Regularized Outcome Weighted Subgroup Identification for Differential Treatment Effects

• Activity Number: 372
• Type: Contributed
• Date/Time: Tuesday, August 11, 2015 : 10:30 AM to 12:20 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Yaoyao Xu* and Menggang Yu and Yingqi Zhao and Quefeng Li and Sijian Wang and Jun Shao
• Companies: AbbVie and University of Wisconsin - Madison and University of Wisconsin - Madison and Princeton University and University of Wisconsin - Madison and University of Wisconsin - Madison
• Keywords: Comparative effectiveness ; Regularization ; Subgroup ; Heterogeneity of treatment effectiveness ; Variable selection

Abstract

To facilitate comparative treatment selection when there is substantial heterogeneity of treatment effectiveness, it is important to identify subgroups that exhibit differential treatment effects. Existing approaches model outcomes directly and define subgroups according to interactions between treatment and covariates. Because outcomes are affected by both covariate-treatment interactions and covariate main effects, direct modeling can be hard due to model misspecification. Alternatively one can directly work with differential treatment effect estimation. We propose such a method that approximates a target function whose value directly reflects correct treatment assignment. The function uses patient outcomes as weights rather than modeling targets. Consequently, our method can deal with binary, continuous, time-to-event, and possibly contaminated outcomes in the same fashion. We first focus on identifying only directional estimates from linear rules that characterize important subgroups. We further consider estimation of comparative treatment effects for identified subgroups. We demonstrate the advantages of our method in simulation studies and in analyses of real data sets.