If You Build It Will They Come? The Urgent Need for Implementation Science in Genomic Medicine

As the evidence base for genomic medicine grows, evaluating and implementing genomics to improve health care and population health remains an ongoing challenge. There is no guarantee that “if we build it they will come,” paraphrasing the well-known expression from the movie, Field of Dreams. To improve adoption, uptake and sustainability of genomic medicine applications, and for translation to population health impact, implementation science (IS) must be an integral component of genomics research. This requires formulating implementation science questions based on identified knowledge gaps, and using appropriate instruments (theories, models, and/or frameworks), strategies and outcome measures as part of the research process.

In a recently published systematic review of the literature on IS in translational genomic medicine research, Roberts et al found many scientific gaps and areas requiring improvement, including

• Use of IS frameworks, models, and/or theories,
• Expansion to diverse patient populations and health care settings,
• Incorporation of experimental study designs,
• Incorporation of collaborative processes (i.e. engagement of multilevel key stakeholders), and
• Evaluation of the sustainability of genomic medicine and related interventions.

Similar findings were identified in our recent published review of the NIH portfolio of awarded grants on IS in genomic medicine. Only 42 NIH grants were funded over a 5 year period (~1.75% of investigator-initiated genomics grants). The majority of grants proposed cross-sectional study designs, with clinical settings in primarily white, non-Hispanic study populations. Most grants were in cancer genetic testing for risk assessment.

Moving forward, we should focus on broad priority areas at the intersection of IS and genomic medicine. Increasing awareness of the role of IS in the success of genomics is a first step. While building an evidence base around genomic medicine remains a critical area, it is not sufficient for successful implementation. It is equally important to understand what factors contribute to the success or failure of a genomic applications within a particular setting, and adapt an application to other settings to achieve long term sustainability.

The second step is to formulate pertinent IS questions in genomic medicine. These questions extend beyond whether a genomic application is effective by asking why or why not an application is effective in a given setting. For example, if a universal tumor testing program for Lynch syndrome identification is successful, without incorporating implementation science methods and measures into the study, we are unable to determine the reason for its success. Was the application successful due to a provider educational strategy, a system-level electronic health record reminder tool, or another factor? Aligning IS questions with implementation outcomes will allow us to identify and measure which factors lead to successful implementation and which provide insights into future implementation in genomic medicine efforts.

The third step is to bring together a multidisciplinary team with expertise in implementation science, clinical genetics and other clinical or public health specialty areas, statisticians and epidemiologists, behavioral and social scientists, and economists. After honing the research questions, key elements for a successful study include using appropriate frameworks, models or theories, identifying a strategy or strategies for the implementation process, and identifying appropriate outcome measures (see recent references for IS frameworks and metrics here and here).

The National Institutes of Health’s funding specific to Dissemination and Implementation Research in Health has identified IS within genomic medicine as an important area of research (PAR-16-237, 18-007, and -017). Specifically, funding announcements suggest that research directions in the field should include the following:

Studies of the dissemination and implementation of effective and cost-effective strategies for incorporating genomic medicine, sequence-based diagnostics and therapeutics in clinical care.

Studies testing the implementation and use of genomic information, family history risk information, and/or pharmacogenetic information for improved diagnosis and treatment.

The time is right for a growing intersection of genomic medicine and implementation science. Training opportunities in implementation science are available (e.g., Society for Implementation Research 2017). These opportunities range from in-person to online courses, and from short programs (1-2 days) to full academic programs. As the awareness and interest by the genomic medicine community builds, the opportunity for offering IS training as an adjunct to a genomic-related conference can be considered, for example during the annual meetings of the American College of Medical Genetics and Genomics and the National Society of Genetic Counselors. By integrating implementation science and translational genomic research, we will have the opportunity to fulfill the promise of genomics to benefit individual and population health.