Associations between weekly gestational exposure of nitrogen dioxide and preterm birth in a North Carolina birth cohort, 2003-2015
On this page:
Preterm birth (PTB) is associated with exposure to criteria air pollutants, though variability in the magnitude and consistency of associations exists. We evaluated the associations between weekly gestation exposure to nitrogen dioxide (NO2) with PTB (less than 37 weeks completed gestation) in a North Carolina (NC) birth cohort from 2003-2015 (N=1,367,757). Daily NO2 concentrations from a hybrid model with a spatial resolution of 1 km x 1 km were aggregated to census tract level estimates and linked to residential address at delivery, and then averaged to obtain exposure estimates for each week of pregnancy. Modified Poisson regression with robust errors was used to estimate risk differences (RD) and 95% confidence intervals (CIs) per 10 ppb increase in NO2, adjusted for gestational parent marital status, race/ethnicity, age at delivery, Medicaid status, and month of conception. The associations between NO2 exposure and PTB were generally null throughout pregnancy. RDs for weekly exposure during gestation ranged from -7 (95% CI: -14, 1) to 0 (-6, 5) per 10,000 births. However, when adjusting for estimated PM2.5 (from hybrid model) and ozone (from EPA’s Community Multiscale Air Quality (CMAQ) model) concentrations to account for potential copollutant confounding, there was a consistent pattern of decreased risk of PTB per 10 ppb increase in NO2 exposure in each week of gestation. RDs in copollutant model ranged from -29 (-36, -22) to -10 (-15, -5) per 10,000 births for NO2, 5 (4,6) to 12 (10, 14) per 10,000 births or PM2.5, and -17 (-24, -10) to 4 (-3, 11) per 10,000 births for ozone. While NO2 exposure was not associated with PTB in the single pollutant model, there was decreased risk with PTB when adjusting for other criteria air pollutants.