Sleep Duration and Mortality According to Health Status in Older Adults

OBJECTIVES To examine the association between usual sleep duration and mortality according to physical and mental health status in older adults. DESIGN Prospective study conducted from 2001 to 2008. SETTING Community-based study. PARTICIPANTS Cohort study of 3,820 persons representative of the noninstitutionalized population aged 60 and older in Spain. MEASUREMENTS Sleep duration was self-reported at baseline. Analyses were performed using Cox regression and adjusted for the main confounders. The analyses were then stratified according to numerous indicators of health status. RESULTS During follow-up, 897 persons died. Mortality was higher in those who slept 8 hours (relative risk (RR)=1.34, 95% confidence interval (CI)=1.02-1.76), 9 hours (RR 1.48, 95% CI=1.12-1.96), 10 hours (RR 1.73, 95% CI=1.30-2.29) and 11 hours or more (RR 1.66, 95% CI=1.23-2.24) than in those who slept 7 hours (P for trend <.001). The association between long sleep duration (>= 10 vs 7 hours) and mortality was observed even in persons with good health status: optimal perceived health, good cognitive function (Mini-Mental State Examination score > 27), no depression, quality of life better than the cohort median (Medical Outcomes Study 36-item Short Form Survey Physical Component Summary score >= 46 and Mental Component Summary score >= 52), and without disability in instrumental activities of daily living. Sleeping 6 hours or less was not associated with higher mortality than sleeping 7 hours in persons with good health status. CONCLUSION Self-reported sleep duration was associated with 7-year mortality in this cohort of older adults, even when adjusted for health status. Further research is needed to determine the mechanisms and clinical implications of these findings.