Twenty-four chronic schizophrenic outpatients with a mean age of 37.21 years +/- 9.96 SD were treated with risperidone (RSP) at the dosage of 2-9 mg/die (mean 4.46 mg/die +/- 1.30 SD, mean 0.06 mg/kg +/- 0.01 SD) for a year. Clinical evaluation was assessed with the Brief Psychiatric Rating Scale (BPRS), Positive and Negative Symptoms Scale (PANSS), Extrapyramidal Side Effects Rating Scale (EPSE) and a checklist for Anticholinergic Side Effects (ACS) at TD, then after 1 (T1), 2 (T2), 3 (T3), 6 (T6), 9 (T9) and 12 (T12) months. RSP and 9-hydroxy-risperidone (SOH-RSP) plasma levels were determined at T12 by the HPLC method. BPRS and PANSS mean values showed a significant improvement during the study. No correlation between RSP dosage (mg/kg) and RSP, SOH-RSP plasma levels or active moiety resulted. A positive correlation between age and active moiety was observed. A positive correlation between RSP and SOH-RSP plasma levels was observed. A curvilinear relationship between active moiety and PANSS improvement (%) was observed. Patients with the higher PANSS amelioration showed RSP + 9OH-RSP plasma levels ranging from 15 to 30 ng/mL. RSP seems to be quite an effective drug. It seems, however, difficult to devise appropriate dose schedules and plasma level determination seems to be necessary in some cases. (C) 2001 iditions scientifiques et medicales Elsevier SAS.