OBJECTIVES: This study sought to determine rates of dual disorders (psychiatric and substance use disorders) in a population of low-income inner-city outpatients, to compare the rates in outpatient mental health and substance abuse treatment settings, and to examine the clinical usefulness of classifying patients with dual disorders into three subtypes. METHODS: A total of 57 low-income urban residents receiving mental health treatment and 73 receiving substance abuse treatment were given semistructured clinical interviews to ascertain lifetime and concurrent DSM-III-R axis I disorders. Patients with dual disorders were classified into subtypes depending on whether their psychiatric or substance use disorder was caused by the comorbid disorder or whether both disorders existed independently. RESULTS: Eighty-three patients had a lifetime history of dual disorders: 34 patients (60 percent) in the mental health settings and 49 (67 percent) in substance abuse treatment. Among the 83 with dual disorders, more than half had experienced symptoms of both disorders within the past year. Each of the disorders was considered primary (that is, no indication was found that one was caused by the other) for 24 patients in the mental health settings (71 percent) and 31 in the substance abuse treatment settings (63 percent). CONCLUSIONS: In each type of treatment setting, nearly two-thirds of the patients met criteria for a lifetime diagnosis of a dual disorder. This high rate of comorbidity did not appear to be attributable to substance use causing psychiatric symptoms, or vice versa. The high rate suggests the need for greater integration of mental health and substance abuse treatment, regardless of setting.