Effect Modification by Ambient Temperature on the Association of Ambient Ozone Exposure with Diet-Controlled and Insulin-Treated Gestational Diabetes Mellitus

Background: Previous studies have explored the association between ambient ozone (O3) exposure and gestational diabetes mellitus (GDM). However, little is known about the association of O3 with different GDM clinical classifications—diet-controlled GDM (GDMA1) and insulin-treated GDM (GDMA2), and whether temperature modifies the associations. Methods: We conducted a retrospective cohort study including 11,491 pregnant women between 2017 and 2023 in Tianjin, China. O3 exposure levels for each participant were assessed using the Tracking Air Pollution in China dataset. Logistic regression models were employed to analyze the associations of trimester-specific O3 exposure with GDMA1 and GDMA2. Distributed lag nonlinear model (DLNM) was used to identify potential critical gestational weeks of O3 exposure. Ambient temperature was categorized using the 10th and 90th percentiles to evaluate the effect modification of extreme temperatures on these associations. Results: No statistically significant association was observed in the fully adjusted trimester-specific models. In the weekly-specific DLNM analyses, we observed 5–12 weeks before pregnancy and 26–28 gestational weeks as sensitive windows where O3 exposure was weakly associated with elevated GDMA1 risk, and 11–25 gestational weeks for GDMA2. Each 10 μg/m3 increase in O3 during pregnancy was associated with small increases in the odds of GDMA1 (28 weeks: OR = 1.019, 95% CI: 1.002, 1.035) and GDMA2 (25 weeks: OR = 1.018, 95% CI: 1.002, 1.035). High temperatures appeared to strengthen the observed association between preconception O3 exposure and GDMA1. Additionally, O3 exposure during pregnancy may increase the risk of GDMA1 among women aged <35 years, preconception obesity, gestational hypertension, and family history of diabetes. Conclusions: This study suggests possible weak weekly-specific associations between O3 exposure and GDM classifications and indicates that ambient high temperature may modify the O3-GDM association. These findings should be interpreted cautiously and require confirmation in future studies.