BACKGROUND: Crack cocaine use undermines adherence to highly active antiretroviral therapy (HAART). This pilot randomized clinical trial tested the feasibility and efficacy of 2 interventions based on the Information-Motivation-Behavioral Skill model to improve HAART adherence and reduce crack cocaine problems. METHODS: Participants were 54 adults with crack cocaine use and HIV with <90% HAART adherence. Most participants were African-American (82%) heterosexual (59%), and crack cocaine dependent (92%). Average adherence was 58% in the past 2 weeks. Average viral loads (VL) were detectable (logVL 2.97). The interventions included 6 sessions of Motivational Interviewing plus feedback and skills building (MI+), or Video information plus debriefing (Video+) over 8 weeks. Primary outcomes were adherence by 14-day timeline follow-back and Addiction Severity Index (ASI) Drug Composite Scores at 3 and 6 months. Repeated measure ANOVA assessed main effects of the interventions and interactions by condition. RESULTS: Significant increases in adherence and reductions in ASI Drug Composite Scores occurred in both conditions by 3 months and were maintained at 6 months, representing medium effect sizes. No between group differences were observed. No VL changes were observed in either group. Treatment credibility, retention, and satisfaction were high and not different by condition. CONCLUSIONS: A counseling and a video intervention both improved adherence and drug problems durably among people with crack cocaine use and poor adherence in this pilot study. The interventions should be tested further among drug users with poor adherence. Video interventions may be feasible and scalable for people with HIV and drug use.