A multi-institutional team including Health Canada researchers has found that prenatal air exposure to specific particulate matter components and early-life ozone is associated with autism spectrum disorder in Ontario children.
Fine particulate matter has been linked to adverse health outcomes, with prenatal and early postnatal exposure associated with neurodevelopmental outcomes including autism spectrum disorder.
Most previous work has focused on fine particulate matter of airborne particles with a diameter of 2.5 micrometers or less (PM2.5), leaving uncertainty about variation in toxic effects among various chemical components and timing of exposure related to sensitive points in pregnancy. A large Southern California cohort study reported associations for several components, including sulfate, and a follow-up study also noted nitrate.
In the study, “Prenatal Exposure to Fine Particulate Matter Components and Autism Risk in Childhood,” published in JAMA Network Open, researchers conducted a population-based retrospective cohort study to examine associations between prenatal and first-year-of-life exposure to specific PM2.5 components, nitrogen dioxide, and ozone with autism diagnoses, and to identify potentially sensitive gestational windows.
Looking at birth through the first five years in Ontario, covering approximately 20 years, yielded 2,183,324 children after exclusions and 19,569 children who received an autism diagnosis.
Postal code modeling
Exposure assessment assigned prenatal concentrations by maternal postal code at delivery. Weekly nitrogen dioxide and ozone and biweekly PM2.5 mass and components were estimated from conception to age 36 weeks, with first-year exposures as annual postal code–level averages weighted by time at each address.
Components included black carbon, dust, ammonium, nitrate, organic matter, sulfate, and sea salt. Models integrated satellite data, chemical transport modeling, land-use regression, and ground monitoring data.
What the data show
Prenatal PM2.5 mass was associated with increased risk when adjusted for first-year averages (HR 1.15), and window-specific signals were driven by sulfate during weeks 23–36 (HR 1.11) and ammonium during weeks 21–34 (HR 1.11), after which PM2.5 mass was no longer associated with autism.
First-year ozone exposure was associated with autism risk with HR 1.09. Weekly models indicated significant windows for PM2.5 during gestational weeks 14 to 32, sulfate during weeks 23 to 36, ammonium during weeks 21 to 34, and ozone during weeks 26 to 30, with reported window-specific HRs of 1.12 for PM2.5, 1.11 for sulfate, 1.11 for ammonium, and 1.03 for ozone.
Black carbon, organic matter, dust, sea salt were not significant after adjustments, suggesting that specific chemical and not general PM2.5 exposure associations.
Patterns found across places and groups
Urban settings showed larger estimated effects for PM2.5, sulfate, and ammonium compared with rural areas. Sex-stratified results indicated larger estimates for male children, with sulfate the only pollutant significantly associated among female children.
Neighborhood patterns suggested more pronounced estimated risks in lower-income and middle-income areas and in areas with higher proportions of racialized and newcomer populations.
Authors conclude that prenatal exposure to specific PM2.5 components, particularly sulfate and ammonium, was associated with autism risk, with sensitive periods in the second and third trimesters.
Postnatal ozone exposure in the first year of life was also associated with risk. Findings point to the potential importance of early-life environmental exposures.