In utero exposure to near-roadway air pollution and autism spectrum disorder in children

Sarah A Carter,Md Mostafijur Rahman,Jane C Lin,Yu-Hsiang Shu,Ting Chow,Xin Yu,Mayra P Martinez,Sandrah P Eckel,Jiu-Chiuan Chen,Zhanghua Chen,Joel Schwartz,Nathan Pavlovic,Frederick W Lurmann,Rob McConnell,Anny H Xiang,Sarah A. Carter,Jane C. Lin,Mayra P. Martinez,Sandrah P. Eckel,Frederick W. Lurmann,Anny H. Xiang

DOI: https://doi.org/10.1016/j.envint.2021.106898

IF: 11.8

2022-01-01

Environment International

Abstract:ImportancePrevious studies have reported associations between in utero exposure to regional air pollution and autism spectrum disorders (ASD). In utero exposure to components of near-roadway air pollution (NRAP) has been linked to adverse neurodevelopment in animal models, but few studies have investigated NRAP association with ASD risk.ObjectiveTo identify ASD risk associated with in utero exposure to NRAP in a large, representative birth cohort.Design, Setting, and ParticipantsThis retrospective pregnancy cohort study included 314,391 mother–child pairs of singletons born between 2001 and 2014 at Kaiser Permanente Southern California (KPSC) hospitals. Maternal and child data were extracted from KPSC electronic medical records. Children were followed until: clinical diagnosis of ASD, non-KPSC membership, death, or December 31, 2019, whichever came first. Exposure to the complex NRAP mixture during pregnancy was assessed using line-source dispersion models to estimate fresh vehicle emissions from freeway and non-freeway sources at maternal addresses during pregnancy. Vehicular traffic load exposure was characterized using advanced telematic models combining traditional traffic counts and travel-demand models with cell phone and vehicle GPS data. Cox proportional-hazard models estimated hazard ratios (HR) of ASD associated with near-roadway traffic load and dispersion-modeled NRAP during pregnancy, adjusted for covariates. Non-freeway NRAP was analyzed using quintile distribution due to nonlinear associations with ASD.ExposuresAverage NRAP and traffic load exposure during pregnancy at maternal residential addresses.Main OutcomesClinical diagnosis of ASD.ResultsA total of 6,291 children (5,114 boys, 1,177 girls) were diagnosed with ASD. The risk of ASD was associated with pregnancy-average exposure to total NRAP [HR(95% CI): 1.03(1.00,1.05) per 5 ppb increase in dispersion-modeled NOx] and to non-freeway NRAP [HR(95% CI) comparing the highest to the lowest quintile: 1.19(1.11, 1.27)]. Total NRAP had a stronger association in boys than in girls, but the association with non-freeway NRAP did not differ by sex. The association of freeway NRAP with ASD risk was not statistically significant. Non-freeway traffic load exposure demonstrated associations with ASD consistent with those of NRAP and ASD.ConclusionsIn utero exposure to near-roadway air pollution, particularly from non-freeway sources, may increase ASD risk in children.

environmental sciences