Study finds no association between neonatal circumcision and autism diagnosis

Public concern over a possible association between neonatal medical male circumcision (NMC) and autism spectrum disorder (ASD) has persisted despite limited and inconsistent evidence. A new prospective cohort study published in JAMA Pediatrics found no evidence that NMC is associated with an increased risk of ASD or autism-related traits in children.¹

Investigators analyzed data from the National Institutes of Health-funded Environmental Influences on Child Health Outcomes (ECHO) Cohort, including 2,771 singleton male children from 14 US cohort sites enrolled between 2003 and 2025. Children were included if medical record data documented circumcision status within 28 days of birth and ASD outcome data were available.

Overall, 1,840 children (66%) underwent neonatal circumcision before hospital discharge, while 931 did not. ASD was identified in 192 children (7%), with a mean age at diagnosis of 3.8 years. Among circumcised boys, 108 (6%) had an autism diagnosis compared with 84 (9%) among uncircumcised boys.

How was autism assessed?

The primary outcome was ASD diagnosis based on either parent report of a diagnosis made by a medical professional or documentation of a diagnosis established using gold-standard clinical assessments.

Secondary outcomes evaluated autism-related characteristics using 2 validated instruments: the Social Responsiveness Scale, Second Edition (SRS-2), which measures autism-related traits, and the Child Behavior Checklist (CBCL) DSM-5 Autism Spectrum Problems subscale, which evaluates autism-related behaviors.

What did the study find?

In unadjusted analyses, circumcision appeared to be associated with lower odds of ASD. However, after investigators adjusted for maternal education, parental age, family history of ASD, and geographic region, the association was no longer significant (adjusted odds ratio [OR], 0.83; 95% CI, 0.59-1.17).

The findings remained consistent after stratification by US Census region, preterm birth, and neonatal intensive care unit admission, with no significant interactions identified.

Investigators also found no association between neonatal circumcision and autism-related traits measured by SRS-2 or autism-related behaviors measured by the CBCL. Adjusted analyses showed no significant differences for continuous or categorical measures on either instrument.

Multiple sensitivity analyses—including exclusion of a cohort enriched for familial ASD, restriction to children born before 2021, exclusion of children with a family history of ASD, and analyses limited to children without neonatal acetaminophen exposure—produced similar findings, supporting the robustness of the results.

What about acetaminophen exposure?

Because speculation has also focused on acetaminophen administered during circumcision, investigators evaluated its use during the neonatal period.²

Among circumcised infants with available procedural data, 4% received acetaminophen during or immediately after circumcision, and approximately 5% of the overall cohort received acetaminophen during the birth hospitalization. Because exposure was uncommon, investigators did not formally evaluate acetaminophen as a mediator of the association between circumcision and ASD.

In the discussion, the authors noted that acetaminophen use was uncommon, "further suggesting that an acetaminophen-mediated pathway linking NMC to ASD is unlikely."

How do these findings compare with prior studies?

The authors noted that earlier studies suggesting an association between circumcision and ASD were subject to important limitations. One ecological analysis relied on country-level data, while a Danish registry study may have been influenced by residual confounding because circumcision in Denmark is performed primarily for religious or cultural reasons. In contrast, hospital-based neonatal circumcision is routinely performed across a broader range of families in the United States.³

What were the study limitations?

The investigators acknowledged several limitations. Insurance status was incompletely captured, leaving the possibility of residual confounding related to access to hospital-based circumcision and ASD diagnosis. The analysis included only hospital-based neonatal circumcisions and may not generalize to procedures performed later in childhood or outside hospital settings. Additionally, the ECHO cohort does not represent all US male neonates, and acetaminophen use outside hospital records was not captured.

What is the clinical implication?

The authors concluded, "In this large, prospective US cohort study, our findings do not support an increased risk of ASD by NMC, whether assessed by parent-reported medical professional diagnosis or standardized measures." They added, "These findings may provide reassurance for families considering NMC and those who have already elected to have NMC for their child. Nonetheless, continued monitoring of both short- and long-term safety remains important for any medical procedure."

References

1. McGrath M, Li X, Jacobson LP, et al; for the ECHO Cohort Consortium. Neonatal medical male circumcision and child autism diagnosis. JAMA Pediatr. Published online August 3, 2026. doi:10.1001/jamapediatrics.2026.3239.

2. D’Antonio F, Flacco ME, Valle LD, et al. Prenatal paracetamol exposure and child neurodevelopment: a systematic review and meta-analysis. The Lancet Obstetrics, Gynaecology, & Women’s Health. Published online January 2026. doi:https://doi.org/10.1016/s3050-5038(25)00211-0

3. Frisch M,SimonsenJ. Ritual circumcision and risk of autism spectrum disorder in 0- to9-year-old boys: national cohort study in Denmark. JR Soc Med. 2015;108(7):266-279. doi:10.1177/ 0141076814565942