Open Access
Translator Disclaimer
June 2011 Analysis of rolling group therapy data using conditionally autoregressive priors
Susan M. Paddock, Sarah B. Hunter, Katherine E. Watkins, Daniel F. McCaffrey
Ann. Appl. Stat. 5(2A): 605-627 (June 2011). DOI: 10.1214/10-AOAS434


Group therapy is a central treatment modality for behavioral health disorders such as alcohol and other drug use (AOD) and depression. Group therapy is often delivered under a rolling (or open) admissions policy, where new clients are continuously enrolled into a group as space permits. Rolling admissions policies result in a complex correlation structure among client outcomes. Despite the ubiquity of rolling admissions in practice, little guidance on the analysis of such data is available. We discuss the limitations of previously proposed approaches in the context of a study that delivered group cognitive behavioral therapy for depression to clients in residential substance abuse treatment. We improve upon previous rolling group analytic approaches by fully modeling the interrelatedness of client depressive symptom scores using a hierarchical Bayesian model that assumes a conditionally autoregressive prior for session-level random effects. We demonstrate improved performance using our method for estimating the variance of model parameters and the enhanced ability to learn about the complex correlation structure among participants in rolling therapy groups. Our approach broadly applies to any group therapy setting where groups have changing client composition. It will lead to more efficient analyses of client-level data and improve the group therapy research community’s ability to understand how the dynamics of rolling groups lead to client outcomes.


Download Citation

Susan M. Paddock. Sarah B. Hunter. Katherine E. Watkins. Daniel F. McCaffrey. "Analysis of rolling group therapy data using conditionally autoregressive priors." Ann. Appl. Stat. 5 (2A) 605 - 627, June 2011.


Published: June 2011
First available in Project Euclid: 13 July 2011

zbMATH: 1223.62163
MathSciNet: MR2840167
Digital Object Identifier: 10.1214/10-AOAS434

Keywords: Bayesian modeling , hierarchical modeling , mental health , multilevel modeling , substance abuse treatment

Rights: Copyright © 2011 Institute of Mathematical Statistics


Vol.5 • No. 2A • June 2011
Back to Top