The causal and independent effect of ozone exposure during pregnancy on the risk of preterm birth: Evidence from northern China.

The concentration of ozone (O3) in the environment is gradually increasing, but there are limited reports on the exposure to O3 during pregnancy on the risk of adverse birth outcomes. Our study aimed to examine the causal and independent effect of O3 exposure during pregnancy on the risk of preterm birth (PTB) and to identify the critical window. Based on the baseline population of the birth cohort in Jinan, northern China, we obtained the individual exposure for each subject during pregnancy of ambient 8-h moving average O3 through the inverse distance weighting model. The effect of O3 exposure during pregnancy on PTB was evaluated through the time-dependent Cox proportional-hazard models. And we assessed the causal relationship by controlling unknown confounding factors using the instrumental variable (IV) analysis, estimated the independent effect by principal component analysis, and identified the critical window period of exposure through the distributed lag model. Among 6501 subjects, 285 mothers delivered prematurely. The median (IQR) of O3 concentration during pregnancy was 109.51 (23.54) µg/m3. The high level of O3 exposure (>173.64 µg/m³) increased the risk of PTB, with HR of 1.92 (95% CI: 1.38-2.66). Furthermore, the HR (95% CI) of the O3 estimated value calculated by the IV (wind speed) on the risk of PTB was 2.63 (1.41-4.88). In addition, the high level of O3 exposure was associated with the risk of PTB in the 13th-18th gestational weeks. Therefore, the high level of O3 exposure during pregnancy may independently increase the risk of PTB, which may be a causal effect. The 13th to 18th week of gestation is a critical window for preventing this risk.