Forty schizophrenic patients were treated with clozapine for 5 weeks. Patients were then classified as treatment responders or nonresponders according to a priori established criteria. After only 1 week of treatment, the responders showed a significant decrease on the Brief Psychiatric Rating Scale (BPRS) total score and the psychosis and tension subscale scores. Higher BPRS scores at baseline and larger improvements in BPRS scores after the first week predicted a more favorable outcome in this trial and yielded good classification accuracy.