Ambient air pollution and childhood obesity from infancy to late childhood: An individual participant data meta-analysis of 10 European birth cohorts

Abstract

Ambient air pollution may contribute to childhood obesity through various mechanisms. However, few longitudinal studies examined the relationship between pre- and postnatal exposure to air pollution and obesity outcomes in childhood. We aimed to investigate the association between pre- and postnatal exposure to air pollution and body mass index (BMI) and the risk of overweight/obesity throughout childhood in European cohorts. This study included mother–child pairs from 10 European birth cohorts (n = 37111 (prenatal), 33860 (postnatal)). Exposure to nitrogen dioxide (NO2) and fine particulate matter with aerodynamic diameter < 2.5 µm (PM2.5) was estimated at the home addresses during pre- and postnatal periods (year prior outcome assessment). BMI z-scores (continuous) and overweight/obesity status (categorical: zBMI≥+2 (<5 years) or ≥+1 (≥5 years) standard deviations) were derived at 0–2, 2–5, 5–9, 9–12 years. Associations between air pollution exposure and zBMI were estimated separately for each pollutant and cohort using linear and logistic longitudinal mixed effects models, followed by a random-effects meta-analysis. The overweight/obesity prevalence ranged from 12.3-40.5 % between cohorts at 0–2 years, 16.7–35.3 % at 2–5 years, 12.5–40.7 % at 5–9 years, and 10.7–43.8 % at 9–12 years. Results showed no robust associations between NO2 exposure and zBMI or overweight/obesity risk. Exposure to PM2.5 during pregnancy was associated with 23 % (95%CI 1.05;1.37) higher overweight/obesity risk across childhood, and higher zBMI and overweight/obesity risk at 9–12 years. Heterogeneity between cohorts was considerable (I2:25–89 %), with some cohort-specific associations; e.g., pre- and postnatal exposure to PM2.5 was associated with lower zBMI across age periods in UK cohorts (ALSPAC and BiB), while postnatal exposure to PM2.5 and NO2 was associated with higher zBMI in one Dutch cohort (Generation R). Overall, this large-scale meta-analysis suggests that prenatal PM2.5 exposure may be associated with adverse childhood obesity outcomes, but provides no evidence to support an effect of postnatal air pollution exposure, although cohort-specific associations were observed.