Abstract
OBJECTIVES: To assess the prevalence and correlates of self-reported hearing loss during the last 2 years of life. DESIGN: Observational cohort study. SETTING: The Health and Retirement Study (HRS), a longitudinal nationally representative cohort of adults aged 50 and older (2000-13). PARTICIPANTS: Older adults (N = 5,895, mean age at death 78, 53% female, 20% nonwhite). MEASUREMENTS: The HRS interview closest to death was used (mean 12.2 months before death). Participants rated their hearing (excellent, very good, good, fair, poor) and indicated whether they used hearing aids. The prevalence and correlates of fair and poor ratings are described, adjusted for age and sex. RESULTS: Thirty-two percent (95% confidence interval (CI) = 31-34%) of all participants and 60% (95% CI = 57-64%) of the 7% of participants who used hearing aids rated their hearing as fair or poor. The prevalence of fair or poor hearing was highest in participants interviewed closest to death (29% 19-24 months before death, 36% 1-6 months before death, P for trend = .01). Correlates of fair or poor hearing during the last 2 years of life included age at death (50-59, 22%; 60-69, 21%; 70-79, 26%; 80-89, 38%; ≥90, 50%), sex (men 35%, women 30%), race and ethnicity (Hispanic 42%, white 33%), wealth (lowest quartile 38%, highest quartile 27%), history of heart disease (yes 38%, no 27%), activity of daily living dependence (yes 42%, no 26%), difficulty taking medications (yes 46%, no 29%), and probable dementia (yes 44%, no 24%). CONCLUSION: Self-reported hearing loss increases during the last 2 years of life and is associated with physical and social vulnerability.