Abstract
Background: Obstructive sleep apnea (OSA) severity often varies considerably from night-to-night, but whether environmental factors play a role is unclear. This study investigated seasonal and temperature-related changes in OSA severity.
Methods: Data were acquired from 70,052 participants with an average apnea-hypopnea index (AHI) ≥ 5 events/hour who used an under-mattress sleep sensor at least 4 times/week between January 2020 and September 2023. Fixed effect models were used to investigate the association between AHI and day of the year, adjusting for geographical location, variation in total sleep time, ambient temperature, and air pollution.
Results: Participants are middle-aged (mean ± SD, 53 ± 13 years), predominantly male (81%), overweight (BMI; 29 ± 6 kg/m2) and have an average of 492 ± 341 nights of data. Mean AHI is 18.0 ± 14.0 events/h and within-subject coefficient of variation is ±51%. AHI is ~5% higher during summer/winter compared to spring/autumn in the northern hemisphere, and 10-15% higher during summer compared to spring in the southern hemisphere. Higher ambient temperature (25th vs. 75th percentiles; 6 vs. 18 degrees Celsius) is associated with a 6.4% (95% CI; 6.3-6.5) increase in AHI. Results are consistent across 23 countries, although the effect of temperature on AHI is larger in Europe vs. the United States or Australia.
Conclusions: Here we demonstrate a seasonal component to OSA severity, partially explained by ambient temperature and seasonal variation in sleep duration. Our findings highlight the need to report data collection months in OSA clinical trials, and further study to uncover the physiology behind seasonal variation in OSA severity are required.
Methods: Data were acquired from 70,052 participants with an average apnea-hypopnea index (AHI) ≥ 5 events/hour who used an under-mattress sleep sensor at least 4 times/week between January 2020 and September 2023. Fixed effect models were used to investigate the association between AHI and day of the year, adjusting for geographical location, variation in total sleep time, ambient temperature, and air pollution.
Results: Participants are middle-aged (mean ± SD, 53 ± 13 years), predominantly male (81%), overweight (BMI; 29 ± 6 kg/m2) and have an average of 492 ± 341 nights of data. Mean AHI is 18.0 ± 14.0 events/h and within-subject coefficient of variation is ±51%. AHI is ~5% higher during summer/winter compared to spring/autumn in the northern hemisphere, and 10-15% higher during summer compared to spring in the southern hemisphere. Higher ambient temperature (25th vs. 75th percentiles; 6 vs. 18 degrees Celsius) is associated with a 6.4% (95% CI; 6.3-6.5) increase in AHI. Results are consistent across 23 countries, although the effect of temperature on AHI is larger in Europe vs. the United States or Australia.
Conclusions: Here we demonstrate a seasonal component to OSA severity, partially explained by ambient temperature and seasonal variation in sleep duration. Our findings highlight the need to report data collection months in OSA clinical trials, and further study to uncover the physiology behind seasonal variation in OSA severity are required.
| Original language | English |
|---|---|
| Article number | 314 |
| Number of pages | 10 |
| Journal | Communications Medicine |
| Volume | 5 |
| Issue number | 1 |
| DOIs | |
| Publication status | Published - 29 Jul 2025 |
Keywords
- Epidemiology
- Population screening
- Respiratory tract diseases