A Semiparametric Bayesian Approach to an Instrumental Variable Model with Right-Censored Time-to-Event Outcome

• Activity Number: 128
• Type: Contributed
• Date/Time: Monday, August 5, 2013 : 8:30 AM to 10:20 AM
• Sponsor: Biometrics Section
• Author(s): Xuyang Lu*+ and Gang Li
• Companies: University of California, Los Angeles and University of California at Los Angeles
• Keywords: instrumental variable analysis ; time-to-event outcome ; semiparametric Bayesian model ; Dirichlet process mixtures ; Mendelian Randomization ; accelerated failure time model

Abstract

The method of instrumental variable (IV) analysis is widely used in economics, epidemiology, and other fields to estimate the causal effects of intermediate covariates on outcomes, in the presence of unobserved confounders or measurement errors in covariates. However, IV methods for a time-to-event outcome with censored data remain underdeveloped. We extend the IV method to time-to-event outcome for linear models with right censored data. We propose a semiparametric Bayesian model with Dirichlet process mixtures for the random errors, in order to relax the parametric assumptions and address heterogeneous clustering problems. A Markov Chain Monte Carlo sampling method is developed for the endogenous variable parameters. Performance of our method is examined by simulation studies. We illustrate our method on the Women's Health Initiative Observational Study and the Atherosclerosis Risk in Communities Study.