Is Tylenol Use During Pregnancy Connected To Autism?

[MUSIC PLAYING] FLORA LICHTMAN: This is Science Friday. I’m Flora Lichtman. If you tuned into the news at all this week, you likely heard this new medical guidance from the Trump administration.

DONALD TRUMP: Acetaminophen– is that OK?

SPEAKER: Yes. Yes.

DONALD TRUMP: Which is basically commonly known as Tylenol, during pregnancy, can be associated with a very increased risk of autism. So taking Tylenol is, uh, not good.

FLORA LICHTMAN: You may have also heard the backlash to this advice from many experts who disagree and worry about the risks of avoiding Tylenol during pregnancy. And then there are concerns coming from the autistic community around the narrative that autism is inherently bad, a scourge that needs to be eliminated. And there are concerns that moms are being blamed for having autistic children. There’s a lot to unpack about genetics, about vaccines.

But today I want to focus on the acetaminophen piece. If I saw these headlines while I was pregnant, I would want the nitty gritty– where this idea came from and what exactly researchers have found. So today, we are digging into the details with an epidemiologist who led one of the largest peer-reviewed studies looking at this exact question– what do we know about the link between acetaminophen and autism? Dr. Brian Lee is an epidemiologist at Drexel University and the senior author on a paper in JAMA about acetaminophen use during pregnancy and children’s risk of autism and other neurodevelopmental outcomes. Brian, welcome to Science Friday.

BRIAN LEE: Hi. Thank you for having me.

FLORA LICHTMAN: So this Tylenol linked to autism claim didn’t come out of nowhere. You have looked at this association yourself. What made you decide to study this in the first place?

BRIAN LEE: Well, acetaminophen is one of the most widely used medications in the world. And I think it is proper to question whether or not a medication during pregnancy is going to be safe. Our team looked at the available evidence. And basically, there are methodological limitations to the existing evidence that our study wanted to overcome.

FLORA LICHTMAN: Before we get into your study, what was the available evidence? Because the fact sheet provided by the administration cites this review article, for example, from the Journal of Environmental Health that looks at over 40 studies and finds over half of them show this association between taking Tylenol and pregnancy and autism. We invited one of those authors on the show, by the way, and she declined. But can you tell me a little bit about this review and what your reservations were about the evidence?

BRIAN LEE: Sure. So this review, it’s important to note it’s just a new look at old existing studies. And they looked at somewhere over 40 studies of acetaminophen use and various neurodevelopmental developmental outcomes. And the conclusion they came up with was that based on the available evidence, acetaminophen use during pregnancy was associated with increased risk of autism. The problem, however, is that association is not causation, right?

FLORA LICHTMAN: Right. OK, let’s talk about the study that you did. What was your approach to this question?

BRIAN LEE: Yea. So our study relied on looking at this amazing data resource of 2.5 million pregnancies, looking at the mothers and their children followed for over 20 years in Sweden. Importantly, we didn’t ask a mom, hey, 20 years ago, what did you take during your pregnancy, right? This was something that we were able to look at midwife interviews at their prenatal visits, where they asked questions about medication use. And we also had prescription drug records available. And then we also had the electronic health record.

But there’s a couple missing pieces here. Like, the reason you take acetaminophen is, you have a headache. You have an infection, fever, pregnancy pain, stuff like that. Is it going to be captured in data sources? Our data, for example, only captures receipt of clinical services. So if you just have a random headache, you’re not necessarily going to go in to the hospital to get that checked out.

FLORA LICHTMAN: So you don’t know why people are taking the Tylenol.

BRIAN LEE: Not necessarily. You can try and adjust for statistical differences with the data you have. But if you don’t have data on something, you can’t adjust for it or account for it, OK? And so these indications for use of acetaminophen– migraines, headache, infection, fever– these sorts of things, incidentally, are associated with increased risk of autism. So there’s already the possibility right there that it’s not necessarily the acetaminophen that is increasing risk, but the reason you’re taking it.

FLORA LICHTMAN: OK, let me see if I can sum that up. So one of the reasons why doing these epidemiological studies to try to untangle this link is, one of the reasons it’s difficult to do them is that you don’t have data on why people are taking the Tylenol. And the reason people might be taking the Tylenol also could be a risk factor for the outcomes that you’re looking at.

BRIAN LEE: Exactly. So it basically comes down to this. What you want is an apples-to-apples comparison of the women who use acetaminophen versus the women who don’t use. But users of medication are going to be different in many different ways from non-users.

And the biggest thing, of course, is, you don’t take medication for fun. You’re taking it for a health reason. And so the women who use acetaminophen are going to be sicker, in some way, than non-users. And so what you’re really doing, then, is an apples-to-oranges comparison. And this makes it challenging to look at the effects of acetaminophen on neurodevelopmental outcomes.

FLORA LICHTMAN: OK, that’s one of the challenges. How do you get around it?

BRIAN LEE: Yeah. And then I should mention the elephant in the room that people don’t often talk about in these studies is genetics. So autism and other neurodevelopmental disorders, they are highly heritable. Interestingly enough, Mom’s genetic risk for neurodevelopmental disorders, like autism and ADHD, has also been linked with greater pregnancy pain, headaches, migraines, more use of pain relief medications, and in fact, more use of acetaminophen.

FLORA LICHTMAN: So if your genetic profile makes you more likely to have a child with autism, because of genetics alone, you may also experience more pain in your pregnancy, that you might treat them with Tylenol.

BRIAN LEE: Exactly. Right? So you can see this is a really complex issue because you’re trying to isolate out the independent effect of acetaminophen, if there is one. But you have all of these other factors at play here. And in these large population-based studies– I should mention that our study has 2.5 million pregnancies. The basis of most of these arguments by the administration are on studies with the largest caps out around, like, 50,000 pregnancies or so.

But the basis of these studies, they usually do not have genetic information. We didn’t either, but we were able to get around this with a statistical approach. I think that’s kind of interesting. So we actually have the entire population of Sweden at our disposal. We have family units where we can go and look for the families where there were one sibling who was exposed to acetaminophen in the womb, and compare their outcomes to their unexposed sibling.

And what happened was, without the sibling control, we saw the exact same statistical association that, oh, acetaminophen supposedly increases risk of autism. But when we did the sibling control analysis, that small statistical association completely flatlined. It disappeared. In other words, there was no evidence to support that acetaminophen causes autism.

FLORA LICHTMAN: OK, so you looked at families where there were multiple siblings. And these were pregnancies where the parent took Tylenol during one pregnancy, but not the other. And then you compared to see, did those kids who were exposed to Tylenol in utero, did they have a higher likelihood of having autism than their sibling? And the answer was no.

BRIAN LEE: Exactly. Because if acetaminophen really does cause autism, then the sibling who is exposed to it should have a higher probability of autism than the sibling who was not exposed, right?

FLORA LICHTMAN: Right. One of the findings of the review study cited by the administration is that it looked like dose mattered, that prolonged use of Tylenol in pregnancy, taking Tylenol for over four weeks resulted in this stronger association with autism. Is that evidence of causation?

BRIAN LEE: This is one of those things where you could potentially look at it two different ways. Like, oh, yeah, if you take it for longer, that’s a clear dose response. But our take on this is that, why are you taking it for longer? It’s because you have more infection. You have more pregnancy pain. You just have more of that health condition. So yeah, in my opinion, it doesn’t strongly support the argument for causality.

FLORA LICHTMAN: How would you counsel a pregnant person in your family? We know there are risks of not treating fevers during pregnancy. And at the same time, anyone who’s been pregnant knows that so much of pregnancy is navigating these uncertainties. So what would you tell them? And do you think there’s any reason for caution around using Tylenol when you’re pregnant?

BRIAN LEE: I think caution is warranted for everything. Just because walking on the street is safe, statistically safe, doesn’t mean you shouldn’t keep your eyes open and watch out for the banana peel. But bringing it back to acetaminophen, the strongest and best evidence available to date doesn’t suggest that acetaminophen causes autism.

But that doesn’t mean, of course, that you start popping acetaminophen like candy. Acetaminophen has a known liver toxicity issue. And so the long story short, the evidence to date suggests no causal effect of acetaminophen on autism. And folks can do with that what they will.

FLORA LICHTMAN: I think that’s the perfect place to leave it. Thank you for joining us today, Brian.

BRIAN LEE: All right. Thank you.

FLORA LICHTMAN: Dr. Brian Lee is an epidemiologist at Drexel University who studies maternal medication use during pregnancy and child neurodevelopmental outcomes. I know this is a conversation that is so personal for so many of us and that there’s so much misinformation around neurodivergence, which is why we want to keep covering this on Science Friday. And we would like you in this conversation. What questions do you have about the science of autism? Call us, 877-4SCIFRI.

[MUSIC PLAYING]

Today’s episode was produced by Rasha Aridi and Dee Peterschmidt. But a lot of folks help make this show happen every single week, including–

JORDAN SMOCZYK: Jordan Smoczyk.

EMMA GOMETZ: Emma Gometz.

VALISSA MAYERS: Valissa Mayers.

SANTIAGO FLÓREZ: Santiago Flórez.

FLORA LICHTMAN: I’m Flora Lichtman. Thanks for listening.

Copyright © 2025 Science Friday Initiative. All rights reserved. Science Friday transcripts are produced on a tight deadline by 3Play Media. Fidelity to the original aired/published audio or video file might vary, and text might be updated or amended in the future. For the authoritative record of Science Friday’s programming, please visit the original aired/published recording. For terms of use and more information, visit our policies pages at http://www.sciencefriday.com/about/policies/