Although it is becoming increasingly more common for clinicians to use genomic data in their practices for disease prevention, diagnosis, and treatment, the process of integrating genomic data into the practice of medicine has been a slow and challenging one. Some of the major barriers impeding the incorporation of new genomic technology into clinical practice are: the difficulty of changing routine medical practices to account for the use of genetic testing, the limited knowledge of patients and providers about genomic medicine, assessing sufficient evidence to support the use of genetic tests, privacy and data security issues, and uncertainty about reimbursement (Manolio et al., 2013).
Genomic medicine programs are currently under way at several academic medical centers and large integrated health systems (Manolio et al., 2013), but it has been challenging to identify which genomic applications have robust evidence supporting their use in the clinic to improve patient outcomes (Dotson et al., 2014). With a constantly evolving evidence base, it is not unexpected that, even with the integration of successful applications, the collection of evidence could continue during genetic test implementation (see Chapter 4). Furthermore, the incorporation of genomic approaches into clinical care is taking place independently at medical centers throughout the country, and practices and health systems could benefit from structured collaboration, knowledge sharing, and an implementation
1The planning committee’s role was limited to planning the workshop. The workshop summary has been prepared by the rapporteurs as a factual account of what occurred at the workshop. Statements, recommendations, and opinions expressed are those of individual presenters and participants and are not necessarily endorsed or verified by the National Academies of Sciences, Engineering, and Medicine. They should not be construed as reflecting any group consensus.
framework to improve the efficiency and effectiveness of practice change.
The field of implementation science may be able to provide insights concerning efficient ways to incorporate genomic applications into routine clinical practice. One definition of implementation science is “the study of methods to promote the integration of research findings and evidence into health care policy and practice”2 (NIH, 2016). The focus of implementation science studies is to identify integration bottlenecks and optimal approaches for a given setting and ultimately to promote the uptake of research findings (WHO, 2014). The overarching goal of the field is to create generalizable knowledge that can be applied across settings to achieve sustained health improvements (Madon et al., 2007). Implementation research can be applied to a variety of fields and issues pertaining to education, civil and criminal justice, social welfare, and child welfare. The tools and approaches of implementation science could help make it possible to more efficiently incorporate genomic advances into common clinical practice.
To explore the potential of implementation science to improve the integration of genomics into medicine, the Roundtable on Translating Genomic-Based Research for Health held a workshop in Washington, DC, on November 19, 2015, titled Applying an Implementation Science Approach to Genomic Medicine.3 The workshop brought implementation scientists together with clinicians, payers, community engagement experts, and health systems leaders who have an interest in genomic medicine. It has been difficult to bridge the gap between discoveries in genomics and positive population health outcomes, observed workshop co-chair Greg Feero, associate editor of the Journal of the American Medical Association, and implementation science may offer an opportunity to close that gap and provide cumulative knowledge that can be adopted and adapted so that “institutions are not forced to reinvent the wheel at each site every time.”
During the workshop, speakers and participants discussed the challenges and opportunities of integrating genomic advances into the clinic through the lens of implementation science, and by doing so they are “ready to talk about implementation in a new way,” according to Roundtable co-chair Sharon Terry, president and chief executive officer of Genetic Alliance. The specific workshop objectives are provided in Box 1-1.
2See http://www.fic.nih.gov/researchtopics/pages/implementationscience.aspx (accessed February 26, 2016).
3The workshop agenda, speaker biographical sketches, statement of task, and registered attendees can be found in Appendixes A, B, C, and D, respectively.
Implementation research is designed to “generate insights and knowledge about implementation processes, barriers, facilitators, and strategies,” explained Brian Mittman, a research scientist at Kaiser Permanente Research. Workshop speakers examined issues that pertain to advancing genomic medicine, including engaging diverse audiences in genomic medicine, gathering evidence during implementation, and using genomic medicine to improve population health. The underlying question throughout the workshop was, “How can implementation science help to address these challenges?”
Following each session’s presentations, a panel of five reactants offered reflections on and extensions of the presenters’ comments from the perspectives of a clinician, a payer, an implementation scientist, a patient advocate, and a health disparity expert. The reactants’ remarks are incorporated throughout this summary to recap some of the most important themes emerging from the workshop.
Engaging Large and Diverse Audiences
One major issue concerning the implementation of genomic medicine is determining which methods will encourage widespread participation from minority or disadvantaged populations. Genome-wide association studies (GWASs) have expanded the understanding of a broad spectrum of human traits and diseases, but many of these studies include relatively small numbers of samples derived from minority groups (Coram et al.,
2015). Small sample sizes from minority populations could result in gaps in the evidence base used in genomic research—and subsequent inequities in clinical care.
Systematic efforts are under way to engage diverse populations in genomics research by enhancing communication and building trust (NHGRI, 2016). Such engagement can help ensure that the implementation of genomic medicine fits the needs of specific populations and also fits within the local context (Joosten et al., 2015). Introducing genomic medicine at both large academic health centers and small community-based practices will be important for reaching diverse patient populations. Community engagement can help reveal gaps in an implementation strategy and provide critical evidence needed to fill those gaps (see Chapter 3). In addition to community engagement, consistent financing strategies and efforts to improve genomic literacy among multiple groups (including clinicians, payers, and patients) could all promote the inclusion of minority and underserved populations in genomic research and medicine.
Gathering Evidence During Implementation
Synthesizing robust evidence is another issue that is important for the integration of genomic innovations into clinical practice. Lengthy time periods between discovery and clinical uptake can be attributed, in part, to a linear research approach which encourages stepwise progression from evidence building to clinical research to implementation research (Glasgow et al., 2003). However, evidence building and clinical research can be done in parallel with implementation research, in what is referred to as hybrid effectiveness-implementation studies (see Chapter 4). The examination of case studies of rapid and successful implementation of genomic applications could yield valuable information about how to optimize the uptake of new genomic medicine approaches.
Implementation Science and Population Health
The integration of genomics into clinical care has the potential to improve public health at the population level, to expand our understanding of human diseases, and to increase genomic literacy. Programs run by state health departments, such as the Public Health Genomics Pro-
gram in Michigan,4 can be examples of how the adoption of genomic applications has the potential to affect population health (see Chapter 5). Certain genomic applications, if implemented in the proper subset of the population, have the ability to save lives, prevent disease, or improve the quality of life for many patients (Green et al., 2015). Implementation science may help to build a common framework and best practices for integrating evidence-based genomic applications into population health programs. Designing a framework and identifying best practices for implementation involving a diverse group of stakeholders could result in a successful plan for incorporating genomics into practice.
Immediately following this introductory chapter, Chapter 2 provides an introduction to implementation science research, with a special focus on its goals, methods, and approaches. The chapter also includes background on the distinctions that set implementation science apart from related fields. Attention is paid to the barriers to implementation, evidence generation during implementation, and gaps in implementation research as they relate to genomic medicine.
Chapter 3 examines specific methods for engaging a large and diverse patient population in the early stages of implementation, where information gathered can be useful for genetics discovery efforts. The case studies of implementation presented in Chapter 3 involve a large regional health care system, a genomic research network in Québec, and a program aimed at enrolling greater numbers of underrepresented minorities into research. This chapter also addresses the issue of identifying the most appropriate time to introduce implementation science into the translation process.
Chapter 4 explores models for gathering evidence as a new technology or practice is being introduced into routine care. The evidence gathered during implementation can include measures of the knowledge and skills of providers, patient acceptance, external incentives, and health outcomes. This chapter features a case study on a proprietary program for health care providers that offers easily accessible information on cancer treatments and clinical trials. In addition, Chapter 4 examines the rap-
4For more information on the Public Health Genomics Program in Michigan, see http://www.michigan.gov/mdhhs/0,5885,7-339-73971_4911_4916_47257---,00.html (accessed February 23, 2016).
id adoption of a novel genomic technology and the benefits and disadvantages of its speedy implementation.
Chapter 5 focuses on effective strategies and infrastructure that can facilitate the implementation of genomic medicine equitably across the population. This chapter describes a multifaceted statewide genetics plan as well as a pan-Canadian biobank that is in the process of expanding from longitudinal cancer studies to other disease fields. A bottom-up approach to improving diabetes diagnoses is also presented, highlighting the value of evidence gathering in strengthening the implementation process.
In Chapter 6, the value of using implementation science in genomics is considered, particularly as it relates to addressing health disparities, improving genomic literacy, and financing genetic approaches in clinical care. Potential ideas from individual speakers for actionable next steps are laid out in Chapter 6.