Associations between weather variables and sleep architecture: A large-scale analysis using consumer sleep data

This study investigated the impact of daily weather conditions on sleep architecture and assessed whether these effects vary according to the severity of suspected sleep-disordered breathing.

We analyzed 236,115 nights of home-based sleep data, collected over one year from 1,067 users of the SleepRoutine® mobile application in South Korea. Each night's data was paired with contemporaneous weather data from Seoul. Apnea-hypopnea index (AHI) values were classified on a nightly basis into low (< 15) and high (≥ 15) groups.

Elevated temperature was associated with decreased sleep efficiency (SE), total sleep time (TST), and deep sleep ratio (all p < 0.001), along with increased sleep latency (SL), and wake after sleep onset (WASO) (all p < 0.001). Higher humidity correlated with lower SE and light sleep ratio (p < 0.001, 0.004), and with longer WASO (p = 0.02, < 0.01). Longer sunshine duration was linked to greater TST, WASO, and REM sleep ratio (all p < 0.01) and to a decrease in sleep latency and light sleep ratio (p < 0.01). Relative to spring, summer demonstrated the poorest sleep quality, whereas autumn and winter showed variable effects. Importantly, temperature and sunshine duration demonstrated significant interactions with AHI group. In nights with higher AHI, weather-associated changes in SE, TST, and WASO were diminished. Seasonal interactions with AHI also significant across multiple sleep stages.

Temperature, humidity, and sunshine duration were significantly associated with sleep structure. These effects may be less pronounced in individuals with sleep-disordered breathing.