Abstract

Importance Emerging evidence suggests that exposure to air pollution affects children's glucose metabolism. However, the underlying mechanisms are not fully understood. Objective To investigate whether body mass index (BMI; calculated as the weight in kilograms divided by the height in meters squared) growth trajectories mediate the association between traffic-related air pollution (TRAP) and insulin resistance. Design, Setting, and Participants As part of the Southern California Children's Health Study, the ongoing Meta-Air2 cohort substudy followed up participants from pregnancy to 24 years of age and examined the mediation role of BMI. Cardiometabolic follow-up was initiated as participants transitioned into adulthood. Data from the Meta-Air2 substudy were collected from November 27, 2018, to May 31, 2023. Exposures The California Line Source Dispersion Model was used to calculate mean childhood exposure to traffic-related total nitrogen oxides (NOx) from pregnancy to 13 years of age by calculating mean monthly estimates. Traffic density within a 300-m buffer around participants' residence was calculated as a secondary outcome. Main Outcomes and Measures Insulin resistance markers were assessed during the recent visit in young adulthood, including the homeostatic model assessment of insulin resistance (HOMA-IR; calculated from fasting glucose and insulin levels) and glycated hemoglobin (HbA(1c)) level. Participants' BMI growth trajectories, including BMI at 13 years of age and accelerated BMI growth, were analyzed as potential mediators. Using fully adjusted PROCESS macro mediation models, their role in mediating the association between traffic-related total NOx exposure and insulin resistance was examined with adjustment for demographic characteristics, smoking status, and parental history of diabetes. Results Among the 282 participants (mean [SD] age, 24.0 [1.7] years), each 1-SD increase in childhood exposure to traffic-related total NOx was associated with a BMI increase of 0.71 (95% CI, 0.29-1.13) at 13 years of age and adult HOMA-IR increase of 0.55 (95% CI, 0.23-0.87). An estimated mediation effect identified BMI at 13 years of age combined with accelerated BMI growth as accounting for 41.8% of the estimated total effect (beta, 0.23; 95% bootstrap CI, 0.01-0.52) between total NOx and HOMA-IR. Similar patterns were observed when exploring traffic density as an exposure or HbA(1c) level as an outcome. Conclusions and Relevance In this cohort study of young adults, the long-term association between TRAP and insulin resistance may be partially explained by higher BMI and accelerated BMI growth from early adolescence into young adulthood. These findings highlight the importance of weight management in children, particularly those residing in highly polluted areas.