Prenatal exposure to organophosphate insecticides at the levels most Americans encounter was not associated with autism related traits in children, according to a federally funded study of 3,339 mother-child pairs published in JAMA Pediatrics.
That qualifier about the exposure level is the load-bearing part of the finding, and it should not be trimmed. The study measured ordinary, largely dietary exposure, the kind that comes from eating fresh produce. It does not address the far higher exposures faced by agricultural workers, pesticide applicators, or families living immediately beside treated fields.
For parents who have encountered claims connecting pesticides on food to autism, this is the largest and most diverse test of that specific question to date, and it came back negative. For farmworker families, the question remains open.
Urine Markers from Pregnancy, Behavior Scores at Age Six
Researchers led by a team at NYU Langone Health analyzed data from mother and child pairs enrolled at 20 medical centers in the National Institutes of Health Environmental Influences on Child Health Outcomes program, a long-running national research effort designed to follow children from before birth.
Exposure was measured directly rather than estimated. Investigators tested urine samples collected during each pregnancy for the metabolites that organophosphate pesticides break down into in the body. Nearly all mothers had detectable levels, a fact that speaks to how routine this exposure is.
Years later, parents completed the Social Responsiveness Scale, a questionnaire that rates social communication difficulties and autism related characteristics on a spectrum rather than as a yes-or-no diagnosis. That design captures children who show many autism traits without a formal diagnosis. Children were assessed at a mean age of 6.4 years.
Some inverse associations appeared in crude and minimally adjusted models. After full adjustment for other factors, those associations weakened and lost statistical significance. Analyses separated by child sex, alternative statistical approaches, multiple sensitivity checks, and adjustment for maternal diet all pointed in the same direction, according to a published summary of the findings.
The Exposure Levels Studied Were Ordinary, Not Occupational
Lead author Haleigh Cavalier, who conducted the work as a doctoral student in NYU Langone Health's departments of population health and pediatrics, said in a university news release that these chemicals do not appear to be a major driver of autism related characteristics at the relatively low exposure levels observed.
She was equally clear about the boundaries. The results should not be interpreted as declaring these pesticides safe, and they do not rule out risks at higher exposure levels or among more vulnerable groups. Organophosphates are already linked in prior research to other neurodevelopmental harms, including IQ loss, cognitive difficulties and attention deficit hyperactivity disorder. Those associations are not overturned by this study, which examined a different outcome.
Cavalier also offered a possible explanation for the null result that is worth sitting with. Outside agricultural and occupational settings, most Americans meet these insecticides through fresh fruit and vegetables. The nutritional benefits of a produce-rich diet during pregnancy may partly offset any effect of residue on brain development.
Null Results From Prospective Cohorts Carry Particular Weight
Readers accustomed to studies that find something may treat a negative result as less newsworthy. In this case, the design argues for the opposite.
Smaller studies on this question have produced inconsistent results, with some finding a connection and others not. Retrospective work examines exposure after an outcome is known, which invites recall bias and selective analysis. A prospective cohort collects biological samples during pregnancy, before anyone knows how a child will develop, which removes that particular failure mode.
Scale and diversity matter too. Pooling 3,339 pairs from 20 sites across the country allowed the team to account for environmental, social, and biological factors that smaller studies could not.
The limitations are stated by the authors rather than buried. Organophosphate metabolites are short-lived in the body and vary with each person's metabolism, so a urine sample captures a narrow window and may not represent exposure across an entire pregnancy. Parent-reported behavior scales are not clinical diagnoses. The study cannot speak to pesticide classes it did not measure. Senior author Akhgar Ghassabian of NYU Grossman School of Medicine said the team plans to examine other pesticides, microplastics, and additional toxins, as well as how autism-linked genes interact with environmental exposures. Funding came from National Institutes of Health grants.
Produce, Diet and the Advice That Has Not Changed
Nothing here changes prenatal guidance. Pregnant people should continue eating fruits and vegetables. Washing produce under running water reduces surface residue, and buying organic is an option for households that can afford it, but neither the study nor its authors suggest reducing produce intake, which would trade a hypothetical risk for a documented nutritional loss.
Families with genuine occupational exposure are in a different position. Agricultural workers, pesticide applicators, greenhouse and nursery staff, and people living adjacent to treated fields face exposures this study did not evaluate. Pregnant workers in those settings can ask an employer about protective equipment, changing clothes and showering before entering the home, and can raise the exposure with a prenatal care provider. Occupational health resources and state agricultural worker safety programs exist in most farming states.
Parents who are worried about a child's development should not use this or any single study as a substitute for evaluation. Concerns about speech, social interaction, eye contact or repetitive behaviors are worth raising with a pediatrician, who can arrange developmental screening and referral. Early evaluation carries real benefit regardless of what caused a difference.
MedicalDaily has previously covered research on other prenatal environmental exposures and will continue to track the ECHO program as further analyses are published.
Key Questions Answered
What did the study find? No significant association between prenatal organophosphate pesticide exposure and autism related traits in children, across 3,339 mother-child pairs from 20 sites in a national research program.
What exposure levels does this cover? Ordinary exposure, mostly through diet. It does not address occupational or agricultural exposure, which is substantially higher and was not evaluated.
Does this mean organophosphate pesticides are safe? No. The lead author explicitly cautioned against that reading. Prior research links these chemicals to other neurodevelopmental effects, including IQ loss and attention deficit hyperactivity disorder.
How was exposure measured? Through metabolites in urine samples collected during each pregnancy. Nearly all participating mothers had detectable levels.
How were autism traits assessed? Using the Social Responsiveness Scale, a parent questionnaire rating social communication difficulties on a spectrum. Children were evaluated at a mean age of 6.4 years. It is a screening tool, not a diagnosis.
Should pregnant people eat less fruit and vegetables? No. Neither the study nor its authors suggest that. Washing produce under running water reduces surface residue, and the nutritional value of produce during pregnancy is well established.
Who funded the research? National Institutes of Health grants supported the work.
