Effectiveness of Biological Drug Testing Among Adolescent Substance Users: A Multiple Group Propensity Score Analysis

• Activity Number: 236
• Type: Topic Contributed
• Date/Time: Monday, August 5, 2013 : 2:00 PM to 3:50 PM
• Sponsor: Health Policy Statistics Section
• Author(s): Megan Schuler*+ and Beth Ann Griffin and Rajeev Ramchand and Daniel Almirall and Daniel McCaffrey
• Companies: Johns Hopkins Bloomberg School of Public Health and RAND and RAND and University of Michigan and RAND
• Keywords: propensity score ; substance abuse

Abstract

This talk will present an overview of a method to obtain average treatment effect estimates via propensity score weighting when there are more than 2 treatment groups. An application comparing effectiveness, relative to no treatment, of 3 treatment modalities (biological drug screening (BDS) only, Motivational Enhancement Therapy-Cognitive Behavioral Therapy (MET/CBT), and MET/CBT combined with BDS) for adolescent substance abuse will be discussed. Data are from a multisite longitudinal observational study of 5186 adolescents. Youth were classified into four treatment groups based on self-reports of received services. Outcomes of interest were measured with the GAIN survey's Substance Frequency Scale and Substance Problems Scale. Weights were generated from propensity scores estimated by generalized boosted models; weighting resulted in comparable groups. Weighted regression analyses estimated the impact of treatment on each outcome. Overall, biological drug screening was associated with a significant decrease in both substance use frequency and problems through 12 months of follow-up. Combining BDS and MET-CBT does not appear to result in more favorable outcomes than BDS alone.