The aim of this study was to investigate the use of psychotropic medication and therapeutic drug monitoring in a high-security psychiatric unit and to compare the doses and serum concentrations both with the recommended intervals and with the doses and serum concentrations in a control group. One hundred thirty-two patients were admitted in the period from January 2000 to December 2005. All available samples were used when comparing serum concentrations and doses with the recommended ranges. For the comparison of doses and serum concentration-to-dose (C:D) ratios with the control group only 1 sample from each patient was used. A total of 459 analyses of 27 different drugs in samples from 8 women and 73 men were included. The median number of therapeutic drug monitoring analyses per patient was 4 (range 1-29). Thirty-seven of the 81 patients (46%) used 2 or more antipsychotics at the same time. Clozapine, lamotrigine, olanzapine, quetiapine, ziprasidone, and zuclopenthixol were often given in doses above the recommended. The serum levels were frequently above those recommended for clozapine, olanzapine, quetiapine, risperidone, ziprasidone, and zuclopenthixol. The serum levels were significantly higher in the study group than in the control group for clozapine, lamotrigine, quetiapine, and zuclopenthixol. The given dose was significantly higher in the study group than in the control group for clozapine, lamotrigine and zuclopenthixol. The C:D ratio was significantly lower in the study group than in the control group for olanzapine but higher for quetiapine. The non-evidence based practice of high-dose polypharmacy with several antipsychotics is widely used in this unit. The use of higher doses in the study group than in the control group was not due to differences in metabolism or adherence to treatment between the 2 groups. The frequent use of therapeutic drug monitoring did not seem to have a great impact on the prescribed doses.