Abstract
Policy evaluation is complex, requiring deep knowledge regarding the policies themselves, the contexts in which they are enacted, the degree to which they are implemented, as well as appropriate data and statistical methods for making causal inferences about policy effects on outcomes. High-quality study design is critical in this research, both for investigators to improve validity of causal inference, and for readers and policymakers to understand, trust, and act on the results. Target trial emulation is an approach to designing rigorous non-experimental studies by “emulating” key features of a clinical trial. Most used outside policy contexts, this approach is also valuable for policy evaluation and can address challenges unique to that context. We discuss how using the policy trial emulation framework to conduct and report on research design and methods supports transparent assessment of threats to causal inference in non-experimental studies intended to assess the effect of a health policy on clinical or population health outcomes. We show how careful design thinking lays a foundation for cutting-edge statistical tools to enable rigorous, high-quality policy evaluation studies, with application to a study of the effects of state medical cannabis laws on opioid prescribing for chronic noncancer pain.