- Background: The first episode of an illness may respond differently to any treatment compared to multiple episodes of the same illness. This study details the treatment response of six first‐episode manic patients who participated in a previously reported study of 139 subjects comparing olanzapine to placebo in bipolar I mania (Tohen M, Sanger TM, McElroy SL, Tollefson GD, Chengappa KNR, Daniel DG. Olanzapine versus placebo in the treatment of acute mania. Am J Psychiatry 1999; 156: 702–709). Methods: Six first‐episode subjects participated in a 3‐week double‐blind, random assignment, parallel group, placebo‐controlled study of olanzapine for bipolar mania. The Young Mania Rating Scale (Y‐MRS), Clinical Global Impression, and Hamilton Depression ratings were administered weekly. Lorazepam as rescue medication was permitted for the first 10 days. Results: Five subjects were randomized to placebo and one to olanzapine. Two subjects (40%) with psychotic mania (who also had their first‐illness episode) were assigned to placebo and responded with greater than 50% reduction in the Y‐MRS score and also remitted in 3 weeks. Another placebo‐assigned subject had a 46% reduction in the Y‐MRS scores, and two placebo‐assigned subjects worsened. The olanzapine‐assigned subject had a 44% reduction in the Y‐MRS score. In contrast, 34 of 69 (48.6%) multiple‐episode olanzapine subjects responded and 14 of 61 (23.0%) of placebo‐treated subjects did. Conclusions: This preliminary data set suggest there may be differences in treatment response between first‐illness episode versus multi‐episode bipolar manic subjects. Larger numbers of subjects with these illness characteristics are needed to either confirm or refute this suggestion.