What the ECHO Study Found on Circumcision and Autism

New research published in JAMA Pediatrics found no evidence that newborn circumcision is associated with autism spectrum disorder, offering a direct counter to claims made by top U.S. health officials. The analysis drew on prospective data from the NIH-funded ECHO cohort, covering 2,771 boys from 14 study sites. Among circumcised boys, 6% had an autism diagnosis, compared with 9% of uncircumcised boys — a gap that disappeared after adjusting for other factors (odds ratio 0.83, 95% CI 0.59-1.17).

The study was prompted in part by comments from Health Secretary Robert F. Kennedy Jr., who said in September 2025 that circumcised boys were twice as likely to develop autism because they received acetaminophen. The remarks followed a White House warning to pregnant women about an unproven link between acetaminophen and autism. Lead author Monica McGrath of the Johns Hopkins Bloomberg School of Public Health said the team began hearing from friends, family and colleagues after media coverage of those statements, which cited studies published more than a decade ago.

Beyond the autism diagnosis finding, the researchers found no association between circumcision and autism-related traits or behaviors, measured using the Social Responsiveness Scale and the Child Behavior Checklist. Most infants in the study received nerve blocks (72%), oral glucose or sucrose (59%) or anesthetic cream (12%) during the procedure; only 4% received acetaminophen during circumcision and 7% in the 30 days afterward.

McGrath said the findings may reassure families who are considering or have chosen circumcision for their son. The study has limitations — acetaminophen data came only from hospital prescription records, the analysis covered hospital circumcisions within 28 days of birth, and residual confounding is possible — but the authors said the results held up in analyses stratified by region, preterm birth and NICU admission.

Advertisement

Why the New Data Undercut the Administration's Claim

What the Numbers Do and Don't Show

The headline comparison — 6% versus 9% — could be misread as evidence that circumcision protects against autism. The study's adjusted analysis argues the opposite interpretation: the crude difference reflects the characteristics of families who choose circumcision, not the procedure itself. After adjustment, the odds ratio of 0.83 carried a confidence interval that crossed 1, meaning no statistically meaningful association. Stratified analyses by region, preterm birth and NICU admission produced inverse or null results, and separate analyses of autism traits and behaviors found no links either.

How a Weak Evidence Base Became Official Messaging

The September 2025 claim from Health Secretary Robert F. Kennedy Jr. rested on two studies that independent expert David Mandell of the University of Pennsylvania described as poorly conducted. The first was an ecological study that used circumcision as a proxy for acetaminophen use — it did not track which individual children were circumcised, took the drug or developed autism, and in some analyses used the proportion of Jewish or Muslim residents as a stand-in for circumcision rates. The second, a Danish registry study, hypothesized that pain and trauma from circumcision cause autism, a theory Mandell said has no evidence behind it. In Denmark, hospital circumcision is rare and largely limited to boys with perinatal complications and NICU stays — known risk factors for autism that the study did not control for.

What This Study Settles — and What It Doesn't

The ECHO analysis is prospective, peer-reviewed and draws on a national NIH-funded cohort, making it a substantially stronger piece of evidence than either study cited in the official messaging. But it is one observational study. Acetaminophen exposure outside the hospital was not captured, the sample skewed white, and only in-hospital circumcisions were included. The result is high-quality reassurance on the specific question of autism risk, not the final scientific word on every mechanism the administration raised.

What Families and Clinicians Should Take From the Findings

This is reassurance-level evidence, not a clinical recommendation. The practical takeaways depend on who is reading.

  • Families weighing the decision: the specific autism concern raised by HHS officials now has a direct, peer-reviewed answer from prospective U.S. data — no association between neonatal circumcision and autism diagnosis, traits or behaviors. The decision itself remains personal; this study addresses only the autism question, not other clinical considerations.
  • Clinicians fielding questions: parents may cite the September 2025 administration claims. The two studies behind them had serious methodological limits — one used circumcision as a proxy for acetaminophen use, the other did not control for NICU admission and perinatal complications that are themselves autism risk factors.
  • What to tell families about the evidence: among 2,771 boys in the ECHO cohort, the crude 6% versus 9% difference disappeared after adjustment (OR 0.83, 95% CI 0.59-1.17), and analyses of autism traits and behaviors were also null.
  • Honest limits: acetaminophen use was tracked only in hospital records, the sample was predominantly white, and only hospital circumcisions within 28 days of birth were included — so this is strong but not definitive evidence.