Acute associations between ambient air pollution and risks of preterm and early-term births: results from 8 states in the United States

Ambient air pollution during pregnancy has been linked to adverse pregnancy outcomes, but evidence of acute exposure and shorter gestation length remains inconsistent. We examined the association between ambient air pollution and preterm (28-36 weeks) or early-term (37-38 weeks) births. Daily concentrations of 12 bias-corrected model-derived pollutants were linked to vital records of 1,085,162 preterm and 3,901,185 early-term singleton live births (2005-2017) across eight U.S. states. Under a time-stratified case-crossover design, odds ratios (OR) were estimated, adjusting for risks among ongoing pregnancies, meteorology, time trends and federal holidays. We estimated cumulative associations up to a 6-day lag using distributed lag models. Risk estimates per interquartile range increase were pooled across states using inverse-variance weighting. We observed positive associations between 0-2 day cumulative exposure to several air pollutants and early-term births, including NO ₂ (OR: 1.0023, 95% CI:1.0010, 1.0037 per 7.1 μg/m ³ ), PM _2.5 (OR=1.0022, 95% CI: 1.0006, 1.0038 per 4.6 μg/m ³ ), PM _2.5 organic carbon (OR= 1.0026, 95% CI: 1.0013, 1.0039 per 1.7 μg/m ³ ) and PM _2.5 elemental carbon (OR=1.0025, 95% CI: 1.0014, 1.0035 per 0.26 μg/m ³ ). Associations with preterm birth were mostly null. Our findings suggest positive associations between short-term air pollution exposure, including PM and major PM _2.5 components, and risks of early-term birth.