On September 22, 2025 the Department of Health and Human Services (HHS) announced that it would be changing the safety labelling of acetaminophen (brand name Tylenol) to highlight the possibility that exposure to the drug in utero could raise the likelihood of a child later being diagnosed with autism.
At a press conference held the same day as the announcement, President Trump addressed pregnant women directly, saying, “You’ll be uncomfortable. It won’t be as easy, maybe. But don’t take it if you’re pregnant. Don’t take Tylenol.”
In contrast, the written announcement from HHS struck a more measured tone, noting that while some studies suggest a link between acetaminophen use during pregnancy and autism, others have found no such association.
Further, experts, including FDA Commissioner Marty Makary, point out that acetaminophen remains the safest drug to manage pain and fevers in pregnancy, and note that taking the drug is often safer than leaving such conditions untreated.
If the recent back-and-forth has you confused, it’s not you: Research on acetaminophen use in pregnancy is complex: the kinds of studies scientists can ethically do on pregnant women don’t allow for randomization — that is, researchers can’t randomly assign some women to take acetaminophen and ask others to avoid it.
In a controlled trial, which is the standard design used to analyze the majority of medications, participants are randomly assigned to a placebo group or a group given the actual drug. Without that randomization, it’s difficult to know whether the outcomes of these acetaminophen studies are due to the medicine itself or to the underlying factors that made study participants more likely to take acetaminophen to begin with, such as conditions that cause pain, genes that cause those conditions, or even growing up in an environment where acetaminophen was commonly used.
To illustrate the complexity of the issue, we’ve broken down the studies the administration pointed to on acetaminophen use in pregnancy below.
Types of Studies Used to Assess Birth Outcomes with Acetaminophen Use
Pregnant women are rarely included in randomized controlled trials. It would be unethical to randomize pregnant women into different groups, then give some of them a medication that researchers suspect could harm their infant, while leaving others with untreated pain or fever, which is known to raise the risk of adverse outcomes during pregnancy.
For that reason, most studies on acetaminophen use during pregnancy are observational. That means researchers observe individuals throughout their pregnancies and give them surveys or biological tests to determine whether they’ve used acetaminophen. Researchers then follow those women and their children to track health outcomes.
Observational studies are very helpful in identifying correlations that warrant further study. The Framingham Heart Study, an ongoing project that has tracked health outcomes since 1948, reported in 1957 that high cholesterol and high blood pressure were linked to a greater risk of heart disease. Later studies probed these connections more deeply, examining the impacts of genes and drugs that reduced cholesterol or blood pressure, controlling diets in animal models, and other aspects.
What’s more, since acetaminophen is used to treat pain and fevers, pregnant women who take it are likely already experiencing those conditions. That can make it difficult to determine whether certain outcomes are associated with acetaminophen exposure itself or by the underlying illness that led women to take it in the first place.
Scientists use a variety of statistical methods to try to tease out other factors — such as maternal or paternal age, conditions that cause pain, and even socioeconomic status — that might influence results. Still, while an observational study can identify likely associations (for example, taking acetaminophen during pregnancy was associated with a higher chance of having a child with ADHD or autism), an observational study cannot prove that acetaminophen caused a diagnosis.
Think of it like this: Autism diagnoses increased at just about the same rate as Home Depot’s stock price over the last 25 years, points out Erika Sanborn, a Sociology Ph.D. candidate at the University of Minnesota and research consultant with autism and ADHD, but few would suggest Home Depot caused autism.
What Research Did the HHS Announcement on Tylenol and Autism Cite?
The fact sheet that accompanied HHS’s press release linked to six studies, three of which found an association between acetaminophen exposure and an autism or ADHD diagnosis, and one of which found an association between acetaminophen exposure and certain other birth outcomes, such as a lower birth weight. Two studies found no association between acetaminophen exposure and autism or ADHD. Here are the details:
ADHD and Acetaminophen Use (February 2019)
Type of Study: Observational. Researchers gave women surveys to ask about regular acetaminophen use during pregnancy and whether their children were later diagnosed with ADHD.
Participants: 8,856 children born between 1993 and 2005
Main Findings: Using acetaminophen more than once or twice a week increased the likelihood of an ADHD diagnosis in childhood from 7.8 out of 100 children to 10.2, a difference of 2.4 per 100 children. This finding is sometimes described as a more than 30% increase in the likelihood of having a child diagnosed with ADHD. But that 30% refers to a relative change, not an absolute one. The absolute increase was 2.4 percentage points. In other words, the risk rose from 7.8% to 10.2%. Because 2.4 is about 30% of 7.8, that’s where the ‘30% increase’ figure comes from.
ADHD, Autism, and Acetaminophen Use (October 2019 )
Type of Study: Observational with a biomarker (a biological measure that represents someone’s health or a condition). In this case, the biomarkers were acetaminophen and its metabolites in umbilical cord plasma collected after birth. Detecting these biomarkers indicates that the infant was exposed to acetaminophen in utero. Researchers analyzed cord plasma after women gave birth and followed children to see if they were later diagnosed with ADHD or autism.
Participants: 996 infants born between 1998 and 2018
Main Findings: Signs of acetaminophen use in the cord plasma at the time of birth were associated with an increased chance of the child being diagnosed with ADHD or autism. The greater the exposure, the greater the likelihood of diagnoses.
ADHD and Acetaminophen Use, Sibling Study (June 2021)
Type of Study: Observational, sibling control study. In hopes of teasing out the impact of genetics (which could underlie conditions that lead women to use acetaminophen during pregnancy and also raise the likelihood of an ADHD diagnosis in a child), the researchers compared the chances of two siblings being diagnosed with ADHD when only one was exposed to acetaminophen in utero. Mothers took two surveys during their pregnancies and one after giving birth, which included questions about the number of days they had used acetaminophen. Researchers followed the children’s medical records until 2017 for a later ADHD diagnosis.
Participants: 26,613 children born between 1999 and 2008 from 12,902 families in Norway.
Main Findings: There was no increased probability of an ADHD diagnosis for children who were exposed to up to 28 days of acetaminophen in utero. While there was an increased likelihood of ADHD diagnosis after 29 days of acetaminophen exposure, study authors suspected that factors other than just acetaminophen exposure played a role.
ADHD, Autism and Acetaminophen Use (April 2022)
Type of Study: Observational with a biomarker. Researchers analyzed the meconium — a newborn’s first stool — for acetaminophen. They suggested that acetaminophen in the meconium likely represented exposure during the second or third trimesters of pregnancy.
Participants: 393 infants born between 2007 and 2009 in Canada
Main Findings: The study did not directly test whether acetaminophen use during pregnancy was associated with an increased likelihood of an autism or ADHD diagnosis. Instead, researchers asked whether the medication was correlated with certain adverse birth outcomes such as preterm births, pregnancy complications, and low or high birthweight. They theorized that these conditions may, in turn, raise the chances of a later neurodevelopmental diagnosis.
Researchers found that acetaminophen detected in meconium was not linked to infants being born small for gestational age, preterm, or to pregnancy complications such as preeclampsia or gestational diabetes. However, it was associated with babies being born slightly earlier — though not preterm — and weighing somewhat less.
ADHD, Autism, Intellectual Disability and Acetaminophen Use (April 2024 )
Type of Study: Observational, sibling control study. The researchers reviewed medical records of children born in Sweden who were exposed to acetaminophen in utero and siblings who were not exposed.
Participants: 2,480,797 children born in 1995 to 2019 in Sweden
Main Findings: There was a slight increase in ADHD, autism and intellectual disabilities among children who had been exposed to acetaminophen in utero in the entire study population, but that increase disappeared in the sibling analysis, suggesting the diagnoses were more likely to be explained by genetic and environmental factors shared within families than by acetaminophen use.
Autism, ADHD, Neurodevelopmental Disorders and Acetaminophen Use (August 2025)
Type of study: Meta-analysis. Researchers did not collect new data for this study. They combined data from eight observational studies on acetaminophen and autism spectrum disorders, which included infants born between 1995 and 2019 and 20 studies on acetaminophen and ADHD diagnoses, including infants born between 1974 and 2020.
It’s important to note that some of the studies mentioned above are included in this analysis.
Main findings: There was an increased likelihood of autism or ADHD in children whose mothers took acetaminophen while pregnant. The more acetaminophen taken, the higher the possibility of a diagnosis, according to some of the studies used.
Of Note: The senior author of this meta-analysis had previously been paid to provide expert testimony in 2023 linking acetaminophen use during pregnancy to neurodevelopmental disorders, before this study was published. The judge in that case ruled to exclude his testimony, noting that he had not published research supporting such a link and had, in fact, authored a 2022 paper cautioning against any changes to clinical practice. Additionally, some researchers question the methods of his 2025 analysis, which counted the results of the Swedish sibling study listed above, both with and without the sibling analysis.
What Do These Different Findings Mean?
Because the findings are inconsistent, and because observational studies make it very difficult to capture details like timing, dose, and even the brand of acetaminophen used (especially since generics can vary and nearly three-quarters of drugs and ingredients are now made overseas, where FDA inspections are harder to maintain) the implications remain uncertain. More research is needed to better understand the association.
Many scientists continue to urge caution about the administration’s new health declaration linking Tylenol and autism — not because the statement is necessarily wrong, but because the science remains unsettled. Jeremy Faust, M.D., an emergency physician at Brigham and Women’s Hospital, captured this sentiment in a September 6 Substack post:
What worries me is that this is not pseudoscience. It’s just not ready for primetime science and… if we get this backwards, we could cause as much or more harm.
Are There Any Alternatives to Taking Tylenol During Pregnancy?
Acetaminophen is used to treat fevers and pain such as headaches. Left untreated, these can have their own adverse effects on pregnancy. According to the Centers for Disease Control and Prevention (CDC), fevers above 101 F require treatment during pregnancy. Untreated fevers raise the risk of neural tube defects, congenital heart defects, oral clefts, neurodevelopmental disorders, and yes, the likelihood of autism.
Non-steroidal anti-inflammatory drugs (NSAIDS) such as ibuprofen and aspirin can also lower fevers and treat pain, but they’re not recommended in pregnancy because they’ve been linked to a higher risk of miscarriage.
Before reaching for acetaminophen, which the HHS report still suggests is the safest option during pregnancy, you can consider these non-drug options.
For a Fever
If the fever is mild, you can rest and hydrate, but if it is higher than 100.4 degrees Fahrenheit, it’s time to call your healthcare provider.
For Pain
Lillie Rosenthal, DO, a New York-based integrative pain management specialist and member of MedShadow’s Health and Medical Advisory Panel, recommends starting with tried and true strategies, including light exercise, stretching, posture modifications, and applying heat or ice to the affected area.
If these strategies do not reduce your fever or make your pain manageable, experts still recommend acetaminophen as the safest option during pregnancy. Pain and fever are signals from your body, and you should feel comfortable discussing them and the best ways to treat them with your healthcare provider, whether or not you are pregnant.
