Association between prenatal exposure to traffic-related air pollution and preterm birth in the PELAGIE mother–child cohort, Brittany, France. Does the urban–rural context matter?

Introduction Evidence has accumulated that exposure to ambient air pollution during pregnancy may influence preterm birth (PTB) in urban settings. Conversely, this relation has barely been investigated in rural areas where individual characteristics (demographic, socioeconomic, and psychosocial factors) and environmental co-exposures may differ. Objective We examined the association between prenatal exposure to traffic-related air pollution and PTB among pregnant women from the PELAGIE mother–child cohort (Brittany, France, 2002–2006) living in urban (n=1550) and rural (n=959) settings. Methods Women’s residences were classified as either urban or rural according to the French census bureau rural–urban definitions. Nitrogen dioxide (NO2) concentrations at home addresses were estimated from adjusted land-use regression models as a marker of traffic-related pollution. Associations between NO2 concentrations and PTB were assessed with logistic regression models. Results Prevalence of PTB was similar among women living in urban (3.2%) and in rural (3.5%) settings. More positive socioeconomic characteristics and health behaviors but more single-parent families were observed among urban women. NO2 exposure averaged 20.8±6.6 µg m−3 for women residing in urban areas and 18.8±5.6 µg m−3 for their rural counterparts. A statistically significant increased risk of PTB was observed among women exposed to NO2 concentrations ≥16.4 µg m−3 and residing in urban areas but not among their rural counterparts. Discussion The results of this study, conducted in a region with interspersed urban–rural areas, are in line with previous findings suggesting an increased risk of PTB associated with higher NO2 concentrations for women living in urban areas. The absence of association among their rural counterparts for whom exposure levels were similar suggests that environmental mixtures and psychosocial inequalities might play a role in this heterogeneity.