Fifteen variables went in. Two came out standing. A study in Psychological Reports found that when you line up the 50 US states, a state’s socioeconomic status and its level of microscopic particle pollution are the strongest predictors of how many of its residents have an autism diagnosis.
Higher average wealth and education, together with more particle pollution, tended to go with higher diagnosis rates. The author is Stewart J. H. McCann, professor emeritus of psychology at Cape Breton University, who works in geographical psychology, the mapping of traits and health outcomes across physical space.
McCann said he got interested because of the federal government’s attention to “rather alarming increases in the reported prevalence of ASD over recent years.”
Under the hood
He drew mainly on 2017 data for all 50 states and used a statistical model of autism prevalence among adults aged 18 to 84. Against it he set 15 state-level predictors. These included racial demographics, average intelligence scores, urbanization, maternal age, education spending per student, uninsured residents, Medicaid enrollment, and the average Big Five personality profile of each state.
On a basic level, nearly all of them correlated with autism rates. States with more autism diagnoses tended to have higher socioeconomic status, higher intelligence scores, more urban populations, older mothers and higher per-pupil spending.
Then came the statistical control. McCann held socioeconomic status constant and checked whether the other 14 variables still mattered. Most did not. Only racial makeup, personality profile, urban population share, air pollution, prenatal-to-3 policies and maternal age survived. Put all six together with socioeconomic status in one model, and only socioeconomic status and air pollution remained as significant, independent predictors.
Together they accounted for 55.7 percent of the variance in autism rates across the states. “I also was surprised by the fact that only SES and air pollution ultimately showed any independent relation to state ASD prevalence rates,” McCann told PsyPost.
As a check, he ran the same predictors against ADHD. State ADHD rates showed the opposite relationship, which suggests the pattern is specific to autism.
What it does not show
The design is cross-sectional, a snapshot at one point in time. McCann was blunt about it: “readers cannot draw inferences regarding cause and effect.” Nothing here says air pollution or wealth causes autism.
There is a plainer explanation to consider. Wealthier states have better health care infrastructure, so parents and adults may find it easier to get a formal evaluation. The study cannot separate a real rise in autism from a rise in diagnosis and awareness. And because the data describe states, McCann cautioned that whether the same relations hold for individuals “cannot be known without replicative research.”
He suggested future work look harder at maternal and paternal age, since older parenthood is closely tied to higher socioeconomic status. The paper is titled “An Exploratory Study of Socioeconomic Status, Air Pollution, and 13 Other Variables as Predictors of U.S. State Autism Spectrum Disorder Rates.”
Consult your doctor for medical advice.
