Case-mix Adjustment for Comparing State Substance Treatment Outcomes

• Activity Number: 199
• Type: Contributed
• Date/Time: Tuesday, August 10, 2004 : 8:30 AM to 10:20 AM
• Sponsor: Social Statistics Section
• Author(s): Gary Huang*+ and Sameena M. Salvucci and Dhirendra Ghosh and Ping Yu and Sue Becker
• Companies: Synectics for Management Decisions, Inc. and Synectics for Management Decisions, Inc. and Synectics for Management Decisions, Inc. and Synectics for Management Decisions, Inc. and Substance Abuse and Mental Health Services Administration
• Keywords: length of stay ; retention ; discharge rate

Abstract

Case-mix adjustment is used in healthcare cost studies and program evaluation to assure comparability across providers. The procedure, however, has not been widely used in substance abuse treatment research. We use discharge data collected in the 2000 Treatment Episode Data Set (TEDS) to explore risk variables that are critical for case-mix adjustment in comparing treatment outcomes (length of stay and planned discharge) across states. In addition to sex and age, the case-mix adjustment models specify and test a number of measures of risk factors, including type of substance used, years of using the substance, and co-occurring disorders. The case-mix adjusted average length of stay and planned discharge rates will be presented state by state. The results of the study will contribute to knowledge about measures of risk factors and statistical techniques specifically relevant to the case-mix procedures used in studying substance abuse treatment outcomes across organizations at levels above the local provider level.