Until recently, conventional antipsychotics were the standard pharmacotherapy for psychosis and behavioral disturbances associated with dementia. This double-blind, placebo-controlled study compared the acute efficacy of the selective serotonin reuptake inhibitor citalopram and the neuroleptic perphenazine with placebo for the treatment of psychosis and behavioral disturbances in nondepressed patients with dementia. Eighty-five hospitalized patients with at least one moderate to severe target symptom (aggression, agitation, hostility, suspiciousness, hallucinations, or delusions) were randomly assigned to receive either citalopram, perphenazine, or placebo under double-blind conditions for up to 17 days. Patients treated with citalopram or perphenazine showed statistically significant improvement on several Neurobehavioral Rating Scale factor scores. Compared to those receiving placebo, only patients treated with citalopram showed significantly greater improvement in their total Neurobehavioral Rating Scale score as well as in the scores for the agitation/aggression and lability/tension factors. Side effect scores were similar among the three treatment groups. Citalopram was found to be more efficacious than placebo in the short-term hospital treatment of psychotic symptoms and behavioral disturbances in nondepressed, demented patients.