Bayesian Estimation of Average Causal Effect with Adjustment for Confounding

• Activity Number: 114
• Type: Topic Contributed
• Date/Time: Monday, August 5, 2013 : 8:30 AM to 10:20 AM
• Sponsor: Health Policy Statistics Section
• Author(s): Chi Wang*+ and Giovanni Parmigiani and Francesca Dominici
• Companies: University of Kentucky and Dana-Farber Cancer Institute and Harvard School of Public Health
• Keywords: causal inference ; Bayesian model averaging ; propensity score

Abstract

Estimating the average causal effect of an exposure on an outcome is a common goal in many observational studies. One challenging issue is how to properly adjust for confounding factors. In this talk, we present a Bayesian approach to identify and adjust for confounders. Our work extends the Bayesian adjustment for confounding (BAC) method to the framework of generalized linear models, which allow for arbitrary types of exposure and outcome. Our method also allows for the inclusion and selection of interactions between exposure and confounders. In simulation studies, we compare our proposed method with the propensity score-based methods.