Project H&ART was a randomized intervention trial for homeless alcohol abusers in Albuquerque, N.M. Interventions were four months in duration and included: a high intensity program (case management plus peer-supervised housing), a medium intensity group (peer-supervised housing only); a housed, and a nonhoused control group. Clients were interviewed at baseline and re-interviewed ten months following program entry to determine substance use, housing stability and employment status. Program graduation rates were about 25% for the three housed groups. The outcome evaluation revealed significant within groups improvements in all of the outcomes, no between groups or racial outcome differences, and more favorable alcohol use and housing stability outcomes among program graduates than dropouts. On follow-up, women in the study had fewer days of alcohol use and had more days of stable housing, but were less likely to be employed, compared with men. We suggest that clients' personal motivation for recovery, rather than program-related factors, were most influential in determining outcomes.