Who Loses Sleep? Daily Time-Use Profiles, Caregiving Context Exposure, and Sleep Deprivation Among U.S. Adults

Background: Approximately one in three U.S. adults experiences insufficient sleep (< 7 hours per night), associated with elevated risk of chronic disease, mental health disorders, and premature mortality. Despite its public health significance, population-level research has rarely examined how daily time structures shape sleep deprivation risk. Caregiving context exposure—living with or providing care to adults with functional limitations—represents an often invisible time demand competing directly with sleep, yet its relationship to daily time-use patterns remains underexamined. This study examined (1) distinct daily time-use profiles among U.S. adults, (2) caregiving context exposure distribution across profiles, and (3) whether profile membership predicts sleep deprivation after accounting for sociodemographic characteristics.

Methods: Data were drawn from the pooled American Time Use Survey (ATUS; 2011–2024; N = 133,787). Latent Profile Analysis (LPA) was conducted using nine behavioral time-use indicators (sleep, eating, relaxation, work, housework, social activities, education, childcare, and adult care). Profile-specific sleep deprivation rates (< 7 hours/day) were estimated via the Bolck–Croon–Hagenaars (BCH) auxiliary variable method to correct for classification uncertainty. Survey-weighted modified Poisson regression examined associations between profile membership and sleep deprivation, adjusting for age, sex, race/ethnicity, education, income, marital status, weekday versus weekend diary day, and survey year.

Results: LPA identified five interpretable profiles (k = 5; LMR p < .001; Entropy = .988): Home-centered (n=86,095, 64.4%; highest relaxation and housework time), Work-centered (n=37,601, 28.1%; highest work time, mean = 9.29 h/day), Community-centered (n=6,575, 4.9%; highest social activity time, mean = 4.96 h/day), Education-centered (n=2,124, 1.6%; highest education/study time, mean = 7.31 h/day), and Care-centered (n=1,392, 1.0%; highest adult care time, mean = 5.33 h/day). Caregiving context exposure ranged from 21.0% (Education-centered) to 55.8% (Care-centered); all Care-centered members engaged in active adult caregiving on the diary day. BCH-corrected sleep deprivation rates differed significantly across all profiles (Wald χ²(4) = 4,065.57, p < .001). In adjusted modified Poisson regression, Work-centered (24.9% sleep deprived; RR = 2.44) and Care-centered adults (21.7%; RR = 2.27) showed the greatest risk of sleep deprivation relative to the Home-centered reference (9.4%), followed by Education-centered (17.1%; RR = 1.78) and Community-centered adults (14.4%; RR = 1.55; all p < .001). Black, non-Hispanic adults showed substantially elevated risk of sleep deprivation compared with White, non-Hispanic adults (RR = 1.49, p < .001).

Conclusion: Adults whose daily time is organized around caregiving or paid work face the greatest risk of sleep deprivation after adjusting for sociodemographic characteristics. Notably, Work-centered and Care-centered profiles carried statistically equivalent risk of sleep deprivation (RR = 2.44 and 2.27, respectively), suggesting that a care-centered day imposes sleep demands comparable to a work-centered one. This pattern was independent of household income, indicating that caregiving-related sleep loss is not confined to lower-income groups. The elevated risk among Black, non-Hispanic adults compared with White, non-Hispanic adults points to structural factors that compound sleep disparities beyond daily time structure. For social work practice and policy, findings support upstream interventions, including accessible respite care, flexible workplace policies, and caregiver sleep health screening— targeting the most time-burdened adults.