Dr. Fox: Welcome to today’s episode of “Healthful Woman,” a podcast designed to explore topics in women’s health at all stages of life. I am your host, Dr. Nathan Fox, an OBGYN and maternal fetal medicine specialist practicing in New York City. At “Healthful Woman,” I speak with leaders in the field to help you learn more about women’s health, pregnancy, and wellness.
Emily, thank you for coming back on the podcast. It’s so nice to hear your voice. I was saying before, I miss you. It’s like we don’t spend enough time together anymore, and it really upsets me.
Emily: I totally agree. I miss you, too. And I’m delighted that we’re getting to spend this time together.
Dr. Fox: It is terrific. And I have to say that in doing some background on this, I stumbled across a podcast that you did with Chelsea Clinton on the same topic. And how good must it feel for you to be both on Chelsea Clinton’s podcast and my podcast.
Emily: I feel really great about that. Although I really enjoyed being on Chelsea Clinton’s podcast, I will say that you are one of my very favorite people to talk to.
Dr. Fox: Thank you. I’m actually curious is what is the highest profile podcast you’ve been on? Hers I assume is pretty high profile. I don’t know. I’m just guessing, but I don’t know who’s…
Emily: Hers is pretty high profile. Chrissy Teigen. I was on Chrissy Teigen’s podcast.
Dr. Fox: Oh, all right. Well, that’s cool. I don’t typically listen to Chelsea’s podcast. I’m not opposed to it, but the only podcast of intellectual worth that I ever listened to was yours until you retired from podcasting. I guess, my own, but other than that, it’s mostly like sports gambling and movies and, you know, total nonsense. That’s how I roll with podcasts.
Emily: My podcast consumption is almost exclusively podcasts about endurance sports. So, just like podcasts…I like to go running, listen to podcasts about running. I mean, that’s, like, my dream. And so, yeah.
Dr. Fox: Wonderful.
Emily: That’s it.
Dr. Fox: Well, good. So, we’re not going to rehash everything that you said on Chelsea’s podcast, although I don’t know how many of our listeners overlap. Honestly, I have no clue. Maybe many, maybe few. Who knows? But either way, you know, obviously, in my own practice, I get asked all the time about Tylenol more so in the past few months than previously, but it always existed. And obviously, it blew up a couple of months ago with the government doing what they did. And I figured…someone was like, you should do a podcast on this. I was like, I don’t know. And then I said, you know what, I’m going to do it. And you were obviously the first person I wanted to talk to about this just because I liked the way you put things together. So, you’re my first call on this.
Emily: Amazing. Well, then let’s talk about it. Let’s try to make people’s lives easier.
Dr. Fox: Hopefully. And if listening to this is painful, have a Tylenol. Okay. So, I guess, let’s start back. Why is this a thing? Like, why are we even talking about Tylenol in pregnancy?
Emily: So, I first encountered this discussion way back in, you know, maybe 2014. And there has been, for a very long time, a research out there in the world suggesting that Tylenol might…Tylenol consumption during pregnancy might be linked to neurodivergence, ADHD, autism in particular, in kids. And so, that’s kind of one of these things that people had put out, again, sort of more than a decade ago in these kind of large correlational cohort studies. And it has been continually discussed. It sort of comes up occasionally at least since then.
Dr. Fox: Yeah. And I think that’s a really important point that I’ve been making with a lot of people that this is not new. Meaning, it’s not like the U.S. government said, you know, we’re going to start looking for things related to autism, hey, we came up with Tylenol. Like, people have been discussing this for a very, very long time and studying it for a very, very long time. And we already had opinions on it and comments on it and statements on it and all this stuff about it. So, there wasn’t anything really new that developed in the research world that led to this, you know, statement about them saying not to use Tylenol in pregnancy. They just sort of rehashed stuff that had been out there for a very, very long time. So, I think that’s important for people to know that this isn’t, like, groundbreaking news flash that happened.
Emily: I agree with that. I think that often when statements like this come out of official bodies, people will come and say, you know, what changed, what was new? And the answer in this case and in many of those cases, but particularly in this one, is nothing. There was no new finding. There was no new anything in particular. This was just a statement based on things that we already knew before.
Dr. Fox: So, we’re going to go into, obviously, the data sort of. And I do want to do not a really deep dive, but, like, a medium…you know, not the shallow end. I’m going to go somewhere in the middle kind of dive on the data. But can you explain the types of studies that were done that, I guess, concluded that there was this correlation or association between Tylenol use in pregnancy and outcomes…you know, we’ll say autism, though it’s more than autism. There’s ADHD, as you said, but we’ll just say autism for simplicity’s sake. Like, what were the studies that were done, is question number one. And number two, if there were so many studies that made this link, why isn’t it proven at this point?
Emily: So, the studies…let’s start with the studies that we sort of have that suggest a link. So, these are what a standard kind of prospective cohort study… So, what we would do in a study like this is have a set of people, of women who were pregnant. And in the kind of best of these studies, they would be asked questions about their Tylenol consumption and many other things while they were pregnant, and then their kids would be followed over time, and we would look for diagnoses of autism or ADHD.
There are versions of this which are less good, like you ask people what they did in the past, that’s not very good. But let’s take, like, the sort of best version of this kind of study, which is we know from data during pregnancy what people were taking and then we see their kids later. And in those studies, if you compare the people who take particularly a lot of Tylenol — and I actually think that’s an important point — in most of these studies, the kind of exposure measure is something like more than 28 days during the pregnancy. So, this isn’t actually Tylenol once for a headache. It’s much closer to, you know, I’m taking it in a kind of consistent way. But you can compare that group to people who didn’t take any, you will see sort of slightly higher rates of autism or ADHD in their offspring.
I say slightly higher because I think in some of these, like, public discussions, it sort of makes it sound like this is increasing the chance of having autism by, like, 100% or something. But actually in even these studies…and we can talk about the problems with them. But even in these correlational studies, the effect is about one percentage point. So, it’s a pretty small increase among the kids whose moms took a lot of Tylenol during pregnancy. And that’s kind of the standard set of methods that we start with.
Dr. Fox: Yeah. And so, again, let’s assume these studies were perfect, which they weren’t. Right? The magnitude that we’re talking about is… Again, if they were 100% correct, if you take Tylenol for a month in a row, you’re increasing your child’s chance of autism by 1%. Right? So, that alone sort of lowers or should lower, like, you know, the, I guess, outcry about this. Right? We’re talking about a high exposure that most people don’t have. And again, obviously, no one wants any risk of autism, but 1% is not the same thing as saying you will get autism, obviously. And that’s the worst case scenario. So, why would those studies not be conclusive that that’s a true statement, that taking a lot of Tylenol increases your risk of autism by 1%?
Emily: Yeah. So, I think that the main reason…there’s two important reasons. So, one is just the set of people who are taking Tylenol versus not are different demographically in some ways that may contribute to autism or importantly to autism diagnosis. So, when we think about, you know, measured autism or ADHD rates in kids, it’s kind of not just a symptomatic measure, it’s also a diagnostic measure. And so, if either of those things…if, like, the taking of the Tylenol is associated with either the chance of diagnosis or with symptoms for some other reason, like, that may be part of what’s going on. So, there’s one just, like, basic demographic differences across groups.
And then a second issue is a question of, you know, what are they taking the Tylenol for? And is that itself potentially linked to this risk? And one very direct thing many people have brought up is, you know, people who are on the autism or ADHD spectrum themselves may have more sensitivities and more…you know, stuff that medical situations that make it more likely that they are taking autism…they are taking Tylenol. And then that, of course, could generate a link because we know there are genetic links in these conditions. So, those are kind of the two things that might be going on that might reflect this relationship being a correlation and not causal.
And let me say one other thing is the fact that this effect is very small, for me, is pretty important in light of those other things. Because what that says is these other factors, these other, like, reasons there might be differences don’t actually have to be very big to explain all of the effect that we see. You know, if we saw that people who take Tylenol for 28 days, their kids are, like, you know, 50 times as likely to have autism, you might be like, okay, look, that seems like a really strong correlation. Like, we really probably want to dig into this. This is like there’s a tiny increase. And so, a lot of things could explain it, including these biases and sort of correlating those factors.
Dr. Fox: Yeah. And also, I mean, typically, like with medications in pregnancy, the ones that are known to be dangerous — and that list is small — the likelihood of an outcome is pretty noticeable, and you find out about it pretty quickly. And people have been taking Tylenol in pregnancy for, you know, whatever, 50 years, whatever it is. And so, if it were something that were that dangerous and so many people take it, it would be much more blatant. Everyone would see it, like, right away. And the fact that it’s like people are arguing about it so subtle and it’s this and it’s that, usually indicates it’s…there’s probably a lot more smoke than fire, so to speak, with that. And I do think that’s also very, very important. Again, we have a long, long history. It’s not a new medication that just came out six months ago, where you have to really wait and figure it out. This has been around for a long time. And if it was really that dangerous, it would be much more obvious.
Emily: Yeah. We would see it. And I will say the other thing is, like, in these sort of public conversations, some of what people are talking about, you know, when kind of Trump and RFK got up and talked about Tylenol and autism, the idea is sort of, like, this is part of what explains the increase in autism over time. But that is actually, like, literally impossible because the usage of Tylenol during pregnancy has gone down over time slightly, and the rates of autism have gone up. So, whether this is, like…whatever you think about this, the sort of link in the data at an individual level, it definitely isn’t anything to do with the trends over time because the changes in Tylenol usage rates have…are in the wrong direction.
Dr. Fox: Right. And so, let’s say we wanted to design a study. The only way we could really design a study, which isn’t going to happen, to find this out appropriately is we would have to take literally, like, 10,000 women or whatever it is, randomly divide them into two groups and say to 5,000 of them, I want you to take Tylenol every day for a month. Why? Because. There’s, like, no sin to just take it every day for a month. And the other group don’t take any Tylenol no matter what. And then you follow all those 10,000 kids for rates of autism. That might be a way to actually figure out, does it increase the risk? And if so, by how much? But that’s not going to be done. It’s never been done. It’s never going to be done for many, many, many logistical and ethical reasons. So, it’s not going to happen.
So, okay. So, we’re left with sort of this either what you described, like a prospective study looking forward in time, but not choosing who takes Tylenol and who doesn’t, or looking back in time, what we call a retrospective study, which are all flawed, all of these limitations. So, what could we do to try to tease out…and I know it’s been done. Like, what are the things we can do to try to tease out whether this little spike in autism we see in the Tylenol group is related specifically to the Tylenol or due to, like, those other factors? Maybe it’s the condition for which she’s taking the Tylenol, maybe it’s other demographic things or whatever it is. How would you sort of sort that out from a data perspective?
Emily: Yeah. Let me say on the first question of, like, could you run a randomized trial? I think, in principle, actually, you could, although you would need to do it as an encouragement design. So, basically, take a bunch of people, tell…like, half of them, tell them, like, really don’t take Tylenol. And the other half, maybe you don’t tell them that, and you’d probably get some difference in Tylenol usage, which you could then potentially link. Part of what makes that infeasible among…you might get pushed back from an ethics review, but I think the other thing that makes it infeasible is you would need such a large sample size, it would be, like, basically impossible.
So, in my view, the kind of best empirical approach for this is to use a sibling design. That’s basically a study in which you have siblings, same mom. Typically, we can’t usually ensure that’s the same dad, but it’s, like, the same mom. And then they took Tylenol in one pregnancy and not another. And the sort of two reasons I really like this kind of study is, you know, one, it holds constant a lot of factors that are important, specifically everything about the mom. So, you’re then kind of comparing within sibling groups. So, any of these concerns we have, like, you know, the moms who take Tylenol are different from the moms who don’t, like, that’s kind of taken care of. So, I think that’s the sort of…a lot of the biases is removed.
And the second thing is, in a really well done version of the study, you can usually show the effects without looking within siblings. So, just doing the kind of naive comparison across families and then looking within family. And that can give you a sense of, like, okay, am I seeing the effect when I look across families and then not within families? And that’s a really good way to test for whether the effects you’re seeing across families are in fact about these other factors.
And so, that actually has been done in the case of Tylenol, using very large sample sizes. This study from Nordic countries, it’s, like, millions of people. And what you see there is, you know, when you compare across families, they see relationships between Tylenol and autism. And when you look within family, you don’t see that. So, it actually really looks like all of the impacts that we’re seeing in these cross-sectional studies are really just about differences across the moms in other factors. And if you control for those, you just don’t see this relationship showing up in the data.
Dr. Fox: So, it’s so interesting because those types of studies have been done with other things that have correlations. So, for example, there were studies like that done… I know that there’s this question about, like, how long should you wait between pregnancies? And some of the data showed that waiting shorter was bad and longer is good. And all the data pointed that way. But then when they looked within a family, right, a mom has, let’s say, you know, three kids. And between one and two, she waited a certain amount. And between two and three, she waited a different amount. And then it made no difference because same mom, same family, it really didn’t matter. And so, it kind of debunked that.
There have actually been really good studies about this related to breastfeeding. If you breastfeed one child, but not the other child, so, like, is the one that got breastfed going to be smarter and healthier, you know, all the great things? No. They seem to be very similar and sort of lowers sort of, again, the heat on how medically important breastfeeding is. There’s other benefits, obviously.
And I was actually trying to think of, like, a good analogy to this for people who are sort of, like, trying to figure out what the hell’s going on. And I thought of it this morning as I was going to Starbucks. Right? Plug for Starbucks. So, I drink iced coffee pretty much every morning. And I was thinking, someone, let’s say, would design a study and look at people who drink iced coffee versus people who drink hot coffee every morning. Let’s say a study came out and said, we found that people who drink iced coffee are four times as likely to develop skin cancer than people who drink hot coffee. Whoa. Like, that seems like a big deal.
And then someone might say, well, how do you know it’s the coffee? Like, that doesn’t make sense. Maybe it’s people who drink iced coffee live in hotter places and are more exposed to the sun and sun causes skin cancer. Okay. So, what would you do? Just repeat the study in the State of Florida. Just look at Florida and see if there’s a difference. Then you find that there’s not a difference in the State of Florida. So, if you run it in the U.S., you see that it’s true. And then you run it in the State of Florida and it’s not true. And, really, you find out it’s just the sun.
And the same thing is probably true here. I mean, if you look at, like, why would Tylenol be linked? There’s so many potential causes of autism. It can’t be as simple as Tylenol. It just doesn’t make sense, especially when, you know… Like, I’m on the other end. Everyone’s worried about autism. And they’re like, okay, if I get a vaccine while I’m pregnant, it’s going to increase the risk of autism. My baby gets a vaccine, it’s going to increase the risk of autism. If I take Tylenol, it’s going to increase the risk of autism. If I get too many ultrasounds, it’s going to increase the risk of autism. If my baby gets circumcised, that’s another good one. They’re more [crosstalk 00:18:14] autism. I’m like, how many things cause autism? Like, really? Like, is that even possible?
And the answer is, no, it’s not possible that all these things do it. It’s probably very complicated. So much of it is probably genetic, epigenetic, you know, all these things that come into it that we don’t understand yet. And I think it’s almost…like, when I view our understanding of autism now is probably the view of, like, people who died of infections in 1880 before they figured out these things called bacteria. And, like, once they figured it out, it all made sense. But before then, they had all these, like, wacky theories about it that were totally just wrong because they just didn’t have the science yet. That’s how I feel we are right now, unfortunately, with, like, what causes autism.
Emily: Yeah. I mean, I think it’s…there’s an added layer on top of that that I think it’s very hard to really, like, I don’t know, like, parse and get to the bottom of, which is this question of diagnosis. So, even, like, the germ theory is an interesting parallel. But at least there, you sort of, like, it was well defined if someone was dead or not. That’s, like, a well-defined thing, which we all agree. Like, for human…all of human history, like, that’s a particular defined outcome.
A lot of what’s going on with autism right now is that we have changed the diagnostic criteria and the set of things that fall in that category. And so, we are…when you say, like, you know, diagnoses rates have gone up in the past 25 years, which is definitely true, like, some share of that, you know, something probably between 80% and 100% of that is a result of changes in the diagnostic criteria, not changes in the symptom profile.
And so, there’s…it seems, though, that, like, there’s some amount left, but it’s actually not as much as we would think. So, now, we’re trying to explain a sort of smaller difference. And there are all of these different explanations. I think you’re right. Probably almost none of those is at all important. But because people don’t know what is the explanation, like, everything is a possibility. And most things are very hard to rule out.
Dr. Fox: Yeah. It is tough stuff. And, you know, I was really…I was disappointed by the fact that our government sort of dropped this on everybody. Because, listen, it’s one thing to say things that’s, like, wrong. People say things and they’re wrong. It happens. Like, it’s not necessarily the end of the world. But the problem is the consequences now… So, let’s say you have someone who’s pregnant. And she has pain, she has a headache, she has a fever. So, not only she’s deciding, like, what do I do about it? She’s like, well, if I take Tylenol, which is what the doctor’s saying I should take, now, I’m giving my baby autism. Right?
So, A, that’s just wrong and you’re causing, like, horrible stress in someone, which is unnecessary. B, in addition to that stress, she’s suffering from whatever it is that caused her to need the Tylenol, whether it’s pain or fever or whatever. And the third thing… All right. So, those are both, like, just bad. Right? Straight up bad that you’re causing people to suffer. But the third thing, which could be even, like, counterproductive, like, the…like, just make it worse is let’s say it’s not the Tylenol that’s associated with autism or causes autism. But what if fever causes autism, which is much more plausible than Tylenol, right?
I mean, if I had to pick, like, if I had to put…you know, going back to sports gambling. If I had to put my money on which one of these two, if it’s one of them causes autism, I would say fever is much worse for a developing brain than Tylenol. And so, now, let’s say that’s the case. Not only are you not, like, solving a problem, you’re causing a problem because now people are gonna walk around with fevers. And it’s just… I don’t know. It’s really, really… It’s tough.
Emily: I think the other thing I would add to that, and this is, you know, sort of from the vantage point that I see the most is, you know, there are a lot of people whose kids have been diagnosed with autism, who did take Tylenol during pregnancy. And when this was announced, the number of people I heard from who were like, is this my fault? Like, I can’t believe I…you know, I can’t believe I did this. And for them, like, even if there was a link, actually, they would not be helped by hearing about it. Right? Like, that’s nothing they can do now. But also since there is not, now, you are just, like, making people’s lives worse because now they are not only sort of struggling with whatever they’re dealing with, with the diagnosis, but now they feel like it’s their fault for no reason. And I think that’s…you know, we often don’t think about that as a real, like, cost for people, but it really, really is.
Dr. Fox: It’s a cost for people themselves. And then you also get into, like, people could be judgy. Right?
Emily: Totally.
Dr. Fox: You know? And they’re like, oh, your child has autism. I bet you took Tylenol.
Emily: [crosstalk 00:22:59].
Dr. Fox: It’s really, really unfortunate. And, you know, on the medical side, when people ask us…and this is not just, you know, Nate Fox giving his opinion, but this is, like, what the American College of OBGYN says, it’s what Society for Maternal-Fetal Medicine says. All the people who sort of do this for a living and are at the top of the food chain in these fields have basically said the same thing, which is there’s nothing new. The data’s been out there for a long time. None of the data that is good actually suggests that Tylenol causes autism. Obviously, we don’t take things for the hell of it when you’re pregnant. And so, if…you know, you don’t just take it because it tastes good, but if you have a reason to take it, this is the medication that is known to work, and actually has a very large safety profile. And this is what you should take.
And that’s sort of the advice we’re giving to people. But, you know, there’s less trust of doctors, a lot more trust of TikTok, and it’s tough. And I don’t know what the solution is to all of that, but it’s an unfortunate situation. And I do think that people need to know that…it doesn’t even suggest it, what they say the data proves.
Emily: Exactly. I mean, I think…but I think you’re right that the sort of crisis of trust is not helped by…it’s already an issue with people trusting TikTok instead of people. But it is even worse when then the government is also saying that.
Dr. Fox: Yeah. Going back to the Chelsea thing because I listened to it. I agree with you. I believe that RFK Jr. believes that this is true. Right? I think that he actually…he believes this. Trump obviously has no idea either way. That’s pretty clear. He’s, like, just given a piece of paper, read this. He’s like, fine. But still, just because you believe something is true, unless you have real good data to support that belief, you really shouldn’t be dropping it on the country, so to speak.
Emily: I totally agree. Yeah. I mean, I think we have to try to tell people the things that are true, but certainly the current moment has not supported that.
Dr. Fox: Wow. Crazy. Wow. Well, Emily, thank you for coming on. What’s up for you next? What are you up to? Sorry, listeners. You know, [inaudible 00:25:11]… Everyone’s following you all over the place, but you know, what big projects do you have? World domination? What are you doing?
Emily: Exactly. World domination. No. I mean, I think we’re spending a lot of time on parent data, trying to really make sure that we have all the information people need. And then, honestly, I’ve spent a lot of time in the last six months, you know, kind of trying to respond in the moment to some of this misinformation and sort of be there for people when they need answers.
Dr. Fox: To me, I got…I have to comment on one more thing that I heard on the podcast. It was the funniest thing. You were mentioning, you know, that there’s all this stuff parents are, like, just bombarded with horrible things that they’re doing to their children and ruining them in every which way possible. And you were saying that someone sees on TikTok or Instagram that if they rush the kid out of the house to catch the bus, they’re going to cause their child…
Emily: Hurried child syndrome.
Dr. Fox: They’re going to cause the child to have anxiety. And you were talking about it and how that’s, you know, debunked and not true, blah, blah, blah. And I said the ironic thing about that is that the fact that we’re looking at TikTok and Instagram, that’s what’s causing the anxiety. Like, the thing that causes anxiety is telling me that something else is going to cause anxiety.
Emily: My favorite thing about that particular thing was my daughter, who was, like, 14, I was, like, explaining it to her. And she was like, but the reason I’m anxious…she was like, but the reason I’m anxious is that you’re anxious, and you’re rushing me out the door because you’re anxious. And of course, I’m going to have anxiety because I’m related to you.
Dr. Fox: It’s all your fault. Everything.
Emily: That’s what it is.
Dr. Fox: It comes back to you.
Emily: It’s all my fault. That’s parenting. It’s all your fault.
Dr. Fox: A hundred percent true.
Emily: And there it is.
Dr. Fox: There we go. Well, thank you so much. I really appreciate you taking the time.
Emily: Great. Always great to talk to you.
Dr. Fox: Thank you for listening to the “Healthful Woman” podcast. To learn more about our podcast, please visit our website at www.healthfulwoman.com. That’s healthfulwoman.com. If you have any questions about this podcast or any other topic you would like us to address, please feel free to email us at hw@healthfulwoman.com. Have a great day.
The information discussed in “Healthful Woman” is intended for educational uses only. It does not replace medical care from your physician. “Healthful Woman” is meant to expand your knowledge of women’s health and does not replace ongoing care from your regular physician or gynecologist. We encourage you to speak with your doctor about specific diagnoses and treatment options for an effective treatment plan.
