Parenteral benzodiazepines or antipsychotics are often used to manage acute agitation in emergency department (ED) settings in which alternative strategies have failed or are not feasible. There are scant data comparing parenteral medication regimens. We aim to determine the efficacy and safety of intravenous droperidol or olanzapine as an adjunct to intravenous midazolam for rapid patient sedation. We undertook a randomized, double-blind, placebo-controlled, double-dummy, clinical trial in 3 EDs (August 2009 to March 2011). Adult patients (n=336) requiring intravenous drug sedation for acute agitation were randomized to receive a saline solution (control), droperidol (5 mg), or olanzapine (5 mg) bolus. This was immediately followed by incremental intravenous midazolam boluses (2.5 to 5 mg) until sedation was achieved. The primary outcome was time to sedation. Secondary outcomes were need for "rescue" drugs and adverse events. Three hundred thirty-six patients were randomized to the 3 groups. Baseline characteristics were similar across groups. The differences in medians for times to sedation between the control and droperidol and control and olanzapine groups were 4 minutes (95% confidence interval [CI] 1 to 6 minutes) and 5 minutes (95% CI 1 to 6 minutes), respectively. At any point, patients in the droperidol and olanzapine groups were approximately 1.6 times more likely to be sedated compared with controls: droperidol and olanzapine group hazard ratios were 1.61 (95% CI 1.23 to 2.11) and 1.66 (95% CI 1.27 to 2.17), respectively. Patients in the droperidol and olanzapine groups required less rescue or alternative drug use after initial sedation. The 3 groups' adverse event profiles and lengths of stay did not differ. Intravenous droperidol or olanzapine as an adjunct to midazolam is effective and decreases the time to adequate sedation compared with midazolam alone.