McGinty EE, Seewald NJ, Bandara S, et al. Scaling Interventions to Manage Chronic Disease: Innovative Methods at the Intersection of Health Policy Research and Implementation Science. Prev Sci. 2024;25(Suppl 1):96-108. doi:10.1007/s11121-022-01427-8
This article emphasizes the need to study policy implementation better. The authors suggest that more research using existing implementation methodologies is necessary when scaling evidence-based interventions. Using the ACA Medicaid waiver behavioral health home context, which seeks to aid individuals with serious mental illness to better navigate the US health system, as a case example, this article details the consensus of an expert forum on methodological considerations when studying policy implementation. The forum aimed to answer the primary question, what research methods can be used to examine which policy implementation structures and strategies are needed to achieve policy goals? Three main methodological approaches to studying policy implementation were recorded. The first is methods that aim to describe or document. These methods include qualitative research, survey research, document review, case studies, descriptive analyses of implementation, and dimension reduction approaches, e.g., latent class analysis. These methods characterize policy implementation and guide hypothesis development and future analysis. The second category is methods that aim to predict or understand. This includes regression approaches, machine learning approaches, and systems science methods. These methods aim to determine which implementation strategies and/or structures predict or relate to policy outcomes. They cannot be used to make causal inferences. To assess causal inferences researchers should use methods that examine causal links, like randomized experiments, and nonexperimental approaches e.g., difference-in-differences, or configurational analysis. The additional methods suggested by the expert forum include effect modification methods that aim to determine if the policy implementation structures and strategies alter the policy outcomes, causal mediation approaches to address implementation mechanisms, and methods to characterize uncertainty in science systems, using systems models. Overall, the forum called for greater use of existing methodologies to better bridge implementation science and effective policy outcomes.