Free-at-Last is a drug treatment program in East Palo Alto, California, a low-income predominantly African American community in Silicon Valley. In this research, a cohort of treatment clients was compared to a random sample of community residents. Both groups used drugs. Two-way analysis of variance was used to identify factors that predicted the number of drugs used, controlling for client or community sample status. Significant predictors turned out to be perception of race discrimination, ever selling drugs, contact with the police, the number of relatives who died suddenly as a juvenile, ever having thoughts of suicide, and marital status. Path analysis was used to show the relationship between predictors of the number of drugs used for treatment clients; a second path was done for community clients. By comparing each path analysis, we were able to show how treatment clients arrived at significantly higher drug use than peers in the community.