Title: Evaluating risk-adjustment methodologies for patients with mental health and substance abuse disorders in the veterans health administration

Authors: Amy K. Rosen, Cindy L. Christiansen, Maria E. Montez, Susan Loveland, Priti Shokeen, Kevin L. Sloan, Susan L. Ettner

Addresses: Center for Health Quality, Outcomes and Economic Research, Bedford VAMC (152), 200 Springs Road, Bedford, MA 01730, USA. ' Center for Health Quality, Outcomes and Economic Research, Bedford VAMC (152), 200 Springs Road, Bedford, MA 01730, USA. ' Center for Health Quality, Outcomes and Economic Research, Bedford VAMC (152), 200 Springs Road, Bedford, MA 01730, USA. ' Center for Health Quality, Outcomes and Economic Research, Bedford VAMC (152), 200 Springs Road, Bedford, MA 01730, USA. ' Center for Health Quality, Outcomes and Economic Research, Bedford VAMC (152), 200 Springs Road, Bedford, MA 01730, USA. ' VA Puget Sound HCS, Seattle Division (116-DDTP), 1160 South Columbian Way Seattle, WA 98108, USA. ' David Geffen School of Medicine at UCLA, Division of General Internal Medicine and Health Services Research, 911 Broxton Plaza, Room 106, Box 951736, Los Angeles, CA 90095-1736, USA

Abstract: Although the difficulties in applying existing risk-adjustment measures to mental health populations are increasingly evident, the need for adequate risk-adjustment methodologies continues to increase due to the ongoing pressures of the need to constrain costs and allocate resources equitably across key population sub-groups. The performance of several risk-adjustment measures in predicting total healthcare costs and Mental Health/Substance abuse (MH/SA) costs was compared with a national sample of patients with MH/SA disorders receiving healthcare services in the Department of Veterans Affairs (VA). Differences in the performances of the models in predicting both total and MH/SA costs were small; mean absolute prediction errors and predictive ratios did not demonstrate any clear ranking of model performance. Inequitable allocation of resources may result when models that have been developed specifically for general patient populations are applied to unique populations with different healthcare needs.

Keywords: risk adjustment measures; mental health; substance abuse; case-mix; behavioural healthcare; veterans health administration; healthcare costs; resource allocation.

DOI: 10.1504/IJHTM.2006.007872

International Journal of Healthcare Technology and Management, 2006 Vol.7 No.1/2, pp.43 - 81

Published online: 29 Sep 2005 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article