Associations of long-term exposure to ambient air pollution and road traffic noise with sleep health in UK Biobank

Background: Evidence about associations of pollutants with sleep is limited, and most of studies focused on individual sleep behaviors, neglecting their interrelation. We aimed to assess the relationship between ambient air pollution and road traffic noise with overall sleep health. Methods: The study included 378,223 participants from the UK Biobank. Including five sleep behaviors (chronotype, sleep duration, insomnia, snoring, and daytime sleepiness) to construct overall sleep pattern. Ambient air pollution exposure was estimated using Land Use Regression model. Road traffic noise exposure was estimated using a simplified version of the Common Noise Assessment Methods model. Using multinomial and binary logistic regression models to identify the associations between pollutants with overall and individual sleep behaviors, respectively. Results: Participants were derived in three sleep patterns: healthy (n = 140,490), intermediate (n = 220,627), and poor (n = 17,106). After adjustment for potential confounders, compared with the lowest quartile of PM2.5, the highest quartile had higher odds of intermediate and poor compared to healthy sleep pattern [OR (95% CI) for poor: 1.28 (1.21-1.36); for intermediate: 1.11 (1.09-1.14)]. We observed similar relationships for PM10, PM2.5 absorbance, PMcoarse, NOx, and NO2. In unadjusted model, compared with low exposure of Lnight, high Lnight exposure had higher odds of intermediate and poor compared to healthy sleep pattern [OR (95% CI) for poor: 1.13 (1.06-1.20); for intermediate: 1.03 (1.00-1.06)]. However, such associations disappeared after further adjustment for potential confounders. Conclusions: Long-term ambient air pollution is associated with overall sleep health. Road traffic noise itself is weakly associated with overall sleep health.