Abstract Purpose Although olanzapine may have advantages over other second-generation antipsychotics (SGAs) regarding longer time to treatment discontinuation among chronically ill patients, little evidence has been provided for the comparative effectiveness of SGAs in the acute phase. We aimed to determine if any of four SGAs were more effective in treating newly admitted acute schizophrenic patients. We performed a rater-blinded, randomized controlled trial of four SGAs in 15 psychiatric emergency sites. Eligible patients were 18–64 years old and met diagnostic criteria for schizophrenia, acute schizophrenia-like psychotic disorder, or schizoaffective disorder. A final total of 78 patients were randomly assigned by means of sealed envelopes to receive risperidone (3–12 mg/day; n = 20), olanzapine (10–20 mg/day; n = 17), quetiapine (300–750 mg/day; n = 20), or aripiprazole (12–30 mg/day; n = 21), with follow-up at 8 weeks. The primary outcome measure was all-cause treatment discontinuation. Results Overall, 37% (29/78) of patients discontinued the study medication before 8 weeks: 25% for risperidone; 12% for olanzapine; 55% for quetiapine; and 52% for aripiprazole. Time to treatment discontinuation for any cause was significantly longer in the olanzapine group than in the quetiapine ( p = 0.006) or aripiprazole ( p = 0.008) groups, but not compared to the risperidone group ( p = 0.32). Time to treatment discontinuation was significantly longer in the risperidone group than in the quetiapine group ( p = 0.048), but not compared to the aripiprazole group ( p = 0.062). However, the rate of p.r.n. intramuscular haloperidol use was significantly higher in the aripiprazole group than in other groups ( p = 0.029). Conclusion Olanzapine and risperidone are superior to quetiapine and aripiprazole for the acute treatment of psychosis in hospitalized patients.