In an earlier 21-day, placebo-controlled trial, ziprasidone was efficacious in improving symptoms of mania and was well tolerated. To confirm these results, a similarly designed 21-day trial was conducted. Inpatients with bipolar I disorder, manic or mixed, were randomized to ziprasidone (40 to 80 mg BID) or placebo. Efficacy rating scales were derived from the Schedule for Affective Disorders and Schizophrenia-Change Bipolar Scale (SADS-CB). SADS-CB-derived Mania Rating Scale (MRS) total score was the primary efficacy parameter. Secondary SADS-CB-derived efficacy parameters included Manic Syndrome and Behavior and Ideation Subscales, Hamilton Depression Rating Scale (HAM-D), and the Montgomery Asberg Depression Rating Scale (MADRS). The Clinical Global Impression-Severity Scale (CGI-S), the Global Assessment of Functioning (GAF), and the Positive and Negative Syndrome Scale (PANSS) were also assessed. Sixty-five placebo and 137 ziprasidone patients were evaluable for efficacy. Baseline-to-endpoint mean changes in MRS scores were -11.1 for ziprasidone and -5.6 for placebo (all patients, last observation carried forward [LOCF]; P < 0.01). Ziprasidone produced significantly greater improvements in Manic Syndrome (P < or = 0.01) and Behavior and Ideation Subscales (P < or = 0.001), CGI-S score, (P < or = 0.001), PANSS Total (P < or = 0.01) and Positive Subscale (P < or = 0.001) scores, and GAF (P < or = 0.001). With ziprasidone, significant improvements were observed from Day 2 onward for MRS and CGI-S at all time points except Day 4 for MRS. Treatment-related discontinuations due to adverse events were 5.8% for ziprasidone and 1.5% for placebo (P = 0.20). Ziprasidone was well tolerated, rapidly efficacious, and superior to placebo in improving symptoms and global illness severity in these inpatients with acute bipolar mania, both manic and mixed episodes.