We assessed the changes of plasma haloperidol concentrations and clinical responses repeatedly up to 4 weeks after coadministration or discontinuation of rifampin in 12 schizophrenic patients taking haloperidol alone (group I) and 5 patients taking haloperidol and antituberculotic drugs (group II). After coadministration of rifampin in group I, daily trough haloperidol concentrations rapidly decreased and reached 63% of baseline level by day 3, 41.3% by day 7, and 30% by day 28. On the other hand, after discontinuation of rifampin in group II, plasma haloperidol concentration increased to 140.7% of baseline level by day 3, 228.7% by day 7, and 329% by day 28. In this study, a 30% or greater change in the clinical rating scale was considered a positive clinical response of the drug interaction. Using this criterion, 50% of the group I subjects responded according to the Brief Psychiatric Rating Scale (BPRS) total score, and 25% responded according to the BPRS subscale for psychiatric symptoms. No positive responses were observed in group II patients. These results strongly suggest that rifampin interacts with the clinical effects as well as the plasma concentrations of coadministered haloperidol, and careful monitoring should be considered when coadministration or discontinuation of rifampin is needed in a schizophrenic patient taking haloperidol.