What we know, and don’t, about the link between painkillers and autism

On Monday, the US government announced two initiatives related to autism. It will update labelling on the common painkiller paracetamol, also called acetaminophen, warning use during pregnancy may increase the risk of autism and ADHD in children. It will also approve a drug called leucovorin for use in some children with autism spectrum disorder (ASD). Neither of these moves is based on robust scientific findings, and they are likely to have minimal effect on autism rates in the US.

The announcement comes after the country’s highest ranking public health official, Robert F Kennedy Jr, pledged in April to identify the causes of autism by the end of September. US President Donald Trump touted the initiatives as “historic steps to confront the crisis of autism” during a press conference.

This framing reveals either a misunderstanding or a misrepresentation of what we know about autism. There is no crisis – rates of autism began to increase rapidly in the 1980s as diagnostic criteria for the condition expanded. Over the past few decades, a higher level of awareness and recognition of the symptoms of autism has also contributed to an increase in autism diagnoses.

As for the cause of the condition, there is little evidence suggesting paracetamol is to blame. It is true multiple studies have suggested paracetamol use in pregnancy may increase the risk of autism in children. A recent analysis of 46 studies found 27 of them identified a significant association between using paracetamol in pregnancy and children having a higher risk of neurodevelopmental conditions such as autism. But just because there is a relationship doesn’t mean paracetamol causes autism – other factors could be at play.

And there are conflicting findings from other studies. For example, a 2024 study of nearly 2.5 million children found a slightly higher risk of autism in those exposed to paracetamol during pregnancy, but after comparing children exposed to the painkiller with those of their unexposed siblings, the effect disappeared. “It was the family history that mattered and not the use of paracetamol,” said Dimitrios Siassakos at University College London in a press statement. “Research has shown that any apparent marginal increase [in autism] as a result of paracetamol in pregnancy tends to disappear when the analyses take into account the factors that matter the most”, such as genetics, which decades of research has indicated is a primary contributor to the condition.

During the press conference, Trump said pregnant women shouldn’t take paracetamol for pain or fever during pregnancy unless they “can’t tough it out”, and they “should talk to their doctors for more information”. However, the US Food and Drug Administration (FDA) acknowledged in a press release a causal relationship between paracetamol and autism hasn’t been established and said in a notice to physicians paracetamol remains the safest widely available painkiller for use during pregnancy. “[It] is the only over-the-counter drug approved for use to treat fevers during pregnancy, and high fevers in pregnant women can pose a risk to their children,” the agency said. Fever during pregnancy has been linked to pre-term birth and defects in brain or spinal cord development in the fetus.

The FDA also announced it will approve the drug leucovorin for people with cerebral folate deficiency, a condition that impairs uptake of vitamin B9 in the brain. Some research indicates upwards of 40 per cent of autistic people may also have this condition, which results in similar symptoms, including communication and sensory processing issues.

Leucovorin, a form of vitamin B9, is already available for treating deficiencies of this vitamin as well as side effects from certain cancer medications. Early evidence suggests it might improve some severe autism symptoms, too. For instance, a 2016 study treated 23 autistic children with language impairments using two doses of leucovorin daily while a separate group of 25 children received a placebo. After 12 weeks, 65 per cent of children receiving leucovorin saw a clinically meaningful improvement in verbal communication compared with 24 per cent of those in the control group.

But so far, all of the studies on leucovorin and autism have been small, and most indicate only modest improvements. A larger clinical trial is underway, the results of which are expected next year. “Any evidence available at this point is exceptionally tentative and would not be considered robust,” said Monique Botha at Durham University in the UK in a press statement. “Whilst medications may help with very specific aspects [of autism], there is no medication or treatment that actively cures or erases autism.”

Following the announcement, the US Department of Health & Human Services clarified in a statement that “while promising, it is important to note that leucovorin is not a cure for ASD and may only lead to improvements in speech-related deficits for a subset of children with ASD”.

Kennedy, with Trump’s backing, has made tackling the “autism epidemic” a pillar of US public health policy. They both portrayed these policy changes as remarkable progress made in a short period of time. But the science suggests neither avoiding paracetamol in pregnancy nor taking leucovorin will significantly affect autism as Trump and Kennedy have claimed. Instead, the most likely outcome of these initiatives is they will sow confusion around how and when to safely treat pain or fever during pregnancy.