Health care is a knowledge-intensive industry.1 The knowledge used by health care professionals to treat patients is derived from the best available research evidence, and there is an understanding that best practices rely on research evidence.2 However, despite its importance, the link between knowledge generation (research) and knowledge application (practice) has been described as tenuous.3,4 The need to strengthen this link has been addressed through various knowledge transfer initiatives that facilitate the flow of information from research findings into practice.5 Some examples include translational research efforts that focus on translating discoveries generated during basic and clinical research to the practice setting,6 as well as researchers publishing in relevant journals and guideline developers using research to formulate guidelines for practice. Practitioners work according to these guidelines and might establish journal clubs to assist in the interpretation and application of research knowledge to their own practice.7 To ensure clinical relevance, the reverse flow—from practice to research—is also important.6 For example, there have been recent insights into the importance of including clinicians5,8 and the patient voice9–11 in research projects.
Past efforts to strengthen the link between research and practice have varied in nature and degree of interaction between the 2 worlds but frequently conceptualize the link between them as linear and unidirectional.2 More recently, the research–practice gap has been reconceptualized as occurring on various levels, with discontinuities between research and practice occurring as a result of epistemic, professional, and organizational issues.2,12 This complexity suggests that creating a viable link between research and practice requires a more dynamic, bidirectional approach.13 Attempts from both sides are necessary to lower the boundaries between 2 distinctly different sociocultural domains.5 One manner of linking research and practice bilaterally is through brokers: professionals who have legitimate access to both settings14 and engage in knowledge management, capacity building, and linkage and exchange activities.15 (In this review, we chose to use the term brokers rather than knowledge broker to showcase the breadth of activities [e.g., organizing continuing professional development workshops, maintaining an extensive network] that such a role encompasses.16)
In terms of linking research and practice, clinician–scientists are considered to be ideally placed to fulfill the broker role. Their importance was flagged in the early 1900s, with Meltzer advocating for medical professionals (clinicians) who are competent in conducting scientific research rather than relying on external researchers to set the research agenda.17 To date, titles such as physician–scientist, clinician–scientist, and nurse–scientist are used to describe doctors, allied health professionals, and nurses working simultaneously in clinical and research contexts, respectively. Many publications exist that highlight the importance of brokers in health care.14 Issues such as the dwindling numbers of clinician–scientists despite their importance,18 the importance of good research mentoring for clinician–scientists,19 and the difficulty of securing research funding20 are discussed in these publications. However, the broker role of clinician–scientists is not explicated as a separate task in these publications. Clinician–scientists’ potential to link bench to bedside is assumed, and their brokering tasks are not described or considered to be separate from their clinical- and research-related work.21 Thus, limited attention has been given to the nature of the broker role. In this review, we sought to answer the question: Which contextual factors and mechanisms contribute to the outcomes achieved by clinician–scientists in linking research and practice?
To provide insight into how clinician–scientists connect research and practice, we first analyzed which outcomes they achieved. We then looked in detail at which context factors influence the execution of the broker role and which mechanisms are induced within these contexts that contribute to the outcomes. We subsequently considered program theories that might assist in the understanding of why contexts and mechanisms yield certain outcomes.
We conducted a realist review, which is a type of review that is well suited to study complex social phenomena by answering questions such as “What works for whom?” “How does it work?” and “In what circumstances does it work?”22 A realist review unpacks context–mechanism–outcome (CMO) configurations, which are chains of reasoning that detail which contexts facilitate the activation of certain mechanisms that contribute to certain outcomes. We judged this to be an adequate method to answer our research question about how clinician–scientists connect research and practice. For the context, we focused on inner and outer context factors.23 Inner context factors refer to characteristics of individuals, such as beliefs about science, and outer context factors refer to situational factors at the organizational level, such as management support. Mechanisms are defined as people’s underlying responses, processes, or manners of reasoning that “operate in particular contexts to generate outcomes of interest.”24(p368) We classified outcomes as individual level or organizational level.
We developed 2 search strings iteratively. The search strings consisted of a combination of the search term clinician–scientist or broker and their synonyms. These terms and their synonyms were also truncated in different ways to assist in finding relevant articles (see Supplemental Digital Appendix 1 at http://links.lww.com/ACADMED/A692 for the full search strings). The first search string was designed to identify articles about professionals who have a title such as clinician–scientist. The second search string was designed to identify articles about professionals who may not have an official title signifying a dual role but who perform both clinical- and research-related tasks. Both search strings were applied to the CINAHL, Embase, Web of Science, PubMed, and PsycINFO databases on 4 separate dates from May to August 2017.
In 2 discussion rounds with all authors, the list of inclusion and exclusion criteria was applied to a selection of 48 articles to calibrate the criteria. Subsequently, titles and abstracts were screened by 2 authors (M.B. and E.d.G.) and discussed. When in doubt about whether a paper should be included or if there was disagreement, the full text was screened and the other authors were consulted. The main objective of the criteria was to identify publications that described the broker activities of clinician–scientists in a manner that was detailed enough to be able to identify the contexts, mechanisms, and outcomes of these—regardless of whether the professionals had the formal title of clinician–scientist. List 1 presents the final inclusion and exclusion criteria.
An overview of all included articles was compiled to provide insight into variations between the articles’ participant populations, study contexts, methods, and study aims related to bridging the research–practice divide. The title and country of publication were listed, as were the level of qualification (e.g., PhD, master’s, bachelor’s) and profession (e.g., nurse, physician) of the clinician–scientists in the study. For each study, the aim, research method, sample size, and our assessment of its rigor were documented. Two authors—one an experienced quantitative researcher (M.-L.E.L.B.) and one an experienced qualitative researcher (N.S.)—assessed the studies using the Joanna Briggs Foundation guidelines.25 They judged the methodological quality of each study and the extent to which bias was addressed using the appropriate critical appraisal checklist. To enhance credibility, we included only empirical studies. Relevance and rigor were not treated as absolute criteria but, rather, as dimensions of fitness for the purposes of this review.26 As is usual for realist reviews, we did not exclude or rank publications based on our assessment of the rigor.26
Data analysis and synthesis
A 3-step approach was taken during the data analysis process.
Step 1. The full text of the articles was read, reread, and discussed by 2 members of the author team (E.d.G. and M.B.) to obtain a comprehensive understanding of the data.
Step 2. E.d.G. and M.B. coded the entire dataset using NVivo 11 (QSR International, Doncaster, Victoria, Australia). To assist in understanding inner and outer context factors, we coded for individual and situational barriers and facilitators. To assist in the identification of mechanisms, we coded for brokering role descriptions and brokering activity descriptions. To assist in gaining an understanding of outcomes, we coded for outcomes of brokering activities. The coding began with codes derived from sensitizing concepts and with the aim of a realist review in mind. We added codes iteratively during the data analysis. We did not use the codes or themes presented in the original research of included qualitative studies.
Step 3. E.d.G. and M.B. discussed the initial coding and subsequently refined the coding by making a distinction between context factors that pertained exclusively to the broker role and those that related to either research- or practice-based activities only, with no element linking the two. Contextual factors that were not exclusively relevant to the broker role (e.g., research funding availability) were excluded from our CMO configurations.
The data synthesis was initially conducted by E.d.G. and M.B. and was subsequently discussed with the full author team. The data synthesis focused on assembling findings into CMO configurations, which a realist review typically seeks to unpack.22 First, we focused on outcomes of interest and then identified mechanisms associated with each outcome.27 The descriptions of brokering activity provided a starting point for this process. We then examined data to determine the relationship between contexts, mechanisms, and outcomes to derive CMO configurations. Next, we clustered mechanisms to identify program theories that are useful in explaining clinician–scientists’ broker role. Program theories are useful in describing why certain outcomes were achieved.28 Thus, these theories provide a diversity of lenses through which the broker role of clinician–scientists can be viewed to gain an understanding of how it functions and how it could be supported.
Finally, we tested the plausibility of each CMO configuration by rereading the full text of the included articles and discussing the configurations among the entire author team.
After duplicates were removed, our initial searches yielded 1,445 articles for the first search string and 796 for the second string (n = 2,241). After title and abstract screening and applying inclusion and exclusion criteria (see List 1), 52 articles remained for full-text review. After full-text review (M.B. and E.d.G.), 41 articles were excluded for assuming the broker role of clinician–scientists and focusing on clinical and/or research activities of clinician–scientists without providing a description of the broker role, and 2 were excluded for their focus on research mentoring rather than brokering. The results and stages of the selection process are presented in the PRISMA29 flowchart in Figure 1. Nine articles were included in the final full-text review.2,30–37
Final Inclusion and Exclusion Criteria for a 2017 Realist Review Examining How Clinician–Scientists Connect Research and Practice
- In the study, both clinical researchers and academic professionals with a clinical role are involved or titles such as clinician–scientist, nurse–scientist, or physician–scientist are used.
- Studies that contribute to the understanding of professional roles and activities that span the academic (research) and clinical field of function (that is, studies that aim to describe the role and work activities of a broker).
- Studies are conducted in a medical professional context: professionalsa or professionals in PhD-level training within the medical context, such as physicians, surgeons, physicists, pharmacists, dentists, nurses, allied health professionals, clinician–scientists, and nurse–scientists. (Residents are included.)
- Unit of analysis should be the individual clinician–scientist and his/her work context.
- Studies with a focus on cultural or psychological elements.
- Studies that clarify factors that facilitate fulfilling the role of broker.
- Studies that present empirical data (i.e., primary research data).
- Studies focusing on the collaboration between researchers and clinicians will be included to ascertain potential broker activities.
- Studies that address the role of mentor to a broker and also reference the broker role of the mentee.
- Studies that were peer-reviewed.
- Studies from any date (the oldest included publication was from 1961 and the newest was from 2017).
- Unit of analysis on the level of a team or organization (indicated by statements such as “teams should” or “organizations could better”). A team is not able to act as a broker.
- When the word “facilitator” or “moderator” is used in studies about a specific research design (e.g., focus groups). Studies are about a different topic, but the word “facilitator” or “moderator” is included in the abstract.
- Studies about professionals who are appointed to be brokers, it is their formal job, and who do not have the same background as professionals from one of the “sides of the boundary.” Probably called a “liaison.”
- Conceptual or theoretical papers, opinion pieces, position papers, commentaries, obituaries, or literature reviews.
- Language of the full article is not English.
- Studies conducted in contexts related to but distinctly separate from the medical context, such as complementary medicine, psychology, psychotherapy, music therapy, art therapy, social work, and social sciences.
- Studies researching training and mentoring of medical, nursing, and allied health students at the bachelor’s or master’s level who are not employed as clinician–scientists (e.g., teaching students research skills to be able to perform a clinician–scientist role).
- Studies focusing on training and mentoring without a clear link to the clinician–scientist role (e.g., training to increase research productivity).
aProfessionals are individuals who have a position as a clinician–scientist, which includes individuals who are employed as clinician–scientists and hold a bachelor-level qualification or higher.
Of the 9 included articles, the nursing profession is represented in 3 of them,30,31,33 whereas physicians are represented in 1.34 The remaining papers represent multiprofessional groups.2,32,35–37 A summary of the characteristics of all of the included articles is presented in Table 1.
Outcomes of brokering
Three outcomes of brokering activities were identified—2 at the organizational level and 1 at the individual level (Chart 1).
The first outcome was a perceived growth in clinically relevant and practically applicable research results.2,31,32,34–36 However, 1 article indicated that practically applicable research still remained sparse.35 The second outcome was a perceived increase in evidence application to improve care.30–32,34–37 These were the organizational-level outcomes.
The individual-level outcome of brokering activity was the professional development of the clinician–scientist in their capacity as a researcher, clinician, and broker.2,30,32,34,36
One article identified an individual-level outcome that does not fall under professional development—that is, a participant reported that he/she perceived his/her relationship with colleagues had changed since becoming a clinician–scientist.32 As a clinician–scientist, he/she felt that he/she no longer belonged to the community of clinicians or to the community of researchers.
We found a number of inner and outer context factors that play a role in harnessing mechanisms to achieve organizational-level outcomes. Of these, some inner context factors reflect the necessity of brokers having multidimensional skills, including networking,32,36 innovation,32 and change management skills.35,37 In addition, clinician–scientists require the ability to prioritize tasks to maintain a career across both fields.35 Necessary outer context factors included management support30,35,36 as well as epistemological and cultural alignment between scientific and clinical work contexts2,31,32,34–36 (Chart 1).
In contrast to organizational-level outcomes, the individual-level outcome requires relatively few inner and outer context factors to activate the mechanisms necessary to achieve the outcome. Inner context factors required to achieve the individual outcome were a strong interest in the research results of others30 and a high level of motivation and capacity to read relevant scientific literature.30 Similar to organizational outcomes, an outer context factor that yields the individual outcome is management support.30,35,36 Yet, in contrast, we did not find that an alignment between the scientific and clinical contexts was required to achieve individual-level outcomes of brokering (Chart 1).
Further, we set out to understand the ways in which context factors would result in outcomes by focusing on supporting mechanisms. Understanding these might provide insight into the specific elements required for interventions that seek to capitalize on the value of brokering for bridging the gap between research and practice. Mechanisms that contribute to the achievement of organizational-level outcomes include the extent to which brokers align their own goals or values with organizational goals,36 balancing economic and scientific interests,34 direct involvement in key decision making in research activity,31,32,35 and direct involvement in key decision making in evidence-based change projects35,37 (Chart 1).
The mechanisms for the individual-level outcome include engaging in reflective practice32,35 and associating habitually with the clinical care setting2 (Chart 1). The mechanisms that feature in both individual- and organizational-level CMO configurations are the performance of boundary-crossing activities35,36 and a strategic focus on networking activities36 (Chart 1).
CMO configurations and program theory
Formulating CMO configurations (labeled as CMO 1–CMO 4 below) and then clustering the mechanisms (Chart 1) assisted in identifying 4 program theories that could be helpful in explaining the broker role of clinician–scientists. Figure 2 presents the overarching realist program theory.
Boundary crossing theory assisted in understanding the presence of both organizational- and individual-level outcomes. Boundary crossing theory stipulates that boundaries between related yet disparate contexts offer learning opportunities for brokers attempting to bridge the discontinuities between the culture, perspectives, and practice of the disparate contexts of research and health care practice.38 Boundary crossing occurs at the intrapersonal, interpersonal, and organizational levels.
At the organizational level, the effectiveness of boundary-crossing activities of clinician–scientists depends in part on the degree of alignment between research and practice.2,31,32,34–36 CMO 1 gives an example of this: A clinician–scientist who works in an organization where there is epistemological and cultural alignment between the scientific and clinical work contexts2,32,34–36 (context) is able to effectively engage in boundary-crossing activities35,36 (mechanism), which contributes to increased evidence application to improve care30–32,34–37 (outcome).
Social network theory offers insights into how brokers occupy central positions in a network through which valuable resources can be accessed. Brokers are often seen as occupying “structural holes”39 in a network, in which they have valuable access and, therefore, hold disproportionate power over the distribution of resources (e.g., knowledge) in a group of people.
CMO 2 gives an example of this as applied to clinician–scientists: A clinician–scientist with good networking skills32,36 and management support30,35,36 (context) could establish collaborations between clinicians and researchers in the writing of research proposals35 (mechanism). This can contribute to an increase in the volume of practically applicable, clinically relevant research results2,31,32,34–36 (outcome).
Communities of practice.
Communities of practice theory stipulates that professionals with a shared domain of practice work together in groups (or communities) in daily work activities.40 The degree to which professionals engage with their community of coworkers either leads to full participation (community membership) or marginality (isolation). Community membership ascribes a level of competence to an individual and allows for participation in key decision-making processes.41
CMO 3 gives an example of this applied to clinician–scientists: A clinician–scientist who is viewed by clinicians as a plausible or competent clinician32 (context) is in a position to be directly involved in key decision making in evidence-based change projects31,32,35 (mechanism), thereby facilitating increased evidence application to improve care30–32,34–37 (outcome).
Diffusion of innovation.
Diffusion of innovation theory stipulates that new practices (innovations) are communicated through social systems within organizations. The adoption of these innovations by colleagues depends in part on the plausibility and competence of those leading change projects.42 Innovativeness requires both an individual with new ideas and an organization that supports these, which then results in new products, services, or processes.42
CMO 4 gives an example of this as applied to clinician–scientists: A clinician–scientist who has an open mindset in allowing practice and policy to inform research2,35 and has management support30,35,36 (context) is able to critically reflect on public health issues and the patient perspective32,35 (mechanism). This places them in a position to increase the volume of clinically relevant and practically applicable research2,31,32,34–36 (outcome).
In this realist review, we aimed to provide insight into how clinician–scientists connect research and practice through an analysis of the outcomes, context factors, and mechanisms in the included articles. Several inner and outer context factors are necessary to facilitate the mechanisms needed to achieve outcomes of interest. These mechanisms can be understood better using boundary crossing, social network, communities of practice, and diffusion of innovation theory, which were helpful in developing an overarching realist program theory (see Figure 2).
In our review, we encountered many papers in which the activities characterizing the broker role of clinician–scientists, who bridge the worlds of practice and research, were not described explicitly; instead, these papers focused only on reporting research or clinical activities or the mentor role. While valuable, these papers did not offer insights specific to the broker role of clinician–scientists. They were, therefore, excluded from this review, yet they contributed to our initial finding that the broker role is frequently assumed without an in-depth understanding of what it really entails.
Our finding of a clinician–scientist who perceived herself as belonging to neither the community of practitioners nor the community of researchers32 is singular yet relates to a known general difficulty experienced by professionals in dual roles.21 Clinician–scientists might not be full members of communities of clinicians or communities of scientists; however, this appears to be an important mechanism by which they can achieve outcomes as brokers. As such, this could provide a target for interventions aiming to support clinician–scientists in their broker role.
Clinician–scientists facilitate bidirectional linkage and exchange15 between the contexts of research and practice. They contribute to the design and development of research that is practically applicable and to the implementation of research results in practice. In the limited number of research papers found in this review that empirically studied how clinician–scientists facilitate these connections, evidence was mostly focused on organizational-level outcomes.2,30–32,34–37 In contrast, individual-level outcomes were mentioned less frequently.2,30,32,34,36
Moreover, we did not find work that has been conducted to explore the impact of alignment (or the lack of alignment) between the scientific and clinical contexts on the broker role of clinician–scientists and its potential effects on the individual-level outcome. Given that boundaries between the scientific and clinical contexts play an important role in the ease with which clinician–scientists can perform their role2 and that crossing these boundaries provides valuable learning opportunities for both contexts, it seems necessary to further explore these boundaries, the actual or perceived alignment (or misalignment) between research and practice, and the resulting broker outcomes. The organization has a role to play in creating a cultural alignment between the scientific and clinical work contexts,2,32,34–36 which is, for the most part, beyond the capacity of a single broker.
The mechanisms found in our review suggest that clinician–scientists who are able to switch between strategic thinking and operational work tasks achieve organizational-level outcomes. An example of this is the ability to balance economic and scientific interests34 while being directly involved in key decisions in conducting research.31,32,35 Additionally, having a strategic focus on their networking activities36 while also collaborating and strengthening ties with partners across boundaries2,35,36 is essential. These are skills that are additional to the skills required for research and the clinical work of a clinician–scientist. We therefore suggest that very specific skills are required to broker a connection between research and practice.
Practical and theoretical implications
The mechanisms we identified offer valuable levers that policymakers and educators can use to support brokers in optimizing their role. One way to support the broker role could, for example, be to include the organization of opportunities for clinician–scientists to actively contribute to “reflection across boundaries” in strategic change projects that explicitly focus on bridging research–practice gaps. Individuals and organizations could gain from the expertise and frequent boundary-crossing activities of these professionals who connect research and practice, which, in many instances, are still disparate sociocultural domains.
Strengths and limitations of the review
Strengths of this review include the comprehensive literature search that was achieved through the use of 2 distinct search strings to capture a wide variety of articles on medical professionals in a dual role. A limitation of this study is that little empirical research has been done on the broker role of clinician–scientists, and, therefore, we had to base our findings on only 9 original research articles. In addition, the findings of the original articles are mainly qualitative in nature and, therefore, could reflect the perceptions and experiences of those participants rather than generalizable facts.
We identified 4 CMO configurations by which clinician–scientists achieve individual- and organizational-level outcomes in brokering a connection between research and practice. As such, our review provides insight into the inner and outer context factors and mechanisms that allow clinician–scientists to connect the worlds of research and practice. Clinician–scientists require additional skills, distinct from the skills they need to conduct research and carry out their clinical duties, to fulfill their broker role effectively, implying that organizational settings, initial training, and continuing professional development for clinician–scientists need to focus explicitly on skills for the brokering role, as well as those needed for research and clinical practice. Based on our findings, we expect that if more attention is paid to learning these skills and management support for the broker role is strengthened, stronger links between research and practice could be forged.
The authors would like to acknowledge the contribution of the 3 anonymous peer reviewers in strengthening this article.
1. Yamazaki T, Umemoto K. Knowledge management of healthcare by clinical-pathways. J Inf Knowl Manage. 2010;9:119–125.
2. Lander B. Boundary-spanning in academic healthcare organisations. Res Policy. 2016;45:1524–1533.
3. Butler D. Translational research: Crossing the valley of death. Nature. 2008;453:840–842.
4. Roberts SF, Fischhoff MA, Sakowski SA, Feldman EL. Perspective: Transforming science into medicine: How clinician–scientists can build bridges across research’s “valley of death”. Acad Med. 2012;87:266–270.
5. Paget SP, Caldwell PH, Murphy J, Lilischkis KJ, Morrow AM. Moving beyond ‘not enough time’: Factors influencing paediatric clinicians’ participation in research. Intern Med J. 2017;47:299–306.
6. Coller BS. Translational research: Forging a new cultural identity. Mt Sinai J Med. 2008;75:478–487.
7. Ter Brugge BPH, Bartelink MEL, Damoiseaux RAMJ, de Groot E. The use of evidence during group meetings of Dutch general practitioners. Educ Prim Care. 2017;28:307–312.
8. Chan R, Gardner G, Webster J, Geary A. Building research capacity in the nursing workforce: The design and evaluation of the nurse researcher role. Aust J Adv Nurs. 2010;27:62–69.
9. Chew-Graham CA. Patient involvement in research - participants or collaborators? Health Expect. 2017;20:369–370.
10. Manafò E, Petermann L, Vandall-Walker V, Mason-Lai P. Patient and public engagement in priority setting: A systematic rapid review of the literature. PLoS One. 2018;13:e0193579.
11. Sheridan S, Schrandt S, Forsythe L, Hilliard TS, Paez KA; Advisory Panel on Patient Engagement (2013 inaugural panel). The PCORI engagement rubric: Promising practices for partnering in research. Ann Fam Med. 2017;15:165–170.
12. Evans S, Scarbrough H. Supporting knowledge translation through collaborative translational research initiatives: ‘Bridging’ versus ‘blurring’ boundary-spanning approaches in the UK CLAHRC initiative. Soc Sci Med. 2014;106:119–127.
13. Rowland P, Ng S. Multiple boundaries: Professional and institutional identities of clinician-scientists. Med Educ. 2017;51:568–570.
14. Bornbaum CC, Kornas K, Peirson L, Rosella LC. Exploring the function and effectiveness of knowledge brokers as facilitators of knowledge translation in health-related settings: A systematic review and thematic analysis. Implement Sci. 2015;10:162.
15. Ward V, House A, Hamer S. Knowledge brokering: The missing link in the evidence to action chain? Evid Policy. 2009;5:267–279.
16. Fernandez RM, Gould RV. A dilemma of state power: Brokerage and influence in the national health policy domain. Am J Sociol. 1994;99:1455–1491.
17. Meltzer SJ. The science of clinical medicine: What it ought to be and the men to uphold it. JAMA. 1909;53:508–512.
18. Yin HL, Gabrilove J, Jackson R, Sweeney C, Fair AM, Toto R; Clinical and Translational Science Award Mentored to Independent Investigator Working Group Committee. Sustaining the clinical and translational research workforce: Training and empowering the next generation of investigators. Acad Med. 2015;90:861–865.
19. Brant JM. Bridging the research-to-practice gap: The role of the nurse scientist. Semin Oncol Nurs. 2015;31:298–305.
20. Rickard CM, Williams G, Ray-Barruel G, et al. Towards improved organisational support for nurses working in research roles in the clinical setting: A mixed method investigation. Collegian. 2011;18:165–176.
21. Kislov R, Wilson P, Boaden R. The ‘dark side’ of knowledge brokering. J Health Serv Res Policy. 2017;22:107–112.
22. Wong G, Greenhalgh T, Westhorp G, Buckingham J, Pawson R. RAMESES publication standards: Realist syntheses. J Adv Nurs. 2013;69:1005–1022.
23. Aarons GA, Hurlburt M, Horwitz SM. Advancing a conceptual model of evidence-based practice implementation in public service sectors. Adm Policy Ment Health. 2011;38:4–23.
24. Astbury B, Leeuw FL. Unpacking black boxes: Mechanisms and theory building in evaluation. Am J Eval. 2010;31:363–381.
25. Joanna Briggs Institute. Critical appraisal tools. http://joannabriggs.org/research/critical-appraisal-tools.html
. Accessed May 7, 2019.
26. Pawson R, Greenhalgh T, Harvey G, Walshe K. Realist review—A new method of systematic review designed for complex policy interventions. J Health Serv Res Policy. 2005;10(suppl 1):21–34.
27. Byng R, Norman I, Redfern S. Using realistic evaluation to evaluate a practice-level intervention to improve primary healthcare for patients with long-term mental illness. Evaluation. 2005;11:69–93.
28. Rogers PJ. Using programme theory to evaluate complicated and complex aspects of interventions. Evaluation. 2008;14:29–48.
29. Moher D, Liberati A, Tetzlaff J, Altman DG; PRISMA Group. Preferred reporting items for systematic reviews and meta-analyses: The PRISMA statement. Ann Intern Med. 2009;151:264–269, W64.
30. Adamsen L, Larsen K, Bjerregaard L, Madsen JK. Danish research-active clinical nurses overcome barriers in research utilization. Scand J Caring Sci. 2003;17:57–65.
31. Logsdon MC, Kleiner C, Oster CA, et al. Description of nurse scientists in a large health care system. Nurs Adm Q. 2017;41:266–274.
32. Kluijtmans M, de Haan E, Akkerman S, van Tartwijk J. Professional identity in clinician-scientists: Brokers between care and science. Med Educ. 2017;51:645–655.
33. Kelly KP, Turner A, Gabel Speroni K, McLaughlin MK, Guzzetta CE. National survey of hospital nursing research, part 2: Facilitators and hindrances. J Nurs Adm. 2013;43:18–23.
34. Wilson-Kovacs DM, Hauskeller C. The clinician-scientist: Professional dynamics in clinical stem cell research. Sociol Health Illn. 2012;34:497–512.
35. Hoeijmakers M, Harting J, Jansen M. Academic Collaborative Centre Limburg: A platform for knowledge transfer and exchange in public health policy, research and practice? Health Policy. 2013;111:175–183.
36. Long JC, Hibbert P, Braithwaite J. Structuring successful collaboration: A longitudinal social network analysis of a translational research network. Implement Sci. 2016;11:19.
37. Ritchie MJ, Parker LE, Edlund CN, Kirchner JE. Using implementation facilitation to foster clinical practice quality and adherence to evidence in challenged settings: A qualitative study. BMC Health Serv Res. 2017;17:294.
38. Akkerman SF, Bakker A. Boundary crossing and boundary objects. Rev Educ Res. 2011;81:132–169.
39. Burt RS. Network disadvantaged entrepreneurs: Density, hierarchy, and success in China and the West. Entrepr Theory Pract. 2018;43:19–50.
40. Wenger E. Communities of Practice: Learning, Meaning, and Identity. 1991.Cambridge, UK: Cambridge University Press.
41. Lave J, Wenger E. Situated Learning: Legitimate Peripheral Participation. 1991.Cambridge, UK: Cambridge University Press.
42. Straub ET. Understanding technology adoption: Theory and future directions for informal learning. Rev Educ Res. 2009;79:625–649.
Contexts and Mechanisms for Organizational- and Individual-Level Outcomes of Brokering Activities, From a 2017 Realist Review Examining How Clinician–Scientists Connect Research and Practice