The lagged effect of air pollution on human eosinophils: a distributed lag non-linear model

Abstract

Objectives The aim of this study was to determine the lag between exposure to air pollutants and changes in human eosinophil counts. Material and Methods This was a retrospective study employing 246 425 physical examination records dated December 2013 – December 2016 from Chengdu, China. The authors determined the prevalence of individuals with eosinophil counts above the normal reference range each day. A distributed lag non-linear model was used to evaluate the lagged effect of each air pollutant on eosinophil counts. The lagged effects of each air pollutant were counted and presented with smoothing splines. Results The effects of air pollutants such as particulate matter (PM 2.5 , aerodynamic diameters <2.5 μm; PM 10 , aerodynamic diameters <10 μm), nitrogen dioxide (NO 2 ) and ozone (O 3 ) were evaluated. In women, the effects of PM 2.5 (RR = 1.154, 95% CI: 1.061–1.255) and PM 10 (RR = 1.309, 95% CI: 1.130–1.517) reached the maximum values on lag day 0. In men, there was no significant effect of PM 2.5 , but significant effects of PM 10 were found for lag days 20–28. The effects of NO 2 and O 3 on eosinophils were not statistically significant for either gender. Conclusions The air pollutants of PM 10 have a significant effect on human eosinophils for both women and men, but with different temporal patterns, with women showing a lag of 0–5 days and men showing a lag of 20–28 days. In addition, PM 2.5 was significant for women with a lag of 0–3 days but it was not significant for men. Int J Occup Med Environ Health. 2020;33(3):299–310

Keywords

• air pollution
• gender difference
• eosinophils
• real-world data
• lagged effect
• distributed lag non-linear model