West Coast Ballroom
Evaluation of health policy interventions for contagious outcomes (307864)
Forrest W. Crawford, Yale UniversityDaniel J. Eck, Yale University
*Olga Morozova, Yale University
Keywords: cost-effectiveness analysis, causal effects, contagion, infectious disease, vaccine trial
Evaluation of population-level impact of health policy interventions relies on the unbiased estimation of individual-level causal effects. Under contagion an individual’s outcome may be influenced by other people’s treatments, outcomes, or both, complicating the definition and estimation of causal effects. Randomized trials in clusters of interacting individuals are often used to estimate causal effects under contagion. A widely accepted approach to evaluate the individualistic effect of an intervention on the person who receives it defines the “direct effect” as the contrast between the potential outcomes under treatment and no treatment, averaged over the conditional distribution of possible treatment assignments to other individuals under a given randomization scheme. Assuming individualistic causal interpretation, health policy researchers use the estimates of direct effect from such trials as a measure of effectiveness in cost-effectiveness analysis or to parameterize simulation models. We show that under contagion, some randomization designs ensure that the direct effect cannot have an individualistic interpretation and can be misleading when used in health policy evaluation.