Quasiexperimental Outcome Studies with Adolescents Suffering From Substance Use Disorders
This symposium begins with a brief primer on propensity score matching, an advanced observational study procedure that, when key assumptions are met, can be used to infer causality. In this primer, we describe this analytic model’s assumptions, procedures, and limitations. Resources are provided to participants that are interested in learning more about the propensity score matching. We then present three propensity score matching studies that address pressing questions, including: whether youth receiving residential treatment have superior outcomes compared to matched controls receiving outpatient treatments, whether African-American youth have superior substance use outcomes when exposed to an optional drug refusal skills training procedure, and whether youth receiving a promising new treatment at one site have equivalent outcomes to matched controls from other sites. Notwithstanding limitations, these studies assist policy makers’ and practitioners’ decisions about how to best direct scarce resources to maximize the public health benefits of adolescent substance abuse treatments.