Pragmatic Randomized Trials

Public health systems studies on the effectiveness and efficiency of programs, policies, and implementation strategies can be designed to integrate randomized trials in pragmatic ways. Although random assignment methods are the gold standard for strong internal validity and determining causality in studies, public health research designs often avoid randomized trials due to study time and cost concerns as well as logistical, legal or ethical barriers. Pragmatic trials can take advantage of real-world settings and reflect realistic policy and program choices and alternatives, and randomly assign research subjects and/or settings to an intervention or control group.  An increasing number of studies on the health and economic impacts of public health programs and population health promotion and prevention delivery system strategies are incorporating randomized evaluations and demonstrating how the results of such randomized impact evaluations can inform policy.

Building on the documented association between successful adoption of a public health policy and barriers within the states’ discretion, pragmatic trial designs can be applied to evaluating state and local public health strategies. Opportunities exist to use state and local public health data sources in evaluating key decisions in practice and policy implementation. Public health system data, gleaned from active and passive surveillance systems and program reporting requirements, have potential including types and levels of staffing, location, target population recruitment and dissemination methods, modifications for specific subgroup populations, implementation timing, and financing and payment mechanisms. Related resources and public health services and systems research projects are listed below.