A0571
Title: Nonparametric sensitivity analysis for unobserved confounding with survival outcomes
Authors: Rui Hu - Shenzhen Technology University (China) [presenting]
Ted Westling - University of Massachusetts Amherst (United States)
Abstract: In observational studies, the observed association between an exposure and outcome of interest may be distorted by unobserved confounding. Causal sensitivity analysis can be used to assess the robustness of observed associations to potential unobserved confounding. For time-to-event outcomes, existing sensitivity analysis methods rely on parametric assumptions on the structure of the unobserved confounders and Cox proportional hazards models for the outcome regression. If these assumptions fail to hold, it is unclear whether the conclusions of the sensitivity analysis remain valid. Additionally, causal interpretation of the hazard ratio is challenging. To address these limitations, a nonparametric sensitivity analysis framework is developed for time-to-event data. Specifically, nonparametric bounds are derived for the difference between the observed and counterfactual survival curves and propose estimators and inference for these bounds using semiparametric efficiency theory. Nonparametric bounds and inference are also provided for the difference between the observed and counterfactual restricted mean survival times. The performance of the proposed methods is demonstrated using numerical studies and an analysis of the causal effect of physical activity on respiratory disease mortality among former smokers.
