Article Abstract Objective: Olanzapine-fluoxetine combination has shown efficacy in the acute treatment of depressive episodes in patients with bipolar I disorder. The present analyses examined the efficacy and safety of longer term treatment with olanzapine-fluoxetine combination or olanzapine monotherapy in a 6-month open-label extension study. Method: 376 patients with DSM-IV bipolar I disorder, depressed, who completed an acute trial entered the open-label study and received 1 week of olanzapine monotherapy (5-20 mg/day). At all subsequent visits, patients could choose between olanzapine monotherapy or olanzapine-fluoxetine combination (6/25, 6/50, or 12/50 mg/day). Three treatment groups were defined retrospectively according to the medication course taken from week 1: olanzapine, olanzapine-fluoxetine combination, or switched. The efficacy measures were the Montgomery-Asberg Depression Rating Scale (MADRS), Clinical Global Impressions-Bipolar Version, and Young Mania Rating Scale. The study was conducted from July 2000 to May 2002. Results: Among patients who started in remission, MADRS total scores did not change significantly from baseline to endpoint in the olanzapine-fluoxetine combination (0.8) or olanzapine (0.3) groups, but increased slightly in the switched (2.3, p = .02) group. For patients who started in nonremission, MADRS total scores decreased significantly in all groups (olanzapine-fluoxetine combination: - 5.7, p = .001; olanzapine: - 11.6, p = .004; switched: - 6.4, p = .015). The majority of patients who entered the study in nonremission achieved remission (MADRS total score < = 12) during the trial (olanzapine - fluoxetine combination: 66.7%, olanzapine: 64.7%, switched: 62.5%). The overall rate of depressive relapse was 27.4%, and the overall incidence of mania emergence was 5.9%. Conclusions: The present findings suggest that long-term treatment with olanzapine-fluoxetine combination may be a useful option for the management of depressive symptoms and carries a low risk of mania emergence.