US study links wildfire smoke in pregnancy to later autism diagnosis, but it does not show cause

A large study from the United States has found a statistical link between wildfire smoke during pregnancy and children later being diagnosed as autistic. It does not show that smoke causes autism, and the researcher who led it says more studies are needed to confirm the finding.

The paper was published online in JAMA Pediatrics on 26 October 2026. According to WHYY reporting carried by WITF, it was led by Stefania Papatheodorou, an obstetrician and epidemiologist at the Rutgers School of Public Health. She worked with researchers from Harvard T.H. Chan School of Public Health, Emory University, Massachusetts General Hospital and other institutions.

What the researchers did

The team used US Medicaid health insurance records for 893,212 mothers and their children, born between 2007 and 2014.

They did not measure smoke in anyone’s home. Instead, they used modelled weekly estimates of the fine particles in wildfire smoke, known as PM2.5. They matched these to each family’s zip code (the US version of a postcode) across the pregnancy. A child counted as autistic or as having ADHD when that diagnosis showed up in insurance claims.

What they found

Of the children in the study, 8,357 had an autism diagnosis in their records and 22,302 had an ADHD diagnosis. By age eight, about 1.9% had been diagnosed autistic.

Higher wildfire smoke exposure across pregnancy was linked with a higher chance of an autism diagnosis. The paper gives this as a “hazard ratio” of 1.30 for each rise of 0.75 micrograms per cubic metre in smoke exposure across pregnancy. WITF describes this as a 30% rise in risk. The link was strongest for exposure in the middle and later months of pregnancy. The researchers found no meaningful link with ADHD.

Reading the numbers carefully

• It is an association, not a cause. WITF reports that, because the study was observational, Papatheodorou said it cannot establish that wildfire smoke causes autism.
• 30% is relative, not absolute. Across the whole study, roughly 2 in 100 children had been diagnosed by age eight. A 30% relative rise does not mean 30 in 100 children.
• The uncertainty is wide. The paper’s confidence range runs from a 7% to a 59% increase.
• It counts diagnoses, not autistic people. A diagnosis only appears in insurance records if a child is assessed, diagnosed and has it recorded in a claim.
• Exposure was estimated, not measured. Smoke levels were modelled for each area, not measured for each person.
• It covers one group. All the families were enrolled in US Medicaid. The study says nothing directly about the UK.

Questioning the language of “risk”

Research like this is often framed around the “risk” of autism, as if autism were a harm to be reduced. WITF’s own report says plainly: “Autism isn’t a disease.” The same report quotes Papatheodorou, speaking about the health risks of wildfire smoke, as saying:

“The population disease burden is going to increase with the increased exposure to fires.”

The quote doesn’t make clear whether she meant autism specifically or the wider health effects of smoke. But when autism research is reported in the language of “risk” and “disease burden”, it comes close to the language of “prevention”, which treats autistic people’s existence as a problem. At Neurodivergent News, we think that framing should be questioned. Autism is a lifelong difference in how people think, sense and communicate.

Wildfire smoke is a real health issue on its own terms. WITF notes that fine particles can settle deep in the lungs and enter the bloodstream, and that breathing them can cause breathing and heart problems. That case doesn’t need autism to be cast as the harm.

This article does not give health advice, so we have left out the pregnancy guidance in other coverage.

What happens next

Papatheodorou said further studies are needed to confirm the association. The paper’s authors say wildfire activity is getting worse. Whatever later research finds, this study doesn’t change the case for understanding and supporting autistic children and adults as they are.

This piece summarises reporting by WITF (WHYY) and JAMA Pediatrics. Read the original articles for the full story.