California voters approved the Substance Abuse and Crime Prevention Act (SACPA), also known as "Proposition 36" in November 2000, which allows adults convicted of nonviolent drug possession offenses to participate in drug treatment in lieu of incarceration. The current study drew on data from one of the first cohorts of offenders to receive treatment under SACPA in order to compare their recidivism rate to that of other criminal justice clients. Participants were a group of 3,748 clients admitted to treatment between July 2001 and December 2001, a group of 688 SACPA clients who had received treatment, a group of 1,178 non-SACPA criminal justice clients, and a group of 1,882 clients who entered treatment without a criminal justice status. Data included questionnaires completed by program directors, the Addiction Severity Index Lite questionnaire, and patient treatment participation data from the California Alcohol and Drug Data System, and arrest records from a statewide database. Results of statistical analyses indicated that, when compared with non-criminal justice clients, criminal justice clients (both non-SACPA and SACPA clients) with severe drug problems were significantly less likely to be admitted to residential programs. The analysis further revealed a drug use severity by treatment modality interaction in which high-severity, outpatient clients were most likely to be rearrested on a drug-related charge during the 12 months following treatment admission. Although the likelihood of rearrest decreased for all treatment groups, the SACPA clients were more likely to be rearrested for a drug-related offense after controlling for the interaction between drug use severity and treatment modality. Several strategies are recommended for improving treatment services for California's severely drug addicted offenders. Future research should examine treatment process data in order to improve intervention strategies. Figure, tables, references