Using more than 12 million nights of wearable sleep data, researchers show that warmer days and nights already reduce sleep across the U.S., with future climate warming expected to deepen losses and widen existing health and social inequalities.
Study: Impact of heat exposure on sleep health and its population vulnerability in the United States. Image Credit: Stock-Asso / Shutterstock
In a recent study published in the journal Environment International, researchers investigated associations between heat exposure and sleep health.
Exposure to high ambient temperatures suppresses the normal reduction in core body temperature, which signals the onset of sleep and entry into deeper sleep stages. Nighttime and daytime heat exposure can alter circadian thermoregulation and disturb the wake–sleep rhythm. Heat-related sleep impairments have been associated with adverse mental and cardiovascular health outcomes.
Quasi-experimental and observational studies have reported associations between higher nighttime and daytime temperatures and reductions in total sleep time (TST) in adults and children. Studies also suggest that sleep quality, including sleep continuity, sleep stages, and macrostructure, is an important predictor of physical and mental health. Nevertheless, how environmental heat impacts multidimensional sleep quality in large populations remains less well understood.
About the study
In the present study, researchers assessed the associations of outdoor heat exposure with multidimensional sleep health using longitudinal data from the All of Us Research Program (AoU) in the United States. AoU commenced in May 2017, recruiting over one million adults, with multimodal data collection through questionnaires, electronic health records (EHRs), genomics, biospecimens, physical measurements, and digital wearables.
The primary outcome was total sleep time. Secondary outcomes included sleep onset timing, sleep continuity, and sleep stage-specific durations. Existing disease status was determined from EHRs. Daily meteorological data, including precipitation, wind speed, maximum relative humidity, and minimum and maximum temperatures, were obtained for the period 1990–2023.
Longitudinal sleep data collected between 2010 and 2022 were linked to gridded meteorological data. Daytime (DTA) and nighttime (NTA) temperature anomalies were calculated as heat exposure metrics.
DTA and NTA were defined as the difference between the observed daily maximum or minimum temperature during the sleep-tracking day and the long-term average daily maximum or minimum temperature from 1990 to 2009, calculated at the ZIP code level.
A multivariate mixed-effects model was used to examine associations between heat exposure (NTA and DTA) and sleep outcomes. Vulnerability to heat exposure was assessed across spatiotemporal factors (month and climate zone), demographic characteristics (age, sex, ethnicity, socioeconomic status), and health-related conditions.
Estimated NTA–sleep duration associations were subsequently combined with projected NTA values from Shared Socioeconomic Pathway (SSP) climate scenarios to project future changes in total sleep time from 2020 to 2099.
Findings
The study included 14,232 individuals with a mean age of 50.5 years, contributing over 12.5 million person-days of sleep duration and onset data and 8.13 million person-days of sleep continuity and stage-specific data. Most participants were female (68.3 percent), White (81.5 percent), and non-Hispanic (89.9 percent). Mean total sleep time was 393.5 minutes, and average sleep efficiency was 91.5 percent.
Mean wake after sleep onset was 50.7 minutes. Average daily durations of deep, light, and rapid eye movement (REM) sleep were 60.9, 258.7, and 82.5 minutes, respectively. Mean nightly temperature anomaly was 0.9 °C, while mean daytime anomaly was 0.75 °C.
Among participants who shared EHR data, 22 percent had cancer, 14.5 percent had cardiovascular disease, 10 percent had depressive disorders, 5 percent had diabetes, and 11.7 percent had obesity.
A 10 °C increase in nighttime and daytime temperature anomalies was associated with 2.63 and 2.19 minutes lower total sleep time, respectively. A 10 °C increase in nighttime temperature anomaly was also associated with a 0.05-minute longer wake time after sleep onset, a 0.03 percentage-point lower sleep efficiency, a 1.66-minute delay in sleep onset, 1.58 minutes less light sleep, 0.93 minutes less deep sleep, and 0.19 minutes less REM sleep. Daytime temperature anomalies showed similar associations, except for non-significant effects on wake after sleep onset and deep sleep.
The strongest associations between nighttime heat exposure and sleep loss were observed during late spring to early summer and late summer to early autumn, as well as in the marine climate zone, where estimated effects were more than twice as large as those observed in other climate zones, a pattern the authors suggest may partly reflect the lower prevalence of household air conditioning in these regions.
Total sleep time decreased by 2.76 minutes per 10 °C increase in nighttime temperature anomaly among individuals aged 40–50 years, approximately 20 percent greater than in those under 40 years. Females experienced a 2.65-minute reduction, about 23 percent greater than males.
Greater sleep losses were also observed among individuals with lower socioeconomic status and among those with obesity, cardiovascular disease, or depression.
Under a high-emissions and high-economic-growth scenario (SSP5-8.5), populations living in mixed, marine, hot, and cold climate zones during 2080–2099 were projected to experience an additional 8.5, 24.0, 11.8, and 8.5 hours of sleep loss per person-year, respectively, compared with 1995–2014.
Individuals residing in marine climate zones were estimated to lose more than two hours of sleep per month between May and October, with the largest reduction occurring in August at approximately 3.4 hours per month.
Conclusions
Overall, 8.5 to 24.0 hours of sleep per person-year were projected to be lost by the end of the century across different U.S. climate zones compared with 1995–2014, with the greatest losses occurring in marine and hot climate zones, particularly during summer months.
Adults aged 40–50 years, females, individuals with lower socioeconomic status, and those with chronic physical or mental health conditions were especially vulnerable to heat-related sleep disruption.
Because outdoor temperature does not fully capture individual indoor heat exposure or adaptive behaviours, such as air-conditioning use, the authors note that future sleep-loss estimates may be conservative.
These findings highlight growing inequalities in climate-related sleep loss and may inform targeted interventions to improve heat adaptation and resilience.
Insufficient sleep may decrease life expectancy