Sixty-eight male and female individuals with both DSM-IV diagnoses of cocaine dependence and major depressive disorder were randomly assigned to one of two medication conditions (placebo vs. 40 mg per day) as part of a double-blind, placebo-controlled clinical efficacy trial of fluoxetine for the treatment of this dual diagnosis. During the 12-week outpatient treatment phase all participants also received individual cognitive-behavioral psychotherapy targeting both cocaine use and depression. Depressive symptoms remitted as a function of time in treatment, with no significant medication effects found. Fewer cocaine positive urines were found during the first 6 weeks of treatment in the placebo group compared with the 40-mg group. Cocaine use and depressive symptoms during treatment were significantly correlated. The findings fail to support the role of fluoxetine for treatment of cocaine use and depression in dually-diagnosed patients.