A Warning We Can’t Afford to Ignore

Published October 1, 2025

This month, the FDA took an unprecedented step—initiating a safety label change for acetaminophen (Tylenol) to warn about a possible link between prenatal use and later autism and ADHD. This isn’t some fringe theory anymore. The U.S. government has looked at the evidence and decided doctors and expectant mothers need to know. Do I think this is the only factor that might cause autism? No. But I think this evidence could point to an important contributor underlying this issue.  You might have noticed from the bottom of my newsletters long before these recommendations were made I have been cautious on acetaminophen and I have good scientific reasons for that. But before we dive into the science, we need to address something more fundamental: the way that our culture talks about autism has become I believe, harmfully dishonest.

The Uncomfortable Truth About Autism

You’ll often hear autism described as simply “a different way of experiencing the world” or “neurodiversity.” Major publications like Nature describe autistic people as showing “differences in social communication and interaction.” This language sanitizes a devastating reality. I reject this framing for the most part. Yes, some individuals with autism lead productive, independent lives, but for many families, autism means a child who will never speak, never use the toilet independently, never live without 24-hour care or at least live independently. To call this merely a “difference in communication” is not just misleading—it’s insensitive to the families living this reality every day. They have had to readjust their hopes and dreams for their child and regularly sacrifice their time and resources to care for their child in the best way possible. (Often doing so with few complaints because of their love for them.)

The truth is that autism prevalence is rising dramatically, and cases are becoming more severe. The latest CDC data shows autism has increased from 1 in 36 children to 1 in 31. For boys, the rate is now an astounding 1 in 20, and in California it’s 1 in 12.5. Even more concerning, cases are getting MORE severe, not less. The percentage of autism cases with higher IQs has actually decreased in recent surveys—nearly two-thirds of children with autism now have severe or borderline intellectual disability.

This isn’t “better awareness.” If it were, we’d expect to see more high-functioning cases flooding the system—but we’re seeing the exact opposite. We’re facing a genuine public health crisis, and we need to take action. One of the most concrete steps we can take right now is addressing possible links – prenatal acetaminophen exposure being one. 

What the Science Shows

The FDA’s decision didn’t come from nowhere. Multiple large studies have found associations between prenatal acetaminophen use and neurodevelopmental disorders, but one stands out as particularly compelling.

The Boston Birth Cohort study, published in JAMA Psychiatry, is the smoking gun. Instead of relying on maternal recall, researchers measured actual acetaminophen levels in umbilical cord blood from 996 births—objective biological evidence that eliminates the “recall bias” problem plaguing most studies. The results were striking: children with the highest acetaminophen exposure were 2.86 times more likely to develop ADHD and 3.62 times more likely to develop autism compared to those with the lowest exposure. This clear dose-response relationship—more acetaminophen equals higher risk—is exactly what you’d expect if the drug were causing harm.

The biological mechanisms make sense too. Acetaminophen can deplete glutathione (a critical antioxidant), disrupt hormone signaling, and trigger inflammation—all during the most vulnerable periods of brain development.

One compelling piece of evidence supporting this connection comes from understanding glutathione’s critical role in brain development. A 2022 animal study published in Frontiers in Psychiatry found that N-acetylcysteine (NAC)—a medication that boosts glutathione levels—prevented autistic-like behaviors in rats exposed to toxins during pregnancy by normalizing glutathione in their brains. Here’s why this matters: if restoring glutathione prevents autism in animal models, then depleting glutathione could cause it. Recall acetaminophen’s primary mechanism of toxicity is glutathione depletion. This creates a troubling picture—we have one drug (NAC) that protects developing brains by increasing glutathione, and another drug (acetaminophen) that depletes the same protective molecule. 

The Swedish Study: Why Size Doesn’t Trump Quality

Critics will point to a massive 2024 Swedish study of nearly 2.5 million children that found essentially no link between acetaminophen and autism. This study deserves attention—it’s large and well-designed in some respects.

But here’s the fundamental problem: the Swedish researchers relied on prescription records and midwife interviews—essentially recall data with enormous gaps. They completely missed over-the-counter acetaminophen use (the vast majority of use), and couldn’t accurately measure how much or how long mothers actually used the drug.

This is the same problem that plagued decades of diet research. Remember when massive observational studies told us low-fat diets were the answer to obesity? Those studies had hundreds of thousands of participants, yet they were fundamentally wrong because they relied on recall data. We’ve learned the hard way that even giant studies are useless when exposure measurement is flawed.

A Simple Risk-Benefit Analysis

Let me put this in the clearest possible terms. We have three scenarios:

Scenario 1: There’s a real link between prenatal acetaminophen and autism. It may not be the only cause, but even if it’s part of the picture—and we issue warnings and mothers avoid unnecessary use—we prevent countless cases of a devastating developmental disorder.

Scenario 2: There’s no link, but we’ve issued warnings anyway. Mothers avoid acetaminophen unnecessarily. The downside? They use alternative pain and fever management. Not ideal, but not catastrophic. We also might send unnecessary funding to further explore this idea and delay finding the real cause.

Scenario 3: There’s a real link, but we do nothing. We allow a preventable cause of autism to continue affecting thousands of children every year.

To me, the risk-reward framework here isn’t complicated. The potential benefit of precaution outweighs the risk of being wrong.

What This Means for You

If you’re pregnant or planning to become pregnant, here’s my recommendation based on the current evidence:

Avoid routine or prolonged acetaminophen use during pregnancy. This means no taking it “just in case” or for minor aches.

When you truly need it—for significant pain or high fever—use the lowest effective dose for the shortest duration possible. Fevers can indeed also cause fetal harm in pregnancy, including neural tube defects and preterm birth. Don’t leave dangerous fevers untreated!

Remember that high fevers during pregnancy carry their own risks. Sometimes acetaminophen is the right choice. The key is intentional, judicious use, not routine consumption.

Some might ask, “Should I take NAC to boost glutathione if I take Tylenol?” This is an intriguing question. As mentioned previously, NAC seems to be protective in animal models of autism. NAC has been studied in human pregnancy for other indications—including preterm labor, recurrent pregnancy loss, and acetaminophen overdose—with a good safety profile (FDA states generally safe with no teratogenic effect). However, to my knowledge, no human clinical trial has specifically evaluated whether NAC can prevent acetaminophen-related neurodevelopmental effects when taken together during pregnancy. I think such a study should be done, given NAC’s established safety.

Overall, for minor discomfort, consider other approaches: rest, hydration, ice or heat therapy, gentle stretching, or consultation with myself or another healthcare provider about pregnancy-safe alternatives.

The Bigger Picture

The acetaminophen issue represents something larger: our willingness to confront uncomfortable truths about what’s happening to our children’s health. The data is clear—autism rates haven’t just risen, they’ve exploded. And severity is increasing, not decreasing. Something has fundamentally changed in our environment, our exposures, or both.

The FDA has now given us clear guidance: there’s enough evidence to warrant caution. ACOG agrees that acetaminophen should be used judiciously in pregnancy. We have objective biological evidence from cord blood studies showing a dose-dependent relationship. The scientific consensus is moving toward precaution.

It’s time we listen—and act accordingly.

Blessings,

Dr. Jeff Ponke, MD

NB – This is the first in a series on autism and potential environmental contributors. In my next newsletter, we’ll explore the connection between folate receptor alpha antibodies (FRAA) and autism—a fascinating area of research showing that some children with autism have antibodies blocking folate from entering their brain, and how folinic acid (leucovorin) supplementation may help restore function in these cases