LAUSR

Understanding normative age-related changes is essential for distinguishing normal from abnormal sleep trajectories. This study examines age- and sex-related variations in sleep and circadian patterns using large-scale data from consumer wearables. One month of sleep data was evaluated from 25,918 Oura Ring users, equally distributed across age groups (spanning 20y to 85y, in increments of 5y and an equal male-to-female ratio (1:1). Standard sleep parameters (total sleep time, time spent awake, percentage of time in ‘light’, ‘deep’, and REM sleep), circadian factors (sleep midpoint, chronotype), and 30-second epoch-to-epoch transition probabilities (a measure of within-night sleep stability) were analyzed via generalized linear model (GLM) regression with age, sex, and age*sex as factors. Overall, women slept longer than men (+16.2 min; p<.000). Total sleep time peaked in the 20–25-year-old group (441±78 min) and declined to 403±68 min in the 80–85-year-old group (p<.000). Deep sleep percentage halved with age, while REM sleep showed smaller decreases beginning after 35 years (p<.000). Light sleep increased steadily, comprising 70–75% of total sleep in the oldest group (p<.000). The sleep midpoint slightly decreases with age, with chronotype showing advancement as people age. Sleep stability declined with age, with deep-to-deep and REM-to-REM transitions decreasing, particularly from 35 years onward (p<.000). Men experienced earlier stability declines around 40–45 years, while in women, these began around 55–60 years. Deep-to-light transitions rose from 3.4% in young adults to 8.4% in older adults (p<.000), with men showing earlier changes. REM-to-awake transitions increased with age, and REM-to-light transitions accelerated after 35 years, becoming most pronounced after 60 years. Significant differences were also observed in the interaction between age and sex on several other aspects of sleep dynamics. This study highlights the impact of aging and sex on sleep dynamics, with wearable data offering powerful insights into previously inaccessible patterns. By leveraging these findings, wearables have the potential to transform our understanding of sleep health and its role in disease progression, paving the way for targeted interventions and personalized solutions.