A1462
Title: Causal vaccine effects in the naturally infected
Authors: David Benkeser - Emory University (United States) [presenting]
Abstract: Establishing the long-term effects of interventions aimed at preventing intermediate outcomes poses challenges. For example, vaccines designed to prevent diarrhea caused by Shigella bacteria in children may also positively impact long-term growth, as Shigella-induced diarrhea is a known cause of growth faltering. However, given the relatively low frequency of Shigella-related diarrhea, the vaccine's marginal causal effect on growth may be too small to detect in a randomized controlled trial. Nevertheless, policymakers are interested in demonstrating the broader benefits of vaccination on growth outcomes. To address this challenge, alternative estimands that enjoy improved power for detecting effects on long-term outcomes in realistic trial settings are proposed. Both principal stratification and interventional causal frameworks are used, and both approaches are demonstrated to yield the same identifying functional under different assumptions. Notably, the principal stratification approach relies on cross-world independence assumptions, whereas the interventional estimand does not. Nonparametric efficient, and doubly robust estimators for these estimands are further derived, leveraging machine learning techniques for nuisance parameter estimation. Through realistic simulations, these estimators are shown to provide clinically meaningful inferences even within the constraints of practical Shigella vaccine trials.
