Favorable Social Conditions Tracked With Longer, More Regular Sleep Over Nearly Three Years
Long-term wearable data from the All of Us Research Program tied more favorable social determinants of health to both longer and more stable sleep, a pairing that has immediate relevance for clinicians who still rely too heavily on self-reported sleep habits. In this cross-sectional analysis, participants with better social and structural conditions slept modestly longer each night and showed less night-to-night variability in sleep duration across a median follow-up of 982 days. The signal was not subtle at the population level: the most favorable SDOH tertile was associated with 5.82 additional minutes of sleep and 4.79 minutes lower sleep-duration variability after full adjustment. Small differences in nightly sleep accumulate across months and years, and variability itself often reflects the chronic strain that fragments circadian regularity, alters sleep opportunity, and erodes restorative rest.
The cohort was large and demographically skewed toward older, highly educated adults. Among 12,403 participants, the mean age was 57.8 years, 70.3% were female, 88.3% were White, and 6.1% were Hispanic. More than 73% held a college degree or higher. Their mean device-estimated sleep duration was 406.1 minutes, or about 6.8 hours, and the mean day-to-day sleep duration SD was 76.9 minutes. Those baseline figures matter because they frame sleep as a long-term behavioral and physiologic exposure rather than a single-night snapshot. Wearables capture the rhythm of sleep across ordinary life, where work demands, caregiving responsibilities, housing instability, transportation barriers, and neighborhood safety all shape bedtime, wake time, and the consistency needed for sleep consolidation.
Economic Stability Drove the Strongest Sleep Signal
The strongest associations emerged in the economic stability domain. In fully adjusted models, economic stability showed the clearest relationship with both mean sleep duration and sleep-duration variability, with example estimates of 10.70 minutes longer sleep and 8.70 minutes lower variability. Among the four domains evaluated—social and community context, economic stability, neighborhood environment, and health care access and quality—economic strain stood out as the most potent correlate of habitual sleep. A person who is constantly deciding between rent, food, medications, and transportation is not just under stress; the body’s sleep system is being forced into repeated compromise, with arousal pathways staying active when they should be downregulated.
The mechanism is clinically plausible. Economic insecurity can lengthen wakefulness through stress-mediated hyperarousal, irregular shift work, crowded or unsafe housing, and inconsistent access to routines that anchor sleep timing. At the same time, sleep becomes less regular when a household’s schedule is dictated by external pressures rather than circadian preference. Wearable-derived variability captures this instability better than a single self-report question ever could. The associations in this study suggest that sleep clinicians should think beyond insomnia symptoms and ask what kind of life circumstances are preventing stable rest in the first place.
Wearable Metrics Captured Patterns Self-Report Often Misses
This analysis is especially notable because it relied on long-term device-estimated sleep rather than recalled sleep duration. Self-report tends to flatten sleep into an average number, while wearables expose the day-to-day scatter that can signal persistent social disruption. Here, mean sleep duration and sleep-duration SD were derived from main sleep episodes recorded across all available days, giving the study a view of habitual sleep behavior across a long span of everyday life. Participants contributed at least 180 days of wearable data, and some had more than 4,000 days of observation. That scale gives the findings a practical edge: the associations reflect ordinary sleep in the context of ordinary living, not a short laboratory window.
The limited subgroup heterogeneity across age, sex, race, and ethnicity suggests that the SDOH-sleep relationship was broadly consistent rather than confined to a single demographic slice. This pattern matters because it points toward a shared structural influence on sleep health, even if the lived expression of disadvantage differs across populations. Social and structural exposures rarely act through one pathway alone; they influence stress hormones, work schedules, home environment, access to care, and the ability to maintain regular routines. Sleep sits downstream of all of those pressures.
The study design still leaves room for caution. Cross-sectional data cannot establish the direction of causality, and people with poorer sleep may also experience worse social functioning, lower work capacity, or more difficulty sustaining economic stability. Even so, the long observation window and objective sleep capture strengthen the argument that social conditions are inseparable from sleep measurement. Clinicians who treat sleep as an isolated symptom risk missing the environmental and economic forces that shape it.
Sleep Health Evaluation Needs a Broader Social History
For bedside practice, the message is straightforward: sleep history should not stop at bedtime, wake time, and caffeine intake. A meaningful sleep evaluation also has to account for housing stability, financial pressure, neighborhood safety, access to transportation, and barriers to care, because these conditions can produce chronically irregular sleep even when a patient is trying to improve sleep hygiene. The study’s findings support a more integrated model in which sleep health is treated as both a behavioral outcome and a social exposure. In that framework, a patient’s variable sleep schedule is not simply a compliance problem; it may be a physiologic trace of structural disadvantage.
The data also sharpen the role of wearables in population health research. Devices can quantify sleep duration and variability over months, which makes them well suited for studying social exposures that unfold slowly. If future work confirms these patterns in more diverse cohorts, wearable sleep metrics could become a useful way to monitor whether social interventions produce measurable physiologic benefit. For now, the evidence points in one direction: better social conditions are linked to longer, steadier sleep, and economic stability appears to carry the greatest weight.