This analysis used pooled data from three randomized, placebo-controlled trials that examined the efficacy and safety of risperidone for the treatment of agitation, aggression, and psychosis associated with dementia in elderly nursing home residents to assess the risk-benefit of the use of risperidone in this population. The efficacy data (risperidone n=722, placebo n=428) were obtained from the Cohen-Mansfield agitation inventory (CMAI) and behavioral pathology in Alzheimer's disease (BEHAVE-AD) total and subscales. Additionally, clinical global impression (CGI) assessments were performed. Subgroup analyses were performed by type of dementia, severity of dementia, presence or absence of somnolence as an adverse event, and presence or absence of psychosis at baseline. Safety assessments included evaluation of treatment emergent adverse events, Extrapyramidal Symptom Rating Scale, ECG and vital signs, and Mini-Mental State Examination (MMSE). The mean dose of risperidone at end point was 1.0 mg/day (0.02 S.E.). The observed mean change at end point was significantly higher for risperidone than for placebo on CMAI total score (-11.8 versus -6.4, respectively; p<0.001), total aggression score (-5.0 versus -1.8, respectively; p<0.001), BEHAVE-AD total score (-6.1 and -3.6, respectively; p<0.001), and psychotic symptoms score (-2.1 and -1.3, respectively; p=0.003). The main treatment effects of risperidone were similar in all subgroup analyses. Additionally, risperidone-treated patients scored significantly better than placebo-treated patients on the CGI scales at end point. The incidence of treatment-emergent adverse events was comparable between risperidone (84.3%) and placebo (83.9%). More patients discontinued due to adverse events in the risperidone-treated group (17.2%) than in the placebo group (11.2%). Differences in adverse event incidences between placebo and risperidone were observed for extrapyramidal symptoms (EPS), mild somnolence and the less common cerebrovascular adverse events (CAE). Risperidone induced neither orthostatic, nor anticholinergic side effects nor falls nor cognitive decline. Of all atypical antipsychotics, risperidone has the largest database of double-blind controlled trials to support its efficacy and safety in the treatment of agitation, aggression, and psychosis associated with dementia. At the recommended doses, risperidone displayed a favorable risk-benefit profile. Risperidone was well tolerated with respect to EPS, somnolence, and anticholinergic side effects in this elderly population. In view of the risk for CAEs, risperidone, should be targeted towards the treatment of those patients in whom psychotic and behavioral symptoms of dementia are prominent and associated with significant distress, functional impairment or danger to the patient.