Ultrafine particulate matter exposure during second year of life, but not before, associated with increased risk of autism spectrum disorder in BKMR mixtures model of multiple air pollutants

• Published in: Environmental research Vol. 242; p. 117624
• Main Authors: Goodrich, Amanda J., Kleeman, Michael J., Tancredi, Daniel J., Ludeña, Yunin J., Bennett, Deborah H., Hertz-Picciotto, Irva, Schmidt, Rebecca J.
• Format: Journal Article
• Language: English
• Published: Netherlands Elsevier Inc 01.02.2024
• Subjects: Air Pollutants - analysis; Air Pollutants - toxicity; Air pollution; Air Pollution - adverse effects; Air Pollution - analysis; Autism; Autism Spectrum Disorder - chemically induced; Autism Spectrum Disorder - epidemiology; Bayes Theorem; BKMR; Case-Control Studies; Child; Environmental Exposure - analysis; Female; Humans; Inositol Phosphates; Mercaptopurine; Mixtures; Nitrogen Dioxide - analysis; Particulate Matter - analysis; Particulate Matter - toxicity; Pregnancy; Prostaglandins E; Ultrafine particulate matter; Pregnancy; Autism; BKMR; Air pollution; Mixtures; Ultrafine particulate matter
• ISSN: 0013-9351; 1096-0953
• DOI: 10.1016/j.envres.2023.117624

Prenatal and early postnatal air pollution exposures have been shown to be associated with autism spectrum disorder (ASD) risk but results regarding specific air pollutants and exposure timing are mixed and no study has investigated the effects of combined exposure to multiple air pollutants using a mixtures approach. We aimed to evaluate prenatal and early life multipollutant mixtures for the drivers of associations of air pollution with ASD. This study examined 484 typically developing (TD) and 660 ASD children from the CHARGE case-control study. Daily air concentrations for NO2, O3, ultrafine (PM0.1), fine (PM0.1-2.5), and coarse (PM2.5-10) particles were predicted from chemical transport models with statistical bias adjustment based on ground-based monitors. Daily averages were calculated for each exposure period (pre-pregnancy, each trimester of pregnancy, first and second year of life) between 2000 and 2016. Air pollution variables were natural log-transformed and then standardized. Individual and joint effects of pollutant exposure with ASD, and potential interactions, were evaluated for each period using hierarchical Bayesian Kernel Machine Regression (BKMR) models, with three groups: PM size fractions (PM0.1, PM0.1-2.5, PM2.5-10), NO2, and O3. In BKMR models, the PM group was associated with ASD in year 2 (group posterior inclusion probability (gPIP) = 0.75), and marginally associated in year 1 (gPIP = 0.497). PM2.5-10 appeared to drive the association (conditional PIP (cPIP) = 0.64) in year 1, while PM0.1 appeared to drive the association in year 2 (cPIP = 0.76), with both showing a moderately strong increased risk. Pre-pregnancy O3 showed a slight J-shaped risk of ASD (gPIP = 0.55). No associations were observed for exposures during pregnancy. Pre-pregnancy O3 and year 2 p.m.0.1 exposures appear to be associated with an increased risk of ASD. Future research should examine ultrafine particulate matter in relation to ASD. •First study to evaluate exposure to ultrafine particulate matter in relation to ASD.•Daily concentrations of five pollutants predicted from chemical transport models.•Used mixtures model to examine joint effect of multiple air pollutants on ASD.•Ultrafine particulate matter in 2nd year of life positively associated with ASD.