BACKGROUND Olanzapine, a thienobenzodiazepine, is one of the relatively new atypical antipsychotic drugs. The lowest threshold of effective olanzapine plasma levels in inpatient treatment is assumed to be 9 ng/ml. Very little is known about the plasma concentration in patients at various oral doses of olanzapine or about the clinically relevant interactions with co-medications. METHODS In 71 schizophrenic patients (age 32.6 +/- 12.1, range 18-63 years; 31 women, 40 men), plasma olanzapine levels were assessed in 377 tests by high-performance liquid chromatography (HPLC) with electrochemical detection. Fifty-six of these plasma levels were assessed while patients were receiving olanzapine as monotherapy; otherwise, the plasma levels were assessed with the patients receiving various co-medications. RESULTS The mean daily oral dose of olanzapine was 17.5 mg (SD = 7.0, range 5-40 mg), and the mean olanzapine plasma concentration was 54.2 ng/ml (SD 37.8 ng/ml, range 1.2-208 ng/ml). The plasma concentration of olanzapine increased linearly with the daily oral dose (r = 0.64, p < 0.001). A multiple variance analysis considering age and sex as covariables showed a significant difference in the dose-corrected plasma levels of olanzapine among 40 smokers and 31 non-smokers; age and sex did not affect the dose-corrected plasma levels. However, women received a significantly lower daily dose of olanzapine under routine clinical study conditions. No differences could be detected among the dose-corrected plasma concentration of those patients who were taken off olanzapine because they did not respond (n = 14) or because of side effects (n = 5) and those who were discharged while still on olanzapine. Under the co-medication with fluvoxamine, significantly higher dose-corrected olanzapine plasma concentrations were found than with olanzapine monotherapy, whereas significantly lower dose-corrected olanzapine plasma concentrations were detected under lithium and trimipramine co-medication. Under co-medication with amitriptyline, benperidol, carbamazepine, flupentixol, and lorazepam, the dose-corrected olanzapine plasma concentrations were no different than the plasma levels under olanzapine monotherapy. CONCLUSIONS The relevance of therapeutic drug monitoring is emphasized with respect to the data presented and to the literature. Future studies should examine, in particular, the effects of a wider range of co-medications in a larger patient sample.