Seasonal variation in sleep apnea and other sleep parameters from diagnostic polysomnography data, 2003-2024

Seasonal variation in daylength, temperature, and humidity may influence the sleep cycle. We investigated how sleep apnea, measured by the apnea-hypopnea index (AHI) and five other sleep parameters vary across seasons.

Our cross-sectional analysis used diagnostic in-laboratory polysomnography studies from adults without major diseases (3965 female; 2886 male) between 2003 and 2024 at a tertiary hospital in North Carolina, USA. We assessed seasonality separately in males and females using regression models with periodic spline terms and adjustment for age, body mass index, and calendar time.

We show that AHI is higher in males than females throughout the year and exhibits a single yearly cycle in both sexes (Likelihood Ratio Test [LRT]: p = 0.049, 0.058, respectively) with the peak for males in March and the nadir in early September; respective extremes in females occur about a month later. Sleep efficiency exhibits two cycles per year in both males and females (LRT: p = 0.003, 0.051, respectively). In males, the peaks are in April and September with troughs in early January and June, with respective extremes in females about a month later. Total sleep time shows no evidence of seasonality in females (LRT: p = 0.91) but two cycles per year in males (LRT: p = 0.005). We find no strong evidence of seasonality for the proportions of N3 sleep, REM sleep, or N2 sleep for either sex.

Seasonal patterns for AHI and sleep efficiency are common to both sexes, whereas patterns for other sleep parameters differ between sexes or lack convincing evidence of seasonality.