Online Program Home
My Program

Abstract Details

Activity Number: 407 - Novel Methods for Causal Inference in Health Policy
Type: Contributed
Date/Time: Tuesday, July 30, 2019 : 2:00 PM to 3:50 PM
Sponsor: Health Policy Statistics Section
Abstract #304928
Title: Causal Inference Under Interference in Dynamic Therapy Group Studies
Author(s): Susan Paddock* and Bing Han and Lane Burgette
Companies: NORC at the University of Chicago and RAND Corporation and RAND Corporation
Keywords: causal inference; interference; prognostic score

Group cognitive behavioral therapy is a common treatment modality for behavioral health conditions. Patients enter and exit therapy groups on an ongoing basis, leading to a type of dynamic group. We apply the Rubin Causal Model framework to define the causal effect of high versus low session attendance of group therapy at both the individual patient and peer levels. There are several challenges to address to identify causal effects in this setting, such as interference, the interrelatedness of dynamic group participation, and the observational nature of the data. The dynamic therapy group setting motivates a unique causal inference scenario, as the treatment statuses are completely defined by the structure inducing interference, i.e., the attendance record of all therapy sessions. We propose a strategy to identify individual, peer, and total effects of attendance on patient outcomes, and balance the observed treatment status groups on prognostic score strata.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2019 program