Neuropsychological predictors of self-neglect in cognitively impaired older people who live alone.

OBJECTIVE: The authors examined the accuracy of certain neuropsychological tests in the prediction of harm resulting from self-neglect in cognitively impaired seniors who lived alone. METHODS: The study included 130 participants, aged 65 and older, who scored less than 131 on the Dementia Rating Scale. Neuropsychological tests were administered at baseline, resulting in eight predictive scores. Informants and primary care physicians provided information about harm that occurred to the participants during the 18-month prospective follow up. An incident was defined as harmful if it occurred as the result of self-neglect or disorientation and resulted in physical injury or property loss or damage and required emergency interventions. Proportional hazard regression analysis was conducted to examine the predictive relationship between the eight neuropsychological tests and time to incident harm with age, sex, education, the Charlson Comorbidity Index, and the Mini-Mental State Examination included in the model as covariates. RESULTS: Twenty-seven participants experienced harm during the 18-month follow-up period. A proportional hazards model indicated that three neuropsychological tests, which measured recognition memory, executive functioning, and conceptualization, were independent risk factors for harm. CONCLUSIONS: These findings provide insight into why harm occurred in these cognitively impaired elders who lived alone. They also support the ecologic validity of these tests and suggest directions for the development of intervention strategies for harm prevention.