OBJECTIVE: To prospectively compare atypical and typical antipsychotics on time to all-cause medication discontinuation, an important effectiveness measure in the usual care of patients with schizophrenia. METHODS: Participants (N=1704) were initiators on oral atypical or typical antipsychotics (low, medium, or high-potency) in a three-year naturalistic study of schizophrenia. Medication groups were compared on time to all-cause medication discontinuation during the one-year following medication initiation. Statistical analysis used Kaplan-Meier and Cox proportional hazard models. RESULTS: Patients treated with atypical antipsychotics had longer time to medication discontinuation compared to patients receiving low, medium, or high-potency typical antipsychotics (odds ratio= 1.4, 1.5, 1.9; p= 0.044, 0.004, <0.001, respectively). Among atypical antipsychotics, clozapine and olanzapine-treated patients had a significantly longer time to medication discontinuation than patients receiving low, medium, or high-potency typical agents. Risperidone and quetiapine-treated patients had a longer time to medication discontinuation compared to only high-potency typicals. Ziprasidone did not significantly differ from low, medium, or high-potency typical agents. Further, only clozapine- and olanzapine-treated patients had a significantly longer time to medication discontinuation compared to perphenazine, a medium-potency typical antipsychotic. CONCLUSIONS: In usual care of schizophrenia patients, atypical antipsychotics appear to be superior to typical antipsychotics (regardless of potency level), and to significantly differ in treatment effectiveness.