Title: Bayesian machine learning for causal inference with multiple treatments and multilevel censored survival outcomes
Authors: Liangyuan Hu - Rutgers University (United States) [presenting]
Jiayi Ji - Rutgers University (United States)
Joseph Hogan - Brown University School of Public Health (United States)
Abstract: Despite numerous recent advances in causal inference, the literature for handling data with multiple treatments and multilevel censored survival outcomes is sparse. Here we develop a way to use Bayesian Additive Regression Trees, a likelihood-based machine learning modeling technique, to draw causal inferences about the effects of multiple treatments for clustered observational survival data. This approach will provide substantial modeling flexibility for a data structure for which few off-the-shelf causal inference methods are available. We further develop a flexible and interpretable sensitivity analysis framework to handle the no unmeasured confounding assumption, respecting the multilevel survival data structure. Our approach addresses unmeasured confounding at both cluster and individual levels and incorporates uncertainty about unidentified model components formally into the analysis. The operating characteristics of our proposed method are examined via an extensive simulation. We demonstrate the developed methods via a case study evaluating the survival effects of three popular types of treatments for high risk localized prostate cancer using the national cancer database.