Predictors of nursing home placement in Parkinson's disease: a population-based, prospective study.

To examine the rate and predictors of nursing home placement in patients with Parkinson's disease. Four-year prospective study. A population-based study in western Norway 178 community-dwelling subjects with Parkinson's disease. Main outcome measure was the time from baseline to nursing home admission. Baseline evaluation of motor symptoms (Unified Parkinson's Disease Rating Scale, UPDRS), cognition (clinical dementia interview, Gottfries, Brane & Steen dementia scale, and Mini-Mental State Examination), depression (clinical interview and the Montgomery & Asberg Depression Rating Scale), and psychotic symptoms (UPDRS Thought Disorder item) were performed. Forty-seven patients (26.4%) were admitted to a nursing home during the 4-year study period. Institutionalized patients were older, had more advanced Parkinson's disease with more severe motor symptoms and impairment of activities of daily living, were cognitively more impaired, were more often living alone, and had more hallucinations than those who continued to live at home. Duration of disease, levodopa dose, and gender distribution did not differ between the two groups. A Cox proportional hazards linear regression analysis showed that old age, functional impairment, dementia, and hallucinations were independent predictors of nursing home admission. Both motor and neuropsychiatric symptoms contributed to institutionalization, but the presence of hallucinations was the strongest predictor. This finding indicates it is possible that effective treatment of hallucinations may reduce the need for institutionalization in patients with Parkinson's disease.