Twenty-nine cocaine-dependent male veterans without other drug dependence completed a double-blind controlled, randomly-assigned study examining the efficacy of bromocriptine versus placebo in the management of cocaine abstinence symptomatology. Serum prolactin (PL) and growth hormone (GH) levels were obtained prior to and after the study was completed. Patients were seen daily and completed several self-report questionnaires, including the Symptom Checklist-90-Revised, the Beck Depression Inventory, and a Cocaine Craving Report. The patients were also asked to rate a variety of cocaine withdrawal symptoms. Overall, there did not appear to be any advantage to receiving bromocriptine versus placebo during the first 3 weeks following cocaine use cessation with the possible exception of changes in activity and appetite level. The placebo group showed a statistically significant increase in activity level during the first week in treatment and a significant increase in appetite throughout the study. Patients in both groups showed significant improvement in the other areas assessed, with improvement appearing to progress according to length of treatment. Hyperprolactinemia or abnormal GH levels were not found in this patient sample as a group. Thirty-four of the original 63 patients dropped out of the study. Seventeen received bromocriptine, and 17 received placebo. There was no significant difference between drug groups in incidence of retaining patients in treatment. The high dropout rate may reflect the difficulty incurred in retaining cocaine-dependent patients in treatment.