Canadian researchers say prenatal exposure to sulphate and ammonium particles may increase autism risk – but experts warn the findings show correlation, not causation.
Exposure to specific components of fine particulate air pollution during pregnancy may raise a child’s risk of developing autism, according to Canadian researchers.
They have found that fine particulate matter, particularly its sulphate and ammonium components, is linked to increased autism spectrum disorder (ASD) diagnoses in early childhood, with the second and third trimesters emerging as critical periods of vulnerability.
The researchers said their study added to growing evidence that air pollution, particularly its chemical makeup, could influence neurodevelopmental outcomes.
Their findings have met mixed reactions from experts around the world, including in Australia, some of whom warn the findings show correlation, not causation.
Sulphate typically originates from industrial and power generation sources, while ammonium stems from agriculture, sewage and vehicle emissions.
The population-based study, published in JAMA Network Open, analysed more than 2.1 million births in Ontario between 2002 and 2022, linking detailed air pollution estimates with administrative health data.
The researchers found that prenatal exposure to sulphate and ammonium – key chemical components of fine particulate matter (PM 2.5) – was associated with a statistically significant increase in ASD diagnoses by age five years.
The findings held even after accounting for total PM 2.5 levels and postnatal exposure.
Children whose mothers were exposed to higher concentrations of sulphate and ammonium during the second and third trimesters were more likely to be diagnosed with ASD, suggesting these periods are particularly sensitive to environmental insults, the researchers wrote.
Postnatal ozone (O₃) exposure was also linked to elevated risk, although the association with prenatal ozone disappeared after adjustment for early-life exposure.
“Fine particulate matter (PM 2.5) is a well-established air pollutant linked to a wide range of adverse health outcomes,” the researchers wrote.
“Despite regulatory advances, PM 2.5 exposure remains a public health concern, particularly in urban areas with substantial industrial and vehicular emissions.
“Prenatal and early postnatal exposure to PM 2.5 is of particular concern, as it has been associated with neurodevelopmental outcomes, including autism spectrum disorder (ASD).
“Clinical and animal studies suggest several biological pathways through which PM 2.5 may influence neurodevelopment, including epigenetic modifications, pro-inflammatory responses, and oxidative stress.
“These mechanisms are thought to contribute to structural and functional changes in the developing brain.
“However, most epidemiological studies have focused on total PM2.5 mass, often overlooking potential variability in toxic effects among its chemical components.”
The cohort included more than two million mother-infant pairs with complete residential and health insurance data. ASD diagnoses were identified using a validated administrative algorithm with a follow-up period extending to five years of age. About 0.8% of children in the study were diagnosed with ASD, with boys affected roughly three and a half times more often than girls.
The associations were strongest among urban populations and more pronounced in lower-income and racially diverse neighbourhoods, underscoring persistent environmental inequities.
The authors acknowledged several limitations, including potential exposure misclassification due to reliance on residential postal codes and limited sensitivity of ASD case identification. Nevertheless, the large sample size, long follow-up, and fine temporal resolution of exposure estimates lent weight to the findings, they said.
“These findings underscore the potential importance of early-life environmental exposures and reinforce the need for public health strategies to reduce air pollution, particularly in urban and socioeconomically disadvantaged communities,” the researchers concluded.
Among the experts unattached to the research who weighed in on the findings was Dr Rachel Moseley, principal academic in Psychology at Bournemouth University in the UK.
“This paper shows a correlation between prenatal exposure to sulphate and ammonium components and childhood autism diagnoses,” she said.
“There is absolutely no evidence within the paper to suggest that the former caused the latter.
“It could potentially cause great worry to the general population if they thought this paper demonstrated a causal effect, having already been exposed to inaccurate claims around Tylenol and vaccines as causes of autism.
“Reports like this contribute to the very harmful myth that autism is increasing in prevalence: robust evidence from many studies indicates that this is not the case, but rather more autistic people, especially women and adults, are being recognised and diagnosed (which is an excellent thing, both on a personal level but also on a socioeconomic one, since late diagnosis is associated with poorer health and suicidality).
Related
“The wording and framing of the article, talking about ‘autism risk’, also contributes to harmful narratives in which autism is presented such as a disease, something negative to be eradicated or avoided; we know that these kind of messages are felt as deeply detrimental by autistic people and their families, who feel they’re living in a world where they’re unwanted and unacceptable.
“It is deeply disappointing that authors continue to contribute to these harmful narratives rather than following responsible guidelines in the way they speak about autism – thus harming the community they’re studying.”
Professor Anne-Louise Ponsonby, division head for Early Brain Science and head of the Neuroepidemiology Research Group at the Florey Institute for Neuroscience and Mental Health in Melbourne, said the paper highlighted “the importance of early life environment in autism”.
“Autism is often a multifactorial condition, where multiple factors, both environmental and genetic combine to cause the condition,” she said.
She said an important next step would be to look at the combined effect of multiple air pollutant chemicals, including those generated inside the home.
“Further, this work reinforces the importance of considering manufactured chemicals more generally to consider the combined effect of total chemical load,” Professor Ponsonby said.
“Consideration needs to be given that chemicals are not only inhaled but absorbed through the skin and also ingested. If chemicals are leading to similar biological effects in the body, then their additive or supra-additive effects will be important to consider.”
“We are investigating these issues in neurodevelopmental work at the Florey Institute, together with other Australian collaborating organisations.
“A key focus is the possible contribution of plastic chemicals in early life, and this work is supported by the Minderoo Foundation. Our work indicates it is important to consider not only combined chemical effects but also child genetic vulnerability to these environmental exposures.”
Dr Chloe Brimicombe, a climate scientist and public engagement manager at the Royal Meteorological Society in the UK, said the study showed correlation between air pollution exposure and potential cognitive development towards neurodivergence and autism.
“This study is not confirming air pollution exposure causes autism in children. This is because we do not know the biological pathways where air pollution could change neurodevelopment in this way,” she said.
“We do know that negative impacts on cognitive function are associated with air pollution exposure but most of this research is for an older age category.
“The study highlights important next steps around introducing socioeconomic variables alongside air pollution to suggest why an individual might be neurodiverse vs neurotypical, another approach is called life course analysis where we model air pollution over different phases of life.”
