Two different methods for assessing psychopathology in opiate addicts were compared as predictors of long-term treatment outcome: (1) categorical psychiatric diagnosis using the Schedule for Affective Disorders and Schizophrenia--Lifetime Version and the Research Diagnostic Criteria and (2) global rating of psychiatric impairment using the Psychiatric Severity scale of the Addiction Severity Index (ASI). Follow-up interviews were completed 2.5 years after treatment seeking in 76% of a sample of 361 opiate addicts. Five dimensions of treatment outcome were assessed, including current functioning, psychosocial adjustment, substance use impairment, legal problems, and medical disability. Most lifetime psychiatric disorders with a prevalence of greater than 10% were significantly related to the outcome dimensions of current functioning and/or psychosocial adjustment and were unrelated to substance use impairment, legal problems, and medical disability. The ASI Psychiatric Severity rating more robustly predicted poorer functioning in the same two areas and less severe legal problems. While controlling for ASI Psychiatric Severity, the only Research Diagnostic Criteria diagnosis that remained significantly related to treatment outcome was major depression, suggesting that, as regards their prognostic characteristics, the other diagnoses are accounted for by a global underlying severity dimension.