An international multi-cohort investigation of self-reported sleep and future depressive symptoms in older adults.

Poor subjective sleep is associated with future depression in older adults, but there is limited consensus on which sleep features have the strongest associations. Moreover, composite scores incorporating multiple features may better represent sleep burden than individual sleep items. We analyzed older adults (age ≥ 60) without clinically relevant depressive symptoms from a multi-cohort United States sample (US; N = 4826) and the Netherlands' Rotterdam Study (RS; N = 3663), with the goal of identifying individual and composite sleep features that are associated with future clinically relevant depressive symptoms 3-6 years later. Sleep-related daytime symptoms (Risk Ratio [95% CI] 2.10 [1.58, 2.80] in US; 2.10 [1.40, 3.14] in RS) and difficulty falling asleep (1.87 [1.49, 2.35] in US; RS = 1.90 [1.50, 2.43] in RS) were the strongest individual sleep features. Moreover, the combination of these features was most impactful (3.32 [2.33, 4.73] in US; 3.19 [2.64, 3.86] in RS), providing the largest effect size with the fewest number of items. Future studies should assess whether screening tools incorporating these features, paired with targeted sleep treatment, could reduce rates of incident depression in older adults. Examining mechanisms underlying these associations could improve the effectiveness of sleep-related treatments in older adults.