The monitoring of plasma olanzapine concentrations has been found to be an important and useful tool for optimizing psychiatric treatment. The present study investigated the effect that clinical factors, such as smoking and age, and functional polymorphisms of UGT1A4, CYP1A2, and CYP2D6 genes have on plasma olanzapine concentration, as well as the effects of plasma olanzapine concentrations on Japanese schizophrenic patients' clinical symptoms. The subjects included 51 chronic schizophrenic patients whose symptoms were not controlled with chronic conventional antipsychotics and therefore were switched to olanzapine. Male smokers had a significantly lower olanzapine concentration-dose ratio and olanzapine/4'-N-desmethyl olanzapine ratio (which reflects CYP1A2 activity) than male nonsmokers and female nonsmokers. The results of a 2-way analysis of covariance showed that smoking had the main effect, rather than gender or age. The functional gene polymorphisms that were studied had no effect on the plasma olanzapine and metabolite concentrations. An improved total Brief Psychiatric Rating Scale (BPRS) score was not correlated with the plasma olanzapine concentration, but individual BPRS scores related to improvement of suspiciousness, hallucinations, and blunted affect were significantly correlated with plasma olanzapine concentration. Clinical factors, especially smoking, were more important modulators of olanzapine metabolism than the functional genotypes. Long-term olanzapine treatment with adequate plasma olanzapine concentrations could be more effective in improving some symptoms than treatment with conventional antipsychotics.