A major new study from Denmark, one of the largest of its kind, has found no evidence to support claims that taking acetaminophen during pregnancy increases a child’s risk of autism, delivering a robust challenge to recent political rhetoric that has altered medical practice.
The Definitive Data: A Study of 1.5 Million Children
Researchers in Denmark leveraged the country’s comprehensive national health registries to track every child born over a 25-year period, from 1997 to 2022—a cohort totalling more than 1.5 million individuals. Within this population, they identified 31,098 children who had been exposed to acetaminophen, known as paracetamol in the UK and often branded as Tylenol in the US, while in the womb. The study, published in JAMA Pediatrics, concluded that autism was diagnosed in 1.8% of children exposed to the medication and 3% of those who were not, finding no statistically significant link.
The scale and design of the research are central to its authority. By using a national registry, the study avoided the recall bias that can affect smaller studies and provided immense statistical power. Crucially, its methodology aligns with other major international research, including large-scale studies from Sweden. Those studies found that any marginal associations initially observed disappeared when comparing siblings, strongly suggesting that genetic or shared environmental factors, not the medication, explain any perceived link.
This growing scientific consensus is echoed in broader analyses. An umbrella review of existing systematic reviews published in the BMJ concluded that the quality of evidence supporting a link between prenatal acetaminophen use and neurodevelopmental conditions is “low to critically low.” Experts consistently point to confounding factors—such as the underlying health conditions or genetics of the pregnant person, which might lead to both higher pain levels and a predisposition to autism—as the likely explanation for any observed association.
Political Claims and a Measurable Impact on Care
This scientific backdrop stands in stark contrast to recent statements from the highest levels of US government. In September 2025, the Trump administration announced that the US Food and Drug Administration (FDA) would initiate a label change for acetaminophen to warn of a potential association with autism. Former President Donald Trump himself repeatedly cautioned against its use, stating at a press conference, “If you’re pregnant, don’t take Tylenol… It’s not good,” and claiming it was associated with a “very increased risk of autism.”
This political messaging had an immediate and measurable effect on patient behaviour. A subsequent study published in The Lancet found that in the weeks following the September announcement, orders of Tylenol for pregnant women in emergency departments dropped by 16%. At its lowest point three weeks after the announcement, the order rate fell by 20%. The study’s co-author, Dr Jeremy Faust of Mass General Brigham and Harvard Medical School, confirmed “words are affecting behaviour,” noting there was no similar decline in orders for non-pregnant women or for other medications like IV fluids or opioids.
At the same September press conference, officials also promoted leucovorin, a B vitamin, as a treatment for autism. That claim has since unravelled; one of the largest studies on leucovorin’s effectiveness for autism traits has been retracted due to data inconsistencies, and the American Academy of Pediatrics does not recommend its routine use for autistic children due to limited evidence.
Global Reassurance and the Risks of Untreated Illness
In response to the political claims, health authorities and professional bodies worldwide have moved to reaffirm the drug’s safety profile during pregnancy. The UK’s Medicines and Healthcare products Regulatory Agency (MHRA) states there is no evidence linking paracetamol to autism, a position supported by the Royal College of Obstetricians and Gynaecologists (RCOG). The NHS continues to advise it as the first-choice pain reliever for pregnant women.
Similarly, the European Medicines Agency (EMA) has stated paracetamol can be used if clinically needed and sees no new evidence requiring changes to EU recommendations. In the US, while the FDA explores a label change, it acknowledges a causal relationship has not been established and that contrary studies exist. Major professional organisations, including the American College of Obstetricians and Gynecologists (ACOG) and the American Academy of Pediatrics (AAP), continue to recommend acetaminophen as the preferred medication for pain and fever in pregnancy, expressing concern that misleading claims are causing harm.
Experts emphasise that the risks of avoiding necessary medication are significant. “Pain is, in itself, a problem,” said Dr Faust, who also noted that untreated fever and pain during pregnancy can lead to serious complications, including miscarriage, malformations, and preterm delivery. The alternatives are limited; non-steroidal anti-inflammatory drugs like ibuprofen are generally not recommended due to bleeding risks and potential placenta issues, while opioids carry risks of dependence and neonatal withdrawal. This leaves acetaminophen, when used as directed, as the safest available option.
This scientific position was underscored in late 2024 when a US judge dismissed over 500 lawsuits against acetaminophen manufacturers, ruling that the plaintiffs’ expert witnesses lacked sufficient scientific evidence to support the alleged link to autism.
