Matching Clients to Alcohol Treatments Using 16 Client Characteristics Simultaneously: A Cluster Analytic Approach with Research and Clinical Implications

Abstract

Project MATCH sought to identify client characteristics that could be used to select treatments for specific clients to increase the effectiveness of psychosocial treatments. Results based on examining matching variables one at a time were deemed disappointing by the Project MATCH investigators. In this secondary data analysis, we present analyses examining 16 matching variables simultaneously through cluster analysis in the outpatient arm. While null results were found for the first cluster (n = 380), there was a longer time to first drink for members of the second cluster (n = 275) receiving Twelve Step Facilitation (TSF) compared to clients receiving cognitive behavioral therapy or Motivational Enhancement Therapy (MET). For the third cluster (n = 297) clients had a longer time to first drink if receiving cognitive behavioral therapy or TSF compared to MET. Additional analyses show that these cluster assignments can be adequately approximated by using nine of the 16 matching variables. It is hoped that these results rekindle interest in the benefits of client-treatment matching.