A prospective community-based pilot study of risk factors for the investigation of elder mistreatment.

PURPOSE To identify risk factors for the investigation of elder abuse, neglect, self-neglect, exploitation, and abandonment in a population-based observational cohort of community living elders. STUDY POPULATION Population-based sample of 2,812 community-living men and women in New Haven, Connecticut who were over age 65 in 1982. METHODS Matching process whereby cohort members who were investigated by Connecticut's State Ombudsman on Aging in 1985 or 1986 were identified. ANALYSIS Relative risks for ombudsman investigation in 1985 or 1986 were calculated based on risk factors status at baseline interview in 1982. RESULTS Sixty-eight (2.4%) members of the cohort received investigation. Features at cohort entry significantly associated with investigation in multiple logistic regression included: requiring assistance with feeding (Adjusted OR 3.5, 95% CI 1.2, 11.7), being a minority elder (Adj. OR 2.3, 95% CI 1.4, 2.8), over age 75 at cohort inception (Adj. OR 1.9, 95% CI 1.1, 3.1), and having a poor social network as defined by a social network index (Adj. OR 1.7, 95% CI 1.0, 2.7). When stratified by race, requiring assistance with feeding was associated with ombudsman investigation in minority elders (Adj. OR 10.8, 95% CI 2.8, 40.5) but not non-minority elders (Adj. OR 1.1, 95% CI 0.5, 7.5). CONCLUSION Functional disability, minority status, older age, and poor social networks were associated with investigation for elder mistreatment in this prospective, community-based population of men and women over the age of 65.