Accelerometer-Based Sleep-Wake Patterns May Help Predict Dementia Risk

Accelerometer-derived sleep-wake cycle patterns were associated with a higher risk of future dementia and modestly improved dementia prediction beyond traditional risk factors, with predictive value comparable to APOE genotype when added to an age-only model. In this large prospective cohort study involving more than 53,000 older adults from the UK Biobank and nearly 4,000 participants from the Whitehall II study, researchers analyzed 36 objectively measured sleep-wake metrics collected through wrist accelerometers. There were 2 major patterns linked to dementia risk: one characterized by lower levels and diversity of daytime physical activity, more sedentary behavior, and greater transitions from activity to rest during the day; the other marked by extreme sleep durations, prolonged nighttime wakefulness, reduced transitions from wake to sleep, and earlier waking times. During follow-up, both patterns were significantly associated with incident dementia, and adding these measures to models containing sociodemographic, behavioral, and health-related factors modestly but significantly improved dementia prediction. The findings suggest that wearable-device measures of sleep and activity rhythms may become scalable tools for early identification of individuals at elevated risk for dementia. Source: https://jamanetwork.com/