The sample consisted of the first 1,088 cases in the ongoing Persistent Effects of Treatment Study for Adolescents (PETS-A), a longitudinal study that is examining the psychosocial outcomes for youths admitted to 1 of 7 geographically dispersed treatment centers. The youths in the sample entered 1 of 4 outpatient (n=511) or 3 residential (n=577) drug treatment programs between 1998 and 2001. The core intake and follow-up participant interviews used at all sites consisted of the Global Appraisal of Individual Needs, a standardized clinical assessment. Independent variables measured were gender and ethnicity, current substance abuse treatment, treatment setting, legal system contact, baseline mental health severity, and baseline physical health severity. Dependent variables were mental health problems, physical health problems, mental health service use, and medical service use. High levels of mental health problems were found at both intake and follow-up, but few received mental health treatment. There were lower rates of medical problems, and more than half received services. Females in residential settings with more current and baseline distress were more likely to receive services. Ethnicity, baseline behavioral problems, and whether or not currently in substance abuse treatment did not predict service use. Females in residential treatment were also more likely to receive medical treatment. Although these findings require replication and validation, they suggest that more could be done in the context of substance abuse treatment to link youths with mental health services. 4 tables and 18 references