Objective: To examine the relationship between psychotic symptoms and subsequent physically aggressive behavior in outpatients with Alzheimer's disease.
Design: This observational study used a time series design to measure the presence and frequency of physically aggressive behavior over a 52-week period. Multiple regression models were used to evaluate delusions and hallucinations, assessed at baseline, as predictors of physical aggression, controlling for demographic and clinical variables.
Setting and participants: Of 315 consecutive eligible outpatients from the Rush Alzheimer's Disease Center, referred for evaluation for dementia, 270 (86%) participated in the study. All participants met NINCDS/ADRDA criteria for probable Alzheimer's disease; Mini-Mental State Examination scores ranged from 0 to 27, with a mean of 14.9 (SD = 5.6).
Measurements: Psychotic symptoms, previous episodes of physical aggression, and demographic variables were measured at baseline through a structured interview with an informant living in the same household. Physically aggressive behavior was measured sequentially at 1-week intervals over a period of 52 consecutive weeks and also through structured informant interviews.
Results: A total of 75 persons had one or more episodes of physical aggression during the 52 weeks of observation. The presence of delusions significantly predicted the presence and frequency of physical aggression. Of participants with high rates of physical aggression (> 1 episode/month), 80% had delusions. This effect was robust, even after controlling for the effects of other clinical variables. By contrast, hallucinations did not reliably predict episodes of physical aggression.
Conclusions: These data suggest that delusions, but not hallucinations, predict physically aggressive behavior in persons with Alzheimer's disease. In light of the persecutory nature of most delusional ideation in Alzheimer's disease, delusions may be associated with distortions in the perception of threat in common social situations.