Inequalities in exposure to the air pollutants PM2.5 and NO2 in Australia

Abstract

Abstract Vulnerable subpopulations may be exposed to higher levels of outdoor air pollution than the rest of the population. Due to the potential for this to exacerbate their existing health burden, concerns about disparities in subpopulations’ air pollution exposure have motivated international public health researchers to examine this topic. In Australia, such research is lacking to date, despite heterogeneity in air pollution at multiple spatial scales across the continent. This study aimed to investigate disparities in exposure to two health-relevant outdoor air pollutants: particulate matter <2.5 μm (PM2.5) and nitrogen dioxide (NO2). We used national land-use regression models to estimate annual average concentrations of PM2.5 and NO2, and area-level census data on ethnicity, age and socio-economic status (SES) to calculate the bivariate associations between each census-derived variable with the concentration of air pollutants. We also used multivariable models including specific measures of SES as covariates to assess to what extent associations were explained by SES. Associations were calculated separately for rural and urban areas using generalised additive models which accounted for spatial autocorrelation. Bivariate results showed significant nonlinear associations (p < 0.001) between vulnerable subpopulations and pollutant concentration. These associations suggested that areas with greater socio-economic disadvantage, a higher proportion of ethnic minorities, and elderly people are exposed to higher concentrations of PM2.5 and NO2, although differences in the magnitude of exposure were small overall. Our multivariable models showed that the associations between ethnic minorities and pollutant concentration appear to be substantially affected by area-level SES. Our results suggested that these vulnerable subpopulations are inequitably exposed to PM2.5 and NO2. While the magnitude of differences in exposure were generally small, the predicted differences in exposure among vulnerable subpopulations could contribute to a potentially avertable health burden at a population-level.