Prenatal Wildfire PM2.5 Exposure Linked to Higher ASD Risk in Medicaid Cohort

Researchers analyzed Medicaid data from 2007 to 2014, with follow-up through childhood, for 893,212 mother-child pairs with continuous enrollment and a stable residential ZIP code during pregnancy. Weekly wildfire-attributable PM2.5 exposure was estimated at 1-km² resolution and aggregated by ZIP code. Incident autism spectrum disorder (ASD) and attention-deficit/hyperactivity disorder (ADHD) were identified using validated claims-based algorithms. Cox models and distributed lag nonlinear models were used to assess associations, including differences by pregnancy timing.

An interquartile-range increase in cumulative wildfire PM2.5 exposure during pregnancy (0.75 µg/m³) was associated with higher ASD risk (hazard ratio, 1.30; 95% CI, 1.07-1.59), but not ADHD risk (hazard ratio, 1.03; 95% CI, 0.91-1.17). ASD associations were strongest from midpregnancy through late gestation. By age 8 years, cumulative incidence was 1.9% for ASD and 10.4% for ADHD.

This observational study identifies an association, not evidence that wildfire smoke causes ASD. The cohort was limited to children enrolled in Medicaid, and exposure was estimated by residential ZIP code. The findings do not establish that any specific exposure-reduction strategy changes neurodevelopmental outcomes.