Online Program

Return to main conference page

Friday, January 12
Fri, Jan 12, 10:30 AM - 12:15 PM
Crystal Ballroom B
Instrumental Variables and Treatment Effect Heterogeneity

Individualized Treatment Effects with Censored Data via Fully Nonparametric Bayesian Accelerated Failure Time Models (304271)

*Nicholas Henderson, Sidney Kimmel Comprehensive Cancer Center, Johns Hopkins University 
Thomas A Louis, Bloomberg School of Public Health, Johns Hopkins University 
Gary L Rosner, Sidney Kimmel Comprehensive Cancer Center, Johns Hopkins University 
Ravi Varadhan, Johns Hopkins University 

Keywords: Bayesian Nonparametrics; Ensemble Methods; Heterogeneity of Treatment Effect; Interaction; Personalized Medicine; Subgroup Analysis

Individuals often respond differently to identical treatments, and characterizing such variability in treatment response is an important aim in the practice of personalized medicine. In this article, we describe a non-parametric accelerated failure time model that can be used to analyze heterogeneous treatment effects (HTE) when patient outcomes are time-to-event. By utilizing Bayesian additive regression trees and a mean-constrained Dirichlet process mixture model, our approach offers a flexible model for the regression function while placing few restrictions on the baseline hazard. Our non-parametric method leads to natural estimates of individual treatment effect and has the flexibility to address many major goals of HTE assessment. Moreover, our method requires little user input in terms of tuning parameter selection or subgroup specification. We illustrate the merits of our proposed approach with a detailed analysis of two large clinical trials for the prevention and treatment of congestive heart failure using an angiotensin-converting enzyme inhibitor.