No single environmental factor causes autism, but a growing body of research has identified several exposures during pregnancy and early life that increase the likelihood of a child being diagnosed. Current CDC data puts autism prevalence at about 1 in 31 children, and while genetics play a major role, environmental influences appear to explain a meaningful portion of that risk, particularly when they interact with genetic susceptibility.
Air Pollution During Pregnancy
Prenatal exposure to fine particulate matter, the tiny particles released by vehicle exhaust, industrial emissions, and burning fuels, is one of the most studied environmental risk factors. A large study published in JAMA Network Open found that exposure to specific components of air pollution during the second and third trimesters increased autism risk by about 11%. The sensitive windows were narrow: sulfate exposure between weeks 23 and 36 of gestation and ammonium exposure between weeks 21 and 34 carried the clearest associations.
The timing matters because those weeks overlap with critical periods of brain development, when neurons are migrating into position and forming early connections. Ozone exposure during weeks 22 to 30 showed a smaller but still measurable increase in risk. These findings don’t mean that breathing city air guarantees a diagnosis, but they do suggest that sustained exposure to polluted air during pregnancy nudges the odds upward, particularly for genetically susceptible children.
Maternal Infections and Immune Activation
When a pregnant person’s immune system mounts a strong response to an infection, the inflammatory signals can cross the placenta and affect fetal brain development. This process, broadly called maternal immune activation, has been repeatedly linked to higher rates of autism in offspring. Population-based studies show that women hospitalized for infections during pregnancy are more likely to have children later diagnosed with autism.
The type of infection matters less than the intensity of the immune response. Both viral and bacterial infections can trigger the effect. Some viruses, like Zika, can also cross the placenta directly and interfere with how neurons migrate and form connections. Animal research reinforces the link: when researchers simulate viral or bacterial infection in pregnant animals using compounds that trigger immune responses without actual infection, the offspring show altered social behavior and repetitive actions that mirror features of autism. Autoimmune conditions and chronic inflammatory states during pregnancy carry similar associations.
Pesticides and Chemical Exposures
Living near agricultural areas where pesticides are applied has been tied to increased autism risk. A population-based study published in the BMJ found that prenatal exposure to several common pesticides, including glyphosate, chlorpyrifos, diazinon, malathion, and permethrin, was associated with higher odds of an autism diagnosis. The risk was especially pronounced for autism with co-occurring intellectual disability, and exposure during the first year of life added further risk on top of prenatal exposure.
Industrial chemicals also play a role. Lead exposure during pregnancy, even at levels once considered safe, has measurable effects on neurodevelopment. A study tracking mothers and children found that doubling blood mercury levels in late pregnancy was associated with a 31% increase in the relative risk of autistic behaviors at age five. Mercury in cord blood showed a similar pattern. These heavy metals interfere with brain development at the molecular level, disrupting how genes are activated and silenced during the formation of neural circuits.
How Environmental Factors Change Gene Activity
One of the clearest ways environmental exposures increase autism risk is through a process called DNA methylation, a chemical modification that controls whether specific genes are turned on or off. Think of it as a dimmer switch on your genes. Environmental chemicals can shift that dimmer in ways that alter brain development without changing the DNA sequence itself.
Lead, for example, causes changes in the methylation patterns of over 1,200 genes during the period when stem cells are becoming brain cells, with most of those genes becoming more active than they should be. The industrial chemical BPA reduces the activity of key enzymes that maintain normal methylation patterns in developing brain tissue. PCBs, a class of industrial pollutants that persist in the environment, dramatically lower the levels of both these enzymes and the chemical building blocks needed for proper methylation. In the brains of people with autism, researchers have found the opposite pattern: elevated activity of these same enzymes and higher overall methylation levels. This suggests a feedback loop where early chemical exposure disrupts the system, and the brain overcorrects in ways that alter its development.
Medications During Pregnancy
Certain medications taken during pregnancy have been linked to increased autism risk, though the picture is complicated. The anti-seizure medication valproate carries the strongest evidence. Among children exposed to valproate in utero, the absolute risk of an autism diagnosis is about 4.4%, roughly double the rate seen in unexposed children of mothers with epilepsy. For childhood autism specifically, the risk nearly tripled compared to unexposed children.
Antidepressants, particularly SSRIs taken during the second and third trimesters, have shown an association with autism in some studies, with one meta-analysis finding an 87% increase in risk. But this finding comes with major caveats. Mothers taking antidepressants may be more likely to seek developmental evaluations for their children, inflating detection rates. There’s also the possibility that the underlying depression or anxiety, rather than the medication itself, contributes to the risk. The scientific community has not reached consensus on whether SSRIs are a direct factor or a marker for other influences.
Folic Acid and Protective Factors
Not all environmental influences increase risk. Folic acid supplementation before and during pregnancy is one of the strongest protective factors identified so far. An NIH-funded study found that women who took folic acid, a multivitamin, or both before becoming pregnant were 61% less likely to have a child diagnosed with autism. That protective effect climbed to 73% when supplementation continued through pregnancy.
The CDC recommends that all women of childbearing age take 400 micrograms of folic acid daily, primarily to prevent neural tube defects, but the autism data adds another reason to follow that guidance. Folic acid is a key ingredient in the methylation process described above, helping to maintain normal gene regulation during fetal brain development. When methylation goes wrong due to chemical exposures, adequate folic acid may help buffer the effect.
Why It’s Rarely One Factor Alone
The research consistently points to a model where environmental exposures interact with genetic vulnerability. A child with certain genetic variants may be more sensitive to air pollution, pesticide exposure, or nutritional deficiencies during pregnancy than a child without those variants. This explains why two pregnancies with similar environmental exposures can produce very different outcomes.
Most of the risk increases identified in research are modest on their own, typically in the range of 10% to 30% higher odds per individual factor. But these exposures don’t happen in isolation. A pregnancy involving air pollution, nutritional gaps, and chemical exposure simultaneously may carry compounding risk. The timing of exposure also matters enormously. The second and third trimesters appear to be the most sensitive windows for many of these factors, coinciding with rapid brain growth and the formation of neural circuits that underlie social behavior and communication.