Using Multiple Control Groups to Address Unobserved Biases in Comparative Effectiveness Research

• Activity Number: 384
• Type: Invited
• Date/Time: Tuesday, August 2, 2011 : 2:00 PM to 3:50 PM
• Sponsor: International Chinese Statistical Association
• Author(s): Sharon-Lise Theresa Normand*+ and Frank Yoon and Haiden Huskamp and Alisa Busch
• Companies: Harvard Medical School and Harvard Medical School and Harvard Medical School and McLean Hospital
• Keywords: causal inference ; quasi-experiments ; fine balancing ; propensity scores ; testing in order

Abstract

Studies of policy interventions typically do not involve randomization. Adjustments, such as matching, can remove the bias due to observed covariates, but residual confounding remains a concern. In this paper we introduce two analytical strategies to bolster inferences using observational data. First, we use a design with multiple comparison groups by identifying how the study groups may differ and selecting a second comparison group on this source of difference. Second, we match subjects using a strategy that finely balances the distributions of key categorical covariates and stochastically balance on other covariates. An observational study of the effect of parity on the severely ill subjects enrolled in the Federal Employees Health Benefits (FEHB) Program illustrates our methods. Comparison subjects, who did not receive parity benefits, were enrolled in matched, private plans on the basis of location and plan type. We use a second comparison grroup of FEHB subjects who were followed before parity was enacted in 2001; they are no different than those who received parity benefits, because they are from the same population.