During a 2.5-year follow-up study of opioid addicts, we found that cocaine abuse had become an increasing and major problem through 1983. Cocaine abuse had only minimally declined during the follow-up period despite treatment, and the number of opioid addicts with at least weekly cocaine abuse had doubled. The clear effect of methadone maintenance treatment in reducing opioid abuse was not evident for cocaine abuse. During the follow-up period, more cocaine use was reported by the methadone-treated subjects than by those undergoing detoxification only. Prognostically, cocaine users were more likely to be nonwhites and men. Subjects who increased their cocaine use during the follow-up period were more likely to have depressive disorders and more likely to be found among methadone- and "drug-free"-treated subjects than among detoxification subjects. Thus, among methadone- and drug-free-treated subjects, depression appeared to be a risk factor for escalating cocaine abuse; this risk factor may benefit from specific interventions.