Before they could walk or talk, dozens of children born in New York City had their first encounter with chlorpyrifos recorded in a vial of cord blood. Years later, when those same children lay still inside an MRI scanner, the insecticide’s signature was still legible: the kids with the highest prenatal exposure showed measurable differences in the structure of their developing brains. That finding, published in the Proceedings of the National Academy of Sciences by researchers at Columbia University’s Center for Children’s Environmental Health, remains one of the most concrete pieces of evidence linking a widely used pesticide to lasting changes in human brain anatomy. And as of June 2026, the chemical has not left the American food supply.
What the brain scans actually showed
The Columbia team, led by environmental health scientist Virginia Rauh and neuroradiologist Bradley Peterson, enrolled pregnant women in upper Manhattan and the South Bronx and collected umbilical cord blood at delivery. They then divided the children into exposure groups based on measured chlorpyrifos levels in those samples. When roughly 40 children from the highest- and lowest-exposure groups underwent structural MRI between the ages of six and eleven, the resulting images revealed notable differences in cortical surface area across multiple brain regions.
The affected areas were not random. Enlargements and reductions clustered in the frontal and parietal cortices, regions that support attention, working memory, and language processing. In several zones, the higher-exposure group showed focal enlargements of the cortical surface that the authors interpreted as disrupted pruning or abnormal neuronal migration during fetal development. Because the study was prospective, with exposure measured at birth and imaging performed years later, the timeline supports a link running from prenatal pesticide contact to childhood brain structure, though it cannot prove causation on its own.
The design had real strengths: an objective biomarker (cord-blood concentration rather than self-reported diet), blinded image analysis, and standardized MRI protocols. Its main limitation was sample size. With about 20 children per group, the study had enough statistical power to detect the patterns it reported but not enough to parse finer questions, such as whether boys and girls were affected differently or whether specific trimesters of exposure mattered most.
A regulatory path that keeps reversing
Chlorpyrifos has been on a regulatory roller coaster for more than two decades. The EPA banned residential use in 2000 after concerns about childhood poisoning. Agricultural spraying continued. In August 2021, the agency revoked most remaining food tolerances, effectively barring the chemical from crops destined for human consumption (86 FR 48315). That decision drew on years of epidemiological evidence, including the Columbia cohort data.
The revocation did not hold. Agricultural interests challenged it in court, and in November 2023 the U.S. Court of Appeals for the Eighth Circuit vacated the rule. The EPA subsequently published a Federal Register notice in early 2024 formally reinstating tolerances for chlorpyrifos on certain crops. The agency’s own 2020 Proposed Interim Decision had already identified 11 specific crop uses it considered defensible, including soybeans, wheat, and several fruit and nut crops.
Several states have moved independently. California, Hawaii, and New York enacted their own bans on chlorpyrifos between 2019 and 2021, meaning exposure risk varies by geography. But for families in states without additional restrictions, the federal reinstatement means the pesticide can legally appear on food they buy.
The FDA’s Pesticide Residue Monitoring Report for fiscal year 2023, the most recent available, continues to sample domestic and imported commodities for organophosphate residues. Chlorpyrifos has not vanished from the detection record, though the report measures residues on food, not levels in human tissue.
The gaps that still matter
The Columbia MRI study documents structural brain differences, but it stops short of drawing a direct line from those differences to specific scores on cognitive or behavioral tests administered at the same ages. Cortical changes in regions tied to attention and language are suggestive, and separate publications from the same cohort have reported associations between chlorpyrifos exposure and lower IQ and working-memory deficits. Yet no single paper has formally linked the imaging findings to those functional outcomes in the same children at the same time point.
A second gap is biomonitoring. No publicly available federal dataset provides individual-level chlorpyrifos measurements for pregnant women collected after the 2021 tolerance revocation and subsequent reinstatement. The CDC’s National Health and Nutrition Examination Survey (NHANES) tracks some organophosphate metabolites, but reporting lags and changes in analyte panels make it difficult to assess whether prenatal exposure has meaningfully declined in the post-2021 period.
Perhaps the most consequential gap is the disconnect between the neuroscience and the regulation. The EPA’s risk assessments and the Columbia imaging data have not been formally reconciled. Whether the 11 retained crop uses produce exposure levels high enough to reach the cord-blood concentrations associated with brain changes is a question current public documents do not answer. Without that bridge, families are left to estimate their own risk with incomplete information.
Individual biology adds another layer of uncertainty. Genetic variation in paraoxonase 1 (PON1), an enzyme that helps metabolize organophosphates, means some people break down chlorpyrifos far more slowly than others. Nutritional status and simultaneous exposure to other chemicals can also shift the equation. The imaging study did not fully characterize these modifiers, so defining a single “safe” threshold for every pregnancy remains out of reach.
How to weigh the evidence
For readers sorting through conflicting headlines, it helps to think in tiers. The top tier is the primary neuroimaging research: prospective design, objective biomarker, quantifiable brain measurements. It is far more reliable than retrospective surveys or studies that estimate exposure from questionnaires alone. Corroborating it are additional publications from the Columbia cohort and from other research groups that have linked prenatal organophosphate exposure to neurodevelopmental outcomes, though not all use the same metrics.
The second tier is the federal regulatory record. EPA notices, court rulings, and FDA monitoring reports tell you what the government has permitted or prohibited. They are authoritative on policy status but cannot, by themselves, tell a family whether a specific apple or bag of flour carries a dose that matters for a developing fetus.
The third tier is interpretive commentary that tries to bridge the science and the regulation. Claims about whether current legal limits adequately protect fetal brain development are judgment calls, not settled facts. The American Academy of Pediatrics stated in a 2012 policy report that children face “unique susceptibility” to pesticide exposure and called for stronger protections, a position the organization has not retracted. But translating that broad stance into advice for a specific pregnancy still requires exposure data that do not yet exist in published form.
What families and doctors can do now
Expectant parents who want to reduce organophosphate exposure have several concrete options. Washing and, where practical, peeling produce lowers surface residues. Choosing organic versions of crops known to carry higher pesticide loads (the Environmental Working Group’s annual “Dirty Dozen” list is one commonly referenced guide) can further reduce dietary intake. For families in agricultural areas, keeping windows closed during spraying seasons and removing work clothes before entering living spaces limits take-home exposure.
Clinicians can fold the current evidence into prenatal counseling without overstating it. It is accurate to say that higher prenatal chlorpyrifos exposure has been associated with measurable differences in brain structure in school-age children, and that the chemical remains in limited legal use on certain U.S. crops. It is equally important to note that the precise impact on any one child’s cognition or behavior cannot yet be predicted from exposure data alone.
For families with unavoidable occupational exposure, pediatricians may consider prioritizing developmental screening in early childhood so that any emerging delays in language, attention, or executive function can be caught and addressed early. Early intervention services, where available, have their own evidence base for improving outcomes in children with neurodevelopmental risk factors.
The picture, as of mid-2026, is one of partial but genuinely concerning knowledge. The science connecting higher prenatal chlorpyrifos exposure to detectable brain differences in children has held up across multiple analyses of the same cohort and is consistent with findings from other research groups. What the record still lacks is a definitive exposure threshold below which harm can be ruled out, and a clear accounting of how the reinstated crop uses translate into fetal doses. Until those pieces fall into place, the most honest guidance is also the most frustrating: reduce exposure where you can, screen early, and keep watching the science.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
