Generalizing about Public Health Interventions: A Mixed-Methods Approach to External Validity

Abstract

Public health researchers and practitioners are calling for greater focus on external validity, the ability to generalize findings of evidence-based interventions (EBIs) beyond the limited number of studies testing effectiveness. For public health, the goal is applicability: to translate, disseminate, and implement EBIs for an impact on population health. This article is a review of methods and how they might be combined to better assess external validity. The methods include (a) better description of EBIs and their contexts; (b) combining of statistical tools and logic to draw inferences about study samples; (c) sharper definition of the theory behind the intervention and core intervention components; and (d) more systematic consultation of practitioners. For population impact, studies should focus on context features that are likely to be both important (based on program theory) and frequently encountered by practitioners. Mixed-method programs of research will allow public health to expand causal generalizations.