BACKGROUND: This study examines the effects of an empirically supported psychosocial treatment, discriminative behavior therapy (DBT), combined with the selective serotonin reuptake inhibitor (SSRI) fluoxetine, for the treatment of BPD. METHODS: This was a randomized, double-blind, placebo-controlled study of patients with BPD identified by the DSM-IV axis II disorder SCID lasting 12 weeks. All individuals received individual and group DBT. of the 20 individuals who completed treatment, 9 were randomly assigned to receive up to 40 mg daily doses of fluoxetine and 11 were randomly assigned to receive placebo. Individuals were assessed at baseline and at 10 and 11 weeks based on self-reported depression, anxiety, anger expression, schizophrenia, and gross functioning. This study was conducted between January 1998 and February 2000. RESULTS: No significant group differences were seen between any of the pre- and post-treatment measures for the time-group interaction effect. However, within the DBT/placebo subgroup, there was a significant pre/post-treatment difference in the direction of progression based on all measures. Within the DBT/fluoxetine subgroup, no significant pre/post-treatment differences were seen. CONCLUSION: The results suggest that the addition of fluoxetine into effective psychosocial treatment does not provide any additional benefit. Further studies with larger samples are pending.