• Change from baseline to end of treatment (Week 8) in Psychosis Cluster Score of Pathology from the Behavioral Pathology in Alzheimer's Disease (BEHAVE-AD) Rating Scale and Clinical Global Impression (CGI).
  

• Change in BEHAVE-AD total score and subscales (other than Psychosis Cluster subscale) from baseline; improvement in CGI scores during treatment; incidence of adverse events throughout study.
  

Dementia is frequently observed in the elderly, often associated with psychotic symptoms such as delusion or hallucinations, or with behavioral disturbances such as aggressive behavior, wandering, and aimless behavior induced by the psychotic symptoms. This is a double-blind, placebo-controlled study of the effectiveness and safety of risperidone (taken twice daily over 8 weeks) in the treatment of psychotic symptoms in patients with Alzheimer's disease. Assessments of effectiveness include: Behavioral Pathology in Alzheimer's Disease (BEHAVE-AD), a scale used for global assessment of symptoms associated with dementia; the Psychosis Cluster Scale of BEHAVE-AD, a subscale that assesses paranoid and delusional ideation as well as hallucination; and Clinical Global Impression-Change (CGI-C), a measure of an improved or aggravated condition. Safety evaluations include the incidence of adverse events throughout the study; physical examinations, electrocardiograms (ECGs), laboratory tests (hematology, biochemistry, urinalysis), and assessment of extrapyramidal symptoms at specified intervals. The study hypothesis is that treatment with risperidone shows greater improvement in psychotic symptoms, as measured by the BEHAVE-AD psychotic cluster score, in patients with Alzheimer's disease, as compared to placebo. In addition, it is hypothesized that risperidone is well tolerated. Risperidone tablets (0.25 mg or 0.50 mg) or placebo tablets taken orally twice daily. Total daily dosage of 0.5mg on Day 1, 1.0mg on Days 3-5, and 1.5mg (maximum dose) on Days 5-13. Optimum dose maintained during Weeks 3-8 of treatment.Dose may be increased or decreased at investigator's discretion.