Accounting for Uncertainty in Confounder and Effect Modifier Selection When Estimating Average Causal Effects in Generalized Linear Models

• Activity Number: 599
• Type: Topic Contributed
• Date/Time: Wednesday, August 3, 2016 : 2:00 PM to 3:50 PM
• Sponsor: Biometrics Section
• Author(s): Chi Wang* and Francesca Dominici and Giovanni Parmigiani and Corwin Zigler
• Companies: University of Kentucky and Harvard T.H. Chan School of Public Health and Dana-Farber Cancer Institute/Harvard T.H. Chan School of Public Health and Harvard T.H. Chan School of Public Health
• Keywords: Average causal effect ; Bayesian adjustment for confounding ; Confounder selection ; Treatment effect heterogeneity

Abstract

Confounder selection and adjustment are essential elements of assessing the causal effect of an exposure or treatment in observational studies. We propose and evaluate a Bayesian method to estimate average causal effects in studies with a large number of potential confounders, relatively few observations, likely interactions between confounders and the exposure of interest, and uncertainty on which confounders and interaction terms should be included. Our method is applicable across all exposures and outcomes that can be handled through generalized linear models. In this general setting, estimation of the average causal effect is different from estimation of the exposure coefficient in the outcome model due to non-collapsibility. We implement a Bayesian bootstrap procedure to integrate over the distribution of potential confounders and to estimate the causal effect. Our method permits estimation of both the overall population causal effect and effects in specified subpopulations, providing clear characterization of heterogeneous exposure effects that may vary considerably across different covariate profiles. Method application is demonstrated using the US Medicare brain tumor data.