의료현장에서 교육혁신의 전파: 훈련 vs 사회적네트워크 (Soc Sci Med. 2010)

의료현장에서 교육혁신의 전파: 훈련 vs 사회적네트워크 (Soc Sci Med. 2010)

Disseminating educational innovations in health care practice: Training versus social networks

Erik Jippes a,*, Marjolein C. Achterkamp b, Paul L.P. Brand a,c, Derk Jan Kiewiet d, Jan Pols e, Jo M.L. van Engelen f

a Postgraduate School of Medicine, Wenckebach Institute, University Medical Centre Groningen, University of Groningen, the Netherlands

b Marketing, Faculty of Economics and Business, University of Groningen, the Netherlands

c Princess Amalia Children’s Clinic, Isala Klinieken Zwolle, the Netherlands

d Loyalis Business Development, the Netherlands

e Wenckebach Institute, University Medical Centre Groningen, University of Groningen, the Netherlands

f Product Development and Strategy, Faculty of Economics and Business, University of Groningen, the Netherlands

도입

Introduction

보건의료서비스 기관에서 혁신은 핵심적 이슈이다. 이것이 그렇게 된 이유는 '가능한 과학지식 활용의 실패' , '변화하인 인구집단으로 인한 비용 증가', '의료 테크놀로지', '의료 과오', '헬스케어 시스템에서 조직 그 자체' 등등 다양하다. 많은 혁신 프로젝트가 성과 달성에 실패한다. 혁신을 유도하거나 죽이는 여섯 가지 힘이 있다.

Innovation in health service delivery and organization has become a central issue. The reasons for it becoming so range from failure to use the available scientific knowledge (Richardson, 2001), to rapidly rising costs due to changing demographics and medical technology (Hartman, Martin, McDonnell, & Catlin, 2009), medical errors (Berwick, 2003), and the very organization of the health care systems themselves (Kuttner, 2008). Many innovation projects often fail to meet expectations. There are six forces which seemto drive or kill innovations:

• players (friends and foes),

• funding,

• policy,

• technology,

• customers, and

• accountability (Herzlinger, 2006).

헬스케어 조직에서 많은 혁신은 training course와 같은 교육을 통해서 도입된다. T&E에 들어가는 비용은 상당하다. 미국에서 평균적으로 헬스케어조직은 매년 training에 15만달러 이상을 투입한다. 헬스케어에서 annual direct training expenditure per FTE 는 $862로, 전체 profit의 12%이다

Many of the innovations in health care organizations are implemented by following a training course or other kind of education. The expenditure incurred for training and education is considerable. In the USA an average health care organization’s (500–999 FTE) annual training expenditure exceeds $150,000 (Controller’s Report, 2006). The average annual direct training expenditure per FTE in health care in the USA is $862, which constitutes on average 12% of profit (Corporate Training & Development Advisor, 2008).

이론과 가설

Theory and hypotheses development

Training and Education (T&E) : Teach-the-teacher training의 효과

Training and education: the effect of Teach-the-Teacher training

medium to large effect 가 있다고 요약된 바 있다. 또 다른 리뷰에서는 T&E가 개인/팀/조직/사회에 긍정적인 효과를 준다고 나타났다.

This has been summarized in a meta-analysis showing a medium to large effect for training and education when using a composite measure of Kirkpatrick’s eval- uation criteria (i.e., reaction, learning, behavior, and results) (Arthur, Bennett, Edens, & Bell, 2003; Kirkpatrick & Kirkpatrick, 2006). Another review showed, that training and education lead to important benefits for individuals and teams, organizations and society (Aguinis & Kraiger, 2009).

Meyers and Sivakumar 는 T&E를 조직의 혁신에 영향을 미치는 핵심 요인으로 보았다. Training은 친숙도를 높이고 기술적 능력을 향상시켜서 긍정적인 분위기와 태도를 형성한다. Training은 specialization을 가능하게 하며, 지식기반을 확대시키고, 지식의 교환을 촉진하고, 혁신을 유도한다. Training은 professionalism, boundary-spanning활동, 새로운 방식에 대한 개방성을 높여준다.

Meyers and Sivakumar (1999) identified training and education as key factors influencing organizational innovation and imple- mentation. Training can create a positive climate and attitude by increasing familiarity and technical competence. Training leads to more specialization which, in turn, can lead to a broader knowledge base, stimulate the exchange of ideas and foster innovation. Training can also lead to more professionalism, more boundary- spanning activities and increased openness to new methods and ways (Meyers & Sivakumar, 1999).

Steinert 등은 교수개발프로그램의 효과성을 보았음. 참가자들에게 가치를 인정받았지만, 학생이나 레지던트의 평가결과가 언제나 참가자의 인식과 같은 것은 아니었고, 조직의 변화나 학생의 학습 변화는 흔히 연구되는 것이 아니었다.

Steinert et al. (2006) conducted a review of the effectiveness of teaching faculty development initiatives in medical education. Faculty development activities appeared to be highly valued by the participants, who also reported changes in learning and behavior. However, student/ resident evaluations did not always reflect the behavioral changes that the participants perceived, and changes in organizational practice and student learning have not been investigated very frequently since (Steinert et al., 2006).

최근의 장기 연구를 보면, 의사들의 didactic skill이나 teaching ability를 높이기 위한 코스들이 지식을 향상시키고 행동을 바꾸어 임상학습분위기를 향상시켰다.

Recently a long-term controlled study showed that Teach-the- Teacher courses, aimed at improving the didactic skills or teaching abilities of doctors, significantly increased doctors’ didactic knowl- edge and teaching behavior, and led to improvements in the clinical learning climate (Rubak, Mortensen, Ringsted, & Malling, 2008).

사회적 네트워크와 강한 매듭, 약한 매듭

Social networks and the effect of strong and weak ties

사회적 네트워크가 혁신의 전파에 중요하다. 다음의 방식으로 작동함

Social networks are assumed to play an important role in the diffusion and dissemination of innovations. Social networks influence diffusion by

• (1) 혁신에 대한 사회적 construction과 협상을 위한 의사소통의 채널 functioning as channels for communica-tion, social construction and negotiation of the innovation,
• (2) 혁신의 관측가능성을 높임 by increasing the observability of the innovation and, therefore,
• (3) 새로움이나 불확실성을 제거하여 위험에 대한 인식을 낮춰줌 by reducing the perceived risk by eliminating novelty or uncertainty for potential adopters of the outcome of the innovation(Greenhalgh, Robert, Macfarlane, Bate, & Kyriakidou, 2004; Larsen& Ballal, 2005; Rogers, 2003).

 

사회적 관계와 사회적 네트워크는 헥스케어 혁신의 지속가능성에 결정적이다.

Social relationships and socialnetworks are critical for the sustainability of health care innovation(Sibthorpe, Glasgow, & Wells, 2005). 

사회적 네트워크를 통한 혁신의 전파는 여러분야에서 연구됨

Diffusion of innovations through social networks has been studied from a number of perspectives in a variety of fields and disciplines (Greenhalghet al., 2004; Wejnert, 2002), for example, in the diffusion of family planning (Boulay, Storey, & Sood, 2002) and health campaigns (Valente & Fosados, 2006).

Greenhalgh 등은 사회적 네트워크의 개념적 프레임워크가 특정 health technologies에 적용되었지만, 보건의료전문직의 사회적 네트워크에 대해서는 잘 안되었다고 함.

Greenhalgh et al. (2004) concluded that, although the conceptual framework of social networks had been extensively applied to the adoption of particular health technologies, the empirical literature on the social networks of health professionals as related to the diffusion of innovations in service delivery and organization (as opposed to health technologies) was extremely sparse.

사회적 네트워크는 다음과 같이 정의됨

A social network can be defined as a finite set of actors and the relationships defined between them(Wasserman & Faust, 1994).

 

SNA를 적용함에 있어서, actor는 개인/그룹/사업체/전체 조직/국가가 될 수 있다.

In the application of Social Network Analysis (SNA), generally speaking actors can be interpreted as discrete individuals, or groups of individuals, business units, entire organizations or even as countries.

SNA를 혁신의 전파에 사용할 때, 이 actor의 관계적 매듭은 정보의 교환, 의사소통, 친선과 신뢰를 구성한다.

In the application of SNA to the diffusion of innovation, the actors are individuals, groups and business units (intra-organizational) or organizations (inter-organizational), and the relational ties consist of the exchange of information, communication, friendship or trust.

 

actor와 relational ties를 규명하는 것에 대한 두 관점: 강한 관계매듭의 강도, 약한 관계매듭의 강도

The configuration of the actors and the relational ties they have with each other – or the structure of the social network itself – can influence the diffusion of innovation in several ways. Two perspectives emerge in the literature: the ‘‘strength of strong relational ties’’ and the ‘‘strength of weak relational ties’’ (Tenkasi &Chesmore, 2003). 

첫 번째 관점은 주로 '동질성homophily'에 대한 인식에 기반한다. 다음과 같이 정의함.

The first perspective is mainly based on the notion of homo- phily. Homophily is defined by Rogers (2003) as

 

‘‘the extent to which two or more individuals

who interact are similar in certain attributes, education, social status and the like.’’

 

서로 동질성을 가지는 사람들끼리는 '감염효과contagion effects'가 발생한다. 한 개인이 다른 사람에게 행동/태도/신념을 전파시키고, 혁신의 전파가 빨라진다. 동질성과 커뮤니케이션은 서로가 서로를 강화시킨다. 커뮤니케이션이 많을수록, 또는 '강한매듭'이 강할수록 더 동질성을 가질 가능성이 높다. 강한 관계매듭은 instruction과 feedback의 기회이기도 하며, 결국 successful adoption을 가져온다. 사람 간 contact와 커뮤니케이션이 혁신의 observability를 높여주고, 결국 novelty 또는 uncertainty를 제거하여 perceived risk를 줄여준다. 커뮤니케이션이 자주 일어날수록 potential risk가 감소하고, diffusion and adoption이 늘어난다. 다수의 강력한 매듭으로 이뤄진 사회적 네트워크는 복잡한 조직 안에서 혁신에 필요한 복잡한 정보를 주고받는데 최적의 조건이고, high quality idea의 origin이다.

Between people who are more homophilous, contagion effects occur: An individual adapts his behavior, attitude and beliefs to those of others, which then enhances the diffusion of innovations. Homophily and communication reinforce each other: The more communication there is between members – or the stronger the tie between the actors – the more likely they are to become homophilous (Rogers, 2003). Strong relational ties also provide more opportunities for instruction and feedback, which can in turn enhance successful adoption (Tenkasi & Chesmore, 2003). Interpersonal contacts and communication increase the observability of the innovation and therefore reduce the perceived risk by eliminating novelty or uncertainty for the potential adopters of the outcome of the innovation (Greenhalgh et al., 2004; Rogers, 2003). More frequent communication decreases potential risk and results in higher diffusion and adoption. Social networks with a larger number of strong ties – or dense networks – create optimal conditions for the exchange of the complex information necessary for innovation in complex organizations (Hansen, 2002) and for the origin of high quality ideas (Bjo¨ rk & Magnusson, 2009).

약한 관계매듭도 장점이 있다. 외부와의 약한 매듭(또는 structural hole)은 새로운 혁신을 발견하여 네트워크의 외부로부터 그것을 capture할 수 있게 해준다. structural hole에 걸쳐있는 네트워크를 보유한 사람들은 다양한 정보에 일찍부터 접근하여, 좋은 아이디어를 발견할 경쟁상의 잇점을 획득하고, 혁신에 대해 빨리 접근할 수 있다. 정보를 퍼트리는데 약한 매듭이 종종 더 중요하곤 한데, 왜냐하면 그것이 아니었다면 서로 단절되어있어을 그룹 사이에 다리를 놓아주고, 다른 contact와 resource에 대한 access를 높여주기 때문이다.

Weak relational ties also have their advantages. External (weak) ties (or structural holes) allow new innovations to be identified and captured from outside the network. Individuals whose networks span structural holes have early access to diverse information, which provides them with a competitive advantage by seeing good ideas and having early access to innovations. Weak ties are often more important in spreading information or resources because they tend to serve as bridges between otherwise disconnected groups and to facilitate access to different contacts and resources (Burt, 2004; Granovetter, 2005).

West and Barron 는 사회적 네트워크를 비교함(의사,간호사)

West and Barron (2005) studied the social networks of clinical directors in medicine and directors in nursing. The

• 의사가 더...former have significantly denser, more cohesive and more horizontal social networks than the latter and

• 두 그룹 모두 중요한 문제를 논의할 때 전문성/성별/연령/Seniority가 비슷한 사람과 논의하였는데, 의사가 이런 성향이 더 강하였다.
  both groups tend to discuss important professional matters with others who are similar in terms of profession, gender, age, and seniority, with clinical directors being more extreme in this regard (West & Barron, 2005).

 

약 처방의 전파에 대해 연구함. link and contact가 많을수록 신약을 빨리 사용함.

Coleman, Katz, and Menzel (1966) studied the diffusion of a prescription drug Gammanym among 125 physicians in four American Midwestern communities. They found the more links and contacts a physician was involved in, or the stronger the ties a physician had, the more likely he or she was to be an early user of Gammanym. Physicians who were more isolated in the network adopted the drug consid- erably later.

The impact upon the integrated physicians was quick and strong, while the impact upon isolated physicians was slower and weaker, thoughnot absent (Colemanet al., 1966).

 

퍼포먼스와 매듭의 강도와는 상관 없다는 연구

A recent study onprescribing behavior of General Practitioners (GPs) in Italy found no significant relationship between the strength of GPs’ ties (as measured by degree centrality) and their performance (meeting adrugexpendituretarget) (Fattore, Frosini, Salvatore, &Tozzi, 2009).

 

강한 매듭과 약한 매듭이 둘 다 중요함.

For the successful diffusion of innovations, both strong and weak relational ties seem to be necessary. Weak ties are necessary to acquire new ideas and strong ties are necessary for subsequent implementation (Burt, 2004; Reagans & McEvily, 2003).

방법

Methods

네덜란드에서 PGME의 혁신

Background: innovations in postgraduate medical training in the Netherlands

We used data from the innovations in the postgraduate medical specialist training programs in the Netherlands to test our hypotheses. The Netherlands has 33 postgraduate training programs (for example, Surgery and Pediatrics) following medical school and they take place in eight university and about 60 non- university teaching hospitals. Postgraduate training has consisted mainly of ‘‘learning on the job.’’ Residents (medical specialists intraining) work under the supervision of a team of qualified medical specialists and learn by reflection on experiences. In this program, neither the method nor the frequency of feedback is structured. Evaluation of the progress of residents is, therefore, rather informal.

In 2004, the Royal Dutch Medical Association (KNMG-CCMS) introduced competency-based education in postgraduate training throughout the Netherlands (Scheele et al., 2008).

• Traditionally resi- dents had been trained according to a pre-defined input (for example, number of operations, number of months in clinical consultation and practice). Assessment was limited to checking whether these numbers were met.

• In competency-based education, medical specialists are now trained according to certain competencies: medical expert, collaborator, communicator, professional, health advocate, manage- ment and scholar (Frank & Danoff, 2007). The periodic assessment which takes place now focuses (using a variety of methods) on knowledge and skills possession relevant to clinical practice.

Key innovations introduced by the Royal Dutch Medical Asso- ciation were the use of the Mini Clinical Evaluation Exercise (Mini- CEX) and the use of structured competency-based feedback (Royal Dutch Medical Association (Centraal College Medisch Specialisten), 2004).

• The Mini-CEX is a method of assessing competencies in real- life clinical practice. It consists of a short observation of a resident demonstrating clinical skills, and is carried out by a qualified medical specialist using a pre-defined scoring format, followed by a structured feedback conversation (Norcini, Blank, Duffy, & Fortna, 2003).

• The method and frequency of the structured feedback are outlined. As a result, medical specialists are expected to adopt a novel structured feedback format.

The Mini-CEX and structured feedback were to be adopted and implemented by all teams of medical specialists that train residents. Our study focuses on the dissemination process of structured feedback within teams of medical specialists. The innovation could originate either from outside or from within the group. Ethical approval was deemed unnecessary for this study.

표본

Sample

The medical specialties of Obstetrics & Gynecology (O&G) and Pediatrics were the first in the Netherlands to implement the innovations in their curriculum(In-VIVO Project, 2006). Data were gathered in 2007 from four O&G departments and five Pediatrics departments in the Netherlands.

자료 수집

Data gathering

The medical specialists and residents received both a structured and validated questionnaire (see below).

종속변수

Dependent variable

적응적 행동: 구조화 피드백

Adoptive behavior: structured feedback

We used the ‘‘structured feedback’’ given by medical specialists to residents as the dependent variable. Structured feedback is based on ‘‘Pendleton’s rules’’ (Pendleton, Schofield, Tate, & Havelock, 2003) and consists of the following components:

• (1) the feedback is structured 

• (2) the medical specialist gives the resident the opportunity to give his/her opinion 

• (3) the medical specialist provides positive points 

• (4) the medical specialist provides specific points for improvement 

• (5) the medical specialist provides the feedback in a ‘‘safe’’ way.

독립변수

Independent variables

교육자교육 트레이닝

Teach-the-Teacher training

Many medical specialists in our sample had followed a Teach- the-Teacher course which was aimed at improving the didactic skills or teaching abilities of the participants. The training consisted of three sequential two-day courses. Registration for the second and third courses was dependent upon successful completion of the first course.

• The introductory course comprised training in structured feedback, training in the Mini-CEX, and the basics of adult learning.

• The second course comprised training in daily educational practice, which includes organizing day-to-day training for residents and adapting the training to the learning styles of the residents.

• The third course included training in peri- odic interviews for the formative and summative assessment of residents.

 

 

사회적 네트워크 분석

Social network analysis: preparation of data for the social network independent variables

각자 자신의 과에서 의사소통 intensity를 평가함. 의사소통은 다음과 같이 정의하였고, 평가는 다음의 척도로 함.

We used SNA techniques to measure the social network inde- pendent variables. Medical specialists rated their communication intensity with their fellow medical specialists in their own departments. The communication was specified

• ‘‘as communica- tion in the past half year about the introduction of innovations, new methods or procedures, or new developments related to the work situation.’’

The rating was on a six-point scale, ranging from

• ‘‘never,’’ to ‘‘less than once a month,’’ ‘‘once in three weeks,’’ ‘‘weekly,’’ ‘‘daily,’’ or ‘‘more than once daily’’ (also used by Kratzer, 2001).

 

The resulting data was analyzed using UCINET VI (Borgatti, Everett, & Freeman, 2002). The answers given by the respondents resulted in a directed valued graph and a matrix. ‘‘Directed’’ means that the relational tie (in this case, communication) of one person to another is either present or not. ‘‘Valued’’ means that the relational tie can range between ‘‘never’’ and ‘‘more than once daily.’’ Graphs and matrices are useful techniques in SNA to represent social networks. In order to test the hypotheses, the data needed to be transformed into an undirected dichotomous matrix (or maximum a symmetric matrix). We used the symmetrizing method to convert the directed matrix into an undirected one and to correct for missing network data. This meant that the highest rating of communication intensity between two persons was used or, in the case of missing network data, the rating from one person.

 

To dichotomize the valued matrix (ranging from1 to 6), we recoded the scores as follows. The values one and two were recoded into zero, which means there is no communication. The values three, four, five and six were recoded into one, which means there is a communication relationship between medical specialists.

강한 매듭의 강도: degree centrality

Strength of strong ties: degree centrality

강력한 매듭이 있는 사람은 네트워크 상에서 더 중심에 위치한다. 세 가지 중심도centrality가 있다. (degree, betweenness and closeness.).

Persons that have stronger ties to others are more central in the social network (Wasserman & Faust, 1994). Three centrality measurements can be distinguished: degree, betweenness and closeness.

 

Degree centrality 란 네트워크상에서 가장 눈에 띄는visible 사람이다. 많은 여러 사람들과 direct contact가 많은 사람이며, 강한 매듭을 가진 사람이다.

Degree centrality refers to persons who are the most visible in the network; these are persons who have a large degree of direct contact or are adjacent to many other persons and have strong ties with other people. Since this index captures direct or strong ties, we calculated it for every medical specialist and used it to test Hypothesis 2 (see Appendix 1 for the calculation of this index).

약한 매듭의 강도: betweenness centrality

Strength of weak ties: betweenness centrality

관계중심도(Betweenness centrality)는 두 actor 사이의 communication path에 있는 사람이다. 이 사람이 중요한 이유는 서로 접촉이 없는 두 사람 사이의 정보교환을 통제하기 때문이다. 반드시 strongly tied 되어 있어야 할 필요는 없다. 대신, 다수의 사람들과 다수의 weak ties를 가지고 있어야 하며, 서로 단절된 그룹 사이에서 정보전달의 bridge 역할을 해야 한다. 

Betweenness centrality refers to individuals who are literally on the communication paths between two other actors. These actors are central because they potentially control information between two non-adjacent persons (Wasserman & Faust, 1994). These persons are not necessarily strongly tied to other people. On the contrary, they have a lot of weak ties with a lot of people and serve as bridges for spreading information and resources between otherwise disconnected groups.

 

아래와 같은 정보를 말하는 것인데, 다른 말로 하면 두 사람 사이의 가장 짧은 경로에 위치하는 비율이다. 따라서 이 index는 indirect or weak ties를 대표한다.

This index represents the ratio of the number of times an actor is on the geodesics of other actors to the maximum amount possible. In other words, it represents the relative proportion that an actor is on the shortest path between two persons; therefore this index represents an actor’s indirect or weak ties. We used the standardized index to test Hypothesis 3 and calculated this centrality measurement for every medical specialist (see Appendix 1 for the calculation of this index).

 

강한 매듭과 약한 매듭: closeness centrality

Strength of strong and weak ties: closeness centrality

친밀성 중심도(Closeness centrality)는 빠르게 다른 사람과 상호작용할 수 있는 사람이고, 정보의 의사소통에 있어 매우 생산적이다. 이 closeness centrality 가 높은 사람은 네트워크에서 great "reach"를 가지고 있다. closeness centrality 는 strong ties와 weak ties를 모두 가진 사람이며, 이 index는 다음과 같이 계산된다. distance가 감소할수록 이 centrality가 증가한다.

Closeness centrality refers to persons who can quickly interact with all others; these actors can be very productive in communi- cating information to the other persons in the network (Wasserman & Faust, 1994). Persons with high closeness centrality have a great ‘‘reach’’ across the network. Closeness centrality can be viewed as persons who have both strong ties (high direct contacts) and weak ties (a lot of indirect ties). The index is the inverse of the sum of the distances from actor i to all other actors. As distances decrease the centrality index increases. This index captures both direct or strong ties and indirect or weak ties, since distances can be short (direct or strong ties) or long (indirect or weak ties). We standardized this index and used it to test Hypothesis 4 (see Appendix 1 for the calculation of this index).

인터뷰: 사회적 네트워크 독립변수의 validation

Interviews: validation of the social network independent variables

사회적 네트워크와 개개인의 centralities를 확인하고자 인터뷰 시행.

모든 프로그렘 디렉터가 확인된 사회적 네트워크에 강력하게 동의하였다.

All program directors of the different departments inour sample were interviewed to validate the social network and the individual centralities of the medical specialists found. Overall, the program directors strongly agreed with the social networks found.

통제변인

Control variables

성별 Gender

남성과 여성의 사회적 네트워크는 여러 측면에서 다른데, 특히 life stage와의 관계에서 특히 다르다.

It has been widely recognized that social networks among men and women differ in complex ways, particularly in relation to life stage (Antonucci, 2001). Other studies have confirmed there is gender difference in social networks (Kunst & Kratzer, 2007).

연령 Age

연령은 사회적 네트워크에 영향을 준다. 더 나이가 많을수록 더 크고 오래된 네트워크를 가지며, 덜 geographically proximal하다.

Age can influence social networks. Older people tend to have larger and older networks which are less geographically proximal (Ajrouch, Blandon, & Antonucci, 2005). Age difference in network structure may reflect differing roles and possibilities according to life stage.

태도 Attitude

혁신의 도입과 전파에 태도와 동기부여가 중요하다.

Attitude and motivation seemto be just as important in innovation adoption and implementation in health care as well (Garcia-Goni, Maroto, & Rubalcaba, 2007).

근무시간(고용형태) Hours of employment (part-time versus full-time employment)

파트타임으로 고용된 보건의료전문직의 숫자가 많다. 이것의 영향은 불분명한데, inventor들 사이에서 part-time과 full-time사이에 유의한 차이가 없다고 밝힌 것도 있다. 연령, 성별, 교육수준, 추구하는 형신의 종류도 비슷했다. flexible working 은 혁신을 drive하는 것이 아니라 혁신의 결과였다.

An increasing number of health care professionals have part- time appointments. The influence of part-time employment on innovation is unclear. Weick and Martin (2006) found no significant differences between part-time and full-time ‘‘inventors.’’ They seemed to be similar in terms of age, gender, educational level, and the types of inventions they pursued (Weick & Martin, 2006). Storey, Quintas, Taylor, and Fowle (2002) looked into the effect of flexible employment contracts on product and process innovations. It turned out that flexible working was found to be a consequence rather than a driver of innovation (Storey et al., 2002).

근무기간 Length of employment in the organization

연구된 바는 적다. 더 오래 한 조직에서 일할수록...다음에 대한 점수가 낮다.

Relatively few studies have addressed length of employment in relation to innovation. Decker, Wheeler, Johnson, and Parsons (2001) found the longer a person worked in an organization, the more negative the scoring on

• job satisfaction,

• the effect of budget adjustments on individual job-related stress,

• the quality of indi- vidual performance, and

• department morale (Decker et al., 2001).

 

반대로, 조직은 자원과 능력을 build on and maintain해야 한다.

On the other hand, the resource-based theorist would argue that organizations must build on and maintain the resources and capabilities needed to compete (Grant, 2001). Based on this it can be argued that length of employment actually has a positive influence on innovation.

 

 

 

Results

Independent t-test

From the t-test results it follows that only closeness centrality causes significant differences (p <.01) in the average adoptive behavior (Table 2 and Figs. 1–5).

 

 

 

Regression analysis

 

 

Discussion

Teach-the-Teacher training course로는 영향이 없엇다. 이 전 연구결과와도 부합하는 것(학생/레지더느의 평가는 참가자(교수)가 인식하는 행동변화와 일치하지 않는다)이다. 비록 Teach-the-Teacher training course 가 지식과 스킬의 향상은 시켜줄 수 잇으나, 그 자체로 혁신을 성공적으로 도입하기에는 충분하지 못한 것이다. 반대로 사회적 네트워크(closeness centrality)의 강력한 영향을 확인하였다.

No effect was found from a two to six day Teach-the-Teacher training course. This is in agreement with previous findings from a systematic review that found that student/resident evaluations did not always reflected the behav- ioral changes in teaching abilities participants perceived after following a faculty development program (Steinert et al., 2006). Although Teach-the-Teacher training can improve didactic knowl- edge and skills (Rubak et al., 2008), this by itself is apparently not enough to adopt the innovation successfully. On the other hand, we found a strong effect for social networks, with a strong association of closeness centrality to adoptive behavior and a moderate effect of degree centrality for adoptive behavior

나이도 중요했다. 나이가 증가함에 따라서 adoptive behavior가 감소하였다. 레지던트가 더 젋은 medical specialist들로부터 더 그러한 행동을 발견한 것이다. 젊은 사람이 더 structured feedback에 친숙할 수 있다.

Age was also important. With increasing age, medical specialists seem to be less likely to show adoptive behavior. It could be plausible that residents identify more with younger medical specialists. It is also possible that younger medical specialists are more familiar with structured feedback because their own medical training was already more oriented towards this innovation.

네트워크 분석

Network analysis

강한 매듭과 약한 매듭을 모두 가진 사람이 새로운 structured feedback을 더 적절히 활용하는 경향이 있었고 Burt의 연구와도 부합함.

In this study, the medical specialists with both strong and weak ties were more likely to properly use the new structured feedback technique. This is in agreement with Burt (2004) who stated that both strong and weak ties were necessary.

• 약한 매듭은 초기에 외부에서 혁신을 발견하는데 도움
  Weak ties are necessary for capturing innovations from outside the network and providing early access to diverse knowledge and resources,

• 강한 매듭은 혁신을 적용하는데 도움 strong ties for implementing the innovations.

 

Herzlinger 는 아군 혹은 적군이 driver or killer라고 하였는데, 아군과 적군은 혁신을 전파하는 사회적 네트워크를 구성한다.

It is also in agreement with Herzlinger (2006) who identified players (friends and foes) as a key driver or killer for innovations in health care. Friends and foes form the social networks which distribute innovations.

CC가 모델에 들어가기 전까지는 DC가 유의하게 기여함. 이는 DC가 별 영향이 없다는 Fattore의 연구와도 부합한다. 종합하면 강한 매듭이 약한 매듭보다 더 중요함.

We found a significant contribution for degree centrality (strong ties) to the regression model until closeness centrality (strong and weak ties) was added. This is in line with Fattore et al. who found no significant relationship between a GPs degree centrality and performance (Fattore et al., 2009). The interaction effect between degree centrality and closeness centrality was not found in the bivariate t-test which legitimates the use of the more compre- hensive multivariate regression analysis as an additional test. The interaction effect makes sense, since closeness centrality captures both strong and weak ties. One might expect the same interaction effect for betweenness centrality (weak ties). However, since this effect was not found, it can be concluded that strong ties are more important here than weak ties.

첫째, 약한 매듭만 많은 경우 정보과부하에 걸릴 수도 있고, 정보를 전달하는것은 시간이 드는 일이다. 자신에게 혁신을 도입할 시간이 부족했을 수도 있다. 둘째, 의사들은 기본적으로 동질적인 집단이다. 동질성과 의사소통은 서로를 강화시킨다. 의사소통이 많을수록, 즉 강한 매듭이 많을수록 더 동질화된다. 따라서 강한 매듭이 더 중요한 역할을 했을 것이다. 셋째, 이번 연구의 대상인 구조화된 피드백 기술은 복잡한 정보를 전달해야 한다. 약한 매듭은 단순한 정보의 전달에 적절하고, 강한 매듭은 복잡한 정보의 전달에 적절하다. 넷째, 이미 T-the-T course 가 있었기 때문에 새로운 구조화된 피드백 테크닉에 대해서 들어봤을 것이고 ,이러한 경우 약한 매듭이 별로 필요하지 않았을 수 있다.

There are a couple of explanations for these findings.

• Actors with more weak ties could experience information overload. Passing information along to others could be time-consuming. These actors are more oriented towards passing information along and have less time to adopt the innovation themselves properly (Kratzer, 2001). Since we measured proper adoption of the innovation and not first contact, this explanation could be plausible.

• Second, medical specialists are a relatively homophilous group; they are similar in educational background, job and social status (West & Barron, 2005). Homophily and communication reinforce each other: The more communication there is between members – or the stronger the tie between actors – the more likely they are to become homophilous (Rogers, 2003). So we could expect strong ties to play an important role in the adoption of innovations within the social networks of medical specialists.

• The third explanation could lie in the fact that the innovation studied – feedback technique – contains relatively complex information. Weak ties are more suitable for conducing relatively simple information and strong ties for diffusing complex information (Hansen, 2002).

• An important final explanation could be the connection with the Teach-the-Teacher training. Since medical specialists who followed the Teach-the-Teacher training course heard, saw and learned the new structured feedback tech- nique, this innovation had already been introduced into the departments. In other words, there were no weak ties needed anymore to penetrate the departments.

사회적 네트워크에서 강한 매듭과 약한 매듭이 T&E보다 중요하다. 경영자의 관점에서 혁신을 전파하기 위해서는 적극적으로 사회적 네트워크에 engage시키는 것이 중요할 것이다. 이미 opinion leaders, gatekeepers, and lead users의 중요성을 알고 있다. SNA를 사용하여 key individual을 알 수 있었다. 이들이 누군지 알게 되면 혁신의 전파에 활용할 수 있다.

We can draw the following conclusions. The most important factors influencing the diffusion of the new structured feedback technique among medical specialists are the strong and weak ties they have within their social networks. These seem to be more important than training and education. From a managerial point of view, it could be worthwhile to actively engage and compose social networks to disseminate innovations among health care profes- sionals. We already know the importance of opinion leaders, gatekeepers, and lead users in innovation processes. In our paper we showed that it was possible to identify these key individuals using Social Network Analysis. After identification of these indi- viduals, they can be harnessed for the dissemination of innovations. They can be incorporated in change initiatives, help to overcome resistance among their colleagues, and follow training and educa- tion on new health technologies and innovations. With regard to the innovation studied in this paper, it could make sense to incor- porate medical specialists who have both strong and weak ties in Teach-the-Teacher courses. This may lead to effective and efficient dissemination to other medical specialists.

한계와 제언

Limitations and suggestions for further research

This study had a number of limitations.

• First, we focused on the dissemination of innovations within teams of medical specialists regardless of whether the innovation originated inside or outside a specific team. The dissemination process within the teams could be influenced by social networks that medical specialists might have with other individuals outside their own team. These individuals range from medical doctors, to nursing staff, manage- ment, educationalists, and management consultants, as well as other support personnel. It would be interesting to examine how these networks are composed and what the effects are on the dissemination of health care innovations.

• Second, the study was carried out at the level of the individual, and not at the depart- mental level. For example, while we looked at individual degree centralities in departments, all individual degree centralities can also be aggregated into a group degree centralization index. This network level perspective could reveal important insights into the impact that the network structure has on effective diffusion and adoption of innovations in health care. Hospitals are characterized by high specialization, which leads to many informal social networks. Further studies are needed to investigate how these networks interact and howtheir compositioncan facilitate effective innovation.

• Third, this study had a relatively small sample size (n ¼81) composed of two medical specialties. It would be inter- esting to examine whether the same conclusions could be drawn from a larger sample size that included more medical specialties. The inclusion of more departments would also make it possible to conduct hierarchical linear modeling. This might improve the model by accounting for the nested structure of the data and by adding departmental level independent variables.

• Fourth, the negative findings for Teach-the-Teacher training need to be inter- preted with some caution. We did not work with an experimental design and we did not measure the effectiveness of the Teach-the- Teacher training with multiple criteria (e.g., Kirkpatrick’s criteria of reaction, learning, behavior and results (Kirkpatrick & Kirkpatrick, 2006)). The primary endpoint here was the degree of adoptive behavior by medical specialists as assessed by their residents. Furthermore, the effectiveness of the Teach-the-Teacher training on the teaching behavior of the study participants was not studied.

• Fifth, we measured the social network relationships for new developments in the departments. To generate more richness in the nature of the social networks, further research might include different kinds of relationships (for example, collaboration, trust, and advice relationships) along with variables which can explain the social relationships found (for example, physical proximities and the personal characteristics of the respondents). A mixture of quantitative and qualitative techniques would be preferable in order to measure these variables.

• Finally, the study was limited to innovation in medical training. Although training and health care delivery are interwoven and the new structured feedback tech- nique can have a direct impact on health care delivery, clinical errors and patient safety, we need to be cautious in generalizing the findings fromthis study to innovation in health care as a whole.

혁신의 분류: 복잡-단순, 과정-결과, 점진적-급진적.

이번 구조화피드백은 복잡한-과정-점진적 혁신이었다.

The structured feedback technique can be defined as a complex incre- mental process innovation;

• complex, because adopters are asked to learn new non-medical knowledge and skills and integrate these into daily practice.

• Process innovations are new elements intro- duced into an organization’s production or service operations in order to produce a product or render a service (Baregheh, Rowley, & Sambrook, 2009). The structured feedback technique improves the learning process of the resident and, therefore, improves the process of health care delivery.

• The focus of incremental innova- tions is on the renewal and improvement of existing products or services and technologies (Baregheh et al., 2009). The new struc- tured feedback technique is a renewal of and an improvement on the existing medical specialty training programs.

 

혁신의 유형에 따라 다른 영향을 보는 것도 흥미로울수도 있음.

It would be interesting to examine the effects that network tie strength and social network structures have on different types of innovations (simple vs. complex, product vs. process, and imitative vs. radical) in medical education and primary health care processes.

 

 

 

Greenhalgh, T., Robert, G., Macfarlane, F., Bate, P., & Kyriakidou, O. (2004). Diffusion of innovations in service organizations: systematic review and recommenda- tions. The Milbank Quarterly, 82(4), 581–629.

Herzlinger, R. (2006). Why innovation in health care is so hard. Harvard Business Review, 84(5), 58–66.

Aguinis, H., & Kraiger, K. (2009). Benefits of training and development for indi- viduals and teams, organizations, and society. Annual Review of Psychology, 60(1), 451–474.

Rubak, S., Mortensen, L., Ringsted, C., & Malling, B. (2008). A controlled study of the short- and long-term effects of a Train the Trainers course. Medical Education, 42(7), 693–702.

Sibthorpe, B., Glasgow, N., & Wells, R. (2005). Emergent themes in the sustainability of primary health care innovation. The Medical Journal of Australia, 183(10 Suppl.), S77–S80.

Wejnert, B. (2002). Integrating models of diffusion of innovations: a conceptual framework. Annual Review of Sociology, 28(1), 297–326.

 

---

Soc Sci Med. 2010 May;70(10):1509-17. doi: 10.1016/j.socscimed.2009.12.035. Epub 2010 Feb 12.

Disseminating educational innovations in health care practice: training versus social networks.

Jippes E1, Achterkamp MC, Brand PL, Kiewiet DJ, Pols J, van Engelen JM.

Author information

• 1Postgraduate School of Medicine, Wenckebach Institute, University Medical Centre Groningen, University of Groningen, the Netherlands. e.jippes@wenckebach.umcg.nl

Abstract

Improvements and innovation in health service organization and delivery have become more and more important due to the gap between knowledge and practice, rising costs, medical errors, and the organization of health care systems. Since training and education is widely used to convey and distribute innovative initiatives, we examined the effect that following an intensive Teach-the-Teacher training had on the dissemination of a new structured competency-based feedback technique of assessing clinical competencies among medical specialists in the Netherlands. We compared this with the effect of the structure of the social network of medical specialists, specifically the network tie strength (strong ties versus weak ties). We measured dissemination of the feedback technique by using a questionnaire filled in by Obstetrics & Gynecology and Pediatrics residents (n=63). Data on network tie strength was gathered with a structured questionnaire given to medical specialists (n=81). Social network analysis was used to compose the required network coefficients. We found a strong effect for network tie strength and no effect for the Teach-the-Teacher trainingcourse on the dissemination of the new structured feedback technique. This paper shows the potential that social networks have for disseminating innovations in health service delivery and organization. Further research is needed into the role and structure of socialnetworks on the diffusion of innovations between departments and the various types of innovations involved.

Copyright (c) 2010 Elsevier Ltd. All rights reserved.

PMID:

 

20199840

 

DOI:

 

10.1016/j.socscimed.2009.12.035

[PubMed - indexed for MEDLINE]