Abstract: Aim: to examine the efficacy of two adolescent drug abuse treatments: individual cognitive behavioral therapy (CBT) and multidimensional family therapy (MDFT). Design: A 2 (treatment condition) x 4 (time) repeated-measures intent-to-treat randomized design. Data were gathered at baseline, termination, 6 and 12 months post-termination. Analyses used latent growth curve modeling. Setting: Community-based drug abuse clinic in the northeastern United States. Participants: A total of 224 youth, primarily male (81%), African American (72%), from low-income single-parent homes (58%) with an average age of 15 years were recruited into the study. All youth were drug users, with 75% meeting DSM-IV criteria for cannabis dependence and 13% meeting criteria for abuse. Measurements: Five outcomes were measured: (i) substance use problem severity; (ii) 30-day frequency of cannabis use; (iii) 30-day frequency of alcohol use; (iv) 30-day frequency of other drug use; and (v) 30-day abstinence. Findings: Both treatments produced significant decreases in cannabis consumption and slightly significant reductions in alcohol use, but there were no treatment differences in reducing frequency of cannabis and alcohol use. Significant treatment effects were found favoring MDFT on substance use problem severity, other drug use and minimal use (zero or one occasion of use) of all substances, and these effects continued to 12 months following treatment termination. Conclusion: Both interventions are promising treatments. Consistent with previous controlled trials, MDFT is distinguished by the sustainability of treatment effects.