Environmental correlates of neuropsychiatric symptoms in nursing home patients with dementia

Objective The aim of the study was to estimate the influence of environmental correlates of neuropsychiatric symptoms in nursing home patients with dementia. Methods 1289 patients in 56 Dementia Special Care Units (SCUs) in the Netherlands were assessed using the Neuropsychiatric Inventory–Nursing Home version (NPI–NH) and the Cohen‐Mansfield Agitation Inventory (CMAI). Potential correlates of the neuropsychiatric symptoms studied were gender and age of the patient, dementia severity, psychoactive medication use and environmental correlates such as the number of patients per unit or per living room, staff size/patient ratio, time spent on care activities and presence of a walking circuit. Multilevel logistic regression analysis was used to estimate the relative contribution of the different factors in explaining neuropsychiatric symptoms variability. Results The prevalence of neuropsychiatric symptoms differed between SCUs, also after correcting for patient factors. Patient‐related factors explained 7–21% of the total variance of neuropsychiatric symptoms. Intra Class Correlation (the correlation of symptoms of patients within SCUs) ranged from 0.039 (psychosis) to 0.163 (apathy). The variance in SCU symptom prevalence (variance between SCUs) accounted for 3.5–14.8% of the total variance in neuropsychiatric symptoms. In SCUs of which the staff spent more time on care activities did the patients show lower levels of apathy. Other environmental correlates did not predict neuropsychiatric symptoms. Conclusion The substantial variation of prevalence rates between SCUs in combination with the clustering of symptoms within SCUs is strong evidence for the environmental contribution to neuropsychiatric symptoms and the multifactorial nature of behaviour in dementia. Copyright © 2009 John Wiley & Sons, Ltd.