Ambient air pollution and acute respiratory infection in children aged under 5 years living in 35 developing countries

Background: Evidence from developed countries suggests that fine particulate matter (<= 2.5 mu m [PM2.5]) contributes to childhood respiratory morbidity and mortality. However, few analyses have focused on resource-limited settings, where much of this burden occurs. We aimed to investigate the cross-sectional associations between annual average exposure to ambient PM2.5 and acute respiratory infection (ARI) in children aged <5 years living in low- and middle-income countries (LMICs). Methods: We combined Demographic and Health Survey (DHS) data from 35 countries with gridded global estimates of annual PM2.5 mass concentrations. We analysed the association between PM2.5 and maternal-reported ARI in the two weeks preceding the survey among children aged <5 years living in 35 LMICs. We used multi-variable logistic regression models that adjusted for child, maternal, household and cluster-level factors. We also fitted multi-pollutant models (adjusted for nitrogen dioxide [NO2] and surface-level ozone [O-3]), among other sensitivity analyses. We assessed whether the associations between PM2.5 and ARI were modified by sex, age and place of residence. Results: The analysis comprised 573,950 children, among whom the prevalence of ARI was 22,506 (3.92%). The mean (+/- SD) estimated annual concentration of PM2.5 to which children were exposed was 48.2 (+/- 31.0) mu g/m(3). The 5th and 95th percentiles of PM2.5 were 9.8 mu g/m(3) and 110.9 mu g/m(3), respectively. A 10 mu g/m(3) increase in PM2.5 was associated with greater odds of having an ARI (OR: 1.06; 95% CI: 1.05-1.07). The association between PM2.5 and ARI was robust to adjustment for NO2 and O-3. We observed evidence of effect modification by sex, age and place of residence, suggesting greater effects of PM2.5 on ARI in boys, in younger children, and in children living in rural areas. Conclusions: Annual average ambient PM2.5, as an indicator for long-term exposure, was associated with greater odds of maternal-reported ARI in children aged <5 years living in 35 LMICs. Longitudinal studies in LMICs are required to corroborate our cross-sectional findings, to further elucidate the extent to which lowering PM2.5 may have a role in the global challenge of reducing ARI-related morbidity and mortality in children.