 같은 장치에 초점을 맞춘다. 이 토론에 이어, 학생들은 그들 자신의 [창의적이고 독특한 성찰을 장려]하기 위해 고안된 [개방형 프롬프트에 대한 개별 응답]을 작성한다. 마지막으로, [글을 쓴 후]에, 학생들은 그들의 [작품을 서로 공유]할 기회를 갖는다. 예를 들어 윌리엄 카를로스 윌리엄스의 짧은 시 '아티스트'(부록 1 참조)에 초점을 맞춘 세션에서 학생들은 등장인물들의 관계와 설정에 대한 질문을 곰곰이 생각하고 토론한다. 그 시에 나오는 사람들은 누구인가요? 그들은 서로에게 누구인가? 내레이터는 누구입니까? 무슨 일이야? 분위기가 어때요? 다음으로, 그들은 ["예상치 못한 아름다움의 순간에 대해 써라"라는 메시지에 대답]한다. 10분 이내로 글을 쓴 후, 학생들은 그들이 쓴 것을 짝을 지어 읽거나 더 큰 그룹과 함께 읽도록 하고, 공유된 글에 응답한다. 모범적인 학생 포트폴리오 응답은 부록 1을 참조하십시오.

To heighten students’ observational and descriptive capacities, small-group discussions focus on such devices as form, time, metaphor, and mood. Following this discussion, students write individual responses to open-ended prompts designed to encourage their own creative, unique reflections. Finally, after writing, students have the opportunity to share their work with one another. For example, during a session focused on “The Artist,”22 a short poem by William Carlos Williams (see Appendix 1), students ponder and discuss questions about the relationships of the characters and the setting. Who are the people in the poem? Who are they to one another? Who is the narrator? What is going on? What is the mood? Next, they respond to the prompt “Write about a moment of unexpected beauty.” After writing for no more than 10 minutes, students are invited to read what they wrote, either in pairs or with the larger group, and respond to the shared writing. See Appendix 1 for an exemplar student portfolio response.

학생들은 매 학기마다 [1~2개의 서사 의학 항목]과 [시그니처 리플렉션]을 교수진(멘토 또는 멘토)에게 제출해야 하지만, 학생이 FCM 교수진을 초청하지 않는 [한 대부분의 항목은 비공개]로 유지된다. 멘토 피드백은 평가적이지 않고, 대신 학생들이 배우고 있는 것을 모델링하는 관찰과 설명의 유사한 면밀한 읽기 방법에 의존한다. (이러한 방법은 다른 곳에서 자세히 설명합니다.) FCM 교수진-멘토들은 1학년 학생들의 남은 의과대학 기간 동안 동행하며, 그들의 서면 피드백은 매 학기 개별 멘토링 회의의 기초를 제공한다.
Students must submit 1 to 2 narrative medicine entries and their Signature Reflection to their faculty–mentor each semester; however, most entries remain private unless a student invites the FCM faculty–mentor to provide commentary. Mentor feedback is not evaluative but instead relies on similar close reading methods of observation and description that model what the students are learning. (These methods are described in detail elsewhere.3) FCM faculty–mentors accompany their first-year students through their remaining years of medical school, and their written feedback provides the basis for individual mentoring meetings each semester.

시그니처 리플렉션
Signature Reflections

P&S 포트폴리오의 중요하고 혁신적인 구성 요소 중 하나는 [시그니처 리플렉션]입니다. 매 학기가 끝날 때마다 학생들은 20분간의 강의실 시간을 갖고, 포트폴리오 전체 항목을 읽는다. 그리고 나서, 그들은 그들 자신의 이전 글들에 면밀한 관찰과 묘사의 동일한 서술적 의학 도구들을 적용하도록 격려하는 프롬프트에 대한 응답으로 '시그니처 리플렉션'을 쓰는데 10분에서 15분을 보낸다. 이러한 방식으로, [시그니처 리플렉션]은 학생들이 [자기 인식을 위한 노력의 증거]를 찾기 위해, [자신의 글을 조사]하고, [의대에서 경험한 것에 대한 가설]을 세우고, [자신에 대해 알아차리는 것에 대한 결론을 도출]할 수 있는 기회를 제공한다. 이 연습은 학생들이 [그들이 쓴 것을 돌아봄]으로써 그들 자신에 대한 더 큰 이해를 얻을 수 있는 메커니즘을 제공한다.
One important and innovative component of the P&S portfolio is the Signature Reflection. At the end of each semester, students have 20 minutes of classroom time to read their entire collection of portfolio entries. Then, they spend 10 to 15 minutes writing a Signature Reflection in response to a prompt that encourages them to apply the same narrative medicine tools of close observation and description to their own previous writings. In this way, the Signature Reflection provides an opportunity for students to plumb their own writing for evidence of their efforts at self-recognition, to frame hypotheses about what they have experienced in medical school, and to draw conclusions about what they notice about themselves. This exercise offers a mechanism by which students may gain a greater understanding of themselves by looking back at what they have written.

학생 글쓰기 분석
Analysis of Student Writing

2015년 1월, 우리는 P&S 포트폴리오가 학생들의 [성찰 능력]과 [전인적 전문직 정체성의 구성]에 미치는 영향을 이해하기 위해 시그니처 리플렉션을 분석했다.
In January 2015, we analyzed Signature Reflections to understand the effect of the P&S portfolio on students’ reflective capacity and on their construction of a holistic professional identity.

샘플
Sample

이러한 1학기 시그니처 리플렉션에 사용된 글쓰기 프롬프트는 "자신이 상상하는 의사에 대해 쓰라; 자신의 글에서 자신에 대해 무엇을 알아차렸는가?"였다 160명의 1학년 학생들 중 4명의 학생들이 본 연구에 참여하지 않기로 결정하였고 24명의 Signature Reflections는 기술적인 문제로 포트폴리오 웹사이트를 통해 접근할 수 없었다. 나머지 132명 중 97명(73%)은 '글에서 자신에 대해 무엇을 알아차렸는가'라는 안내문 2절에 구체적으로 응답했다. 구체적으로, 자신의 글을 읽음으로써 ['시간이 지남에 따라 자신의 정체성이 변화했는지(그리고 그렇다면 어떻게 변화했는지)'를 어떻게 인식했는지] 대한 학생들의 인상을 포착하기 위해, 우리는 리뷰를 이러한 97개의 시그니처 리플렉션으로 제한했다. 이전에 Signature Reflection이 분석된 모든 학생들은 교직원들이 그들의 정체불명의 글을 읽도록 하는 것에 동의했다. 이 프로젝트에 대한 윤리적 승인은 컬럼비아 대학교의 기관 검토 위원회에 의해 승인되었다.  
The writing prompt used for these first-semester Signature Reflections was, “Write about the doctor you imagine yourself becoming; what did you notice about yourself in your writing?” Of the class of 160 first-year students, 4 students opted not to participate in this research and 24 Signature Reflections were not accessible through the portfolio website because of technical issues. Of the remaining 132 students, 97 (73%) specifically responded to the second clause of the prompt, “What did you notice about yourself in your writing?” To capture students’ impressions of, specifically, how they perceived, through reading their own writing, whether their own identities had changed over time (and if so, how), we limited our review to these 97 Signature Reflections. All students whose Signature Reflection was analyzed previously agreed to let faculty read their deidentified writings. Ethical approval for this project was granted by the institutional review board of Columbia University.

분석과정
Process of analysis

분석을 시작하기 위해 17명의 FCM 멘토가 [60개의 식별되지 않은 시그니처 리플렉션 샘플]을 검토하고 그룹별로 공통 단어와 개념을 식별했다. 6명의 연구자로 구성된 소규모 그룹은 이러한 개념을 서술 의학, 성찰적 글쓰기, 전문적 정체성 형성 문헌의 맥락 안에 두었다. 이러한 일반적인 단어와 개념은 초기 코드북의 기초를 형성했으며 포트폴리오, 특히 시그니처 리플렉션이 학생들의 성찰적 역량과 전문적 정체성 구축에 어떻게 기여했는지에 대한 새로운 개념적 표현을 구성할 수 있게 했다. 그런 다음 [나머지 37개의 시그니처 리플렉션]에 대한 코드북의 코드를 지속적으로 적용하고 검토했으며, 지속적인 비교 방법을 사용하여 코드과 코드북을 반복적으로 수정했다.23 팀이 최종 코드북에 합의했을 때, 우리 중 두 명(B.K., L.F.)은 Dedoose 소프트웨어를 사용하여 97개의 Signature Reflection 문서에 코드의 수정된 버전을 적용했습니다. 6명으로 구성된 이 팀은 매주 만나 최종 코딩에 대해 논의하고 발생하는 모든 질문을 해결했습니다. 97개의 모든 시그니처 리플렉션에 코드를 적용한 후에는 코드화된 데이터를 조사하여 새로운 패턴과 두드러진 테마를 식별한 다음 포트폴리오의 기여도에 대한 새로운 이해를 바탕으로 초기 개념 표현을 수정했다.
To initiate the analysis, 17 FCM mentors reviewed a sample of 60 deidentified Signature Reflections and, as a group, identified common words and concepts. A smaller group of 6 researchers placed these concepts within the context of the narrative medicine, reflective writing, and professional identity formation literatures. These common words and concepts formed the basis of our early code book and allowed us to construct an emergent conceptual representation of how the portfolio in general, and the Signature Reflections in particular, contributed to students’ reflective capacity and professional identity construction. We then continued to apply and review codes from the code book for the remaining 37 Signature Reflections, iteratively revising the codes and code book employing the constant comparison method.23 When the team reached agreement on a final code book, two of us (B.K., L.F.) applied revised versions of the codes to all 97 Signature Reflection documents using Dedoose software (SocioCultural Research Consultants, University of California, Los Angeles, with the William T. Grant Foundation). The six-person team met weekly to continue to discuss the final coding and to address any questions that arose. Once we had applied codes to all 97 Signature Reflections, we examined the coded data to identify emergent patterns and salient themes, and then we revised our initial conceptual representation based on our emergent understanding of the contributions of the portfolio.

시그니처 리플렉션 테마
Signature Reflection Themes

학생들이 Signature Reflections에서 설명한 것처럼 P&S 포트폴리오에 대한 반복 분석을 통해 우리는 두 가지 중요한 주제를 발견했다: 인정과 투쟁. 우리는 아래에서 이러한 주요 테마와 6개의 하위 테마를 설명하고, 각각 학생의 시그니처 리플렉션의 예시적인 인용문으로 설명합니다.
Through our iterative analysis of the P&S portfolio, as described by students in their Signature Reflections, we detected two overarching themes: Recognition and Grappling. We describe these main themes and their six subthemes—and illustrate each with exemplary quotations from students’ Signature Reflections—below.

인식
Recognition

서사의학 기법은 [자신과 다른 사람들의 이야기]에 대한 세심한 주의를 촉구하기 위해 고안되었으며, 실제로 이것은 우리 학생들의 글에서 명백했다. 우리는 학생들이 [[자신의 특성]과 [환자의 관점]에 대한 높은 인식]을 설명하기 위해 "인식"을 선택했다. 인식의 두 가지 하위 주제인 [자기 인식]과 [공감]은 학생들의 시그니처 리플렉션에 자주 등장했다.
Narrative medicine techniques are designed to foster close attention to the stories of the self and of others,18 and indeed this was evident in our student writings. We chose “recognition” to describe students’ descriptions of their heightened awareness of their own attributes and of patients’ perspectives. Two subthemes of recognition—self-awareness and empathy—appeared frequently in students’ Signature Reflections.

자각.
Self-awareness.

"당신의 글에서 당신은 당신 자신에 대해 무엇을 알아차렸습니까?"라는 글을 쓰는 프롬프트가 주어졌을 때 학생들이 그들 자신의 특성이 그들의 직업적 정체성 구축에 어떻게 영향을 미칠지에 대해 종종 쓴다는 것은 놀라운 일이 아니었다. 어떤 학생들은 구체적인 성격 특성을 명명하고 그들의 성격이 어떻게 글에 표현되는지 관찰한 반면, 다른 학생들은 포트폴리오가 어떻게 그들의 자아 인식을 향상시켰는지에 대해 언급했다. 이를 설명하기 위해 한 학생은 다음과 같이 썼다:
Given the writing prompt, “What did you notice about yourself in your writing?” it was not surprising that students often wrote about how their own attributes would influence their professional identity construction. Whereas some students named specific personality traits and observed how their personality was represented in their writing, others commented on how the portfolio enhanced their awareness of self. To illustrate, one student wrote:

학기 내내, 이전의 글들을 통해 보았듯이, 나는 정말로 의사가 되고 싶은 나 자신의 감정과 동기를 이해하려고 노력했고, 내 인생에서 나를 이 지경까지 이끈 모든 삶의 경험들과 사람들에 대해 반성했다.
Throughout the semester, as I saw through my previous writings, I really tried to get in touch with my own feelings and motives for wanting to become a doctor and reflected on all the life experiences and people that have led me to this point in my life.

공감.
Empathy.

학생들은 종종 [공감적인, 초기 의대생의 자아]를 인식했고, 이러한 인식이 시그니처 리플렉션의 가치 있는 배당금이라고 보고했다. 이를 설명하기 위해 한 학생은 다음과 같이 썼다:
Students frequently recognized their empathic, early-medical-school selves and reported this awareness to be a valuable dividend of the Signature Reflection. To illustrate, one student wrote:

하지만 이번 학기 포트폴리오를 읽으며 환자들과 공감하고 적절한 관심을 줄 수 있는 능력을 잃지 않을 것이라는 희망을 갖게 되었습니다. 첫 학기 동안 저는 환자들을 위해 작은 관대한 행동을 하는 것이 매우 의미가 있다는 것을 배웠습니다.
However, reading my portfolio entries from this past semester has given me some hope that I won’t lose the ability to empathize with patients and give them proper attention. During both of my [first semester] clerkships I learned that going out of my way to perform small acts of generosity for patients can be very meaningful.

학생 시그니처 리플렉션은 [다양한 관점에 대한 인식]을 보여주고 [환자의 이야기를 전달함]으로써 공감을 표현했다. 다른 학생은 다음과 같이 말했다: 
Student Signature Reflections expressed empathy by showing an awareness of multiple perspectives and by relaying patient stories. Another student commented:

의학 교육 과정에서 이러한 것들을 잊는 것은 매우 쉬울 수 있지만, 제가 쓴 글을 돌아보면…. 나는 아직도 '환자'를 [의학적 '문제']라기보다는 [사람, 독특하고 중요하고 가치 있는 이야기와 관점, 삶의 경험을 가진 개인]으로 우선시하고 있다는 사실에 안심이 된다
In the course of medical training, it can be very easy to forget these things—but looking back on my writing.… I am reassured by the fact that I still view “patients” first as people, as individuals with unique, important and worthwhile stories, perspectives, and life experiences, rather than as medical “problems.”

이 학생의 의견은 시그니처 리플렉션이 지원하는 종적 관점의 중요성과 가치를 강조했다. 이 연습에 참여하면서 학생은 "잊기가 매우 쉬울 수 있다"는 것을 깨달았다.
This student’s comment highlighted the importance and value of the longitudinal perspective supported by the Signature Reflections; in participating in this exercise, the student realized that “it can be very easy to forget.”

그래플링
Grappling

또 다른 흥미롭지만 놀랍지 않은 주제는 학생들이 시간이 지남에 따라 [변화와 불확실성에 대해 논평하는 것]과 관련이 있었다. 우리는 이전 포트폴리오 항목의 내용에서 마주친 변화나 놀라움으로 학생들의 레슬링에 대한 설명을 포착하기 위해 "드잡이Grappling"을 가장 적합한 용어로 선택했다. 그래플링은 다음의 경우에 명백했다.

  • (1) 학생들이 그들 내부의 변화에 대해 토론할 때, 
  • (2) 감정적 거리와 연민의 균형을 맞추는 것에 대한 그들의 걱정에서, 
  • (3) 그들이 질문을 하거나 경이로움을 나타낼 때, 또는 
  • (4) 그들이 불안감을 표현했을 때. 

이러한 의견은 시그니처 리플렉션을 작성하는 활동이 반사 능력을 촉진하는 메커니즘을 설명하는 것처럼 보였다.
Another intriguing, but not surprising, theme pertained to students commenting on change and uncertainty over time. We chose “grappling” as the most fitting term to capture students’ descriptions of their wrestling with changes or surprises they encountered in the content of their earlier portfolio entries. Grappling was evident

  • (1) when students discussed the changes within themselves,
  • (2) in their worries about balancing emotional distance with compassion,
  • (3) when they asked questions or exhibited wonder, or
  • (4) when they expressed anxiety.

These comments appeared to elucidate a mechanism by which the activity of writing a Signature Reflection both fostered—and was fostered by—reflective capacity.

내부 변화.
Internal change.

많은 학생들이 1학기 포트폴리오 글을 통해 [관찰한 관점의 변화]에 대해 글을 썼다. 위에서 인용한 세 학생과 같은 많은 학생들이 긍정적인 교훈을 얻었다고 생각하는 변화에 주목한 반면, 다른 학생들은 부정적인 것으로 인식하는 변화에 직면했다. 학생들은 자신감, 공감, 기술의 변화에 주목했다. 한 사람은 다음과 같이 말했다:
Many students wrote about the changes in their perspectives that they observed through their first-semester portfolio writings. Whereas many students, such as the three whose quotes appear above, noted changes that they considered positive lessons learned, others confronted changes that they perceived as negative. Students noted changes in their confidence, empathy, and skills. One commented:

지난 학기에 내가 쓴 글을 돌아보면 하나의 주제가 떠오른다. 그것은 불편하거나, 다르고, 혼란스럽거나, 불확실한 느낌을 받는 것 중 하나이다. 그것은 그 불편함, 다른 것, 혼란 또는 불확실성에서 이해를 향해 나아가기 위해 노력하는 것 중 하나이다.
Looking back over what I wrote this past semester, a theme emerges. It’s one of being okay with feeling uncomfortable, different, confused, or uncertain. It’s one of striving to move from that discomfort, otherness, confusion, or uncertainty towards an understanding.

또 다른 사람은 다음과 같이 썼다:
Another wrote:

저는 제가 쓴 글의 어둠에 호기심이 있습니다. 왜냐하면 저는 제 자신을 꽤 긍정적인 사람으로 보고 있기 때문입니다. 그리고 저는 여전히 콜롬비아 의과대학과 같은 기관에 있는 것에 대해 대단히 감사하게 생각합니다. 제 첫 작품에 담긴 [설렘과 긍정적인 전망이 이렇게 많이 사라졌다는 사실]에 놀랐다. 나는 학교를 다니고 새로운 곳에 자신을 통합하는 것은 힘든 일이고 약간 우울해지기 쉽다고 생각한다.
I’m rather intrigued about the darkness of my writing because I see myself as a fairly positive person, and I still feel immensely grateful to be at an institution like Columbia for medical school. I was surprised that the excitement and positive outlook contained in that first piece has worn off to such a large extent. I guess school and integrating yourself into a new place is a lot of hard work and it’s easy to get a little down.

실제로 위의 학생들은 'emerging theme'를 파악했을 뿐만 아니라, 이에 대한 놀라움과 불편함을 표현했다. 이러한 놀라움과 그에 따른 명백한 불편함은 이 메타 반사 연습에 의해 독특하게 촉진될 수 있다. 
Indeed, the students above not only identified an “emerging theme” but also expressed surprise and discomfort in doing so. This surprise, and the evident discomfort that ensued, may be uniquely facilitated by this meta-reflection exercise.

이분법: 감정적 거리와 동정심의 균형을 맞추는 것.
Dichotomies: Balancing emotional distance and compassion.

학생들은 시그니처 리플렉션을 쓰는 것이 어떻게 그들이 갈등이나 "이분법"을 인지할 수 있게 해주는지를 표현했다. – [문화에 동화되거나 전문가가 되는 것]과 [자기를 유지하는 것] 사이, [건강한 감정적 거리를 유지하는 것]과 [동정심을 보이는 것] 사이. 한 학생은 이러한 인식을 "장벽"을 지나치는 것으로 설명했다: 
Students expressed how the act of writing the Signature Reflection allowed them to perceive conflicts or “dichotomies”—between acculturating or becoming a professional and maintaining self; between maintaining a healthy emotional distance and showing compassion. One student described this awareness as seeing past “the barrier”:

하지만, 제가 쓴 글에서 장벽을 무너뜨린 순간들이 있었습니다. 그리고 [이] 나온 것은 본질적으로 불협화음인 이분법이었습니다. 내가 잠재적으로 배울 것들에 대한 흥분이지만, 내가 더 이상 관계를 맺지 못할 것들에 대한 두려움(내가 배울 것들 때문에).…
Yet, there were moments in my writings where I brought down the barrier and [what] came out was a dichotomy, dissonant in nature. Excitement about the things I would potentially learn, but fear for the things I could no longer relate to (due to the things I will learn).…

또 다른 학생은 그녀가 지식 습득에 초점을 맞춘 것과 대조적으로 그녀의 초기 이상주의를 설명했다:
Another student described her initial idealism in contrast to her focus on knowledge acquisition:

그녀를 데려온 날들을 묘사해야 했던 우리의 [첫 번째 글쓰기 과제]는 [내가 의대에 있는 최종 목표(사람들을 돕는 것)]을 [어떻게 종종 잊게 되는지] 생각하게 만들었다. 나는 하얀 코트 의식의 흥분이 내 혈관을 타고 흐르고 난 후에 그것을 썼다. 학기가 계속 진행되면서 나는 때때로 생화학 교과서의 페이지들에 대한 감정을 잊곤 한다는 것을 알아차렸다. 그것은 실제로 환자들을 돕는 것과는 거의 관련이 없는 것처럼 보였다.

Our very first writing assignment where we had to describe the days that brought us her[e] made me think about how I would often forget about the end goal of being in medical school—to help people. I wrote it after the hyper of the white coat ceremony was coursing through my veins. As the semester progressed onwards I noticed that sometimes I would forget that sentiment to the pages of biochemistry textbooks that seemed to have little connection to actually helping patients.

학생들은 그들이 [어떻게 변화하고 있는지 혹은 어떻게 변화가 불가피할 것이라고 예측하는지]를 인식하면서 [두려움과 흥분]을 모두 묘사했다. 또한, 후자의 인용문에서 제시된 바와 같이, 우리는 포트폴리오 항목을 검토하는 행위가 일상생활에 너무 깊이 짜여져 일상생활에서 감지할 수 없는 갈등과 경쟁적 요구에 대한 학생들의 인식을 뒷받침했을 수 있다고 제안한다.
Students described both fear and excitement as they recognized how they were changing or how they predicted that change would be inevitable. Furthermore, as suggested by the latter quote, we propose that the act of reviewing their portfolio entries may have supported students’ recognition of conflicts and competing demands that were so deeply woven into their daily lives as to be undetectable day-to-day.

궁금증과 의문.
Wonder and questioning.

학생들은 종종 [어려운 질문들을 조사]하고, 그들의 [과거에 대해 궁금해]하고, [미래에 자신을 투영]하기 위해 시그니처 리플렉션 연습을 사용했다. 이러한 시그니처 리플렉션은 호기심, 상상력, 그리고 시간의 앞뒤로 자신의 위치를 바꿔놓았다. 예를 들어, 한 학생은 다음과 같이 생각했습니다:
Students often used the Signature Reflection exercise to examine thorny questions, to wonder about their past selves, and to project themselves into the future. These Signature Reflections relayed curiosity, imagination, and the repositioning of self back and forth in time. For example, one student pondered:

나의 오래된 글을 되돌아보고 시간의 거리를 두고 보는 것은 정말 흥미롭다… 나는 비록 내가 더 이상 그 사람이 아니더라도 그 글을 쓴 사람을 좋아한다. 그 여자가 어떤 의사가 될 지는 잘 모르겠어요, 왜냐하면 저는 이미 그 여자에게서 어느 정도의 해리를 느끼고 있기 때문인 것 같아요. 나는 그녀가 훌륭한 의사였을 것이라고 생각한다: 약간 순진하지만, 믿을 수 없을 정도로 낙관적이다. 걱정하는. 배려하는. 희망적이다.
It’s really interesting to look back on my old writing and see it with the distance of time … I like the person who did the writing, even if I’m no longer exactly that person. I am not sure what kind of doctor that woman is going to become, I guess because I already feel a level of dissociation from her. I think she would have been a great doctor: a little naive, but incredibly optimistic. Concerned. Considerate. Hopeful.

다른 고려 사항:
Another considered:

저는 지난 학기부터 제가 쓴 글을 읽으면서 [환자로 하여금 의사들이 원하는 만큼의 정보를 공개하지 못하게 하는 의사소통의 장벽]과 다른 문제들에 특히 주의를 기울였다는 것을 깨달았습니다. 언어와 문화의 차이, 시간의 제약, 신체적, 정서적 불편함 등 말이죠. 그런데도 이 글을 쓰면서 나는 다른 의사들을 미행하거나 내가 직접 인터뷰를 진행하면서 목격한 '불완전한' 의사소통에 대한 나의 집착이 궁금해진다. 나는 또한 왜 "완벽한" 의사소통은 친밀하든 그렇지 않든 간에 항상 정보를 드러내는 것에 대한 완전한 공개와 완전한 편안함을 수반한다고 생각하는지 궁금하다.
I’ve come to recognize in reading my writing from the past semester that I’ve been particularly attentive to communication barriers and other issues that prevent patients from disclosing as much information as clinicians might hope: language and cultural differences, time constraints, and physical and emotional discomfort, among many others. And yet, as I write this, I wonder about my fixation on the “imperfect” communication I have witnessed while shadowing other doctors or conducting my own interviews. I wonder, too, why I seem to think that “perfect” communication always involves full disclosure and complete comfort with revealing information, whether intimate or not.

일부는 Signature Reflection을 ["시간의 거리를 두고" 볼 수 있는 가치 있는 특성을 기록하고 문서화하기 위한 시금석]으로 사용했습니다. 바로 위의 인용문의 저자처럼, 학생들에게, [자신을 관찰하고 트렌드를 인식하는 행위]는 그들이 가지고 있는 가정, 그들의 가치, 그들의 직업적인 목표, 그리고 심지어 의학의 실천에 대한 의문으로 이어졌다.
Some used the Signature Reflection as a touchstone to note and document valued characteristics that they were able to see “with the distance of time.” For other students, such as the author of the quotation immediately above, the act of observing themselves and recognizing trends led to questioning the assumptions they held about themselves, their values, their professional goals, and even the practice of medicine.

불안.
Anxiety.

학생들은 자신의 글에서 [불안감]을 자주 관찰했고, 의학교육의 무수한 요구로 인한 '불확실성', '가짜', '답답함' 등에 대해 언급하기도 했다. 이를 설명하기 위해 한 학생은 다음과 같이 썼다: 
Students frequently observed anxiety in their own writing, often commenting on the “uncertainty,” “faking,” and “frustration” caused by the myriad demands of medical education. To illustrate, one student wrote:

이번 학기 내 글에서 반복되는 주제는 불안이었다. 나는 디즈니랜드의 문을 뛰어다니는 아이처럼 의대에 입학했다. 5개월 후, 그 흥분은 불확실성, 불안, 그리고 두 번째 추측의 홍수로 압도되었다. 
A recurring theme in my writing this semester has been anxiety. I entered medical school like a kid running through the gates of Disneyland. Five months later, that excitement has been overwhelmed by a flood of uncertainty, insecurity, and second-guessing.

또 다른 의견은 다음과 같다: 
Another commented:

제가 "아웃사이더가 되어라"에 대한 제 글을 읽을 때, 저는 여전히 제가 흰 코트를 입을 때마다 속이는 것처럼 얼마나 많이 느끼는지 생각해요. 내가 아는 것보다 아는 것이 더 많은 척을 하고 있다는 것. 나는 일상 생활에서 많은 경우에, "당신이 해낼 때까지 그것을 속일 수 있다"는 것을 알고 있다. 하지만 내가 의사가 되기 위한 훈련에서는, 위험이 더 큰 것 같다.
When I read my piece on “Being an outsider,” it strikes me how very much I still feel like I’m faking every time I put my white coat on. That I am pretending that I know so much more than I do. I know that in many instances in day-to-day living, one can “fake it till you make it.” But in my training to be a doctor, it seems like the stakes are higher.

하지만 한 학생이 더 관찰했습니다:
Yet one more student observed:

그러나 내가 알아차린 것은 단절이다. 나는 연결과 느낌 그리고 배려에 대해 쓰지만, 나의 일상적인 활동은 관련이 없다. 나는 매일 대부분의 시간을 의료행위를 할 수 있는 사실fact들을 연구하는데 보낸다. 물론 그것은 필요하지만, 나는 여전히 궁금하다: 우리는 정말 오랜 추적 동안 그것들을 무시함으로써 우리의 목표를 진전시키는가? 나는 시간을 생산적으로 사용하지 못하면 괴로워한다. 기차에 앉아 있는 5분도 책을 읽고, 공부하고, 의사소통하는 등 생산적이어야 한다. 하지만 인내심 있는 학생이 될 수 없는데 어떻게 인내심 있는 의사가 될 수 있을까요? 
What I notice however, is a disconnect. I write about connecting and feeling and caring, but my daily activities are unrelated. I spend most of each day studying the facts that allow one to practice medicine. It is necessary of course, but I still wonder: Do we really further our goals by ignoring them during the lengthy pursuit? I get frustrated easily when I don’t use my time well. Even five minutes of sitting on a train has to be productive—reading, studying, communicating. Yet how will I be a patient doctor when I can’t be a patient student?


학생들은 자신의 성찰을 검토하면서 [직업적 목표, 가족, 친구, 개인적 행복과 같은 여러 요소와 증가하는 교육적 요구의 균형을 맞추는 것]에 대해 언급하면서 일상생활에서 얼마나 어려움을 겪고 있는지에 주목했다. 이러한 학생들에게, 시그니처 리플렉션은 이전에 그들 자신이나 그들의 멘토에 의해 주목받지 못했거나 검토되지 않은 그들 내부의 불안이나 불안에 대한 인식으로 표면화되었을 수 있다.
In reviewing their reflections, students noted the extent to which they were grappling within their everyday lives, as they commented on balancing multiple factors such as professional goals, family, friends, and personal well-being with increased educational demands. For these students, the Signature Reflection may have surfaced a recognition of a disquiet or angst within themselves that lay previously unnoticed or unexamined either by themselves or their mentors.

논의
Discussion

[주기적인 메타 성찰(Signature Reflection과 같은)로 보강된 필수적인 종단적 서사적 포트폴리오]는 학생들에게 인식과 그래플링에 형태를 부여할 수 있는 연습과 공간을 제공한다. 시간적 거리를 둔 상태에서, 학생들은 그들의 과거에 대한 반성을 할 수 있다; 그들은 그들의 성장과 변화를 감시하고 고려할 기회를 가진다. 이러한 성찰적 과정은 일부에게는 잠재적으로 불편하지만 궁극적으로 전체적인 직업적 정체성의 구축을 촉진할 수 있다. 한 학생이 이 아이디어를 잘 포착했습니다: "제가 쓴 것을 돌이켜보면… 주제가 하나 나타납니다…. 그것은 그 불편함, 타자성, 혼란, 불확실성에서 이해를 향해 나아가기 위해 노력하는 것 중 하나이다"
A required, longitudinal narrative portfolio, augmented by periodic meta-reflection (such as our Signature Reflection), offers students the practice and space to give form to recognition and grappling. With the distance of time, students are able to reflect on their past reflections; they have the opportunity to monitor and consider their growth and change. This reflective process, though potentially uncomfortable for some, may ultimately foster the construction of a holistic professional identity. One student captures the idea well: “Looking back over what I wrote …, a theme emerges.… It’s one of striving to move from that discomfort, otherness, confusion, or uncertainty towards an understanding.”

이러한 시그니처 리플렉션에서 표현된 [인식과 고심recognition and grappling]은 놀라운 일이 아닙니다. 실제로, 다른 사람들은 창의적인 글쓰기가 어떻게 [자기 인식과 공감의 발달]을 지원하는지 우아하게 묘사했다. 게다가, 불확실성과 갈등과 같은 의대생의 직업 정체성과 관련된 주제들은 1950년대 초에 르네 폭스에 의해 주목되었다. 그 이후로 교육자들은 의대생들의 정체성, 가치, 신념 통합의 중요한 조절자와 같은 주제들을 계속해서 확인해 왔다. 우리의 연구 결과는 우리의 [서사적 포트폴리오와 시그니처 리플렉션]이 학생들이 이전에 다른 연구자들이 보고한 [변화하는 정체성, 갈등 및 불확실성]에 대한 주제를 스스로 [발견하고, 해결하고, 인식할 수 있도록 함]으로써 이러한 중요한 주제를 보완하고 확장한다. 다른 사람들이 학생들의 전문적 정체성 구축 경험을 포착하기 위해 녹음된 받아쓰기와 인터뷰를 사용한 반면, 우리는 대신 학생들이 직접 만든 해석에 의존해왔다. 게다가, 우리는 우리의 메타 반사 운동이 특히 강력한 방식으로 이 과정을 촉진한다고 믿는다: 학생들이 자신의 [시그니처 리플렉션을 쓰는 것]은 [자신이 누구인지, 그리고 자신이 누구가 되고 있는지에 대한 인식]을 [표면화하고 영구적인 형태로 포착]할 수 있게 해준다. 이러한 신체적 표현은 그들의 현재와 미래의 자아, 그들의 동료와 교수진-멘토, 그리고 더 광범위한 의학 교육 공동체에 의해 검토될 수 있다. 따라서, 시그니처 리플렉션 자체는 성찰적 글쓰기에 관한 학술 문헌에 기여한다.
The recognition and grappling expressed in these Signature Reflections are not surprising. Indeed, others have elegantly described how creative writing supports the development of self-awareness and empathy.6,24–26 Furthermore, themes related to medical student professional identity, such as uncertainty and conflict, were noted by Renée Fox27 as early as the 1950s. Since that time, educators have continued to identify such themes as important modulators of medical students’ integration of their identities, values, and beliefs.10,15,28 Our findings complement and extend these important themes by suggesting that our narrative portfolio and Signature Reflection allow students to discover, grapple with, and recognize for themselves the themes of changing identity, conflict, and uncertainty that other researchers have previously reported. Whereas others have used audiotaped dictations and interviews to capture students’ experience of professional identity construction,29–31 we have relied instead on interpretations created by students themselves. Moreover, we believe that our meta-reflective exercise promotes this process in a particularly powerful way: The students’ writing of their Signature Reflection allows their perceptions of who they are and who they are becoming to surface and be captured in permanent form. This physical representation can be examined by their present and future selves, their peers and faculty–mentors, and the broader medical education community. Thus, the Signature Reflections themselves contribute to the scholarly literature on reflective writing.

P&S 포트폴리오는 다음과 같은 서술적 의학 과정을 통해 [인식과 그래플링이 일어날 수 있는 성찰 공간]을 만들었다: 

  • 호기심을 유발하는 서사적 텍스트의 면밀한 읽기;
  • 개별 프로세스를 준수하는 프롬프트를 사용한 개방형 쓰기; 그리고 
  • 학생들이 그들의 직업적 정체성을 인식하고, 형태를 부여하고, 구성에 참여하는 것을 돕는 메타 성찰. 

이 과정을 통해 FCM 교수진은 질문을 하고 형태를 갖추기 위해 참석함으로써 지원적이고 비판단적인 멘토링을 제공한다. 자유로운 개인 맞춤형 질문은 학생들의 반응을 촉진하고 갈등, 질문, 불확실성을 표현하고 토론할 수 있는 장을 제공한다. 우리는 서사 의학 포트폴리오 커리큘럼의 캡스톤 과제인 시그니처 리플렉션이 의대 과정에서 전개되면서 학생들이 불확실성과 씨름하도록 장려하는 모델 역할을 한다고 믿는다.
The P&S portfolio created a reflective space wherein recognition and grappling could occur through the following narrative medicine processes:

  • close reading of narrative texts that encourage curiosity;
  • open-ended writing using prompts that honor individual processes; and
  • meta-reflection that helps students notice, give form to, and engage in the construction of their professional identity.

Throughout this process, FCM faculty provide supportive, nonjudgmental mentoring by asking questions and attending to form. The open-ended, personalized questions foster responses from students and provide a forum for expressing and discussing conflicts, questions, and uncertainties. We believe that the capstone assignment of our narrative medicine portfolio curriculum, the Signature Reflection, serves as a model to encourage students to grapple with uncertainty as it unfolds over the course of medical school.

비록 우리의 데이터 소스가 실제 학생들의 글이었지만, 인터뷰나 포커스 그룹을 통해 학생들에게 우리의 발견을 확증하도록 요청하면 신뢰성이 향상될 것이다. 또한 학생의 73%만이 시그니처 리플렉션 프롬프트에 포괄적으로 응답했다. 그 결과, 우리는 (반사 공간의 개방성을 유지하면서) 글쓰기 프롬프트와 지침을 보다 명확하게 만들었다.
Although our data source was actual student writings, asking students via interviews or focus groups to corroborate our findings would enhance trustworthiness. Also, only 73% of students responded comprehensively to the Signature Reflection prompt; as a result, we have made writing prompts and instructions more explicit (while still maintaining the openness of a reflective space).

우리의 목표는 학생들의 평생 성찰 기술을 육성하는 것이지만, 학생들의 1학기 시그니처 리플렉션에 대한 이러한 단면 분석은 학생들의 포트폴리오 참여의 장기적 결과를 평가할 수 없다. 이러한 초기 Signature Reflections로부터 배운 내용에 고무되어, 우리는 이러한 성찰적 과정을 통해 학생들이 얻는 이익이 시간이 지남에 따라 증가할 것으로 기대한다. 우리는 현재 의대 첫 3년 동안의 글쓰기 항목을 포함하도록 분석을 확장하고 있으며, 다음 분석에서는 종적 정체성 구축에 더 중점을 둘 계획이다. 앞으로 P&S 포트폴리오가 멘토링 과정과 의대생들의 공동체 의식에 미치는 영향을 연구하고자 한다.
Although our aim is to foster lifelong reflective skills in our students, this cross-sectional analysis of students’ first-semester Signature Reflections does not enable us to assess the long-term consequences of students’ participation in the portfolio. Encouraged by what we learned from these early Signature Reflections, we expect students’ benefits from this reflective process to increase over time. We are currently extending our analysis forward to include writing entries from the first three years of medical school, and we plan for our next analysis to focus more closely on longitudinal identity construction. Going forward, we hope to study the effect of the P&S Portfolio on the mentoring process and on medical students’ sense of community.

우리는 시그니처 리플렉션에 의해 구체화된 "거울" 연습이 우리 기관을 훨씬 뛰어넘는 유용성을 가지고 있다고 믿는다. 여기에 제시된 [비평가적 성찰적 글쓰기 포트폴리오]와 같은 [성찰적 공간] 내에 둔다면 때, 다른 의학 교육자들도 유사한 [메타 성찰 프로세스]를 채택하여 학생들의 [전인적 직업 정체성 구축]을 장려하고, 학생들로 하여금 [자신이 누구가 되고 있는지]에 대한 [종적 관점]을 제공할 수 있다.
We believe that the “mirror” exercise embodied by the Signature Reflection has utility well beyond our institution. When situated within a reflective space such as the nonevaluative reflective writing portfolio presented here, other medical educators can adopt a similar meta-reflection process to encourage students’ construction of a holistic professional identity and to provide them with a longitudinal perspective on who they are becoming.


 

Acad Med. 2018 Jun;93(6):888-894. doi: 10.1097/ACM.0000000000002102.

Looking Back to Move Forward: First-Year Medical Students' Meta-Reflections on Their Narrative Portfolio Writings

Affiliations collapse

Affiliation

1H. Cunningham is assistant professor of pediatrics, Department of Pediatrics, Columbia University Medical Center, New York, New York; ORCID: http://orcid.org/0000-0003-1686-0497. D. Taylor is associate professor of medicine, Department of Medicine, Columbia University Vagelos College of Physicians and Surgeons, New York, New York; ORCID: https://orcid.org/0000-0002-1999-0388. U.A. Desai is assistant professor of medicine, Center for Family and Community Medicine, Columbia University Medical Center, New York, New York. S.C. Quiah is education, assessment, and evaluation specialist, Center for Education Research and Evaluation, Columbia University Medical Center, New York, New York. B. Kaplan is a second-year medical student, University of North Carolina School of Medicine, Chapel Hill, North Carolina. L. Fei is a second-year medical student, Tulane University School of School, New Orleans, Louisiana. M. Catallozzi is assistant professor of pediatrics and population and family health, Columbia University Medical Center, New York, New York. B. Richards is professor of pediatrics and director of education research and scholarship, University of Utah School of Medicine, Salt Lake City, Utah. D.F. Balmer is associate professor of pediatrics, Department of Pediatrics, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania. R. Charon is professor of medicine, Department of Medicine, Columbia University Vagelos College of Physicians and Surgeons, New York, New York; ORCID: http://orcid.org/0000-0002-6003-5219.

PMID: 29261540

PMCID: PMC5976514

DOI: 10.1097/ACM.0000000000002102

Free PMC article

Abstract

The day-to-day rigors of medical education often preclude learners from gaining a longitudinal perspective on who they are becoming. Furthermore, the current focus on competencies, coupled with concerning rates of trainee burnout and a decline in empathy, have fueled the search for pedagogic tools to foster students' reflective capacity. In response, many scholars have looked to the tradition of narrative medicine to foster "reflective spaces" wherein holistic professional identity construction can be supported. This article focuses on the rationale, content, and early analysis of the reflective space created by the narrative medicine-centered portfolio at the Columbia University Vagelos College of Physicians and Surgeons. In January 2015, the authors investigated learning outcomes derived from students' "Signature Reflections," end-of-semester meta-reflections on their previous portfolio work. The authors analyzed the Signature Reflections of 97 (of 132) first-year medical students using a constant comparative process. This iterative approach allowed researchers to identify themes within students' writings and interpret the data. The authors identified two overarching interpretive themes-recognition and grappling-and six subthemes. Recognition included comments about self-awareness and empathy. Grappling encompassed the subthemes of internal change, dichotomies, wonder and questioning, and anxiety. Based on the authors' analyses, the Signature Reflection seems to provide a structured framework that encourages students' reflective capacity and the construction of holistic professional identity. Other medical educators may adopt meta-reflection, within the reflective space of a writing portfolio, to encourage students' acquisition of a longitudinal perspective on who they are becoming and how they are constructing their professional identity.

 

실제 임상실습 상황에서 교육자의 행동: 관찰 연구와 체계적 분석(BMC Med Educ, 2019)
Educators’ behaviours during feedback in authentic clinical practice settings: an observational study and systematic analysis 
Christina E. Johnson1* , Jennifer L. Keating2 , Melanie K. Farlie3 , Fiona Kent4 , Michelle Leech5 and Elizabeth K. Molloy6 

 

배경
Background

역량 기반 교육 및 프로그램 평가와 연계된 현대 임상 교육은 역량 평가를 위한 Miller의 프레임워크에서 최고 수준을 목표로 하는 작업장의 일상적인 작업에 대한 평가 및 피드백에 초점을 맞추고 있습니다[1,2,3]. 피드백은 학습과 성과에 가장 강력한 영향을 미치는 요소 중 하나입니다 [4,5,6,7,8]. 학습자가 다른 실무자의 비판, 추론, 조언 및 지원으로부터 이익을 얻을 수 있는 기회를 제공합니다. 이러한 협업을 통해 학습자는 성과 목표가 무엇이고 이러한 표준에 도달하는 방법에 대한 이해를 높일 수 있습니다[9, 10]. ['On the run' 또는 비공식 피드백]은 환자 치료를 제공하는 도중에 발생하는 짧은 피드백 조각을 의미합니다. [공식적인 피드백 세션]은 일반적으로 상급 임상의(교육자)와 학생 또는 하급 임상의(학습자)가 학습자의 성과를 보다 포괄적인 방식으로 논의하는 것을 의미한다. 공식적인 피드백 세션은 실습기간 중간 또는 실습기간 종료 평가 또는 작업장 기반 평가의 일부로 종종 발생합니다. 그러나 이 모델의 성공은 효과적인 피드백을 제공하는 일상적인 임상의에게 달려 있다. 효과적인 피드백의 어떤 구성 요소가 문헌에서 직장의 감독 관행으로 성공적으로 변환되었는지 그리고 그렇지 않은 구성 요소는 명확하지 않다. [Translation의 격차]에 대한 정보는 전문 개발 교육을 더 잘 목표로 삼거나 품질 피드백 행동을 구현하는 데 장애물을 극복하기 위한 전략을 설계하는 데 사용될 수 있다. 
Modern clinical training, aligned with competency based education and programmatic assessment, is focused on assessment and feedback on routine tasks in the workplace, targeting the highest level in Miller’s framework for competency assessment [1,2,3]. Feedback is one of the most powerful influences on learning and performance [4,5,6,7,8]. It offers the opportunity for a learner to benefit from another practitioner’s critique, reasoning, advice and support. Through this collaboration, the learner can enhance their understanding of what the performance targets are and how they can reach those standards [9, 10]. ‘On the run’ or informal feedback refers to brief fragments of feedback that occur in the midst of delivering patient care. A formal feedback session typically refers to a senior clinician (educator) and student or junior clinician (learner) discussing the learner’s performance in a more comprehensive fashion. Formal feedback sessions often occur as a mid- or end-of-attachment appraisal or as part of a workplace-based assessment. However the success of this model relies on everyday clinicians providing effective feedback. It is not clear which components of effective feedback have been successfully translated from the literature into supervisory practice in the workplace, and which have not. Information on gaps in translation could be used to better target professional development training, or to design strategies to overcome impediments to implementing quality feedback behaviours.

병원에서 진정한 피드백을 [직접 관찰]한 연구는 드물다. [관찰 연구]는 일상적인 임상 교육에서 실제로 일어나는 일에 대한 주요 증거를 제공하기 때문에 매우 가치가 있다. [직접 관찰]은 활동을 관찰하는 연구자나 비디오 관찰을 통해 달성할 수 있다. 우리는 이전의 몇 가지 직접 관측 연구만 확인했다. 이는 공식 또는 비공식 피드백(외래 클리닉, 병동 또는 종합 시뮬레이션 임상 시나리오에 따른)을 포함하는 소수의 전문 분야(내부 또는 가족 의학)에서 주니어 학습자(의대생 또는 주니어 레지던트)를 포함한다. 추가적인 단일 연구에는 공식적인 애착 중간 또는 종료 피드백 동안 물리치료 학생들이 포함되었다[19]. [관찰 연구의 부족]은 시간 소모적인 특성, 바쁜 일정에 끼워진 피드백 회의와 일치하도록 관찰자나 비디오 녹화를 배치하는 것의 어려움 또는 관찰되거나 기록될 참가자의 침묵과 관련이 있을 수 있다. 이러한 연구들은 일반적으로 교육자들이 성과의 특정 측면에 대해 논평을 하고, 중요한 개념을 가르치고, 학습자가 어떻게 개선할 수 있는지 설명하거나 보여준다고 보고했다. 그러나 교수자가 말하는 시간이 대부분이고, 학습자에게 자기 평가를 요청하지만 응답하지 않으며, 수정 의견을 피하고, 일상적으로 실행 계획을 만들지 않습니다. 그러나 이러한 발견은 더 이상 현재의 관행을 반영하지 않을 수 있다. 또한 병원 환경에서 일하는 임상 교육자와 학습자의 다양성을 포착한 연구는 없었다.
Studies involving direct observation of authentic feedback in hospitals are rare. Observational studies are highly valuable, as they provide primary evidence of what actually happens in everyday clinical education. Direct observation can be achieved either by researchers observing the activity or via video-observation. We identified only a few previous direct observation studies: these involved junior learners (medical students or junior residents) in a few specialties (internal or family medicine) involving formal or informal feedback (in outpatient clinics, on a ward, or following summative simulated clinical scenarios) [11,12,13,14,15,16,17,18]. An additional single study involved physiotherapy students during formal mid- or end-of-attachment feedback [19]. The scarcity of observational studies may be related to the time consuming nature, difficulty in arranging observers or video recording to coincide with feedback meetings slotted into busy schedules, or the reticence of participants to be observed or recorded. These studies reported that typically educators make comments on specific aspects of performance, teach important concepts, and describe or demonstrate how the learner can improve. However educators tend to speak most of the time, ask the learner for their self-assessment but then do not respond to it, avoid corrective comments and do not routinely create actions plans. However these findings may no longer reflect current practice. In addition, no study captured the diversity of clinical educators and learners that work in a hospital environment.

따라서 우리는 직장 기반 학습 환경에서 [현대 교육자의 피드백 관행을 검토]하기 위해 자체 녹화된 비디오를 통해 병원 교육에서 진정한 공식 피드백 에피소드를 직접 관찰하기 시작했다. 이는 기회를 명확하게 하고 전문 개발 교육의 설계를 알릴 수 있다. 호주에서는 의료 전문 교육이 병원에 집중되어 입원 병동과 외래 진료소를 통합하고 있으며, 주요 전문 외래 환자 센터는 드물고 가족 의료 진료소는 상대적으로 작다. 바람직한 피드백 요소가 직업에 특정되지 않기 때문에 병원에 존재하는 다양성의 특징인 다양한 참가자를 모집했다. 우리는 완전한 피드백 상호작용을 포착하기 위해 공식적인 피드백 세션을 목표로 삼았다. 그런 다음 고품질 피드백을 위해 권장되는 관찰 가능한 일련의 교육자 행동을 사용하여 교육자 피드백 관행의 구성을 분석했다(표 1 참조) [20]. 이를 통해 덜 구조적이고 더 탐색적인 접근법이 사용되었던 이전 연구와 달리 포괄적인 행동 지표 세트를 사용하여 수집된 첫 번째 데이터 세트를 체계적으로 분석할 수 있었다. 이 프레임워크는 학습자의 참여, 동기 부여 및 개선을 지원함으로써 학습자의 결과를 향상시키는 것으로 간주되는 25가지 개별 관찰 가능한 교육자 행동을 개략적으로 설명한다(표 1 참조)[20]. 우리 팀의 이 초기 출판물은 학습자 결과를 향상시키기 위해 경험적 정보에 의해 입증된 교육자의 역할의 뚜렷한 요소를 식별하기 위한 광범위한 문헌 검토로 시작하여 관찰 가능한 행동으로 운영되고 전문가와 함께 델파이 프로세스를 통해 개선되는 방법을 설명했다.
Therefore we set out to directly observe authentic formal feedback episodes in hospital training, via self-recorded videos, to review contemporary educators’ feedback practice in workplace-based learning environments. This could then clarify opportunities and inform the design of professional development training. In Australia, health professions training is concentrated in hospitals, integrating both inpatient wards and outpatient clinics; major dedicated specialist outpatient centres are rare and family medicine clinics are relatively small. We recruited a range of participants, characteristic of the diversity present in hospitals, as desirable feedback elements are not profession specific. We targeted formal feedback sessions to capture complete feedback interactions. We then analysed the composition of educators’ feedback practice using a comprehensive set of observable educator behaviours recommended for high quality feedback (see Table 1) [20]. This enabled a systematic analysis of the first set of data gathered using a comprehensive set of behavioural indicators, in contrast to previous studies in which less structured and more exploratory approaches were used. This framework outlines 25 discrete observable educator behaviours considered to enhance learner outcomes by engaging, motivating and assisting a learner to improve (see Table 1) [20]. This earlier publication by our team described how these items were developed, starting with an extensive literature review to identify distinct elements of an educator’s role substantiated by empirical information to enhance learner outcomes, then operationalised into observable behaviours and refined through a Delphi process with experts.

방향 및 프로세스
Orientation and Process
1. 관찰된 성과를 기준으로 합니다.
1. Based on observed performance
교육자의 논평은 관찰된 성과에 기초했다.
The educator’s comments were based on observed performance
2. 적시 피드백
2. Timely feedback
교육자는 가능한 한 빨리 성과에 대해 논의하겠다고 제안했다.
The educator offered to discuss the performance as soon as practicable
3. 피드백 목적 명확
3. Feedback purpose clear
교육자는 피드백의 목적이 학습자의 성과 향상을 돕는 것이라고 설명했습니다.
The educator explained that the purpose of feedback is to help the learner improve their performance
4. 비판적이지 않은 분위기 조성: '도움을 주기 위해 여기에 있다.'
4. Establish a non-judgmental atmosphere: ‘here to help’
교육자는 기술을 개발하는 동안 일부 측면이 개선될 것으로 기대되며 교육자는 비판이 아니라 도움을 주기 위해 여기에 있다고 지적했다.
The educator indicated that while developing a skill, it is expected that some aspects can be improved and the educator is here to help, not criticise
5. 피드백 프로세스를 명확히 하여 학습자가 무엇을 기대해야 하는지 알 수 있도록 합니다.
5. Clarify feedback process, so learner knows what to expect
교육자는 피드백 토론을 위한 의도된 프로세스를 설명했습니다.
The educator described the intended process for the feedback discussion
학습자 중심의 초점
Learner-centred Focus
6. 대화 장려
6. Encourage dialogue
교육자는 학습자가 대화형 토론에 참여하도록 권장했습니다.
The educator encouraged the learner to engage in interactive discussions
7. 학습자의 우선순위를 찾습니다.
7. Seek learner’s priorities
교육자는 학습자에게 관찰 및 피드백 토론을 위한 학습 우선순위에 대해 질문하고 응답했습니다.
The educator asked the learner about their learning priorities for the observation and feedback discussion, and responded to them
8. 학습자가 '스스로 해결'하도록 유도합니다.
8. Encourage learner to ‘work it out for themselves’
교육자는 피드백 토론 중에 학습자가 문제와 가능한 해결책을 고려하도록 권장했습니다.
The educator encouraged the learner to consider the issues and possible solutions during the feedback discussion
9. 학습자가 한계를 은폐하기보다는 학습에 집중하도록 유도합니다.
9. Encourage learner to focus on learning, rather than trying to cover up limitations
교육자는 학습자가 해결책을 개발하는 데 도움을 줄 수 있도록 학습자가 어려움을 논의하고 성과에 대해 질문하도록 권장했습니다.
The educator encouraged the learner to discuss difficulties and ask questions regarding the performance so the educator could help the learner to develop solutions
10. 학습자의 정서적 반응을 인정하라
10. Acknowledge learner’s emotional response
교육자는 학습자가 표현한 감정을 인식하고 적절히 대응하였다.
The educator acknowledged and responded appropriately to emotions expressed by the learner
11. '진심으로 관심을 갖는다'
11. ‘Best interests at heart’
교육자는 학습자에 대한 존경과 지원을 보여주었다.
The educator showed respect and support for the learner
성능 분석
Performance Analysis
12. 자기평가의 가치를 명확히 한다.
12. Clarify the value of self-assessment
교육자는 학습자가 자기 평가의 이점에 대해 무엇을 이해하고 명확하게 설명하는 데 도움을 주었는지 물었습니다.
The educator asked what the learner understood about the benefits of self-assessment and helped clarify
13. 학습자 자기 평가
13. Learner self-assessment
교육자는 학습자에게 학습자 성과와 목표 성과 간의 주요 유사점과 차이점을 식별하도록 요청했습니다.
The educator asked the learner to identify key similarities and differences between the learner’s performance and the target performance
14. 목표 성과 및 추론 명확
14. Target performance and reasoning clear
교육자는 학습자에게 목표 성과의 주요 특징을 명확히 설명하고 그 이유를 설명했습니다.
The educator clarified with the learner key features of the target performance and explained the reasoning
15. 명확한 성과 차이를 포함한 교육자 평가
15. Educator assessment, including clear performance gap
교육자는 학습자의 성과와 목표 성과 간의 유사점과 차이점을 학습자에게 명확히 설명하였다.
The educator clarified with the learner similarities and differences between the learner’s performance and the target performance
16. 교육자는 몇 가지 중요한 문제에 대한 의견을 말합니다.
16. Educator comments on a few, important issues
교육자의 의견은 성과 향상을 위한 주요 문제에 초점을 맞췄다.
The educator’s comments focused on key issues for improving the performance
17. 구체적인 사례('무슨 일이')
17. Specific instance (‘what happened’)
먼저 교육자는 중립적인 언어를 사용하여 학습자가 한 일(행동, 결정 또는 행동)과 그 결과를 설명했습니다.
First the educator described, using neutral language, what the learner did (action, decision or behaviour), and the consequences
18. 교육자의 관점 명확 ('왜 중요한가')
18. Educator’s perspective clear (‘why it matters’)
교육자는 학습자가 우려하는 이유를 포함하여 학습자의 행동에 대한 관점을 명확하게 설명했습니다.
The educator clearly explained their perspective on the learner’s actions, including the reason for their concern
19. 교육자는 학습자의 관점을 탐구한다('왜' 학습자는 자신이 했던 대로 행동했는가).
19. Educator explores learner’s perspective (‘why’ the learner acted as they did)
교육자는 학습자의 관점과 추론을 탐구하여 학습자의 행동(예: 학습자가 무엇을 하려고 했는지, 고려된 옵션/어려움에 직면한 어려움)의 근거를 밝힙니다.
The educator explored the learner’s perspective and reasoning to reveal the basis for the learner’s actions (e.g. what was the learner trying to do and options considered/ difficulties encountered)
20. 사람이 아닌 행동에 집중하라 ('is'가 아닌 'did'에)
20. Focus on actions, not the person (‘did’ not ‘is’)
교육자의 의견은 개인적 특성이 아닌 학습자의 행동에 초점이 맞춰졌다.
The educator’s comments were focused on the learner’s actions not personal characteristics
실행 계획
Action Plan
21. 학습 우선순위 선택: 학습자에게 가장 유용한(중요하고 관련성 있는) 것
21. Select learning priorities: most useful (important and relevant) for the learner
교육자는 학습자가 개선할 성과의 몇 가지 주요 측면을 선택할 수 있도록 도와주었습니다.
The educator helped the learner to select a couple of key aspects of the performance to improve
22. 실행 계획을 개발하라: 어떻게 해야 하는가!
22. Develop the action plan: how to do it!
교육자는 학습자가 성과를 개선할 수 있는 방법을 연구하고 성과를 달성하기 위한 실질적인 단계를 명시할 수 있도록 도와주었습니다.
The educator helped the learner to work how they could improve their performance and specify the practical steps to achieve it
23. 학습자가 계획을 이해하는지 확인합니다.
23. Check the learner understands the plans
교육자는 학습자가 학습 목표와 실행 계획을 이해했는지 확인하고, 이를 자신의 말로 요약하도록 요청했습니다.
The educator checked if the learner understood their learning goals and action plan, by asking them to summarise it in their own words
24. 학습자가 근거를 이해하고 있는지 확인합니다. '왜 더 나은지'
24. Checks the learner understands the rationale: ‘why it’s better’
교육자는 학습자가 학습 목표와 실행 계획의 근거를 이해했는지 확인했습니다.
The educator checked if the learner understood the rationale for their learning goals and action plan
25. 피드백의 영향을 검토할 기회 계획
25. Plan opportunities to review the impact of the feedback
교육자는 학습자가 진행 상황을 검토할 수 있는 후속 기회에 대해 학습자와 논의했습니다.
The educator discussed with the learner possible subsequent opportunities for the learner to review their progress
  1. Reproduced with permission from Johnson et al. BMC Medical Education (2016) 16:96 Identifying educator behaviours for high quality verbal feedback in health professions education: literature and expert refinement

 

우리는 학습자 중심의 패러다임을 강력하게 지지하지만, 교육자들이 학습자들이 안전함을 느끼고 참여하며 성공적으로 기술을 향상시킬 수 있는 여건을 조성하기 위해 영향력 있는 위치에 있기 때문에 피드백에서 교육자의 역할에 초점을 맞추기로 결정했다. 우리는 특정 피드백 에피소드가 관련된 개인, 맥락 및 문화에 의해 형성된다는 것에 동의하지만, 학습자의 동기와 성과를 향상시킬 수 있는 능력을 촉진하는 전략은 여전히 관련이 있다. [권장 피드백 행동]은 로봇 방식으로 구현하기 위한 것이 아니라, 상호 작용 전반에 걸쳐 가장 유용한 측면의 우선순위를 지정하여 특정 상황에 맞게 조정된다. 품질 피드백의 핵심 부분에는 목표 성과를 명확히 하고, 이 목표와 비교하여 학습자의 성과를 분석하며, 개선을 위한 실질적인 단계를 개략적으로 설명하고, 진행 상황을 검토하는 방법을 계획하는 것이 포함됩니다 [4, 9, 21]. 가장 중요한 주제는 안전한 학습 환경 내에서 동기 부여 촉진[22,23,24,25], 능동적 학습[26,27,28] 및 협업[29,30,31,32]을 포함한다[10,33,34].
While we strongly endorse a learner-centred paradigm, we have chosen to focus on the educator’s role in feedback because educators are in a position of influence to create conditions that encourage learners to feel safe, participate and work out how to successfully improve their skills. We agree that specific feedback episodes are shaped by the individuals involved, the context and the culture, however strategies to promote a learner’s motivation and capability to enhance their performance remain relevant. Recommended feedback behaviours are not intended to be implemented in a robotic fashion but tailored to a particular situation by prioritising the most useful aspects throughout the interaction. The core segments of quality feedback include clarifying the target performance, analysing the learner’s performance in comparison to this target, outlining practical steps to improve and planning how to review progress [4, 9, 21]. Overarching themes include promoting motivation [22,23,24,25], active learning [26,27,28] and collaboration [29,30,31,32] within a safe learning environment [10, 33, 34].

연구 질문
Research question

본 연구에서 다루어진 연구 질문은 다음과 같다.
The research questions addressed in this study were:

  • 병원 실무 환경에서 임상 교육자가 공식 피드백 세션에서 보여주는 행동은 무엇입니까?
  • 이러한 행동이 공개된 피드백 권장 사항과 얼마나 밀접하게 일치합니까?
  1. What behaviours are exhibited by clinical educators in formal feedback sessions in hospital practice settings?
  2. How closely do these behaviours align with published recommendations for feedback?

방법들
Methods

연구 개요
Research overview

이 관찰 연구에서 선임 임상의(교육자)는 병원 환경에서 일상적인 임상 작업을 수행하는 주니어 임상의 또는 학생(학습자)을 관찰한 다음 후속 공식 피드백 세션 동안 자신들을 비디오로 촬영했다. 고품질 피드백에서 권장되는 일련의 교육자 행동을 기반으로 체크리스트를 사용하여 각 비디오를 분석했습니다(표 1 참조) [20].
In this observational study, senior clinicians (educators) observed junior clinicians or students (learners) performing routine clinical tasks in a hospital setting and then videoed themselves during the subsequent formal feedback session. We analysed each video using a check-list based on the set of educator behaviours recommended in high quality feedback (see Table 1) [20].

피드백 비디오는 2015년 8월과 2016년 12월 사이에 호주에서 가장 큰 대도시 교육 병원 네트워크 중 하나 내의 여러 병원에서 캡처되었다. 윤리 승인은 보건서비스(참고문헌 15,233L)와 대학 인간연구윤리위원회(참고문헌 2,015,001,338)에서 얻었다.
The feedback videos were captured at multiple hospitals within one of Australia’s largest metropolitan teaching hospital networks between August 2015 and December 2016. Ethics approval was obtained from the health service (Reference 15,233 L) and university human research ethics committees (Reference 2,015,001,338).

모집
Recruitment

의학, 간호, 물리치료, 직업치료, 언어치료, 사회복지 분야의 교육자(상급 임상의)와 그들과 함께 일하고 있는 학습자(추가 훈련을 받는 자격을 갖춘 보건 전문가 또는 학생)가 참여하도록 초대되었다. 전단지를 이용한 광범위한 연구 광고, 부서 관리 보조원이 배포한 이메일, 부서 회의에서의 짧은 프레젠테이션 및 보건 서비스 전반에 걸친 직원과의 대면 회의를 통해 광범위한 교육자를 구했다. 참여를 고려하기 위해서는 교육자가 광고에 응답하여 1차 연구원(CJ)에게 연락해야 했다. 교육자가 동의하면, 그들은 그들과 함께 일하는 모든 학습자들에게 전단지를 배포했고, 학습자들이 참여에 관심이 있다면 1차 연구원(CJ)에게 연락하라는 지침이 포함되어 있었다. 참가자들에게 롤링 광고를 통해 다양성을 추구했으며, 건강 직업 및 전문 분야, 성별 및 감독자 경험(교육자) 또는 훈련 수준(학습자) 등 주요 요인을 고려했다. 교육자와 학습자가 모두 동의한 후에는 두 사람에게 조언을 제공하고 일상적인 피드백 세션을 비디오로 녹화할 수 있도록 준비했습니다. 전체 피드백 만남을 기록하되, 시간은 약 10분을 목표로 설정하도록 요청받았지만 피드백 세션을 수행하는 방법에 대한 추가 지침은 제공되지 않았습니다. 참가자들은 비디오 분석에 사용된 고품질 피드백에 대해 권장된 25가지 교육자 행동을 보여주지 않았고, 현재 피드백 관행의 성격을 연구하는 것이 목적이었기 때문에 연구팀으로부터 피드백에 대한 다른 교육도 받지 못했다.
Educators (senior clinicians) across medicine, nursing, physiotherapy, occupational therapy, speech therapy and social work, and their learners (either qualified health professionals undertaking further training or students), who were working with them, were invited to participate. A broad range of educators were sought, via widespread advertising of the study using flyers, emails circulated by unit administration assistants, short presentations at unit meetings and face-to-face meetings with staff across the health service. To be considered for participation, an educator had to contact the primary researcher (CJ), in response to the advertisement. Once an educator consented, they distributed flyers to any learners working with them, with instructions to contact the primary researcher (CJ) if the learners were interested in participating. Diversity was sought by rolling advertising to participants, with consideration of key factors including health profession and specialty, gender and supervisor experience (educators) or training level (learners). Once an educator and a learner had both consented, the pair were advised and they made arrangements to video a routine feedback session. They were asked to record an entire feedback encounter and aim for a duration of approximately 10 minutes but were not given any additional instructions regarding how to conduct the feedback session. Participants were not shown the set of 25 educator behaviours recommended for high quality feedback used to analyse the videos nor given any other education on feedback from the research team, as the aim was to study the nature of current feedback practices.

동의하는 참가자들은 직장 기반 평가 또는 첨부 파일 종료 성과 평가와 관련된 다음 예정된 공식 피드백 세션에서 자신을 비디오로 녹화하기 위해 스마트폰이나 컴퓨터를 사용했다. 이 비디오는 이후 암호로 보호된 온라인 드라이브에 업로드되었으며 참가자들은 자신의 복사본을 삭제하라는 지시를 받았습니다. 비디오는 무작위 번호 생성기를 사용하여 번호가 매겨졌고 (이미지를 제외한) 비디오에는 개인 식별 정보가 포함되어 있지 않았다.
Consenting participants used a smart phone or computer to video-record themselves at their next scheduled formal feedback session related to either a workplace-based assessment or end-of-attachment performance appraisal. This video was subsequently uploaded to a password protected on-line drive and participants were instructed to delete their copy. The videos were numbered using a random number generator and the videos (other than the images) contained no personal identifying information.

비디오 분석
Video analysis

평가자 그룹은 감독 및 피드백에 대한 광범위한 경험이 있는 상급 교육/교육 연구 역할의 모든 보건 전문가(의료 2명, 물리치료 4명)였다. 각 평가자는 각 비디오를 독립적으로 분석하고 관찰 결과를 고품질 피드백을 위해 권장되는 25가지 교육자 행동과 비교했다(표 1 참조) [20]. 각 교육자 행동은 0 = 보이지 않음, 1 = 어느 정도 또는 가끔만 수행됨, 2 = 일관성 있게 수행됨으로 평가되었다.
The group of raters were all health professionals (two medical, four physiotherapy) in senior education/educational research roles with extensive experience in supervision and feedback. Each rater analysed each video independently and compared their observations with the set of 25 educator behaviours recommended for high quality feedback (see Table 1) [20]. Each educator behaviour was rated 0 = not seen, 1 = done somewhat or done only sometimes, 2 = consistently done.

사전 시범 연구에서, 우리는 그 기구를 사용하여 세 개의 비디오를 평가했다. 그런 다음 우리는 등급에 대해 논의하고 항목의 해석과 등급 척도의 사용에서 차이를 식별하기 위해 만났다. 일치를 장려하고 항목의 의미를 명확히 하는 전략이 개발되었다. 특히 '행동 2: 시기적절한 피드백을 확인했다.'는 교육자가 가능한 한 빨리 성과를 논의하자고 제안한 것은 관찰할 수 없었기 때문에 제외되었다. '행동 10: 학습자의 감정적 반응을 인정합니다'에 대해서, 교육자는 학습자가 표현한 감정을 인정하고 적절히 대응하였으며, 우리는 이것이 다음과 같은 상황에서 '2'(당연히 수행됨)로 평가되기로 결정했다. 

  • i) 학습자 감정의 암묵적 또는 명시적 지표(예: 불안 또는 방어)가 감지되고 교육자가 이를 인정하고 이에 주의를 기울인 경우 또는 
  • ii) 교육자와 학습자 사이의 이러한 정서적 균형이 교육자가 단서를 읽고 그에 따라 행동할 것을 요구한다고 가정한 것처럼, 만남 내내 정서적 균형이 관찰되었다.

결과적으로 총 항목 점수는 0에서 48 사이가 될 수 있습니다.

In a preparatory pilot study, we rated three videos using the instrument. We then met to discuss ratings and to identify differences in interpretation of items and the use of the rating scale. Strategies to encourage concordance and to clarify item meaning were developed. In particular we identified that Behaviour 2: Timely feedback: The educator offered to discuss the performance as soon as practicable was not observable, so it was excluded. For Behaviour 10: Acknowledge learner’s emotional response: The educator acknowledged and responded appropriately to emotions expressed by the learner, we decided that this would be rated as ‘2’ (consistently done) in the following situations

  • i) if implicit or explicit indicators of learner emotion (such as anxiety or defensiveness) were detected, and the educator acknowledged, and attended to this, or
  • ii) if emotional equilibrium was observed throughout the encounter, as we assumed that this emotional balance between educator and learner required the educator to be reading cues and acting accordingly.

Subsequently the total item score could range from 0 to 48.

데이터 분석
Data analysis

그 데이터는 두 가지 관점을 제공했다. 

  • i) 개별 교육자의 관행: 고품질 피드백에서 권장되는 행동 중 몇 가지가 각 비디오에서 관찰되었으며 
  • ii) 교육자 그룹 전체에 걸친 관행: 어떤 행동이 일반적으로 수행되었는지. 

비디오에서 볼 수 있는 각 개별 교육자의 실습을 특성화하기 위해, 각 항목의 점수를 평가자들 사이에서 평균화한 다음 합계하여 총점을 주었다. 전체 교육자 그룹에서 특정 교육자 행동이 얼마나 자주 관찰되었는지 설명하기 위해, 각 항목에 대한 평균 점수와 표준 편차를 모든 비디오에 걸쳐 계산했다[35]. 평가자 간 신뢰성을 평가하기 위해 Spearman's rho를 사용하여 각 비디오에 대한 총 점수를 검사자 쌍 간의 일치성에 대해 평가했다.
The data provided two perspectives

  • i) on an individual educator’s practice: how many of the behaviours recommended in high quality feedback were observed in each video and
  • ii) across the whole group of educators: which behaviours were commonly performed.

To characterise each individual educator’s practice seen in a video, the scores for each item were averaged across assessors and then summed to give a total score. To describe how commonly specific educator behaviours were observed amongst the whole group of educators, the mean score and standard devation for each item was calculated across all the videos [35]. To assess inter-rater reliability, total scores for each video were assessed for concordance between examiner pairs using Spearman’s rho.

결과.
Results

5개가 제외된 후 36개의 피드백 비디오를 분석할 수 있었다. 2개는 불완전하기 때문에(스마트폰 메모리 부족), 3개는 녹음의 기술적 오류 때문에(오디오 불분명, 시간 경과 포맷 사용, 참가자가 보이지 않음).
Thirty-six feedback videos were available for analysis after five were excluded: two because they were incomplete (insufficient smartphone memory) and three because of technical errors with recording (audio unclear, time-lapse format used, participants not visible).

비디오 참가자
Video participants

34명의 교육자가 참여했으며, 주요 특성(보건 직업 및 전문 분야, 감독자 경험 기간 및 성별)에 걸쳐 다양성을 갖추고 있습니다. 4명의 간호사, 4명의 물리치료사, 26명의 상급 의료진(3명의 마취사, 3명의 응급의사, 2명의 방사선사, 1명의 소아과 의사, 6명의 의사, 3명의 정신과 의사, 3명의 산부인과 의사, 1명의 안과의사, 4명의 외과의사)이 있었다. 여성 교육자는 18명(52.9%), 남성 교육자는 16명(47.1%)이었다. 교육자 경력이 5년 이하인 사람은 14명(41.2%), 6~10년 이상인 사람은 11명(32.3%), 10년 이상인 사람은 9명(26.5%)이었다.
Thirty-four educators participated, with diversity across key characteristics (health profession and specialty, length of supervisor experience and gender). There were four nurses, four physiotherapists and 26 senior medical staff (three anaesthetists, three emergency physicians, two radiologists, one paediatrician, six physicians, three psychiatrists, three obstetrician-gynaecologists, one opthalmologist and four surgeons). There were 18 (52.9%) female and 16 (47.1%) male educators. Fourteen (41.2%) educators had 5 years or less educator experience, 11 (32.3%) had six to 10 years and 9 (26.5%) had more than 10 years.

35명의 학습자가 주요 특성(건강 직업 및 전문 분야, 훈련 수준 및 성별)에 걸쳐 다양하게 참여했습니다. 수강생은 9명(25.7%), 사후자격 5년 이하 임상 9명(25.7%), 사후자격 6년 이상 임상 15명(42.9%), 상급 임상 2명(5.7%)이었다. 학습자는 23명이 여성(65.7%), 12명(34.3%)이었다. 모든 참가자들은 각각의 교육자들과 같은 보건 직업과 전문 분야에서 왔다.
Thirty-five learners participated with diversity across key characteristics (health profession and specialty, training level and gender). There were 9 (25.7%) students, 9 (25.7%) clinicians who were five years or less post-qualification, 15 (42.9%) clinicians 6 years or more post-qualification and 2 (5.7%) senior clinicians. Twenty-three learners were (65.7%) female and 12 (34.3%) were male. All participants were from the same health profession and specialty as their respective educators.

피드백 세션은 11개(30.6%)의 비디오에서 첨부파일 중간 또는 종료 평가와 25개(69.4%)의 비디오에서 특정 과제(절차적 기술, 임상 평가, 사례 토론 또는 발표 등)와 관련이 있었다. 피드백 세션 중 11회(30.6%)는 대학 또는 전문 의과대학 등 기관의 공식 피드백 양식을 사용하였으며, 대부분 첨부파일 중간 또는 첨부파일 종료 평가였다. 대부분의 평가는 형식적이었지만 일부는 프로그램적 평가 원칙에 따라 종적 훈련 프로그램의 구성요소로서 종합적이었다[3].
The feedback session was related to a mid- or end-of-attachment assessment in 11 (30.6%) videos and to a specific task (such as a procedural skill, clinical assessment, case discussion or presentation) in 25 (69.4%) videos. An official feedback form from an institution such as a university or specialist medical college was used in 11 (30.6%) of the feedback sessions, most of which were mid- or end-of-attachment assessments. Most of the assessments were formative but some were summative as a component of longitudinal training programs aligned with programatic assessment principles [3].

피드백 중 교육자 행동 분석
Analysis of educator behaviours during feedback

각 비디오는 4~6명의 평가자에 의해 분석되어 총 174개의 평가 세트를 제공했다(두 평가자에 의한 프로젝트의 예상치 못한 시간 제한 분석). 누락된 데이터는 드물었습니다(0.2%의 누락).
Each video was analysed by four to six raters providing a total of 174 sets of ratings (unexpected time constraints on the project limited analysis by two raters). Missing data were uncommon (0.2% ratings missing).

평가자 간 신뢰성
Inter-rater reliability

비교를 위한 데이터를 최대화하기 위해 모든 비디오를 분석한 평가자(4/6)에 대해 총 점수에 대한 평가자 간 신뢰성 범위를 계산했다. Spearman's rho는 0.62–0.73이었다. 나머지 두 평가자는 36개 동영상 중 10개(28%)와 21개(58%) 등급을 매겼으며, 평가자 간 신뢰도 분석에는 포함되지 않았다.
To maximise data for comparison, the inter-rater reliability range for total scores was calculated for raters (4/6) who analysed all the videos: Spearman’s rho was 0.62–0.73. The other two raters rated 10 (28%) and 21 (58%) of the 36 videos and were not included in the inter-rater reliability analysis.

개별 교육자의 피드백 실무
Individual educator’s feedback practice

개별 교육자의 관행과 각 비디오에서 권장되는 교육자 행동 중 몇 가지가 관찰되었는지에 대해 자세히 알아보기 위해, 우리는 각 비디오에 대한 총 점수(관찰된 각 행동에 대한 등급의 합, 모든 평가자에 걸쳐 평균)를 계산했다. 총 점수는 최소 5.7점(11.9%)에서 최대 34.2점(71.3%)까지 다양했으며, 교육자들의 평균 점수는 최대 48점에서 22.5점(46.9%, SD 6.6)이었다. 보다 상세한 분석(표 2 참조)을 통해 사실상 모든 교육자(88%)가 총점이 10점에서 30점 사이인 것으로 나타났다. 비록 서로 다른 특성에 걸쳐 성과를 비교하는 것이 우리의 의도는 아니었지만(비교를 가능하게 하기 위해 각 그룹에 충분한 표본 크기가 필요함), 건강 전문가, 경험 및 성별이 점수 범위에 걸쳐 상당히 고르게 분포되어 있는 것으로 보였다.
To learn more about individual educator’s practice and how many of the recommended educator behaviours were observed in each video, we calculated a total score (sum of rating for each observed behaviour, averaged across all assessors) for each video. Total scores ranged from a minimum of 5.7 (11.9%) to a maximum of 34.2 (71.3%), with a mean score across educators of 22.5 (46.9%, SD 6.6), from a maximum possible score of 48. More detailed analysis (see Table 2) revealed virtually all the educators (88%) had a total score between 10 and 30. Although it was not our intention to compare performance across different characteristics (which would require sufficient sample sizes for each group, to enable comparisons), there seemed to be a fairly even spread of health professions, experience and gender across the score ranges.

전체 교육자 그룹에 걸쳐 특정 교육자의 행동 빈도
Frequency of specific educator behaviours across the whole group of educators

모든 참가자들 사이에서 구체적인 피드백 행동이 얼마나 자주 관찰되었는지 탐구하기 위해, 우리는 모든 비디오에서 각 행동에 대한 평균 평점 점수를 계산했다. 표 3은 각 행동에 대한 등급 평균(SD)을 가장 많이 관찰된 것부터 가장 적게 관찰된 것까지 순위를 매겼다. 일부 행동은 거의 모든 비디오(최고 평균 등급 1.75, 행동 10)에서 관찰된 반면, 다른 행동은 매우 드물게 관찰되었다(최저 평균 등급 0.05, 행동 25).

To explore how often specific feedback behaviours were observed amongst all participants, we calculated the mean rating score for each behaviour across all the videos. Table 3 displays the rating mean (SD) for each behaviour, ranked from most to least often observed. Some behaviours were seen in almost every video (highest mean rating 1.75, Behaviour 10) while others were very infrequently observed (lowest mean rating 0.05, Behaviour 25).

가장 흔하게 관찰되는 교육자 행동 중에서(상위 3위: 평균 평점 1.41–2.0), 학습자의 성과에 대한 교육자의 평가와 가장 관련이 있다. 교육자들은 일반적으로 

  • 학습자 성과에 대한 의견을 학습자의 행동(행동 1, 17, 20)과 연결시키고, 
  • 개선을 위한 중요한 측면(행동 16)에 초점을 맞추고, 
  • 학습자의 성과와 목표 성과 간의 유사점과 차이점(행동 15)을 설명하고, 
  • 무엇을 해야 하는지, 그 이유를 명확히 했다(행동 14). 

일반적으로 볼 수 있는 다른 두 가지 행동은 안전한 학습 환경을 만드는 것과 관련이 있습니다. 여기에는 다음이 포함되었다.

  • 존중과 지지를 보여주는 것(행동 11)
  • 학습자가 표현하는 감정에 적절하게 반응하는 것(행동 10)

Amongst those educator behaviours most commonly observed (top third: mean rating score 1.41–2.0), most related to the educator’s assessment of the learner’s performance. Educators commonly

  • linked comments regarding learner performance to the learner’s actions (Behaviours 1, 17, 20),
  • focused on important aspects for improvement (Behaviour 16),
  • described similarities and differences between the learner’s performance and the target performance (Behaviour 15), and
  • clarified what should be done and why (Behaviour 14).

The other two behaviours commonly seen related to creating a safe learning environment. These included

  • showing respect and support (Behaviour 11) and
  • responding appropriately to emotions expressed by the learner (Behaviour 10)

교육자 행동의 중간 계층은 간헐적으로 나타났으며(평균 평점 0.71–1.40) 교육자들은 학습자들이 자신의 생각, 의견, 아이디어를 기여하고 불확실성을 드러내도록 장려하는 것과 관련이 있었다. 이것들은 포함되었다. 

  • 학습자가 대화형 토론에 참여하도록 유도합니다(행동 6). 
  • 스스로 해결하려고 노력한다(행동 8). 
  • 자신의 성과를 분석한다(행동 13). 
  • 행동의 배경에 있는 이유를 밝힙니다(행동 19). 
  • 어려움을 제기하고 질문을 한다(행동 9). 
  • 개선해야 할 가장 중요한 측면을 선택하는 데 참여한다(행동 21). 
  • 실행 계획을 통해 이를 수행하는 실질적인 방법(행동 22).

The middle band of educator behaviours were seen intermittently (mean rating score 0.71–1.40) and related to educators encouraging learners to contribute their thoughts, opinions and ideas, and to reveal their uncertainties. These included

  • encouraging the learner to participate in interactive discussions (Behaviour 6),
  • try to work things out for themselves (Behaviour 8),
  • analyse their own performance (Behaviour 13),
  • reveal the reasoning behind their actions (Behaviour 19),
  • raise difficulties and ask questions (Behaviour 9), and
  • participate in choosing the most important aspects to improve (Behaviour 21) and
  • practical ways to do this through an action plan (Behaviour 22).

교육자 행동의 가장 낮은 범위는 거의 나타나지 않았으며(평균 평점 0–0.7) 주로 피드백 세션의 설정 및 종료와 관련이 있었다

세션을 시작할 때, 안전한 학습 환경을 조성하기 위한 일환으로 권장되는 교육자 행동에는 

  • 피드백의 목적이 학습자의 향상을 돕는 것임을 명시적으로 설명하고(행동 3), 
  • 세션에 제안된 개요를 설명하고(행동 5) 
  • 실수는 학습 과정의 일부이며, 피할 수 없는 것임을 수용한다는 내용이 포함되었습니다. (행동 4). 

세션 결론 또는 마무리의 일환으로, 

  • 권장되는 행동에는 학습 목표와 실행 계획에 대한 학습자의 이해를 확인하는 것(행동 23, 24), 
  • 진행 중인 학습을 촉진하기 위해 진행 상황을 검토할 수 있는 향후 기회에 대해 논의하는 것(행동 25)이 포함되었다. 

그 밖에 거의 볼 수 없었던 교육자 행동으로는

  • 학습자의 학습 우선순위를 통합한 교육자(행동 7)와
  • 자기평가의 가치에 대한 학습자의 이해를 증진시키는 교육자(행동 12)가 있었다.

The lowest band of educator behaviours were rarely seen (mean rating score 0–0.7) and primarily related to the set up and conclusion of a feedback session.

At the start of the session, as part of creating a safe learning environment, the recommended educator behaviours included

  • explicitly explaining that the purpose of the feedback was to help the learner improve (Behaviour 3),
  • describing the proposed outline for the session (Behaviour 5), and
  • stating their acceptance that mistakes are an inevitable part of the learning process (Behaviour 4).

As part of the session conclusion or wrap-up, the recommended behaviours included

  • checking a learner’s understanding of the learning goals and action plan (Behaviours 23, 24), and
  • discussing future opportunities to review progress, to promote ongoing learning (Behaviour 25). 

The other educator behaviours that were rarely seen included the educator

  • incorporating the learner’s learning priorities (Behaviour 7) and
  • promoting the learner’s understanding of the value of their self-assessment (Behaviour 12).

논의
Discussion

교육자의 피드백 관행에 대한 이 연구에서, 우리는 개별 교육자의 관행과 교육자 그룹 전체에서 구체적인 권장 행동이 얼마나 자주 관찰되는지 모두에서 상당한 차이를 발견했다. 이는 병원 기반 교육에서 공식 피드백 에피소드 동안 '현재 발생하는 일'에 대한 귀중한 통찰력을 제공한다. 이러한 통찰력은 상당한 영향을 미칠 수 있는 잠재력을 가진 교육자 개발에 대한 향후 연구의 기회를 명확히 한다. 또한 권장 행동은 교육자가 이러한 품질 표준을 이해하고 제정하는 데 도움이 될 수 있는 특정 전략의 레퍼토리를 제공한다.
In this study of educators’ feedback practice, we found considerable variation in both an individual educator’s practice and how frequently specific recommended behaviours were observed across the group of educators. This provides valuable insights into ‘what currently happens’ during formal feedback episodes in hospital-based training. These insights clarify opportunities for future research into educator development with the potential for substantial impact. Furthermore the recommended behaviours offer a repertoire of specific strategies that may assist educators to understand and enact these quality standards.

교육자 그룹 전체에서 관찰된 특정 권장 행동의 빈도
Frequency of specific recommended behaviours observed across the group of educators

우리는 교육자들이 일상적으로 학습자의 성과에 대한 평가를 제공하고 과제가 어떻게 보여야 하는지 설명하는 것을 발견했지만, 학습자들에게 자기 평가나 실행 계획의 개발을 간헐적으로 요청했을 뿐이다. 이는 학습자의 성과에 대해 [교육자의 분석이 지배적인 문화]를 반영하는 것으로 보인다[36]. 이러한 결과는 이전의 관찰 연구와 피드백 양식의 결과와 일치한다[11, 12, 17, 19, 37, 38, 39, 40]. 이는 이러한 누락이 수 년 전에 마지막으로 보고된 이후 임상 환경에서 일반적인 피드백 관행이 거의 동일하게 유지되었음을 시사한다.

We found that educators routinely gave their assessment of the learner’s performance and described what the task should look like, but only intermittently asked learners for self-assessment or development of an action plan. This seems to reflect a culture in which the educator’s analysis of the learner’s performance predominates [36]. These findings echo those from earlier observational studies and feedback forms [11, 12, 17, 19, 37,38,39,40]. This suggests that typical feedback practice in the clinical setting has remained much the same since these omissions were last reported years ago.

[자기 평가]는 자기 조절 학습 및 평가 판단의 핵심 요소이며, 이는 성찰, 독립 학습 및 성취를 촉진합니다 [28,29,30]. 학습자 자기 평가를 위한 초대장은 학습자에게 자신의 작업을 먼저 판단하고 자신이 가장 도움을 받고 싶어하는 것이 무엇인지를 나타낼 수 있는 기회를 제공합니다 [33, 41, 42]. 자기 평가는 학습자가 자신의 성과를 교육자보다 훨씬 더 높게 평가할 경우 교육자에게 부정적인 정서적 반응과 교육자의 의견에 대한 거부 가능성을 경고할 수 있다[43]. 자가 평가는 학습자가 관찰된 성과와 원하는 성과 표준에 대한 전문가의 이해를 바탕으로 이해를 교정함으로써 평가 판단력을 향상시킬 수 있는 기회를 제공한다. 학생 피드백 리터러시에 대한 최근 연구는 학습자가 피드백을 평가하고 해석하고 활용할 수 있도록 지원하기 위해, 양질의 작업 특성에 대해 논의할 수 있는 기회를 전략적으로 설계하는 것의 중요성을 강조했다[45].
Self-assessment is a key component in self-regulated learning and evaluative judgement, which promotes reflection, independent learning and achievement [28,29,30]. Invitations for learner self-asssessment provide learners with the opportunity to judge their work first and indicate what they most want help with [33, 41, 42]. Self-assessments can alert the educator to the potential for a negative emotional reaction and rejection of the educator’s opinion if the learner rates their performance much higher than the educator [43]. Self -assessment offer opportunities for learners to enhance their evaluative judgement by calibrating their understanding against an expert’s understanding of the observed performance and the desired performance standards [4, 44]. Recent work on student feedback literacy has highlighted the importance of strategically designing opportunities for learners to make judgements and discuss characteristics of quality work, to assist them to appreciate, interpret and utilise feedback [45].

[실행 계획]이 계속해서 자주 무시된다는 사실도 마찬가지로 심각한 주의를 기울여야 한다. 교육자가 학습자에게 실행 계획을 수립하도록 지원하고 안내하지 않는다면, 학습자는 피드백 정보를 성과 개선으로 전환하는 방법을 스스로 해결해야 하는 어려운 과제를 떠안게 됩니다[21]. 또한, 학습자들이 성과 격차에 대해 들었을 때, 개선 방법을 모른다면, 그들의 고통은 더 악화될 수 있다[46].
The fact that an action plan continues to be frequently neglected similarly warrants serious attention. If educators do not support and guide learners to create an action plan, learners are left with the difficult task of working out by themselves how to transform feedback information into performance improvement [21]. Furthermore, when learners hear about performance gaps, their distress may be exacerbated if they do not know how to improve it [46].

우리의 연구는 또한 이전에 문서화되지 않은 많은 누락된 피드백 기능을 식별했다. 하나는 [학습자의 동기, 이해 및 기술 개발]을 피드백의 초점으로 포지셔닝하는 것이다. 이 문헌은 학습자가 '원하고'(동기), '어떻게 할 줄 아는'(명확한 이해) 때만 변화를 성공적으로 구현할 가능성이 있음을 시사한다.
Our study also identified a number of missing feedback features, which have not been previously documented. One involves positioning the development of a learner’s motivation, understanding and skills as the focal point for feedback. The literature suggests that a learner is only likely to successfully implement changes when they ‘wish to’ (motivation) and ‘know how to’ (clear understanding) [9, 29, 47, 48].

[자기결정이론]은 학습자가 자신의 개인적 가치와 열망에 따라 무엇을 해야 할지 결정할 때 더 높은 성과와 더 높은 웰빙과 연관된 내재적 동기가 촉진된다고 주장한다. 이는 학습자를 의사결정자로, 교육자를 지침으로 하는 권장 교육자 행동에 의해 파악됩니다(표 1: 행동 7, 21, 22 참조). 학습자는 피드백이 신뢰할 수 있고 가치가 있다는 것을 스스로 확신해야 합니다(행동 1, 6, 7, 9, 20, 24) [8, 49, 50]. 교육자와 학습자 사이의 정보, 의견 및 아이디어의 자유로운 흐름은 맞춤형 조언과 좋은 의사 결정을 위한 기초로서 공유된 이해를 창출한다[51]. 또한, [목표 설정 이론]은 학습자의 동기가 성과 격차에 대한 명확한 시각, 학습자에게 구체적이고 달성 가능하며 가치 있는 성과 목표, 그리고 그들의 필요에 따라 실용적이고 맞춤화된 행동 계획에 의해 자극된다고 주장한다(행동 14, 15, 21, 22).
Self-determination theory
argues that intrinsic motivation, which is associated with both higher performance and increased well-being, is promoted when a learner decides what to do, in line with their personal values and aspirations [23,24,25]. This is captured by recommended educator behaviours that position the learner as decision maker and the educator as guide (see Table 1: Behaviours 7, 21, 22). A learner must be convinced for themselves that the feedback is credible and valuable (Behaviours 1, 6, 7, 9, 20, 24) [8, 49, 50]. The free flow of information, opinion and ideas between the educator and learner creates a shared understanding, as a foundation for tailored advice and good decision making [51]. In addition, Goal Setting Theory asserts that a learner’s motivation is stimulated by a clear view of the performance gap, performance goals that are specific, achievable and valuable to the learner, and an action plan that is practical and tailored to suit their needs (Behaviours 14, 15, 21, 22) [22].

최근 피드백의 발전은 학습자가 피드백 정보를 처리하고 활용할 수 있도록 지원할 필요성에 초점을 맞추고 있습니다. 따라서 학습자는 자신의 성과를 '어떻게 향상시킬지' 알게 되어야 한다. 이것은 R2C2 피드백 모델에서 예시되며, 여기에는 학습자가 정보와 정보에 대한 반응을 탐색하고, 스킬 개발을 위한 효과적인 전략을 설계하는 데 도움을 주는 것이 포함됩니다. [사회 구성주의 학습 이론]은 학습자가 다른 사람과의 상호 작용을 통해 새로운 정보의 의미를 만드는 방법을 설명한다[52]. 이러한 능동적 학습을 촉진하기 위해 권장되는 교육자 행동에는 학습자가 자신의 성과를 분석하고 '자신을 위해 해결할 수 있는 문제'(행동 8, 12, 13), 학습자의 어려움이나 질문에 대한 질문(행동 9), 그리고 세션을 끝내기 전에 실행 계획에 대한 학습자의 이해를 확인하는 것이 포함됩니다. (행동 23, 24) [53].
Recent advances in feedback have focused on the need to assist learners to process and utilise feedback information, so they ‘know how to’ enhance their performance. This is exemplified in the R2C2 feedback model, which includes assisting a learner to explore the information, their reactions to it and to design effective strategies for skill development [30, 32, 51]. Social constructivist learning theory describes how a learner makes meaning of new information through interactions with others [52]. To promote this active learning, recommended educator behaviours include encouraging the learner to analyse their own performance and ‘work things out for themselves’ (Behaviours 8,12,13), enquiring about the learner’s difficulties or questions (Behaviour 9) and checking the learner’s understanding of the action plan before concluding the session (Behaviours 23, 24) [53].

본 연구에서 거의 볼 수 없는 효과적인 피드백의 또 다른 특징은, (교육자들이 전반적으로 학습자들에 대한 존중과 지지를 보였음에도 불구하고), 세션을 시작할 때 의도적으로 안전한 학습 환경을 설정하지는 않았다는 것이다. 최근의 문헌은 안전한 학습 환경을 촉진하고 교육 동맹을 구축하는 것의 중요성을 강화했다[34]. 이것은 교육자와 학습자가 확립된 관계가 없을 때 특히 중요한 전략이 될 수 있으며, 이는 짧은 배치와 학습자를 돌보는 다수의 감독자가 있는 현대 직장 훈련에서 점점 더 보편화되고 있는 것으로 보인다[54]. 과도한 불안은 사고, 학습, 기억에 부정적인 영향을 미친다. 피드백은 본질적으로 심리적으로 위험하다. 학습자의 한계가 노출되면, 교육자로부터 낮은 점수나 비판적 코멘트를 받거나, 학습자의 자아의식을 위협할 수 있다. Carless [10]은 피드백의 강력한 관계적, 정서적, 동기적 영향을 고려하여 신뢰의 중요한 역할을 강조했다. 교육자들은 자연적인 불안감에 대처하기 위해 "실수는 기술 습득 과정의 일부"이며, 비판적이 아니라 도움을 주고자 한다는 점을 분명히 할 수 있다[53]. 또한, 교육자가 피드백 세션에 대한 프로세스와 기대를 협상한 경우, 학습자가 앞으로 일어날 일을 모르거나 통제할 수 없을 때 발생하는 불안을 줄일 수 있다[30].

Another feature of effective feedback rarely seen in our study was educators deliberately setting up a safe learning environment at the start the session, although they showed respect and support for learners in general. Recent literature has reinforced the importance of promoting a safe learning environment and establishing an educational alliance [34]. This may be a particularly important strategy when the educator and learner do not have an established relationship, which seems to be increasingly commonplace in modern workplace training with short placements and multiple supervisors attending to learners [54]. Excessive anxiety negatively impacts on thinking, learning and memory [53, 55, 56]. Feedback is inherently psychologically risky; if a learner’s limitations are exposed, this can result in a lower grade or critical remarks from the educator, or threaten a learner’s sense of self [5, 33, 46]. Carless [10] highlighted the important role of trust in view of the strong relational, emotional and motivational influences of feedback. In an attempt to counter the natural anxiety, educators could be explicit that “mistakes are part of the skill-acquisition process” and that they desire to help, not to be critical [53]. In addition, if an educator negotiated the process and expectations for the feedback session, this could reduce the anxiety caused when the learner does not know, or have any control over, what is going to happen [30].

마지막으로 중요한 특징 중 하나는 [학습 활동의 고립]이었다. 우리 연구에서 학습자가 목표 기술을 어느 정도까지 성공적으로 개발할 수 있었는지 검토할 수 있는 시기와 방법에 대해 논의한 교육자는 없었다(행동 25). 이것은 모든 것 중에서 가장 낮은 순위의 행동이었다. Molloy와 Boud[9]는 [피드백 계획을 구현하고 후속 과제에서 진행률을 평가]할 수 있도록 [학습 활동을 연계하여 성과 개발을 촉진하는 것]의 중요성을 강조했다. 감독은 본질적으로 점점 더 단기적이고 단편적이기 때문에, 유사한 과제를 수행하는 평가를 받을 수 있는 또 다른 기회를 의도적으로 계획할 때 학습자와 협력하는 것이 중요한 목표로 보인다.
One final important feature was the isolation of the learning activity. In our study, no educator discused when or how the learner might be able to review to what extent they had been able to successfully develop the targeted skills (Behaviour 25); this was the lowest ranked behaviour of all. Molloy and Boud [9] have emphasised the importance of promoting performance development by linking learning activities, so that feedback plans can be implemented and progress evaluated in subsequent tasks. As supervision is increasingly short-term and fragmented in nature, collaborating with the learner in deliberately planning another opportunity to be assessed performing a similar task seems an important objective.

개별 교육자의 관행
Individual educator’s practice

우리 연구에서 발견된 개별 교육자의 점수 범위는 교육자들이 피드백에 대한 다양한 전문 지식을 가지고 있음을 시사한다. 교육자들은 비디오 분석에 사용된 권장 행동 목록을 보여주지 않았다. 공식적으로 테스트되지는 않았지만, 점수 범위에 걸친 감독자 경험의 확산에 기초하여 더 많은 경험이 더 큰 전문성을 부여한다는 징후는 데이터에 없었다(표 2). 우리는 우리 교육자들의 전문적인 개발 훈련에 대해 묻지 않았다. 잠재적으로 흥미롭긴 하지만, 이 정보는 권장되는 행동과 비교하여 현재 직장 관행을 평가하는 우리의 주요 목표와 밀접한 관련이 있었다. 교육 패러다임이 시간이 지남에 따라 상당히 변화하고 교육자의 행동이 학습자 시절에 사용된 방법을 부분적으로 반영할 수 있다는 점을 고려할 때, [피드백 접근법에서 관찰한 가변성]은 최근 발전에 초점을 맞춘 [전문적 개발이 지속적으로 필요함]을 강조한다.

The range in individual educator’s scores found in our study suggests the educators had variable expertise in feedback. Educators were not shown the check-list of recommended behaviours used in video analysis. Although not formally tested, there was no indication in the data that more experience conferred greater expertise, based on the spread of supervisor experience across the score ranges (Table 2). We did not ask about our educators’ professional development training. Although potentially interesting, this information was tangential to our primary goal of assessing current workplace practice against recommended behaviours. Given that education paradigms have changed considerably across time, and that educator behaviour may partly reflect methods used when they were learners, the observed variability in feedback approaches highlights the need for continuing professional development that focuses on recent advances.

[전문직 간에 현저한 점수 차이가 없다는 것]은 공식적인 만남 내의 피드백 기술이 다르다기보다는 비슷할 수 있다는 것을 시사한다. 따라서 [피드백 리터러시 교육]은 적어도 부분적으로 [보건 전문가 전체의 교육자를 위해 설계]함으로써 상당한 효율성을 얻을 수 있다. 그럼에도 불구하고, 이러한 기술이 비공식적인 피드백 만남 내에서 그리고 다양한 맥락에서 변화하는 정도는 더 많은 연구를 필요로 한다. 진료하는 임상의가 대부분의 [보건 전문 교육(선배 학생과 후배 임상의 모두)을 담당]하지만 [교육 및 훈련 역할에 대한 지정된 기준]은 드물다. 대조적으로 보건 전문가들은 그들의 [임상 기술]에 대해 수년간 훈련을 하고 신중하게 평가받는데 시간을 보낸다.
The lack of striking differences in scores between professions suggests that feedback skills within formal encounters may be more similar than different. Hence feedback literacy training could, at least in part, be designed for educators across the health professions, allowing significant efficiencies. Nevertheless, the extent to which these skills vary within informal feedback encounters and across different contexts requires more study. Practising clinicans are responsible for the majority of health professions training (both senior students and junior clinicians) and yet specified standards for their education and training role are rare. In contrast health professionals spend many years training and being carefully assessed on their clinical skills.

본 연구의 목적은 교육자가 학습자를 참여시키고 동기를 부여하며 학습자가 개선할 수 있는 교육자 행동에 대한 설명을 개발하고, [고품질 학습자-중심 피드백]을 생성하도록 지원하는 것입니다. 일단 임상의가 권고된 행동을 고려할 기회를 갖게 되면 누락된 요소를 실무에 도입하는 것이 상대적으로 쉬울 수 있다. 교육자에게 유용할 수 있는 전략 중 하나는 학습자와 함께 피드백을 비디오로 촬영한 후 목록을 사용하여 자신의 행동을 체계적으로 분석하는 것입니다. 이를 통해 교육자는 성찰적 학습과 목표 설정에도 참여할 수 있습니다 [57, 58]. 또한, 품질 피드백 관행을 재현하는 감독관의 문구나 비디오의 예는 교육자들이 고품질 피드백의 원칙을 새로운 의식으로 변환하는 데 도움이 될 수 있다. 제시한 일련의 행동은 포괄적이지만, [25가지 권장 행동]은 특히 신입 교육자에게 압도적으로 보일 수 있기 때문에, [우선순위를 정하거나 요약]하는 것이 유용할 수 있다.
The aim of our research is to assist educators in generating high quality learner-centred feedback, by developing descriptions of educator behaviours that could engage, motivate and enable learners to improve. It may well be that once clinicians have the opportunity to consider the recommended behaviours, it would be relatively easy for them to introduce missing elements into their practice. One strategy that might be valuable for educators would be to video their feedback with a learner and subsequently use the list to systematically analyse their own behaviours. This would enable educators to also engage in reflective learning and goal setting [57, 58]. In addition, exemplars of supervisors’ phrases or videos re-enacting quality feedback practices may help educators to translate the principles of high quality feedback into new rituals. The set of behaviours is comprehensive however it could be useful to prioritise or summarise them, as 25 recommended behaviours may seem overwhelming, especially to new educators.

장점과 한계
Study strengths and limitations

우리 연구의 강점은 '실제로 무슨 일이 일어나는지'를 밝히고 역량 평가를 위한 밀러의 프레임워크의 최상위 수준을 목표로 하기 위해 일상적인 임상 실습에서 실제 피드백 에피소드의 자체 녹화 비디오 관찰을 포함한다. 참가자들은 병원 실습의 특징인 다양한 임상 교육자 그룹을 포함했다. 교육자의 피드백 관행은 경험적으로 도출된 포괄적인 25가지 관찰 가능한 교육자 행동 세트를 활용하여 체계적으로 분석되었다. 
Strengths of our study include self-recorded video-observations of authentic feedback episodes in routine clinical practice, to reveal ‘what actually happens’ and target the top level of Miller’s framework for competency assessment. Participants involved a diverse group of clinical educators, characteristic of hospital practice. The educators’ feedback practices were systematically analysed utilising an empirically derived, comprehensive set of 25 observable educator behaviours.

우리의 연구에는 여러 가지 한계가 있다. 36명의 참가자 중 작은 표본은 단일 보건 서비스 기관에서 추출한 것이지만, 여러 병원이 있는 호주에서 가장 큰 것 중 하나이다. 참가자들은 연구에 자원했고(이로 인해 자원하지 않은 사람들보다 더 강한 기술을 가진 교육자와 학습자의 일부가 되었을 수 있다), 참가자들은 자신의 성과를 기록하여 잠재적으로 우리의 데이터를 지나치게 낙관적으로 만들었다. 이러한 요인들은 우리의 연구 결과의 일반화 가능성을 제한한다. 피드백 중 교육자 행동 평가에 교육자 행동 설명을 적용할 때, 비율 일관성에 약간의 변화가 있었다. 이것에 대한 한 가지 이유는 교육자의 행동 묘사에 대한 다른 해석일 수 있다. 향후 연구에서는 관찰 가능한 행동과 지원 정보에 대한 설명을 세분화하고, 평가자 간의 합의를 최적화하기 위한 추가적인 관행과 논의를 수행하는 데 관심을 기울일 것이다. 
There are a number of limitations to our study. The small sample of 36 participants were from a single health service, although it is one of the largest in Australia with multiple hospitals. Participants volunteered (which may have resulted in a subset of educators and learners with stronger skills than those who did not volunteer) and participants recorded their own performances, potentially making our data overly optimistic. These factors limit the generalisability of our findings. In the application of the educator behaviour descriptions to the assessment of educator behaviour during feedback, there was some variation in rater consistency. One reason for this could be different interpretations of the educator behaviour descriptions. In future research, attention will be directed to refining the descriptions of observable behaviours and supporting information, accompanied by additional practice and discussion to optimise consensus amongst raters.

비록 비디오 평가자들이 단지 두 개의 건강 전문가들(의사 두 명과 물리치료사 네 명)을 대표했지만, 이것이 그들의 직업을 넘어 교육자들의 행동에 대한 분석에 영향을 미칠 가능성을 제기할 수 있지만, 우리는 이것을 뒷받침할 수 있는 그럴듯한 주장을 볼 수 없다. 많은 교육자들이 (대학, 병원 또는 전문대학의) 공식 피드백 양식을 사용했습니다. 지침에 따라 이 양식을 완성하려고 하는 것은 교육자들의 행동에 영향을 미쳤거나 인지적으로 상당히 까다로울 수 있기 때문에 교육자들의 주의를 산만하게 했을 수 있다. 그러나 피드백 모범 사례가 학습자 평가 루빅과 동시에 발생할 수 없는 설득력 있는 이유는 없습니다. 또한 교육자-학습자 쌍은 일부 품질 피드백 행동, 특히 기대 설정 및 신뢰 확립과 관련하여 발생했을 수 있는 초기 피드백 대화를 할 수 있었지만 비디오에 캡처되지 않았다. 

Although video raters represented only two health professions (two physicians and four physiotherapists), which could raise the possibility that this might influence their analysis of educators’ behaviours beyond their own profession, we cannot see a plausible argument to support this. A number of educators used official feedback forms (from university, hospital or specialty college). Trying to complete these forms in accordance with their instructions, may have influenced educators’ conduct or may have distracted educators’ attention, as they can be quite cognitively demanding. However, there are no compelling reasons why best practice in feedback could not occur in parallel with any learner assessment rubric. In addition, educator-learner pairs could have had earlier feedback conversations, during which some of the quality feedback behaviours may have occurred, particularly relating to setting up expectations and establishing trust, but were not captured on video.

결론들
Conclusions


우리의 연구는 공식적인 피드백 세션 동안, 교육자들이 일상적으로 학습자의 성과에 대한 분석을 제공하고, 과제가 어떻게 수행되어야 하는지 설명하며, 대화 내에서 존중하고 지지한다는 것을 보여주었다. 이들은 모두 품질 피드백의 가치 있고 권장되는 구성 요소입니다. 그럼에도 불구하고 다른 바람직한 행동은 거의 관찰되지 않았다. 종종 누락되었던 중요한 요소들 

  • 피드백 세션을 시작할 때 의도적으로 안전한 학습 환경을 조성한다(세션의 목적, 기대 및 가능한 구조를 명시적으로 명시함으로써). 
  • 자기 평가를 장려한다.
  • 학습자의 동기 부여 및 이해를 활성화합니다. 
  • 실행 계획을 수립하고, 후속 성과 검토를 계획합니다. 

Our study showed that during formal feedback sessions, educators routinely provided their analysis of the learner’s performance, described how the task should be performed, and were respectful and supportive within the conversation. These are all valuable and recommended components of quality feedback. Nevertheless, other desirable behaviours were rarely observed. Important elements that were often omitted included

  • deliberately instigating a safe learning environment at the start of the feedback session (by explicitly articulating the purpose, expectations and likely structure of the session),
  • encouraging self-assessment,
  • activating the learner’s motivation and understanding,
  • creating an action plan and planning a subsequent performance review. 

이는 학습자가 성과 정보를 이해하고 통합하고 행동하도록 돕는 것의 중요성과 관련하여, [피드백 연구의 많은 발전]이 [일상적인 임상 교육에 영향을 미치지 않았음]을 시사한다. 우리의 연구는 보건 전문 교육 커뮤니티 전반에 걸쳐 교육자 피드백 기술 개발을 위한 귀중한 목표를 명확히 한다. 그러나 이러한 권장 교육자 행동의 구현이 설계된 대로 학습자 결과를 향상시키는지 여부를 조사하기 위해 추가 연구가 필요하다. 

This suggests that many advances in feedback research, regarding the importance of assisting learners to understand, incorporate and act on performance information, have not impacted routine clinical education. Our research clarifies valuable targets for educator feedback skill development across the health professions education community. However further research is required to investigate whether implementing these recommended educator behaviours results in enhanced learner outcomes, as designed.

 

 


BMC Med Educ. 2019 May 2;19(1):129. doi: 10.1186/s12909-019-1524-z.

Educators' behaviours during feedback in authentic clinical practice settings: an observational study and systematic analysis

Affiliations collapse

Affiliations

1Monash Doctors Education, Monash Health and Department of Medical Education, Melbourne Medical School, University of Melbourne, Melbourne, Victoria, Australia. christina.johnson@monashhealth.org.

2Department of Physiotherapy, School of Primary and Allied Health Care, Faculty of Medicine Nursing and Health Science, Monash University, Melbourne, Australia.

3Allied Health Workforce Innovation, Strategy, Education & Research (WISER) Unit, Monash Health, and School of Primary and Allied Health Care, Faculty of Medicine, Nursing and Health Sciences at Monash University, Melbourne, Australia.

4Education Portfolio, Faculty Medicine, Nursing and Health Sciences, Monash University, Melbourne, Australia.

5Monash School of Medicine, Faculty of Medicine, Nursing & Health Sciences, Monash University and Monash Health, Melbourne, Australia.

6Department of Medical Education, Melbourne Medical School, University of Melbourne, Melbourne, Australia.

PMID: 31046776

PMCID: PMC6498493

DOI: 10.1186/s12909-019-1524-z

Free PMC article

Abstract

Background: Verbal feedback plays a critical role in health professions education but it is not clear which components of effective feedback have been successfully translated from the literature into supervisory practice in the workplace, and which have not. The purpose of this study was to observe and systematically analyse educators' behaviours during authentic feedback episodes in contemporary clinical practice.

Methods: Educators and learners videoed themselves during formal feedback sessions in routine hospital training. Researchers compared educators' practice to a published set of 25 educator behaviours recommended for quality feedback. Individual educator behaviours were rated 0 = not seen, 1 = done somewhat, 2 = consistently done. To characterise individual educator's practice, their behaviour scores were summed. To describe how commonly each behaviour was observed across all the videos, mean scores were calculated.

Results: Researchers analysed 36 videos involving 34 educators (26 medical, 4 nursing, 4 physiotherapy professionals) and 35 learners across different health professions, specialties, levels of experience and gender. There was considerable variation in both educators' feedback practices, indicated by total scores for individual educators ranging from 5.7 to 34.2 (maximum possible 48), and how frequently specific feedback behaviours were seen across all the videos, indicated by mean scores for each behaviour ranging from 0.1 to 1.75 (maximum possible 2). Educators commonly provided performance analysis, described how the task should be performed, and were respectful and supportive. However a number of recommended feedback behaviours were rarely seen, such as clarifying the session purpose and expectations, promoting learner involvement, creating an action plan or arranging a subsequent review.

Conclusions: These findings clarify contemporary feedback practice and inform the design of educational initiatives to help health professional educators and learners to better realise the potential of feedback.

Keywords: Effective feedback; Feedback; Formative feedback; Health professions education; Professional development.

피드백 탐색의 예술과 술책: 임상실습 학생의 피드백 요청에 대한 접근(Acad Med, 2018)
The Art (and Artifice) of Seeking Feedback: Clerkship Students’ Approaches to Asking for Feedback
Robert Bing-You, MD, MEd, MBA, Victoria Hayes, MD, Tamara Palka, MD,  Marybeth Ford, MD, and Robert Trowbridge, MD 

 

학습자에 대한 피드백은 의학 교육 연구자에게 중요한 연구 영역으로 남아 있다. 의학 학습자를 위한 피드백에 대한 최근의 범위 검토는 문헌이 광범위하고 상당히 최근이며 고품질의 증거 기반 권고가 부족하다고 설명했다. 의학 학습자에게 제공된 피드백 내용에 대한 문헌을 통합 검토한 결과, 피드백 제공자가 부정적 피드백을 제공하는 것을 꺼리는 낮은 품질의 피드백과 관용 편향을 발견하여 학습자가 수년 동안 불만을 제기해 온 내용을 확인했다. 피드백 전달을 촉진하는 수십 년간의 교수개발 노력에도 불구하고 교수진의 부적절한 피드백 전달은 계속해서 발생하고 있다. 이러한 피드백에 대한 학습자의 지속적인 불만족의 문제는 [피드백 상호작용의 중요한 파트너로서 학습자]에게 관심을 옮겨 [적극적인 대화와 교류]에 대한 [사회구성주의적 관점]을 강조하고 있다. 우리는 "피드백은 말하는 것이 아니"며, 양방향 대화이고, "피드백은 정보의 일방적 전송에 해당한다"는 가정에 이의를 제기할 필요가 있다몰로이와 보드의 의견에 동의한다. 학습자가 찾는 이유와 피드백을 얻는 방법을 이해하는 것은 탐구의 중요한 영역입니다.
Feedback for learners remains an important area of study for medical education researchers.1–3 A recent scoping review on feedback for medical learners described the literature as broad, fairly recent, and lacking in high-quality, evidence-based recommendations.4 An integrative review of the literature on the content of feedback given to medical learners found low-quality feedback and leniency bias, in which feedback providers were reluctant to offer negative feedback,5 confirming what learners have complained about for years.6,7 The inadequate delivery of feedback by faculty has continued to occur despite decades of faculty development efforts promoting feedback delivery.8–10 The problem of this prolonged dissatisfaction of learners with feedback has shifted attention to the learner as a significant partner in the feedback interaction, emphasizing a socio-constructivist view of an active dialogue and exchange.11–13 We agree with Molloy and Boud14 that “feedback is not telling” but, rather, a two-way dialogue, and there is a need to challenge the assumption that “feedback constitutes a one-way transmission of information.” Understanding the reasons learners seek and the ways they obtain feedback is an important area of exploration.15


수의학과 학생들을 대상으로 한 연구에서 Bok 등은 학생들이 피드백을 추구하는 이유를 설명하는 요소들을 설명했다. 

  • 피드백 추구자의 이익(예: 부정적 판단 회피), 
  • 피드백 추구자의 특성(예: 특정 임상 영역에 대한 자신감), 
  • 피드백 제공자의 특성(예: 양호한 의사소통 기술). 

[피드백의 개념적 모델]은 [피드백을 추구하는 동기요인]과 [피드백을 추구하는 행동의 결과]를 강조하는 조직심리학과 사회과학 문헌에 의해 알려졌다. 수의학과 학생들을 대상으로 한 후속 연구에서, 성적이 높은 학생들은 낮은 학생들에 비해 피드백을 추구하는 동기가 더 높고 자기 결정력이 더 높은 것으로 나타났다. 그러나 이러한 연구의 결과는 수의학과 학생들에 초점을 맞춘다는 점에서 제한적이었다.
In a study of veterinary medicine students, Bok et al16 described factors that explained students’ reasons for seeking feedback, including

  • interests of the feedback seeker (e.g., avoidance of negative judgments),
  • characteristics of the feedback seeker (e.g., confidence in a specific clinical area), and
  • characteristics of the feedback provider (e.g., good communication skills).

Their conceptual model of feedback was informed by the organizational psychology and social science literatures, which emphasize the motivational factors for seeking feedback and the outcomes of feedback-seeking behaviors.17–19 A follow-up study, again with veterinary students, revealed that high-performing students were more motivated to seek feedback and demonstrated higher self-determination compared with low-performing students.20 However, these studies’ findings were limited by their focus on veterinary students.

의학교육에서 학생들의 피드백을 추구하는 행동에 관한 문헌과 주민들이 피드백을 추구하기 위해 사용하는 행동을 기술하는 최소한의 문헌에는 차이가 있다. Jansen과 Prins는 피드백을 추구하는 의료인들이 "학습 목표" 지향(새로운 기술과 지식에 대한 욕구) 대 "수행 목표" 지향(자신의 역량에 대한 긍정적인 평가와 부정적인 판단의 회피에 대한 욕구)의 경향을 가지고 있다고 설명했다. 이러한 의료 종사자들은 피드백과 관련된 세 가지 다른 비용을 인식했다. 

  • 자기 프레젠테이션 비용(학습자를 무능해 보이게 하는 피드백); 
  • 자아 비용(학습자의 자기 이미지에 맞지 않는 피드백); 
  • 노력 비용(학습자가 얻기 위해 너무 많은 노력이 필요하다고 느끼는 피드백)

Teunissen 등은 OB/GYN 전공의들이 ["문의inquiry" 접근법을 통해 직접 피드백을 요청]하거나, ["모니터링monitoring" 접근법을 통해 학습 환경에서 다른 사람들의 행동을 관찰함]으로써 피드백을 추구하는 방법을 설명했다. 곤트 외 연구진은 Teunissen 외 연구진과 유사한 피드백 추구 모델을 사용하여 트레이너의 감독 역할(즉, 지지 대 도구적 리더십 스타일)이 피드백 편익과 비용에 대한 외과 레지던트의 인식과 그에 따른 피드백 추구 행동에 영향을 미친다는 것을 발견했다.
In medical education, there is a gap in the literature regarding the feedback-seeking behaviors of students and minimal literature describing the behaviors that residents use to seek feedback.21–23 Janssen and Prins23 described feedback-seeking medical residents as having a tendency toward a “learning goal” orientation (desire for new skills and knowledge) versus a “performance goal” orientation (desire for positive assessments of one’s competence and avoidance of negative judgments).24–26 These medical residents perceived three different costs associated with feedback:

  • self-presentation cost (feedback that makes the learner look incompetent);
  • ego cost (feedback that does not fit the learner’s self-image); and
  • effort cost (feedback that the learner feels requires too much effort to obtain).23 

Teunissen et al22 described how OB/GYN residents sought feedback either by asking for it directly via an “inquiry” approach or by observing the behaviors of others in their learning environment via a “monitoring” approach.17 Gaunt et al,21 using a feedback-seeking model similar to that of Teunissen et al,22 found that the trainers’ supervisory role (i.e., a supportive vs. an instrumental leadership style) influenced surgery residents’ perceptions of feedback benefits and costs and their subsequent feedback-seeking behaviors.

의대생들의 피드백 추구 행동에 대해 생각할 때 피드백 과정의 개념화를 전환하면 추가적인 통찰력을 얻을 수 있다. Bowen 등은 피드백의 개념적 모델로 "교육적 동맹"(이러한 목표에 도달하기 위한 목표를 상호 이해하고 책임에 대해 합의하는 과정)을 사용하여 피드백에 대한 의대생들의 인식을 탐구했다. 그들은 "피드백 추구"를 세 가지 피드백 행동 중 하나로 설명했다. 다른 두 가지는 "피드백 인식"과 "피드백 사용"이었습니다. 각각의 행동은 학습자의 신념, 태도, 인식, 관계, 교사 속성, 학습 문화, 피드백 방식에 의해 영향을 받았다. 우리가 아는 한, 연구 당시 의대생들의 피드백 추구 행동을 설명하는 다른 발표된 연구는 없었다.
As we think about the feedback-seeking behaviors of medical students, shifting the conceptualization of the feedback process provides additional insight. Using the “educational alliance” (a process between parties to mutually understand goals and agree on responsibilities to reach these goals)27 as a conceptual model of feedback, Bowen et al1 explored medical students’ perceptions of feedback. They described “feedback seeking” as one of three feedback behaviors; the other two were “recognizing feedback” and “using feedback.” Each behavior was influenced by learner beliefs, attitudes, and perceptions; relationships; teacher attributes; learning culture; and mode of feedback. To the best of our knowledge, there were no other published studies describing medical students’ feedback-seeking behaviors at the time of our study.

Ashford 등은 조직심리학 문헌에서 [개인은 수동적으로 피드백을 받는 사람이 아니라]고 주장했다. 개인은 피드백을 자원으로 보고 다양한 이유와 선호에 따라 피드백을 구합니다. 의학 교육 문헌에서 피드백 추구자와 제공자 간의 능동적 대화, 동맹 또는 심지어 복잡한 춤으로서의 피드백의 새로운 개념 모델은 의대생과 전공의가 피드백 과정을 주도하도록 적극적으로 도와야 한다고 제안한다.
In the organizational psychology literature, Ashford et al28 have argued that individuals are not passive recipients of feedback. Individuals view feedback as a resource and seek feedback based on different reasons and preferences.17 In the medical education literature, the emerging conceptual models of feedback as an active dialogue,12 an alliance,1 or even a complex dance5 between feedback seekers and providers suggest that medical students and residents should actively help drive the feedback process.

의대생들이 선생님과의 피드백 교환에 적극적인 참여자가 될 수 있도록 교육 커리큘럼을 설계함에 있어 의대생들이 자신의 성과에 대한 피드백을 추구하는 이유와 그러한 피드백을 얻기 위한 방법을 조사하는 것이 중요하다고 생각했다. 우리의 의도는 다음과 같은 연구 질문에 초점을 맞추어 Bok et al 16의 작업을 확장하는 것이었다. 

  • (1) 의대생들은 임상실습 시기에 어떤 피드백을 추구하는 행동을 하는가? 
  • (2) 교직원과 학생의 관계는 임상실습 학생들의 피드백 추구 접근법에 어떤 영향을 미칩니까?

In designing an educational curriculum to help medical students be active participants in a feedback exchange with their teachers, we thought it was important to investigate the reasons medical students seek feedback about their performance and how they go about obtaining such feedback. Our intent was to extend the work of Bok et al16 with a focus on the following research questions:

  • (1) What feedback-seeking behaviors do medical students use in the clerkship year? and
  • (2) How does the faculty–student relationship affect the feedback-seeking approaches of clerkship students?

방법
Method

우리는 질적 연구 권고 보고 기준을 준수했다. 
We followed the Standards for Reporting Qualitative Research recommendations.29

설정 및 참가자
Setting and participants

이 연구는 메인주 포틀랜드에 있는 독립적인 학술 의료 센터인 메인 메디컬 센터(MMC)에서 수행되었으며, 이는 터프츠 의과 대학의 2년 지역 분교 캠퍼스입니다. 일반적으로 매년 36명의 3학년 학생들과 36명의 4학년 학생들이 메인 트랙에 등록한다. 임상실습 학생들은 MMC에서 전통적인 6주에서 8주간의 입원 환자 블록 순환이나 MMC 또는 메인 주 전역의 9개 시골 지역 병원 중 하나에서 주로 외래 환자 중심의 종방향 통합 임상실습(LIC)을 이수한다. 시골 지역의 LIC 사이트는 일반적으로 한 번에 두 명의 임상실습 학생을 호스팅한다.
The study was carried out at Maine Medical Center (MMC), an independent academic medical center in Portland, Maine, which is a two-year regional branch campus for Tufts University School of Medicine. Typically, 36 third-year (clerkship) and 36 fourth-year students are enrolled in the “Maine Track” each year. The clerkship students complete either traditional, six- to eight-week inpatient block rotations at MMC or a nine-month, predominantly outpatient-oriented longitudinal integrated clerkship (LIC) at MMC or one of nine rural community hospitals throughout the state of Maine. The rural LIC sites typically host two clerkship students at a time.

2017년 3월, 메인 트랙 임상실습 학생들(n = 35)이 이메일로 초청되어 연구 인터뷰에 참여했다. 우리는 블록 로테이션과 LIC 학생을 거의 동일한 수의 모집하기 위해 목적적인 샘플링 접근법을 선택했다: LIC 학생 8명과 블록 로테이션 학생 1명이 처음 응답한 후, 우리는 반복적인 이메일 초대를 보내 로테이션 학생들의 참여를 증가시켰다. 연구 인터뷰는 2017년 4월~5월에 완료되었으며, 이는 학생들이 임상실습 학년에서 4학년으로 전환한 시기와 일치한다. 학생들은 연구의 기밀성을 알게 되었고, 참여 후 작은 금액의 기프트 카드를 받았다. MMC 기관 검토 위원회는 그 연구를 면제하는 것으로 결정했다.
In March 2017, Maine Track clerkship students (n = 35) were invited by e-mail to volunteer to participate in the study interviews. We chose a purposive sampling approach to recruit approximately equal numbers of block rotation and LIC students: After eight LIC students and one block rotation student responded initially, we sent repeat e-mail invitations to block rotation students to increase their participation. The study interviews were completed in April–May 2017, coinciding with the students’ transition from the clerkship year to the fourth year. The students were informed of the confidential nature of the study, and they received a gift card of a small value after their participation. The MMC Institutional Review Board determined the study to be exempt research.

스터디 디자인
Study design

이러한 질적 연구는 복 등이 강조한 조직심리학 문헌과 교육동맹의 틀에서 영감을 얻었다근거이론은 "연구 과정에서 이론을 생산하기 위해 데이터 수집과 분석 사이의 지속적인 상호작용을 요구하는" 질적 방법이다. 반구조화 면담과 자료분석을 위해 구성주의적 근거이론 접근법을 선택한 이유는 [자료에 대한 귀납적 분석이 특정 이론의 연역적 적용보다 적절하다]고 생각했기 때문이며, [연구자로서의 역할의 배경과 가정, 편향성을 인정]했기 때문이다.

This qualitative study was inspired by the organizational psychology literature,17–19 as highlighted by Bok et al,16 and the educational alliance framework.1,27 Grounded theory is a qualitative method “that calls for a continual interplay between data collection and analysis to produce a theory during the research process.”30 We chose a constructivist grounded theory approach for the semistructured interviews and data analysis because an inductive analysis of the data was thought to be more appropriate than a deductive imposition of a specific theory, and we acknowledged the backgrounds, assumptions, and biases of our roles as researchers.31

면접안내 및 절차
Interview guide and procedure

4개 질문 인터뷰 가이드는 Bok 등이 설명한 세 가지 주요 질문을 기반으로 했다.

  • (1) 임상 작업 수행에 대한 정보를 찾는 이유는 무엇입니까? 
  • (2) 피드백을 구하는 방법에 영향을 미치는 요소는 무엇입니까? 
  • (3) 당신은 당신의 성과에 대한 정보를 어떻게 얻나요? 
  • 교사-학습자 관계가 피드백에 미치는 영향을 강조하는 최근 문헌 때문에 다음과 같은 질문을 추가했다. 
    (4) 교수진과의 관계가 피드백을 구하는 방식에 어떤 영향을 미칩니까?

The four-question interview guide was based on the three main questions described by Bok et al16:

  • (1) Why do you seek information about your performance of a clinical task?
  • (2) Which factors influence the way you seek feedback?
  • (3) How do you obtain information about your performance?
  • Because of the recent literature emphasizing the impact of the teacher–learner relationship on feedback,1,5,27 we added a question:
    (4) How does your relationship with faculty influence the way you seek feedback?

각 인터뷰는 저자 중 한 명(R.B. [MMC 의학부] 또는 V.H. [MMC 가정의학부 교수])이 개인 사무실에서 전화로 진행한 두 번의 인터뷰를 제외하고 진행되었다. 이 저자들은 둘 다 질적 연구와 의학 교육의 피드백에 관한 연구 경험이 있으며, 임상실습에 시절에 참여한 학생들을 위한 지도자나 임살실습 책임자가 아니었다. 개별 인터뷰는 약 30분간 진행되었고 녹음되었다. 현장 노트는 인터뷰할 때마다 면접관이 작성했다. 녹음 테이프는 고용된 전사자에 의해 말 그대로 전사되었고, 학생들은 인터뷰가 전사된 순서에 따라 참가자 식별자 이니셜을 할당받았다. 이러한 이니셜은 결과에 보고된 대표적인 인용구와 함께 포함됩니다. 텍스트는 코딩 및 분석을 위해 정성적 연구 소프트웨어 시스템(NVIVO 10, QSR International, London, 영국)에 업로드되었다. 
Each interview was conducted by one of the authors (either R.B. [chief of the MMC Department of Medical Education] or V.H. [a faculty member in the MMC Family Medicine Department]) in a private office setting, except for two interviews that occurred over the telephone. Both of these authors had experience in qualitative research and in research regarding feedback in medical education, and neither was a preceptor or clerkship director for the participating students during their clerkship year. The individual interviews lasted approximately 30 minutes and were audiotaped. Field notes were taken by the interviewer during each interview. The audiotapes were transcribed verbatim by a hired transcriptionist and deidentified, with students assigned participant identifier initials according to the order in which their interviews were transcribed. These initials are included with the representative quotes reported in the Results. The text was uploaded into a qualitative research software system (NVivo 10, QSR International, London, United Kingdom) for coding and analysis.

데이터 분석
Data analysis

인터뷰 녹취록은 초기 코드를 개발하기 위해 저자 중 두 명(R.B., V.H.)에 의해 독립적으로 검토되었다. 4개의 녹취록을 코딩한 후, 이 두 조사관은 코딩 체계를 검토하기 위해 만났고, 그 후 나머지 모든 녹취록을 각각 독립적으로 코딩했다. 지속적 비교 방법을 사용하여 코드로부터 테마를 개발하였다. 모든 기록의 코딩이 완료된 후, 두 조사관은 이론적 충분성이 달성되었으며(즉, 더 이상의 이론적 정보가 표면화되지 않음) 추가 인터뷰가 필요하지 않다는 데 동의했다. 세 번째 조사관(T.P.)은 모든 녹취록을 검토하여 데이터 분석의 철저성을 보장하고 추가 코드가 필요하지 않음을 확인했다. 다섯 명의 연구팀 구성원 모두가 만나서 코딩에서 주제를 해석하는 것에 대한 합의에 도달했다. 구성원 확인은 데이터 진위 여부를 평가하기 위해 인터뷰에 응한 첫 4명의 학생들에게 각각의 오디오 테이프와 관련 기록을 보내면서 수행되었다. 이 네 명의 학생들은 어떠한 수정이나 추가도 제안하지 않았다.
The interview transcripts were independently reviewed by two of the authors (R.B., V.H.) to develop initial codes. After coding four transcripts, these two investigators met to review the coding scheme, and then each independently coded all the remaining transcripts. Themes were developed from the codes using the constant comparative method.32 After coding of all transcripts was complete, both investigators agreed that theoretical sufficiency33 had been achieved (i.e., no further theoretical information was surfacing) and that no additional interviews were necessary. A third investigator (T.P.) reviewed all of the transcripts to ensure thoroughness of the data analysis and that no additional codes were needed. All five research team members met to reach consensus on interpreting the themes from the coding. Member checking was conducted with the first four students interviewed by sending them their respective audiotapes and associated transcripts to assess for data authenticity.33 No revisions or additions were suggested by these four students.

결과.
Results

14명의 학생(남자 6명, 여자 8명)이 연구에 참여했다. 6명의 참가자는 블록 로테이션 학생이었고, 8명은 LIC 학생이었다. LIC 학생들 사이에서, 7개의 다른 병원 사이트들이 대표되었다.
Fourteen students (six men, eight women) participated in the study. Six participants were block rotation students, and eight were LIC students. Among the LIC students, seven different hospital sites were represented.

우리는 4개의 면접 질문과 14개의 관련 주제(표 1)의 제목으로 임상실습 학생들의 피드백 추구 행동에 대한 연구 결과를 제시한다. 다른 모델과 비교하여 새로운 관점에는 환경 요인, 큐잉, 기교 및 피드백 찾기 훈련이 포함된다.
We present our findings on clerkship students’ feedback-seeking behaviors under the headings of the four interview questions and the 14 associated themes (Table 1). New perspectives, compared with other models,1,16–19,27 include environmental factors, cueing, artifice, and training in seeking feedback.

임상 작업 수행에 대한 정보를 찾는 이유는 무엇입니까?
Why do you seek information about your performance of a clinical task?

목표 지향.
Goal orientations.

학생들은 종종 피드백을 추구하는 것이 그들의 기술을 향상시키거나 새로운 지식을 얻기 위한 수단이라고 묘사했다. 때때로 학생들은 자신의 역량에 대한 긍정적인 평가를 확인하고, 부정적인 평가를 피하기 위해 성과 목표로 피드백을 구한다고 지적했습니다.
Students often described seeking feedback as a means to improve their skills or acquire new knowledge. Occasionally, students indicated that feedback is sought as a performance goal to confirm a positive—and avoid a negative—assessment of their competence.

의대생으로서 실력과 지식을 향상시키는 것이 무엇보다 중요하다고 생각합니다. (CC)
I think first and foremost is to improve my skills and knowledge as a medical student. (CC)
[W]제가 원하는 것만큼 잘 하지 못했을 때 저는 그것을 시도했습니다. 그래서 일주일에 한두 번 남은 기간 동안 부탁을 하기 시작했는데, 그렇게 하지 않는 것이 너무 어리석은 일이었기 때문이다. (NN)
[W]hen I didn’t do as well as I wanted I kind of went for it.… So I started asking once or twice a week for the rest of the year, just because it was such a silly thing not to. (NN)

인식된 이점.
Perceived benefits.

그들의 지식과 기술을 향상시키기 위해 동기부여를 받는 것 외에도, 학생들은 더 나은 의사가 되고, 그들의 성적을 향상시키고, 그들의 성과에 대해 더 좋게 느끼는 것을 포함하여, 피드백을 추구함으로써 얻는 다양한 이점들을 인식했다.
Besides being motivated to improve their knowledge and skills, students perceived a variety of benefits from seeking feedback, including being a better doctor, improving their grades, and feeling better about their performance.

저는 이곳 의대에서의 시간과 제가 거기에 쏟은 모든 자원을 의사의 능력을 향상시키기 위한 시간으로 보고 있으며, 우리가 할 수 있는 가장 중요한 것은 좋은 피드백을 받는 것이라고 생각합니다. (JJ)
I view my time here in medical school and all the resources I’ve put into it as time to improve my ability in terms of doctoring and I think the most important thing we can do is just receive good feedback. (JJ)

 

평가받는 것이 우리 성적의 일부이기 때문에 피드백을 구합니다. 그래서 여러분이 어떤 입장에 서 있는지 아는 것은 좋은 일입니다. 그래서 성적은 피드백을 추구하도록 동기를 부여한다. (II)
I seek feedback because it is part of our grade to be evaluated. So it’s good to know kind of where you stand. So, grades motivate me to seek feedback. (II)

학생들은 피드백을 추구하는 것이 그들에 대한 다른 사람들의 인식된 이미지를 향상시킬 수 있다고 지적했다. 학생은 정보를 받기 위해서가 아니라 단순히 자신이 관련되었다는 것을 보여주기 위해 피드백을 요청할 수 있다.
Students pointed out that seeking feedback may enhance others’ perceived image of them. A student might ask for feedback not for the sake of receiving of information but, simply, to show that he or she was involved.

사람들에게 당신이 engage했다는 것을 알리는 것처럼...더 많은 피상적인 이유들이 있다. (JJ)
There are more superficial reasons … to let people know that you’re engaged. (JJ)

인식된 비용.
Perceived costs.

학생들은 피드백을 찾는 것의 이점 외에도 적극적으로 피드백을 찾는 것을 방해할 수 있는 잠재적인 비용을 언급했다.
Besides the benefits of seeking feedback, students cited potential costs of seeking feedback that could prevent them from actively seeking it out.

[W]기분이 좋지 않을 때는 피드백에 대해 더 긴장하고 피드백을 찾을 가능성이 적다고 생각합니다. (II)
[W]hen you’re not feeling great I think that also makes you more nervous about feedback and makes you less likely to seek feedback out. (II)

학생 특성.
Student characteristics.

학생들은 피드백을 추구하도록 동기를 부여하는 자신의 측면을 설명했다. 편안함을 느끼는 것이 중요해 보였다.
Students described aspects of themselves that motivated them to seek feedback. Feeling comfortable appeared to be important.

내가 더 편안하게 느꼈던 교수진은 일반적으로 피드백을 요청하는 것이 더 편안했다. 제 편안함에 영향을 준 것들은 제가 그 분야에서 얼마나 편안하게 느꼈는지에 대한 것이었습니다. (DD)
The faculty I felt more comfortable with I generally felt more comfortable asking feedback from. Things that influenced my comfort level were sort of broadly how comfortable I felt in the discipline. (DD)

비록 일부 학생들은 피드백을 구함으로써 적극적인 학습자가 되어야 한다고 생각했지만, 그들이 교수들에게 부담이 되는 것처럼 느끼는 것은 때때로 학생들이 피드백을 구하는데 방해가 되었다: "나는 아무도 괴롭히고 싶지 않다." (AA). 학생들은 주로 부정적인 피드백에 적응했고 긍정적인 피드백을 인식하지 못했다. 잘 하고 있지 않다고 생각하는 것은 일부 학생들이 피드백을 찾는 것을 피하도록 유도할 수 있지만, 이러한 상황은 다른 학생들이 피드백을 추구하도록 자극했다.
Although some students thought they should be active learners by seeking feedback, feeling as if they were a burden to faculty sometimes deterred students from seeking feedback: “I don’t want to bother anyone” (AA). Students were mainly attuned to negative feedback and not as cognizant of positive feedback. Thinking one was not doing well might prompt some students to avoid seeking feedback, but this situation provoked others to pursue feedback.

내가 더 힘들어하는 것처럼 느껴지는 로테이션이 피드백을 더 많이 요구한다는 것을 알게 되었습니다. 실제로 이 로테이션에서 조금 더 잘 할 수 있을 것입니다. (EE)
I found that rotations where I felt like I was struggling more I asked for feedback more … I actually would up doing a little better in these rotations. (EE)

피드백 공급자 특성.
Feedback provider characteristics.

학생들은 학생들이 피드백을 찾도록 유도하는 피드백 제공자의 몇 가지 특성을 강조했습니다. 예를 들어, 프롬프트 없이 자발적인 피드백을 제공하는 것입니다. 학생들은 자신의 관찰이나 좀 더 일반적인 의미에서 수집된 그러한 특성을 "내가 좋은 교육자라는 인상을 받는다면"(JJ)이라고 설명했다. 효과적인 피드백 제공자들은 "피드백을 주는 것에 열려있다" (II), "조금 더 차분하다" (BB), "교육에 대해 매우 사려깊다" (CC), "칭찬 샌드위치" (AA)와 같은 기술을 사용했으며 가르치는 것을 즐겼다. 학생들은 또한 그들이 더 잘하도록 밀어준 피드백 제공자들에게 감사했다.
Students underscored several characteristics of the feedback provider that would encourage the student to seek feedback—for example, giving spontaneous feedback without prompting. Students described such characteristics gathered from their observations or from a more general sense: “If I get the impression that they are a good educator” (JJ). Providers of effective feedback were “open to giving feedback” (II), “a little more calm” (BB), “very thoughtful about teaching” (CC), used techniques like a “compliment sandwich” (AA), and enjoyed teaching. Students also appreciated feedback providers who pushed them to do better.

그녀는 정말 수준이 높았어요. [S]그는 그녀가 당신이 더 잘하기를 응원하는 것처럼 느끼게 하면서 어떻게 하면 더 잘 할 수 있는지 말해줄 것이다. (BB)
She really had high standards.… [S]he would tell you how to make it better while all the time making it feel like she was rooting for you to do better. (BB)

학생들은 피드백 제공자의 성격에 대한 인식이 피드백을 구할지 여부와 그 방법에 영향을 미친다고 지적했다. 위협적으로 보이는 피드백 제공자들은 피했다. 학생들이 피드백을 구하도록 유도하는 긍정적 피드백 제공자 특성이 있었던 것처럼, 학생들이 해당 개인에게 피드백을 계속 구하고자 하는 욕구를 위축시키는 부정적 특성이 있었다.

Students indicated that their perception of the feedback provider’s personality had an impact on their decision whether to seek feedback and the ways in which they did so. Feedback providers who appeared to be intimidating were avoided. Just as there were positive feedback provider characteristics prompting students to seek feedback, there were negative characteristics that dampened students’ desire to continue to seek feedback from that individual.

나는 다른 성격 유형으로 사람들이 피드백을 적는 것이 얼마나 더 편안한지 확실히 알 수 있었다. (HH)
I could definitely see with different personality types how people would be more comfortable writing down feedback. (HH)
만약 내가 교수진과 나쁜 임상 경험을 했다면 나는 그들에게 피드백을 구하지 않았을 것이다. (II)
If I had a bad clinical experience with a faculty member I wouldn’t seek feedback from them. (II)

피드백을 구하는 방법에 영향을 미치는 요소는 무엇입니까?
Which factors influence the way you seek feedback?

바쁜 환경.
Busy environment.

몇몇 학생들은 때때로 정신없이 바쁜 임상 환경이 피드백을 추구하지 않는 데 영향을 미치는 강력한 요인이라고 언급했다.
Several students commented that the sometimes-frenetic clinical environment was a strong factor influencing them to not seek feedback.

만약 바쁘다면, 사람들이 방을 날아다니고 있다면. "저, 제가 제대로 했는지 알려주실래요?"같은 말은 안 할거에요 (LL)
If it was busy, if people are flying around the room. It’s crazy. I’m not going to say, “Hey, can you tell me if I did that correctly?” (LL)

타이밍.
Timing.

학생들은 임상 환경이 덜 바쁜 시간, 종종 하루나 한 주의 끝이 피드백을 받기에 더 최적의 시간이라고 설명했다. 학생들은 또한 적절한 비공식 피드백을 얻을 수 있는 추가적인 순간들을 강조했다.
Students described times when the clinical environment was less busy, often at the end of the day or week, as more optimal times to seek feedback. Students also highlighted additional moments when opportune informal feedback could be obtained.

피드백은 항상 하루에 함께하는 시간의 끝에 일어난다. (DD)
Feedback would always happen at the end of our time together on a day. (DD)

단서.
Cues.

학생들은 거주자나 교직원들과 일할 때 신호에 매우 익숙했다. 특정한 신호들은 [학생들이 학생의 이름을 배우거나 학생에게 관심을 보이는 것]과 같은 피드백을 얻도록 격려할 것이다.
Students were very attuned to cues when working with residents or faculty. Certain signals would encourage students to seek feedback, such as learning the student’s name or showing interest in the student.

만약 그들이 방에서 걸어나와 "오, 나는 네가 이런 짓을 했다는 것을 알아차렸다. 여기 당신이 작업할 수 있는 것이 있습니다."이라고 말했다. 그것은 정말로 다른 사람들이 하지 않는 방식으로 그들이 주의를 기울이고 있다는 신호였다. (CC)
If they walked out of a room and said, “Oh I noticed that you did this. Here’s a thing that I think you can work on.” That was really signaling me that they were paying attention in a way that other people were not. (CC)

당신은 당신의 성과에 대한 정보를 어떻게 얻나요?
How do you obtain information about your performance?

탐색 대 모니터링.
Inquiry versus monitoring.

학생들은 종종 다음과 같은 inquiry을 통해 자신의 성과에 대한 피드백을 얻었습니다. "야, 어떻게 된 것 같아? 다음 번에는 어떻게 다르게 해야 할까요?" (AA) 학생들이 묻는 질문은 일반적인 유형의 질문보다 더 구체적이었습니다. "이 특정 상황이 어떻게 진행되었다고 생각하십니까?" [T]주말이 되자 그것은 더 일반적이었다." (II)
Students often obtained feedback about their performance through inquiry—for example, asking: “Hey, how do you think that went? How should I do this differently next time?” (AA). Questions the students asked were typically more specific than a general type of inquiry: “How do you think this specific situation went?… [T]owards the end of the week it was more general.” (II)

학생들이 피드백을 얻기 위해 사용한 질문의 우선순위를 보고했지만, 학생들은 그들이 어떻게 수행하고 있는지 감지하기 위해 환경에서 다른 학습자를 관찰하는 것에 대해 종종 설명하지 않았다. LIC 학생들은 모니터링 할 수 있는 다른 학생이 많지 않았다.
Although students reported a preponderance of questions they used to obtain feedback, they did not often describe observing other learners in their environment to sense how they were performing. LIC students did not have many other students rotating with them whom they could monitor.

간접적 기술
Artifice.

일부 학생들은 "직접적으로 피드백을 요청한 적이 없다"(AA)고 지적했다. 대신, 그들은 피드백을 얻기 위해 "나는 항상 다른 것을 가지고 있고 그들이 피드백에 대해 이야기하도록 속이는 것을 좋아한다"(AA)와 같은 몇 가지 간접적이고 비조회적인 방법을 사용했다. 우리가 "똑똑하고 교묘한 기술: 독창성; 기발한 장치 또는 편법"으로 정의한 이러한 방법들은 [대화 스타터, 피드백 준비, 자기 비판적인 논평 사용]과 같은 기술을 포함했다.
Some students indicated that they “never asked for feedback directly” (AA). Instead, they used several indirect and noninquiry methods to get feedback—for example, “I always have something else and like trick them into talking about feedback” (AA). These methods—which we labeled as “artifice,” defined as “a clever or artful skill: ingenuity; an ingenious device or expedient”34—included such techniques as conversation starters, prepping for feedback, and using self-critical comments.

만약 누군가가 병력청취하는 나를 관찰할 것이라는 것을 안다면, 그들이 나를 관찰할 수 있다면, 나는 그들에게 피드백을 요청할 것이다. (JJ)
If I know somebody is going to be observing me in doing a history or something like that I might ask them for feedback right beforehand if they can observe me, give me feedback afterwards. (JJ)
[Y]여러분은 일종의 자기비하적인 발언으로 이끌 수 있습니다. "저는 정말로 제가 그 모든 것을 잘 하고 있다고 생각하지 않습니다.…" (FF)
[Y]ou can lead with a sort of self-deprecating statement saying, “I really don’t think I’m doing all that well with.…” (FF)
[G]나를 건설적으로 비판하도록 그들을 초대한다. 당신은 "내가 이것을 잘못했다고 생각한다" 또는 "나는 내가 어떻게 이것을 했는지 편안하지 않다"라고 말해야 한다. (AA)
[G]ive them an invitation to constructively criticize me. You have to say, “I think I did a bad job on this” or “I am not comfortable with how I did this.” (AA)

학생들이 사용한 다른 [간접적인 기법]에는 피드백 제공자를 칭찬하고, 문제에 대한 의견이나 접근 방식을 제공자에게 묻고, 피드백 제공자에게 감사하는 것이 포함되었다.
Other indirect techniques that students used involved praising the feedback provider, asking the provider for their opinion or approach to a problem, and thanking the feedback provider.

[T]당신이 아주 잘한다는 걸 알아챘어요. 지식의 기반을 구축하려면 어떻게 해야 합니까? (EE)
[T]his is something I’ve noticed you do very well. What can I do to try to build that foundation of knowledge? (EE)
나는 그것을 종종 "당신은 어떻게 그렇게 했겠어요?"라고 표현한다. 그들은 종종 "이것이 당신이 한 일이고 이것이 내가 그것을 했을 방법이다."라고 채운다. (BB)
I often phrase it as like, “How would you have done that?” They often will end up filling in, “This is what you did and this is how I would have done it.” (BB)
[J]그저 감사하고 고맙다고 말하는 것이 피드백을 피드백이 계속 올 수 있도록 하는 좋은 방법이라고 생각합니다. (LL)
[J]ust saying thank you, being appreciative I think of feedback is a good way to make sure feedback will keep coming. (LL)

비록 학생들이 주로 구두 질문을 통해 피드백을 구했지만, 학생들은 또한 때때로 덜 위협적인 접근법으로 피드백을 얻기 위해 서면 또는 전자 매체를 사용한다고 지적했다.
Although students predominantly sought feedback through verbal inquiry, students also indicated they used written or electronic media to get feedback, sometimes as a less threatening approach.

총괄적 피드백.
Summative feedback.

대부분의 경우, 학생들은 [공식적 피드백 제공 접근법]이 [스스로 피드백을 구하는 것]만큼 도움이 된다고 생각하지 않았다. 공식적인 접근 방식에는 필요한 블록 중간 및 끝 회전 피드백, 3개월 및 6개월 LIC 평가 세션, 그리고 각 임상실습에서 DOC(직접 관찰) 카드 세트의 의무적인 교수진 완료가 포함되었다. (DOC 카드는 역사 체크리스트와 신체검사 요소와 같은 기본 역량을 다루었으며, 뒷면에 긍정적이고 건설적인 피드백을 위한 공간을 가지고 있었다.)

For the most part, students did not find the formal approaches to providing students with feedback to be as helpful as seeking feedback on their own. The formal approaches included required mid- and end-of-block rotation feedback, three- and six-month LIC evaluative sessions, and mandatory faculty completion of a set of DOC (direct observation) cards during each clerkship. (The DOC cards covered basic competencies, such as a checklist of history and physical examination elements, and had space for positive and constructive feedback on the back.)

3개월과 6개월의 피드백 중 일부는 도움이 되지 않았습니다. 일반적으로 "오, 그는 좋은 학생이고, 열심히 공부한다." … [D]제가 하고 있는 구체적인 일에 대해서는 언급하지 않았습니다. (EE)
Some of the feedback at three and six months wasn’t as helpful. Generally, “Oh, he’s a good student, he works hard.”… [D]idn’t mention specific things that I was doing. (EE)
대부분의 DOC 카드는 그다지 도움이 되지 않습니다. 건설적인 피드백의 좋은 루프와는 반대로 대부분의 작업인 것 같습니다. (KK)
Mostly DOC cards are not very helpful … it seems it’s mostly a task as opposed to a good loop of constructive feedback. (KK)

피드백을 찾는 교육.
Training in seeking feedback.

학생들은 [의대생들을 위한 교육]이 피드백을 구하는 데 도움이 될 것이라고 언급했고, 그들은 그러한 프로그램에서 학생들에게 무엇을 말할 것인지에 대한 조언을 제공했다. 학생들을 건설적인 피드백에 둔감하게 하는 것이 유익한 접근법으로 언급되었다.
Students indicated that training for medical students in seeking feedback would be helpful, and they offered advice as to what they would tell students in such a program. Desensitizing students to constructive feedback was mentioned as a beneficial approach.

저는 대학에서 꽤 잔인한 4학년 논문을 썼는데, 거기서 피드백을 많이 교환하고, 피드백을 받아 변화를 만드는 데 정말 익숙해졌어요. (NN)
I had a pretty brutal senior thesis year in college where we did a lot of exchanging feedback … that got me really used to taking in feedback and making changes with it. (NN)
[C]가끔은 조금 아플 때도 있다 그래서 피부가 두꺼워지면서 건설적인 피드백을 찾았던 것 같아요. (LL)
[C]onstructive [feedback] hurt a little bit sometimes. So I think as I got thicker skinned I guess, I looked for constructive feedback. (LL)

교수진과의 관계가 피드백을 구하는 방식에 어떤 영향을 미칩니까?
How does your relationship with faculty influence the way you seek feedback?

관계 역학.
Relationship dynamics.

학생들은 피드백을 추구하는 행동에 영향을 미치는 [피드백 제공자와의 관계]의 여러 측면을 설명했다. 학생들은 교직원들이 접근할 수 있을 정도로 충분히 강한 관계를 느껴야 했다.
The students described several aspects of their relationship with feedback providers that influenced their feedback-seeking behaviors. Students had to feel the relationship was strong enough such that faculty seemed approachable.

접근하기 쉬운 사람이 있다면. 피드백을 받는 것이 더 편하다고 느꼈다. 그 사람이 교육 친화적이라는 것을 알았다면… 당신은 그들로부터 피드백을 받고 싶을 것이다. (II)
If somebody is approachable. That I felt more comfortable getting feedback. If I knew that the person was education friendly … you want to seek feedback from them. (II)

그들이 동료라고 느끼거나 피드백 제공자와 공통점이 있다는 것은 피드백을 찾는 것이 더 쉬운 지원 관계를 촉진했다.
Feeling like they were colleagues or having something in common with a feedback provider facilitated a supportive relationship in which seeking feedback was easier.

[H]나는 그들과 많은 관계를 맺고 있다… 만약 우리가 더 많은 개인적인 것에 대해 이야기하게 된다면… 그런 식으로 피드백에 영향을 준다. (JJJ)
[H]ow much of a rapport I establish with them … if we got to talking about more personal things … in that way it influences the feedback. (JJ)

학생들은 [상호 신뢰감과 관계에 대한 관심caring]이 중요하다고 강조했다.
Students highlighted a sense of mutual trust and caring in the relationship as important.

그들이 실제로 당신이 더 나아지는 것을 보고 싶어한다고 느낄 때가 온다. 그것이 그들의 목표인 것처럼, 당신이 느낄 수 있는 것처럼. 그리고 그들은 당신과 함께 그 일에 관여하고 있어요. (DD)
There comes a point when you feel like they actually want to see you become better. Like that’s their goal, like you can feel it. And that they’re in on that with you. (DD)

학생들은 또한 피드백을 제공할 때 전공의와 교직원들이 그들에게 지나치게 친절하다는 것을 인식했고, 이것은 효과적인 피드백 의사소통을 방해했다.
Students also recognized that residents and faculty were too nice to them when getting around to providing feedback, and this inhibited effective feedback communication.

나는 때때로 [그들이 그렇게 일찍부터 나를 기분 나쁘게 하고 싶지 않아하는구나]고 느꼈다. 그들은 우리의 교사와 학습자의 관계를 바꾸고 싶어하지 않았다. 그들은 피드백을 주고 있는 사람의 감정을 상하게 하고 싶지 않기 때문에 피드백을 주고 싶지 않을 수도 있다. (BB)
I felt sometimes they didn’t want to make me feel bad that early on. They didn’t want to change our teacher–learner relationship. They might not like to give feedback because they don’t want to hurt the feelings of the person they’re giving it to. (BB)

LIC 대 블록 회전입니다.
LIC versus block rotations.

학생들은 학생들이 LIC에 비해 [블록 로테이션에 대한 교수진과의 상호작용이 훨씬 적었고], 블록 로테이션에서는 교수진보다 전공의들이 더 많은 피드백을 제공했으며, 학생들은 전공의들을 교수진이라기보다는 동료로 보았다는 점을 강조했다.
Students emphasized that students had much less interaction with faculty on the block rotations than the LIC, that residents provided more feedback than faculty on block rotations, and that students viewed residents more as peers than as faculty.

학생들이 받는 전공의 교육이 많거나, 오히려 전공의들이 수업을 많이 한다는 점에서, LIC와 블록이 조금 다르다고 생각합니다. (KK)
I think the block is a little bit different than the LIC in that there is a lot of resident education that the students get, or rather the residents do a lot of teaching. (KK)

LIC의 긴 기간과 현장에서의 적은 학습자는 교직원과의 더 나은 관계를 촉진하고 LIC 학생들에게 향상된 피드백 기회를 제공했다. 몇몇 학생들은 약 3~4개월 동안 교수진과의 관계가 피드백을 추구하는 행동을 촉진할 수 있을 정도로 견고하다고 지적했다.
The long duration of the LIC and fewer learners at the sites facilitated better relationships with faculty and enhanced feedback opportunities for LIC students. Several students indicated that approximately three to four months into the LIC, their relationships with faculty were solid enough to facilitate feedback-seeking behaviors.

분명히 LIC 대 블록 실습의 차이가 있었다. LIC에서 제 경험의 대부분은 외래 환자였는데, 이는 오늘날의 상황과는 다른 흐름을 가지고 있습니다. 입원 환자에 비해 환자 상황은 예측하기 어렵습니다. (EE)
There definitely was a LIC versus block difference. In the LIC most of my experience was outpatient, which has a different flow to the day … you know when people are going to be available, versus inpatient things are bit more unpredictable. (EE)
3개월쯤 되면 그들이 나를 꽤 잘 알고 내가 무엇을 잘하는지 알고 있다고 생각한다고 말할 것이다. (AA)
I’d probably say by three months I think they knew me pretty well and knew what I was good at. (AA)

논의
Discussion

우리의 정성적 연구 결과는 임상실습 연도의 의대생들의 피드백 추구 행동을 설명하고 임상실습 학생들의 피드백 추구 행동에 영향을 미치는 교수-학생 관계의 측면을 강조함으로써 의학 교육 문헌에 추가된다(표 1). 임상실습 학생들은 피드백을 추구하는 다양한 이유와 동기를 가지고 있었고, 종종 피드백의 인식된 [이익]과 [비용]을 저울질했다. 그들은 피드백을 찾는 기술을 향상시키면서 피드백 추구 접근법을 빠르게 진행되고 종종 혼란스러운 임상 학습 환경에 적응시키는 방법을 배웠다. 우리의 가장 참신한 발견은 학생들이 필요한 정보를 얻는 목표를 달성하기 위해 간접적이거나 비조회적인 피드백 추구 기술을 사용했다는 것입니다.
The findings of our qualitative study add to the medical education literature by elucidating the feedback-seeking behaviors of medical students in the clerkship year and by highlighting aspects of the faculty–student relationship that have an impact on clerkship students’ feedback-seeking behaviors (Table 1). Clerkship students had multiple reasons and motives for seeking feedback, and they often weighed the perceived benefits and costs of feedback. They learned how to adapt their feedback-seeking approaches to the fast-paced and often-chaotic clinical learning environment, as they advanced their skills in seeking feedback. Our most novel finding was that students also used indirect or noninquiry feedback-seeking techniques, which we labeled as “artifice,” to achieve the goal of getting the information they needed.

우리의 연구 결과는 조직 심리학 문헌과 교육 문헌의 피드백과 관련된 이론적 모델을 뒷받침한다. 임상실습 학년의 의대생들은 그들의 [지식과 기술을 향상시키기 위해] 피드백을 구했고, 그들의 역량에 대한 [긍정적인 판단을 받고 부정적인 판단을 피하기]를 원했다. 이러한 동기는 각각 [학습 목표 지향]과 [수행 목표 지향]을 반영한다. 피드백을 추구하는 것은 참여를 보여줌으로써 자신의 이미지를 홍보하는 것을 포함하여 학생들에게 [다양한 이점]을 가져다 주었다. 그러나, 학생들은 그들의 자아, 감정, 그리고 자기 표현에 대한 손상과 같은 [피드백 탐색의 비용]을 확인했다. 학생들이 피드백의 편익과 비용을 처리하는 방법은 규제 초점 이론에 의해 설명될 수 있는데, 이는 개인이 ["향상" 또는 "예방" 초점]을 가지고 있음을 시사한다. 의료 교육 현장 환경에서 규제 초점이 어떻게 예측 가치를 갖는지 설명했다. 다른 이론으로 Watling과 Harrison 등은 피드백에 대한 학생들의 인식과 반응이 종합 평가와 혼동되는 형태적 피드백에 의해 왜곡될 수 있다고 강조했다.
Our findings support theoretical models regarding feedback in the organizational psychology literature17–19 and the educational literature.35–37 Medical students in their clerkship year sought feedback to improve their knowledge and skills, while wishing to receive positive and avoid negative judgments about their competence. These motives reflect learning goal and performance goal orientations, respectively.24–26 Seeking feedback reaped various benefits for students,16 including promoting one’s image by showing engagement.28 However, students identified costs in seeking feedback, such as damage to their egos, feelings, and self-presentation.17,28,38 The ways in which students processed the benefits and costs of feedback may be explained by regulatory focus theory, which suggests that individuals have either a “promotion” or “prevention” focus.39,40 How regulatory focus has predictive value in medical education workplace settings has been delineated.41 As a different theory, Watling42 and Harrison et al43 have emphasized that students’ perceptions and reactions to feedback may be distorted by their confusing formative feedback with summative assessment.

의대생들이 피드백을 구하는 방식은 수의대 학생들과 완전히 비슷하지는 않았다. 우리 학생들은 주로 [탐구 방법]을 사용하고 [모니터링 방법]의 사용을 많이 설명하지 않은 반면, 수의학 학생들은 두 가지 방법을 모두 주요 전략으로 사용하였다. 비록 두 분야의 학습자들이 물질적으로 다를 수 있지만, 더 가능성 있는 설명은 학습 환경의 차이이다. 또한, 바쁜 임상 학습 환경에서 수의학과 학생과 의대생 모두 피드백을 구할 수 있는 최적의 시기를 결정할 수 있었습니다. 우리 학생들은 피드백을 찾는 것이 다소 도움이 될 것이라는 신호를 보내는 분위기와 같은 피드백 제공자의 단서를 관찰했다.
The manner in which our medical students went about seeking feedback was not entirely similar to that of veterinary students.16 Our students predominantly used an inquiry method and did not describe much use of a monitoring method,28 whereas veterinary students used both methods as main strategies. The reasons for this are unclear; although learners in the two fields may be materially dissimilar, a more likely explanation is differences in the learning environments. Also, in a busy clinical learning environment, both veterinary students16 and our medical students were able to determine the best timing to seek feedback. Our students observed for cues from the feedback provider, such as mood,15 that signaled that seeking feedback would be more or less beneficial.

이 연구에서 우리의 독특한 발견은 임상실습 학생들이 [직접적인 질문] 외에도 피드백을 얻기 위한 전략을 형성함으로써 학습 환경에 적응하는 능력이었다. 이 기술은 단순히 [간접적인 질문]을 사용하는 것이 아니라 다양한 [비질문적 전술]을 사용하는 것을 포함한다. 개별 학생들은 [자신에게 가장 잘 맞는 한 두 가지 전략]에 만족하는 것처럼 보였다. 우리는 "인공물"이라는 용어가 암시하는 것처럼 학생들이 의도적으로 조작하거나 기만적이라고 생각하지 않는다. 오히려, 이 중요한 발견은 학생들이 거의 피드백을 받지 못하는 상황에서 개발한 "지혜적인clever" 접근법을 반영한다.
Our unique finding in this study was the ability of clerkship students to adapt to the learning environment by fashioning strategies, besides direct inquiries, to obtain feedback. This artifice was not simply using indirect inquiries28 but also involved employing a variety of noninquiry tactics. Individual students seemed to settle on the one or two strategies that worked best for them. We do not believe that students are intentionally manipulative or deceptive, as the term “artifice” might imply. Rather, this important finding reflects the “clever”34 approaches students have developed in the face of receiving little feedback.

우리의 결과는 의대생의 피드백 추구 행동에 영향을 미치는 상호 관련 측면(예: 목표, 동기, 이점, 비용)이 있는 피드백 제공자와 수신자 간의 복잡한 관계 모델을 지원한다. 교수-학생 관계와 관련하여 추가한 네 번째 인터뷰 질문은 새로운 관점을 제공하기보다는 이전의 결과를 확인시켜 주었다. 그들의 임상실습 학년에 의대생들은 교직원들이 접근하기 쉽다는 것을 느낄 필요가 있다. 그들은 "예의바르고 관대한" 문화가 피드백의 장벽이 된다는 것을 인식한다. 교직원이 지지적이고 배려적이라고 느끼는 학생들과 달리, 교직원에 대한 부정적인 반응을 보이는 학생들은 [피드백 회피 전략]과 [교직원 신뢰도에 대한 인식이 떨어지는 경향]이 있다. 
Our results support the model of a complex relationship5,27 between feedback provider and receiver which has interrelated aspects (e.g., goals, motives, benefits, costs) that affect the feedback-seeking behaviors of medical students. The fourth interview question we added regarding the faculty–student relationship confirmed previous findings rather than providing new perspectives. Medical students in their clerkship year need to feel that faculty members are approachable.1,44 They recognize that a culture of “politeness”45 and leniency5 becomes a barrier to feedback. In contrast to students who feel that faculty members are supportive and caring, students with negative reactions to faculty members46 tend toward feedback avoidance strategies and perceptions of poor faculty credibility.47

놀랄 것도 없이, LIC의 종단적 측면은 학생들이 교수진으로부터 피드백을 구하는 더 강한 관계를 더 쉽게 확립할 수 있도록 한다. 의대생들이 전통적인 입원 환자 블록 로테이션에 대해 충분한 피드백을 얻을 수 있지만, 그러한 피드백은 주로 (전공의보다 로테이션이 잦고 참석이 적은 교직원보다는) 전공의에게서 나온다. LIC 모델에 더 높은 품질의 피드백이 있을 수 있으며, 이는 부분적으로 블록 모델에 비해 피드백 기회가 증가한 것과 관련이 있을 수 있다. 또한 일반적인 입원 환자 블록 회전의 바쁜 임상 환경은 효과적인 피드백에 대한 장벽을 제시한다. 우리는 종단적인 멘토링과 조언을 통합하기 위해 의료 교육 모델을 크게 개혁하라는 어비 외 연구진의 요청을 지지한다.
Not surprisingly, the longitudinal aspect of the LIC facilitates an easier establishment of stronger relationships in which students seek feedback from faculty.48 Although medical students may be able to obtain sufficient feedback on traditional inpatient block rotations, such feedback will be predominantly from residents rather than from faculty members, who transition often and are less present than residents. There may be a higher quality of feedback in the LIC model, which may be partly related to the increased opportunities for feedback compared with the block model. Also, the busy clinical environment of a typical inpatient block rotation presents a barrier to effective feedback. We support the call by Irby et al49 to significantly reform medical education models to incorporate longitudinal mentoring and advising.

실제적인 의미
Practical implications

교수진 피드백 역량을 개발하기 위한 역사적 접근 방식에 의문이 제기되었다. 교수진들이 학습자의 피드백 추구 능력을 증진하도록 권장하고 있지만, [최적의 피드백 교류]를 [공동으로 수행]하는 데 [적극적인 참여자]로서 의대생들에게 힘을 실어주는 데 관심을 기울여야 한다고 생각한다. Crommelink와 Anseel은 훈련 프로그램이 학습 목표 지향을 촉진하기 위해 설계되어야 한다고 제안했다. 우리는 이전에 효과적인 피드백을 받는 의대생을 지도하는 것이 가능하다는 것을 보여주었다. 밀라노 외 연구진은 임상실습 학생들이 피드백을 이끌어내는 방법에 대한 90분간의 워크숍을 거친 후 피드백을 얻는 긍정적인 태도의 변화를 보여주었다.

The historical approach for developing faculty feedback competencies has been called into question.14 Although faculty are encouraged to promote the feedback-seeking skills of learners,50 we believe attention should be given to empowering medical students as active participants in coproducing optimal feedback exchanges.1 Crommelinck and Anseel15 have suggested that training programs be designed to promote a learning goal orientation. We have previously shown that coaching medical students in receiving effective feedback is feasible.51 Milan et al44 have demonstrated changes in positive attitudes toward obtaining feedback after clerkship students underwent a 90-minute workshop on how to elicit feedback.

우리는 또한 교직원들이 단순히 학생들에게 피드백을 요청하고 자주 요청하라고 말하는 것을 넘어서야 한다고 믿는다. Boud와 Moloyy는 피드백 커리큘럼이 피드백 제공뿐만 아니라 자기 규제 촉진, 피드백 추구 기술에서의 연습, 수신 훈련과 같은 요소를 포함해야 한다고 제안했다. 본 연구의 결과와 위에서 강조한 목표지향성 및 교육동맹모형을 바탕으로 [의대생의 피드백 추구능력을 향상시키기 위한 종단적 교육과정의 원형]을 설계하였다(부록 1, 2). 교육과정은 [목표지향성에 대한 개인의 선호도, 탐구 대 모니터링 접근법, 규제 초점에 대한 평가와 자기반성]으로 시작된다. 구성 요소에는 건설적인 피드백에 대한 둔감화와 다른 학생들에게 유용한 것으로 입증된 피드백 기법에 초점을 맞춘 세션이 포함된다. 대화형 활동은 피드백을 받고 반영하는 방법을 강조할 수 있다. 이 커리큘럼의 구성 요소는 실험 연구에 의해 지원되어야 하며(그러나 개념 모델에 의해 잘 부양되고 있음) 일부 내용이나 방법은 논란의 여지가 있을 수 있다는 것을 완전히 인식하고, 우리는 이 프로토타입이 학생들이 적극적인 참가자로서 피드백을 찾는 방법을 개선하는 방법에 대한 논의를 촉진하기를 바란다. 
We also believe that faculty need to go beyond just telling students to ask for feedback and to ask often. Boud and Molloy11 proposed that a feedback curriculum should include elements such as promoting self-regulation, practice in feedback-seeking skills, and training in receiving as well as providing feedback. Based on our findings in this study, and on the goal orientation and educational alliance models highlighted above,26,27 we designed a prototype of a longitudinal curriculum to enhance medical students’ feedback-seeking skills (Appendixes 1 and 2). The curriculum begins with an assessment of and self-reflection on an individual’s predilection to goal orientation,20,52,53 inquiry versus monitoring approaches,38 and regulatory focus.40 Components include desensitization to constructive feedback and a session focusing on the feedback artifice proven to be useful for other students. Interactive activities can emphasize how to receive and reflect on feedback. Fully recognizing that components of this curriculum need to be supported by experimental studies (but are well buoyed by conceptual models11) and that some of the content or methods may even be controversial, we hope this prototype will prompt discussion on methods to improve how students seek feedback as active participants.

 

한계
Limitations

우리 연구의 한계에 대해 반성적인 입장을 취할 때, 우리는 우리의 발견이 구성주의적 근거 이론 접근법의 위험인 우리 자신의 편견과 가정에 의해 제한되고 왜곡될 수 있다고 생각한다. 우리는 우리의 방법론적 과정이 강했다고 생각하지만, 한 가지 맹점은 의대생이나 교수진의 독특성일 수 있다. 우리는 다른 기관에서 두 그룹에 대한 연구가 서로 다른 결과를 초래할 수 있다는 것을 배제할 수 없다. 학생들이 자발적으로 인터뷰를 하였는데, 인터뷰를 하지 않은 학생들이 피드백 추구 접근법에 대한 인식이 다른지는 조사하지 않았다. 두 면접관은 참여 학생들과 직접적인 평가 역할이 없었음에도 학생들의 반응은 면접관의 리더십 역할에 영향을 받았을 수 있다. 인터뷰 가이드가 사용되었지만, 이 주제 영역에서 두 인터뷰어의 연구 경험 또한 우리의 해석을 편향시킬 수 있었다. 또한, 우리의 결론에 기초한 이론적 모델은 의학 교육 환경에서 개발되지 않았기 때문에 이러한 교육적 환경에 맞지 않을 수 있다. 우리는 프로토타입 교육 커리큘럼의 설계를 학생의 인식과 개념 모델에 기초했지만, 그러한 프로그램의 결과(예: 향상된 피드백 추구 기술)는 여전히 불확실하다. 학습 문화는 피드백 훈련 프로그램이 궁극적으로 얼마나 효과적인지에 중요한 역할을 할 수 있다. 
In taking a reflective stance regarding limitations to our study,54 we think our findings could be limited and distorted by our own biases and assumptions, which is a risk with a constructivist grounded theory approach.31 Although we believe our methodologic processes were robust,29 one blind spot may be the uniqueness of our medical student or faculty populations. We cannot rule out that a study of either group at another institution may result in divergent findings. Students volunteered to be interviewed, and we did not investigate whether those who were not interviewed had differing perceptions of feedback-seeking approaches. Even though the two interviewers had no direct evaluation role with the participating students, the students’ responses may have been influenced by the leadership roles of the interviewers. Although an interview guide was used, the research experience of both interviewers in this topic area could also have biased our interpretations. Additionally, the theoretical models for the basis of our conclusions17–19,24 were not developed in medical education environments, so they may not fit this pedagogical environment. We based the design of our prototype educational curriculum on student perceptions and conceptual models,15,44,51 but the outcomes of such a program (e.g., improved feedback-seeking skills) remain uncertain. The learning culture may have a significant role in how effective any feedback training program eventually is.55

결론들
Conclusions

임상실습 학년의 의대생들은 피드백을 추구하는 많은 동기를 가지고 있으며, 피드백을 추구하는 행동을 조정하여 피드백 제공자와의 복잡한 대화 상호작용에 적극적으로 참여한다. 교직원과 학생의 관계는 복잡하고 학생들이 피드백을 얻기 위해 사용하는 접근법에 영향을 미친다. 학생들은 자신의 필요를 충족시키는 구체적인 피드백 정보를 얻는 기술(및 기교)을 점진적으로 개선하지만, 그들에게 훈련을 제공하는 것은 유익할 수 있다. 우리의 프로토타입 커리큘럼은 학생들의 피드백을 찾는 기술을 개발하는 데 도움이 될 수 있다.
Medical students in their clerkship year have many motives to seek feedback, and they adapt their feedback-seeking behaviors to actively participate in an intricate dialogic interaction with their feedback providers. The faculty–student relationship is complex and affects the approaches students use to seek feedback. Students gradually refine the art (and artifice) of obtaining the specific feedback information that meets their needs, but providing them training may be beneficial. Our prototype curriculum may facilitate the development of students’ feedback-seeking skills.

 


Acad Med. 2018 Aug;93(8):1218-1226. doi: 10.1097/ACM.0000000000002256.

The Art (and Artifice) of Seeking Feedback: Clerkship Students' Approaches to Asking for Feedback

Affiliations collapse

Affiliation

1R. Bing-You is professor, Tufts University School of Medicine, Boston, Massachusetts, and vice president for medical education, Maine Medical Center, Portland, Maine. V. Hayes is clinical associate professor, Tufts University School of Medicine, Boston, Massachusetts, and faculty member, Department of Family Medicine, Maine Medical Center, Portland, Maine. T. Palka is clinical assistant professor, Tufts University School of Medicine, Boston, Massachusetts, and faculty member, Department of Psychiatry, Maine Medical Center, Portland, Maine. M. Ford is clinical assistant professor, Tufts University School of Medicine, Boston, Massachusetts, and assistant director of the longitudinal integrated clerkship, Maine Medical Center, Portland, Maine. R. Trowbridge is associate professor, Tufts University School of Medicine, Boston, Massachusetts, and director of the longitudinal integrated clerkship, Maine Medical Center, Portland, Maine.

PMID: 29668522

DOI: 10.1097/ACM.0000000000002256

Abstract

Purpose: As attention has shifted to learners as significant partners in feedback interactions, it is important to explore what feedback-seeking behaviors medical students use and how the faculty-student relationship affects feedback-seeking behaviors.

Method: This qualitative study was inspired by the organizational psychology literature. Third-year medical students were interviewed at Maine Medical Center in April-May 2017 after completing a traditional block rotation clerkship or a nine-month longitudinal integrated clerkship (LIC). A constructivist grounded theory approach was used to analyze transcripts and develop themes.

Results: Fourteen students participated (eight LIC, six block rotation). Themes associated with why students sought feedback included goal orientations, perceived benefits and costs, and student and feedback provider characteristics. Factors influencing the way students sought feedback included busy environments, timing, and cues students were attuned to. Students described more inquiry than monitoring approaches and used various indirect and noninquiry techniques (artifice) in asking for feedback. Students did not find summative feedback as helpful as seeking feedback themselves, and they suggested training in seeking feedback would be beneficial. Faculty-student relationship dynamics included several aspects affecting feedback-seeking behaviors, and relationship differences in the LIC and block models affected feedback-seeking behaviors.

Conclusions: Medical students have many motives to seek feedback and adapt their feedback-seeking behaviors to actively participate in an intricate dialogic interaction with feedback providers. Students gradually refine the art (and artifice) of obtaining the specific feedback information that meets their needs. The authors offer a prototype curriculum that may facilitate students' development of feedback-seeking skills.

 

+ Recent posts