Does Tylenol Increase Autism Risk? A New Study Has Answers

Prenatal exposure to acetaminophen (Tylenol) was not associated with an increased risk of autism in offspring, with or without adjustments for sibling comparisons, nationwide data from Denmark showed.

The study of more than 1.5 million children born between 1997 and 2022 found no statistically significant link between acetaminophen exposure during pregnancy and autism in either general population (HR 1.03, 95% CI 0.95-1.12) or sibling-matched analyses (HR 1.09, 95% CI 0.91-1.27), reported Kira Philipsen Prahm, MD, PhD, of Copenhagen University Hospital Rigshospitalet, and colleagues.

No significant association between exposure and autism emerged in dose-response patterns or in exposure during different trimesters, Prahm and colleagues wrote in a JAMA Pediatrics research letter.

“Overall, our findings showed no association between the use of acetaminophen during pregnancy and risk of autism in offspring,” Prahm said.

“Recent political statements, together with widespread media coverage, have directed attention to a potential link between acetaminophen and autism. Given recent concerns, we found it relevant to make a well-performed study to either confirm or refute a potential association,” Prahm told MedPage Today.

In September 2025, President Donald Trump warned women about using acetaminophen in pregnancy. A 2024 Swedish study had reported a small but statistically significant increase in autism risk among children at the population level that was not seen in sibling-matched analyses, raising questions about residual confounding, Prahm and colleagues said.

Similar findings emerged in a study from Taiwan this year, but parts of the sibling analysis prevented firm conclusions from being drawn. However, a 2025 study from Japan found no link between prenatal exposure to acetaminophen and autism in offspring in either the general population or between-sibling comparisons.

“What we see are studies coming out of multiple different countries that are showing the same thing, which is that when you carefully adjust — or sometimes even when you don’t — for factors, Tylenol doesn’t cause autism,” noted David Mandell, ScD, of the University of Pennsylvania Perelman School of Medicine in Philadelphia, who wasn’t involved with the study.

“A mentor of mine says that science stands on two legs: the weak leg is peer review, and the strong leg is replication,” Mandell told MedPage Today.

“What we look for is the confluence of evidence over multiple studies done by independent groups. So, now we have yet another very strong study in a large sample that finds, even without the sibling adjustment, acetaminophen doesn’t cause autism,” he pointed out.

And notably, even those prior study analyses that did find a statistically significant link found only very small effect sizes, Mandell added.

While some researchers have raised questions about sibling comparisons, others maintain they are valuable to help account for shared confounding variables within a family.

“When we just look at the epidemiology of children with neurodevelopmental disorders, their mothers are much more likely to report infection and headache or migraine during pregnancy,” Mandell observed. “The best way to control for genetic factors or biological factors on the part of the mother is to compare siblings to each other.”

Prahm and colleagues studied all children who were alive at age 1 year and were born from singleton pregnancies in Denmark between January 1, 1997, and July 31, 2022. Prenatal acetaminophen exposure was identified by national prescription records.

Children were followed from age 1 year until July 31, 2023, or until an autism diagnosis, whichever came first. Findings were adjusted for maternal use of other analgesics and other confounding variables.

Of 1.5 million children, 31,098 (2.1%) were exposed to prescription acetaminophen during pregnancy. A total of 554 exposed children (1.8%) were later diagnosed with autism, compared with 44,667 children (3.0%) in the unexposed group.

Women prescribed acetaminophen during pregnancy generally were older (median age 31.2 vs 30.5 years) and had higher parity, higher body mass index, and more comorbidities. They also used more prescription medicine than women who weren’t prescribed acetaminophen.

Study limitations included possible outcome misclassification, although the positive predictive value of an autism diagnosis has been found to be 94%, Prahm and co-authors noted.

“Individual-level information about over-the-counter medication was unavailable; thus, the true exposure level among those with low-level exposure was likely underestimated,” they acknowledged. “However, previous simulation studies of over-the-counter drugs have shown such bias to be largely negligible.”