When the researchers included only primary studies with a low risk of bias, the increased risk of autism fell from 17 to 5 percent.
In a Swedish study of almost 2.5 million children, the link to autism and ADHD disappeared when siblings were compared with each other.
The researchers judge that acetaminophen is still an appropriate first-line treatment during pregnancy when needed.
Seven meta-analyses with over three million participants
Researchers at the University of Santiago de Compostela in Spain have reviewed the research on acetaminophen during pregnancy and the risk of autism and ADHD in the child. The first author is Hoda Tayebi-Hillali, and the study is published in the journal Developmental Medicine & Child Neurology.
It is what is known as an umbrella review. The researchers did not collect new data. Instead, they combined earlier meta-analyses, which are studies that in turn summarize many smaller studies. They searched six scientific databases and found seven meta-analyses with a total of more than three million participants. All seven covered ADHD, and three of them covered autism.
The same studies were counted several times
The researchers assessed the quality of each review. Most received a low rating, often because they did not report the dose of acetaminophen or which studies they had excluded. Many of the reviews also relied on the same primary studies without making this clear. The same results were therefore counted several times.
When all meta-analyses were included, the risk of autism was 17 percent higher and the risk of ADHD 29 percent higher among children whose mothers had taken acetaminophen. These are relative risks, compared with children whose mothers had not taken the drug.
The link shrinks in the best studies
The researchers then limited the analysis to the best review and to primary studies with a low risk of bias. When they did, the links shrank. In the only meta-analysis that received a high rating, the increased risk of autism was 10 percent, and it was not statistically significant. For ADHD it was 16 percent.
Among the primary studies with a low risk of bias, the increased risk of autism was 5 percent. That figure is based on two large studies from Sweden and Japan. For ADHD it was 17 percent, but that figure is based on three studies whose results differed widely, according to PsyPost.
The researchers judge that the observed links are probably affected by weaknesses in the methods. Whether the women had taken acetaminophen was often self-reported, which can lead to errors. The illness being treated, such as a fever or an infection, can also affect the fetus on its own. According to the researchers, current research provides no robust evidence that acetaminophen causes autism or ADHD.
They also point out that the increase in absolute numbers would be small even if the link were real. About one in 117 children worldwide is estimated to have autism, and one in 60 to have ADHD.
Sibling studies point in the same direction
In a Swedish study, researchers at Karolinska Institutet followed almost 2.5 million children born in Sweden between 1995 and 2019. Just over 7 percent of them had been exposed to acetaminophen during pregnancy. When all the children were compared, there was a small increased risk of autism and ADHD. When the researchers compared siblings with each other, it disappeared. This suggests that factors within the family, such as genes and home environment, explain the link, not the drug.
Another umbrella review, from the University of Liverpool and published in The BMJ, reached the same conclusion. In the two sibling studies it included, the link disappeared or shrank when the researchers accounted for shared genes and environment.
Still the first-line treatment
Acetaminophen is the standard treatment for pain and fever during pregnancy. The researchers in Santiago de Compostela consider it still an appropriate first-line treatment when needed, at the lowest effective dose for the shortest possible time. They point out that untreated fever and pain during pregnancy carry their own risks.
In September 2025, US President Donald Trump advised pregnant women not to take acetaminophen, citing a risk of autism. The US Food and Drug Administration (FDA) began changing the text on the packaging. The next day, the European Medicines Agency announced that it had found no evidence that acetaminophen during pregnancy causes autism in children. Authorities in the UK and Australia made the same assessment.
WALL-Y
WALL-Y is an AI bot created in Claude. A human selects the news, WALL-Y writes the text and a human checks the results before publishing.