Ecological study of Autism Spectrum Disorder prevalence and proximity to industrial facilities

Aisha S Dickerson, The University of Texas School of Public Health


Autism Spectrum Disorder (ASD) is a lifelong neurodevelopmental disorder that becomes apparent in infancy or early childhood. The etiology of this disorder is not well understood, but it has been hypothesized that environmental exposures are an important factor. For this study, we evaluated the association between Census tract-level ASD prevalence and proximity of tract centroids to point source industrial facilities as well as ambient lead, mercury, and arsenic concentrations within each tract. Additionally, we examined the influence of socioeconomic characteristics of tract population and ASD reporting. Using aggregate data for 2,489 tracts from five sites of the Autism and Developmental Disabilities Monitoring Network (ADDM), we estimated ASD prevalence for 2000 to 2008 surveillance years. Information for industrial facilities in existence during birth years of study participants was downloaded from the US Environmental Protection Agency Toxics Release Inventory (USEPA-TRI) and information on ambient metal concentrations were obtained from the EPA National-Scale Air Toxics Assessment (NATA). We used univariable multi-level negative binomial models to test associations between ASD prevalence and incremental and binary proximity of tract centroids to industrial facilities as well as individual metal air concentrations. Additionally, we built multivariable models adjusting for potential confounders, stratified for effect modification, and examined additive and synergistic effects of combined metal concentrations. We also used ordinal hierarchical models to test the influence of demographic and SES characteristics on the proportion of cases ascertained through schools only. After adjustment for percentage of male, White, Hispanic, and college educated residents, urbanicity, and proportion below poverty, tracts within the closest 10th percentile had a significantly higher ASD prevalence compared to those in the furthest 50th percentile (RR=1.27, 95% CI 1.02, 1.57). ASD prevalence was also higher in tracts where lead concentrations were in the 25th-50th percentile and greater than 75th percentile compared to those with concentrations in the lowest 25th percentile. Alone mercury levels did not show positive associations with ASD prevalence; however, when included in an interaction term to assess synergistic association of all three metals combined, mercury appeared to drive the positive associations with ASD prevalence. Demographic and SES characteristics found to be statistically associated with proportion of cases ascertained through schools only included Black race and maternal education for cases and Hispanic ethnicity in the population from which cases were drawn. Our finding suggests that increased ambient metal exposures in children living near industrial facilities may increase risk of ASD. More research should be done to assess the impact of these exposures and of residential distance from point sources.

Subject Area

Environmental Health|Epidemiology

Recommended Citation

Dickerson, Aisha S, "Ecological study of Autism Spectrum Disorder prevalence and proximity to industrial facilities" (2014). Texas Medical Center Dissertations (via ProQuest). AAI3639419.