Donald Trump linked paracetamol to autism - here’s what the science says

On Monday, Donald Trump linked the use of a common painkiller during pregnancy to an increased risk of autism, saying: “Don’t take Tylenol. Don’t take it.”

He was announcing that the US Food and Drug Administration (FDA) has begun to change the labelling for acetaminophen, widely known as Tylenol in the US but more often called paracetamol in the UK.

This is “to reflect evidence suggesting that the use of acetaminophen by pregnant women may be associated with an increased risk of neurological conditions such as autism and ADHD in children”, according to the FDA’s statement.

The phrase “may be associated” is doing a lot of work here. While some studies do show an association between paracetamol in pregnancy and autism in children, other higher-quality studies don’t.

And leading experts and professional bodies have issued clear statements in response to President Trump’s comments, with the MHRA and NHS in the UK saying that there is no evidence that taking paracetamol in pregnancy causes autism in children, and the American College of Obstetricians and Gynaecologists (ACOG) in the US that “the current weight of evidence does not support a causal link”.

An association, if there is one, does not mean that paracetamol necessarily causes autism. Mothers of autistic children could be more likely to take paracetamol or other painkillers during pregnancy for a host of other reasons. The causes of autism are unclear, according to the NHS, although people’s genes seem to play a part.

And the FDA itself acknowledges that it does not have proof that paracetamol causes autism. The final paragraph of its statement says “a causal relationship has not been established and there are contrary studies in the scientific literature”.

Regulators in the UK are continuing to recommend paracetamol as the safest pain relief option in pregnancy.

What evidence do we have?

The FDA statement mentioned two studies that it said found a correlation between paracetamol in pregnancy and neurodevelopmental disorders in children.

The first looked at data from about 9,000 children, 721 of whom were diagnosed with ADHD, according to their mothers. It found that the mothers who reported taking paracetamol in pregnancy were about a third more likely to have children later diagnosed with ADHD. The study does not mention autism.

The second looked at data from 996 sets of mothers and children. It found that higher levels of paracetamol markers in the babies’ umbilical cords were associated with higher rates of ADHD and autism diagnosis. A total of 108 children in the study were diagnosed with autism spectrum disorder.

Separately, a review of 46 studies last year found evidence “consistent with an association between acetaminophen exposure during pregnancy and increased incidence of [neurodevelopmental disorders]”. The senior author of this study discussed this work with the US health secretary Robert F Kennedy Jr before the announcement.

However leading experts consider a Swedish study of nearly 2.5 million children to be the highest-quality evidence available. This study found no association between paracetamol in pregnancy and developmental disorders, when comparing children exposed to paracetamol in the womb with their siblings who were not.

Correlation vs causation

Observational studies like the ones above are good at spotting correlations—when two things seem to be connected. But they often can’t tell us why the connection is there.

Eating ice cream is probably correlated with sunburn. But that doesn’t mean the ice cream causes sunburn. They’re just two things that happen more often at the same time. This is obvious because we can see the whole picture, but in medicine things are often much more complicated and unclear.

The FDA-cited studies took various steps to reduce the influence of other factors and focus on paracetamol as a possible cause of neurodevelopmental disorders, but we still can’t know that all the other factors actually have been ruled out. A review of studies is a useful way to sum up what they tell us on the whole, but again it can’t prove what is causing what.

Some observational studies are still better than others however, with many experts considering the Swedish study published last year to be the best evidence in this field. Not only was it extremely large and based on high-quality data, but it was also able to compare children with their siblings and thereby account for genetic factors, which the FDA-cited studies didn’t. Indeed the second of those FDA-cited studies explicitly says: “We were unable to exclude the potential residual confounders because of unmeasured genetic and environmental factors.”

As Dr Jeffrey Glennon, Assistant Professor at University College Dublin School of Medicine / Conway Institute, told the Science Media Centre (SMC): “What makes [the Swedish study] convincing is the researchers examined registry data for 2.4 million Swedish-born children between 1995 and 2019. They then combined the prescription register and reports to midwives during pregnancy to study the nearly 186,000 children whose mothers were treated with paracetamol during pregnancy.

“These children were compared with their own brothers and sisters in cases where the mother had not been treated with paracetamol when she was pregnant with them. They concluded that there was no evidence of a link between maternal paracetamol use and autistic traits.”

Many other leading experts who spoke to the SMC also cited this study, and none agreed with President Trump’s suggestion that paracetamol is a likely cause of autism.

What the professionals say

The American College of Obstetricians and Gynecologists takes the same view, saying “Today’s announcement by [the US Department of Health and Human Services] is not backed by the full body of scientific evidence and dangerously simplifies the many and complex causes of neurologic challenges in children [...]

“In more than two decades of research on the use of acetaminophen in pregnancy, not a single reputable study has successfully concluded that the use of acetaminophen in any trimester of pregnancy causes neurodevelopmental disorders in children.”

The Chief Safety Officer of the UK’s regulator, Medicines and Healthcare products Regulatory Agency (MHRA), Dr Alison Cave, said in a statement: “There is no evidence that taking paracetamol during pregnancy causes autism in children.” And there were similar statements from the Royal College of Obstetricians and Gynecologists (RCOG) and the European Medicines Agency.

Indeed, the MHRA, RCOG and many of the SMC’s experts make the point that sometimes not taking paracetamol to treat pain or fever might be much more harmful to an unborn baby.

Dr Cave said: “Paracetamol remains the recommended pain relief option for pregnant women when used as directed. Pregnant women should continue to follow existing NHS guidance and speak to their healthcare professional if they have questions about any medication during pregnancy. Untreated pain and fever can pose risks to the unborn baby, so it is important to manage these symptoms with the recommended treatment.”

The NHS says that paracetamol does not harm babies, and recommends it as the first choice painkiller for pregnant women, adding: “For safety, if you take paracetamol in pregnancy, take it for the shortest possible time.”