Elderly chronic schizophrenia patients are particularly difficult to treat because of aging-related changes, cognitive impairment, and comorbid physical illness. This article describes a naturalistic, retrospective study of typical antipsychotic treatment versus risperidone for elderly psychotic inpatients. Fifty-one patients, mean age 72.7 + 5.9 years, mean disease duration 33.1 + 12.0 years, who met the DSM-IV criteria for schizophrenia or schizoaffective disorder were treated by risperidone (n = 26) or typical antipsychotic treatment (n = 25) during acute exacerbation and followed up for 18 months. Patients were rated using the clinical global impression (CGI) scale and positive and negative symptom scale (PANSS), and their body weight was recorded at baseline, 6 months, and 18 months. Both treatment groups improved on all rating scales at 18 months. Levels of decrease in PANSS positive and total scores were more prominent in patients treated with risperidone (p < 0.01 and p < 0.05, respectively). The change in CGI scores reached significance only after 18 months and was more pronounced in the risperidone group (p < 0.01). Anti-Parkinsonian medications were used more frequently in the typical antipsychotic group, whereas benzodiazepines were used more frequently in the risperidone group. Body mass index increased minimally after 18 months in the risperidone group (+ 0.3 kg/m2), whereas a larger (+ 1.1 kg/m2), albeit not statistically significant, increase was recorded in the typical antipsychotic group. Emergence of side effects was less frequent in patients treated with risperidone (4/26 vs. 16/25 patients, p < 0.01). The results of this study demonstrate that in elderly chronic schizophrenic patients, switching from typical antipsychotics to risperidone is effective and well tolerated.