All patients were openly treated with risperidone, given once daily in the evening, according to standard guidelines. Doses were gradually increased during an 8-week trial period (increased 1 mg/day per week for the first 3 weeks). Diazepam (mean=13.30 mg/day, SD=6.23) was co-administered to nine patients for agitation, anxiety, or insomnia. The patients’ psychopathology was assessed at baseline (drug-naive state), biweekly, and at endpoint by using the Positive and Negative Syndrome Scale (3). The optimal risperidone dose was defined as the dose at which either symptoms were reduced (at least 20% from baseline) or extrapyramidal side effects emerged. To assess improvement, we calculated the percentage of symptomatic change on the Positive and Negative Syndrome Scale, adjusted for minimum baseline ratings. The mean optimal daily risperidone dose was 2.70 mg/day (SD=0.89). All patients reached their optimal dose before developing extrapyramidal side effects. Four patients developed parkinsonism, and one developed akathisia at a mean daily risperidone dose of 5.20 mg/day (SD=1.60). Thirteen patients (76%) achieved an optimal response with a risperidone dose of up to 3 mg/day, with an average 60% (SD=21%) improvement in total Positive and Negative Syndrome Scale score at the end of the 8-week trial period. Four patients (24%) reached an optimal dose of 4–5 mg/day, with a mean 45% (SD=28%) improvement. None of the patients achieved optimal response at doses higher than 6 mg/day.