Case-crossover study between PM2.5 and hospital admissions for respiratory diseases in Pretoria

Abstract

Background: Particulate matter air pollution has been associated with adverse health effects. Air pollution epidemiology traditionally focuses on the relationship between individual air pollutants and health outcomes (e.g., mortality). To account for potential co-pollutant confounding, individual pollutant associations are often estimated by adjusting or controlling for other pollutants in the mixture. The objective of the study is control for components within the mixture which are highly correlated. Method: An overall study is performed where PM2.5, BC, UV-PM and trace elements were controlled for Tapp and public holidays, stratified for age and sex. A second model, where PM2.5 is also controlled, was performed. Results: In total, there were 15 155 reported hospitalisations for respiratory diseases (J00-J99) over the 34-month period. The increase in respiratory hospital admissions was significantly increased by an increase in total PM2.5 of 2.7% (95% CI: 0.6, 4.9) per 10 µg/m3 increase. respiratory hospital admissions significantly increased for Ca by 4.0 % (95% CI: 1.4% - 6.8%), Cl by 0.7 % (95% CI: 0.0% - 1.4%), Fe by 3.3 % (95% CI: 0.5% - 6.1%), K by 1.8% (95% CI: 0.2 – 3.5) and Si by 1.3 % (95% CI: 0.1% - 2.5%). When controlling for PM2.5, respiratory hospital admissions increased significantly for total Ca by 3.2 % (95% CI: 0.3, 6.1) and for the 0 – 14 age group by 5.2 % (95% CI: 1.5, 9.1). Hospital admissions was positive and significantly increased with an increase in resuspended dust matrix by 2.9 % (95% CI: 0.1% - 5.7%) and for biofuel burning by 1.6 % (95% CI: 0.1% - 3.2%) sources. Conclusion: Controlling for a co-pollutant which is highly correlated with PM2.5 does reduce overestimation, but further studies should include deposition rates and parallel sampling analysis. Keywords: Tapp, PM2.5, sources, respiratory