ATLANTA, July 13 — A fresh analysis of autism prevalence across American states has turned up a correlation that researchers say merits a closer look: children in regions with higher average levels of a specific airborne pollutant are more likely to receive an autism diagnosis. The study, which draws on data from the CDC’s Autism and Developmental Disabilities Monitoring (ADDM) Network and two national health surveys, examined autism spectrum disorder (ASD) prevalence estimates alongside long-term fine particulate matter (PM2.5) concentrations for each state.
Under the hood, the researchers combined ADDM Network data from 2018-2020 for 11 monitoring sites — Arizona, Arkansas, California, Georgia, Maryland, Minnesota, Missouri, New Jersey, South Carolina, Utah, and Wisconsin. For the remaining states, they pulled state-level figures from the CDC’s National Health Interview Survey (NHIS) and the National Survey of Children’s Health (NSCH) for the same period.
The technical reality is that PM2.5 — particles smaller than 2.5 micrometres, small enough to lodge deep in lung tissue — varies enormously by geography. The national average concentration across the study period was 8.4 µg/m³. California’s Central Valley recorded the highest levels, over 15 µg/m³.
Rural Montana saw the lowest, under 5 µg/m³. The state-level numbers offer a telling spread.
California’s ASD rate was 3.2 percent, with a median household income of $84,097 and an average PM2.5 reading of 11.2 µg/m³. Massachusetts posted a 3.1 percent rate, income $89,026, and PM2.5 at 8.5 µg/m³. At the lower end, Mississippi’s rate was 1.9 percent (income $49,111, PM2.5 9.8 µg/m³), and West Virginia’s 2.0 percent (income $48,850, PM2.5 8.1 µg/m³).
The study’s statistical model accounted for 62 percent of the variance in state ASD rates — a figure that suggests the factors examined, including pollution and income, explain a substantial share of the differences but leave a significant portion unaccounted for. As ever, the detail that matters is that correlation is not causation. The study does not claim that PM2.5 causes autism; it identifies an association that warrants further investigation.
The researchers used a model that controlled for some demographic variables, but the interplay of genetics, prenatal environment, socioeconomic factors, and diagnostic access is complex. What to watch next: whether longitudinal studies that track individual children’s exposure during pregnancy and early development can strengthen or weaken this signal.
The ADDM Network’s ongoing surveillance will also be critical for seeing whether the pattern holds as pollution levels shift and diagnostic practices evolve. Consult your doctor for medical advice.
