Translation of medical evidence into practice, policy, and public health improvements refers to the widespread dissemination and adoption of interventions that can have a significant effect on health. Although most interventions designed to improve population health target individuals, access to and delivery of these interventions depends on communities, health care delivery systems (HCDSs), health care professionals, and government agencies. Effective translation has been slow and inconsistent across the spectrum of HCDSs and communities in the United States,1 – 4 and such inconsistencies contribute to the current state of variable and suboptimal population health.5 – 7 Within U.S. HCDSs, challenges with translation are perhaps best exemplified by the finding that dissemination of practice guidelines rarely changes practice.7
Researchers trained in implementation and dissemination science (IDS) are needed to facilitate the translation of evidence into practice. The National Institutes of Health (NIH) defines implementation as “the use of strategies to adopt and integrate evidence-based health interventions and change practice patterns within specific settings” and dissemination as “the targeted distribution of information and intervention materials to a specific public health or clinical practice audience.”8 IDS research aims to examine and promote changes at the intersection of health services (interventions intended to improve health, such as specific treatments, tests, or behavioral strategies), delivery systems (systems that facilitate the delivery of health services, such as hospitals, clinics, or community- or policy-based organizations), and communities (groups of individuals, such as physicians, patients with diabetes, hospital administrators, or policy makers). These sets of actors collaboratively shape the reach, relevance, uptake, and diffusion of interventions.
During the past 30 years, clinical research has been redefined by the application of epidemiology and biostatistics theory and methods to biological, clinical, and population health sciences. This transformation has led to new approaches, methods, and training programs, which have yielded a generation of investigators specializing in relatively new fields such as clinical trials, clinical and social epidemiology, and evidence-based medicine. Recent mandates from stakeholder organizations (e.g., government agencies, payers) that direct researchers to translate knowledge gained from the publicly funded investment in biomedical sciences research into improved population health are similarly transforming IDS research.9
Further, there is a large and growing body of evidence that supports the effectiveness of intervention implementation strategies based on IDS principles.10,11 (For example, the Cochrane Collaboration's Effective Practice and Organisation of Care Group conducts systematic reviews on a wide variety of implementation strategies and tools.12) However, even though IDS researchers require specific skills to help translate evidence into practice, there is a dearth of IDS training programs in the United States—in part because the skill set has not been adequately defined and continues to evolve. A directory of implementation science training programs provided by the March 2011 NIH Conference on Dissemination and Implementation Science identified only 6 “specialized programs” and 14 “general training programs with implementation science as [a] secondary or optional concentration” in the United States.13
To help inform institutional efforts to develop and expand IDS research skills training for health care professionals, we propose in this article a conceptual framework that defines, organizes, and guides specific translational activities. Using this framework, we propose seven domains comprising 12 core competencies for IDS researchers. Finally, we describe the series of IDS-specific courses at the University of California, San Francisco (UCSF) that has evolved from this conceptual framework and supports the attainment of these core competencies.
A Conceptual Framework for Training in IDS Research
The goals and methods of IDS research are distinct from those of conventional clinical research. The IDS approach to improving health care safety, efficiency, equitability, and patient-centeredness focuses on care structures and processes, and it requires sustained engagement with the individuals, communities, and organizations targeted by health interventions.14 To prepare investigators and health professionals for this approach, we believe a new fusion must take place among the clinical and public health disciplines, which provide content and context expertise in specific health-related areas; the population sciences (e.g., epidemiology, biostatistics, health policy), which generate human-subjects-based medical evidence; and the translational disciplines, which we define as the diverse array of IDS-relevant disciplines that provide a theoretical and methodological basis for understanding and changing behaviors (Figure 1).
Given that IDS goals for translating research findings into practice lie at the intersection of a variety of disciplines, it is not feasible for implementation scientists to be expert in all of the relevant theories and methods. This is why successful IDS research draws on multidisciplinary teams. Team science is critical for effective implementation of health care improvement strategies. As a result, IDS training programs—which should include trainees from a wide variety of educational and professional backgrounds—will benefit from an interdisciplinary model for teaching and curriculum development. There is a growing body of research demonstrating the advantages of incorporating interdisciplinary approaches such as team science into training programs. For example, Nair and Finucane15 found that seasoned and early-career researchers who have been trained in interdisciplinary graduate programs are capable of consistently bridging the knowledge and methods of different disciplines and of more effectively tailoring their investigations to local circumstances.
To guide the development and/or expansion of IDS research training curricula for investigators and health professionals, we developed a series of design principles and a conceptual framework. Our iterative, collaborative development process involved reviews of literature on IDS-related research goals and discussions of those goals' implications for training. Below, we describe each of the three design principles and related aspects of the conceptual framework.
Design principle 1: Behavior change among individuals and delivery systems is inherent in the translation of evidence into practice, policy, and public health improvements
As depicted in our conceptual framework (Figure 2), behavior change targets can represent stakeholder organizations, which include government agencies, employers, insurers/payers, and academic institutions; HCDSs, which are organizations that provide health care services or the settings in which such services are delivered (e.g., hospitals, private practices, pharmacy-based clinics); and individuals, who include patients and clinicians practicing within or outside HCDSs. Defining a given translational activity's behavior change target can direct the IDS research team to the relevant theory and current evidence for program planning and development. For example, Shortell and Kaluzny16 have defined some of the theoretical constructs and translational tools that pertain to organizational behavior change among HCDSs. Glanz and colleagues17 have compiled a similar body of knowledge related to behavior change among individuals and social groups (including patients and clinicians).
A key learning objective for IDS trainees involves identifying the range of factors (behavioral, social, ethical, institutional, political, and economic) that shape a particular behavior or organizational process, and then choosing appropriate theoretical constructs to describe and predict a desired change. For example, one of the major lessons that emerged from the application of Prochaska and Velicer's18 “stages of change” theory to smoking cessation trials was that smokers at different stages of change require different interventions. For example, smokers who express motivation and a “readiness to quit” are much more successful with interventions focusing on therapeutic support (e.g., nicotine replacement therapy) than are smokers who have not expressed interest in quitting. By contrast, effective strategies for smokers who have not considered quitting are those that increase their readiness to quit, such as social marketing and motivational interviewing techniques. This theory-driven approach has been successfully employed in behavior change interventions targeting diverse settings, communities, and individuals, including clinicians in quality improvement programs and young adults in social settings.19
Implicit in the concept of behavior change related to the translation of evidence into practice is an operator who sponsors, directs, and/or promotes the proposed behavior change. Depending on the nature of the intervention or research activity, an operator can also serve as the target for behavior change. IDS researchers should be adept at identifying the key players in a translational project because doing so allows them to establish the nature of relationships and the direction of information flow. For example, a stakeholder organization such as Medicare (acting as the operator) can implement a new policy that requires hospitals (the behavior change target) to report the number of their emergency department patients who “left without treatment.” Although the ultimate desired outcome is to reduce the number of patients who leave emergency departments without treatment, from an implementation design perspective, the key behavior of this intervention is hospital compliance with public reporting (and relevant theory might center on organizational behavior change). In another example, a community advocacy group (acting as the operator) could create a media campaign and lobbying effort to convince Medicare (the behavior change target) to monitor and then develop and enact policies (the behavior) that would help decrease crowding in emergency departments. As these examples illustrate, in IDS research there are often multiple and changing actors at different stages of the investigation, intervention development, and implementation; there are no “fixed” assignments of “operator” and “target.”
Defining the operator and the behavior change target for a particular problem can be complex. The IDS research team may need to use qualitative methods to tease out relevant behaviors based on interviews and direct observation of key individuals. Once the key behavior change pathways are delineated, IDS researchers can then begin testing theoretically based constructs or develop them de novo. (For examples of operator and behavior change target relationships, see Appendix 1.)
Using our conceptual framework to define these relationships can also help classify a wide variety of implementation tools and strategies employed in the translation of evidence into practice, policy, and public health (Appendix 1). Examples of tools that are commonly used to directly influence individual behaviors include clinical practice guidelines, decision aids, mass media campaigns, and conditional cash payments. Individual behaviors can also be influenced indirectly. Indirect strategies assume that changing HCDS and stakeholder organizational structures and policies will result in changes in individual (provider or patient) behaviors. These include engineered wellness programs and built-environment enhancements that encourage increased physical activity in work and community settings; measurement and public reporting of quality indicators at the hospital, clinic, health department, or health plan level; financial incentives for HCDSs to achieve certain benchmarks of quality performance (pay for performance); and accreditation policies.
Design principle 2: Engagement with a range of individuals and stakeholder organizations is imperative to achieve effective translation and sustained improvement
Historically, many initiatives to promote healthy behaviors and improve the quality of health care delivery have been implemented without direct input from and engagement with the targeted individuals/communities, or have limited their engagement to individuals/communities who are convenient to access, such as those represented by specific organizations (e.g., professional societies or churches). In this context, we endorse a broad definition of community to include ambulatory care and hospital providers and staff, clinical microsystems (e.g., intensive care units, dialysis centers), public health departments, regulatory agencies, government agencies, employers, insurers, and community-based organizations outside the health arena. The omission of targets' direct input is commonly cited20 as one reason why health care interventions and implementation strategies that succeed in one community setting may be difficult to sustain and disseminate more broadly across other community settings.
To be most effective, IDS researchers must be skilled in building successful and sustainable relationships with community members and organizations. They should integrate a thorough understanding of local context and culture into the design of the research question and intervention. Competencies in community engagement can be initiated in the classroom, but, ultimately, IDS researchers require hands-on experience with relevant community-based organizations and stakeholders.
Design principle 3: IDS research is iterative and benefits from cycles and collaborative/bidirectional relationships
A “one-size-fits-all” approach to spreading successful health care interventions from one setting to another rarely succeeds. A cyclical, rather than linear, approach is necessary because translating evidence into practice requires attention to real-world settings in which many contextual variables will influence the implementation process. With each cycle, the IDS project team identifies barriers to improving outcomes and develops strategies to minimize or work around these barriers. The ongoing evaluation of the implementation process enables the team to assess the effect of specific translational activities. The utility of this approach is exemplified by the iterative, continuous-quality-improvement-inspired plan-do-study-act cycle applied to health care by the Institute for Healthcare Improvement and other organizations,21 the community-engagement practices in community clinic settings such as those described in community-oriented primary care models,22 and the community-based participatory research models adapted from public health.23
The iterative, bidirectional approach we describe applies to implementation of a single translational activity (e.g., represented by a single dyad, such as those shown in Appendix 1) as well as to larger-scale programs that employ multiple dyads across multiple dimensions over time. To illustrate how all three design principles could be applied in a targeted intervention, we provide a case study focusing on an effort to improve chlamydia screening rates (see Supplemental Digital Appendix 1, http://links.lww.com/ACADMED/A75).
Competencies for IDS Trainees
Our design principles and conceptual framework provide a foundation for identifying key knowledge, skills, and abilities for IDS researchers. IDS requirements inherently overlap with those identified by Gebbie and colleagues24 for the field of interdisciplinary research. Their proposed interdisciplinary research competencies, developed using a Delphi process, include the following:
- Use theories and methods of multiple disciplines in developing integrated theoretical and research frameworks
- Integrate concepts and methods from multiple disciplines in designing interdisciplinary research protocols
- Investigate hypotheses through interdisciplinary research
- Draft funding proposals for interdisciplinary research programs
- Disseminate interdisciplinary research results within and outside the discipline
- Author publications with scholars from other disciplines
- Present interdisciplinary research at venues representing more than one discipline24
We integrated these IDS-relevant competencies with our IDS design principles to generate a set of seven IDS domains: team science, context identification, literature identification and assessment, community engagement, intervention design and research implementation, evaluation of effect of translational activity, and behavioral change communication strategies. For each domain, we propose associated IDS competencies that are distinct from those in traditional disciplines (Table 1). Once one of these competencies is identified as a goal in a particular IDS curriculum, course developers can suggest learning activities that are appropriate for that competency. Table 1 provides examples of specific learning activities included in UCSF's IDS courses, which we described below.
UCSF's IDS Curriculum
During 2008–2011, we used our conceptual framework and the competencies outlined above to create an IDS curriculum within UCSF's Training in Clinical Research (TICR) program. In 2010–2011, approximately 20 scholars participated substantively in the IDS curriculum (completing multiple IDS-specific courses and initiating IDS research projects). TICR is based on a foundation of applied epidemiology and biostatistics, and it offers some elective, IDS-oriented courses that predate the IDS curriculum: Clinical Performance and Health Outcome Measurement, Measurement in Clinical Research, Qualitative Research Methods, Decision and Cost-Effectiveness Analysis, Health Disparities Research Methods, and Biomedical Informatics. To this program, we have added five courses specifically geared to IDS training. These courses are summarized below; greater detail and learning objectives are provided in Table 2.
- Introduction to Translating Evidence Into Practice—Theory and Design: Introduces the translation of evidence into practice and policy and includes synopses from the other four courses
- Translating Evidence Into Practice—Individual-Centered Implementation Strategies: Focuses on integrating the components of individual behavior change theories and effective implementation strategies into the design of health care interventions
- Translating Evidence into Practice—System-Centered Implementation Strategies: Introduces theories of organizational behavior and culture in the context of program and policy implementation
- Translating Evidence Into Practice—Community-Engaged Research: Focuses on establishing partnerships with individuals and organizations in specific communities that are targeted for health care interventions
- Translating Evidence Into Policy—Framing Research to Influence Policy: Describes effective strategies for collecting and disseminating research findings to inform and influence the policy-making process
The IDS courses have been integrated into TICR's programs of study, including the two-year master's degree program in clinical research that is intended for advanced predoctoral students, postdoctoral fellows, and faculty members who wish to master clinical research methods and pursue independent research careers. These scholars may enroll in the IDS track, which combines epidemiology and biostatistics courses with a minimum of five courses in IDS theory and methods. (Three of the five must be selected from the IDS-specific courses, and two may be chosen from the IDS-specific courses or the IDS-oriented courses.) Concurrent with their didactic course work, IDS research trainees may opt to be placed with HCDSs (e.g., hospitals, clinics, public health departments) to learn how these community-based organizations prioritize and make decisions about which health care and/or public health interventions to implement. Scholars in the IDS track complete all of the capstone assignments required for the master's in clinical research, including a comprehensive review of the literature on a specific topic, a first-authored oral or poster presentation at a national or international meeting, and a first-authored manuscript submission to a peer-reviewed journal.
We are developing a brief (3–6 months), intensive IDS training experience that is independent of the master's program, in which trainees will learn and apply the principles and skills of IDS through participation in a quality improvement, delivery system innovation, or community health promotion project. These projects will be directed by members of an experienced, multidisciplinary team, and trainees will participate in the project's design, implementation, and evaluation aspects. Through discussions with leaders of HCDSs and public health officers, potential projects will be identified that are high priority to the HCDSs or public health stakeholders and that have high organizational/community commitment and investment.
Our future research will evaluate the impact of these two IDS training programs. We plan to assess the impact of this curriculum on trainees' IDS competencies, as well as the number and type of resulting multidisciplinary collaborations and community partnerships, the number of IDS-related research publications and grant submissions by trainees, and the effectiveness of implementation programs designed and/or coordinated by trainees.
When developing a curriculum, it is essential to identify learners' needs and devise learning objectives and teaching strategies designed to meet these needs. Our conceptual framework for IDS, our design principles for IDS research training, and our interdisciplinary model curriculum represent a break with the dominant educational paradigm in which most courses reside in separate disciplines. Moreover, the competencies we propose here and teach in our IDS courses at UCSF are grounded in interdisciplinary research and team sciences—a departure from the emphasis of standard scientific research training programs. We believe these competencies, along with the conceptual framework that guides them, can serve as the foundation for the development of robust IDS curricula. We encourage other training programs and institutions to use or adapt our design principles, framework, and competencies to examine their existing IDS-related curricula and identify opportunities for growth.
Feedback and comments were graciously provided by Deborah Grady, MD, Jeffrey Martin, MD, MPH, and Chris Ireland from the UCSF Clinical and Translational Sciences Training Program.
Supplemental digital content for this article is available at http://links.lww.com/ACADMED/A75.
1. Lenfant C. Shattuck lecture—Clinical research to clinical practice—Lost in translation? N Engl J Med. 2003;349:868–874.
2. Zerhouni EA. US biomedical research: Basic, translational, and clinical sciences. JAMA. 2005;294:1352–1358.
3. Ogilvie D, Craig P, Griffin S, Macintyre S, Wareham NJ. A translational framework for public health research. BMC Public Health. 2009;9:116.
4. Madon T, Hofman KJ, Kupfer L, Glass RI. Public health. Implementation science. Science. 2007;318:1728–1729.
5. Avorn J, Fischer M. “Bench to behavior”: Translating comparative effectiveness research into improved clinical practice. Health Aff (Millwood). 2010;29:1891–1900.
6. Haynes RB, Hayward RS, Lomas J. Bridges between health care research evidence and clinical practice. J Am Med Inform Assoc. 1995;2:342–350.
7. Davis DA, Taylor-Vaisey A. Translating guidelines into practice. A systematic review of theoretic concepts, practical experience and research evidence in the adoption of clinical practice guidelines. CMAJ. 1997;157:408–416.
9. Sofaer N, Eyal N. The diverse ethics of translational research. Am J Bioeth. 2010;10:19–30.
10. Davis DA, Thomson MA, Oxman AD, Haynes RB. Evidence for the effectiveness of CME. A review of 50 randomized controlled trials. JAMA. 1992;268:1111–1117.
11. Dijkstra R, Wensing M, Thomas R, et al.. The relationship between organisational characteristics and the effects of clinical guidelines on medical performance in hospitals, a meta-analysis. BMC Health Serv Res. 2006;6:53.
15. Nair BR, Finucane PM. Reforming medical education to enhance the management of chronic disease. Med J Aust. 2003;179:257–259.
16. Shortell SM, Kaluzny AD. Health Care Management: Organization Design and Behavior. 5th ed. Albany, NY: Delmar Cengage Learning; 2005.
17. Glanz K, Rimer BK, Viswanath V, eds. Health Behavior and Health Education: Theory, Research, and Practice. 4th ed. San Francisco, Calif: Jossey-Bass; 2008.
18. Prochaska JO, Velicer WF. The transtheoretical model of health behavior change. Am J Health Promot. 1997;12:38–48.
19. Glanz K, Bishop DB. The role of behavioral science theory in development and implementation of public health interventions. Annu Rev Public Health. 2010;31:399–418.
20. Jones L, Wells K. Strategies for academic and clinician engagement in community-participatory partnered research. JAMA. 2007;297:407–410.
21. Berwick DM. Continuous improvement as an ideal in health care. N Engl J Med. 1989;320:53–56.
22. Gavagan T. A systematic review of COPC: Evidence for effectiveness. J Health Care Poor Underserved. 2008;19:963–980.
23. Horowitz CR, Robinson M, Seifer S. Community-based participatory research from the margin to the mainstream: Are researchers prepared? Circulation. 2009;119:2633–2642.
24. Gebbie KM, Meier BM, Bakken S, et al.. Training for interdisciplinary health research: Defining the required competencies. J Allied Health. 2008;37:65–70.
NIH/NCRR 5 KL2 RR024130, UCSF Clinical and Translational Sciences Institute.
Earlier versions of the conceptual framework and curriculum were presented at the second NIH Conference on Dissemination and Implementation Science; January 28–29, 2009; Washington, DC.