Tylenol is not the reason autism diagnoses have exploded

The hysteria about acetaminophen is just more evidence that American trust in medicine is collapsing. I get the fury at the system - believe me, I do - but this isn't helpful.

Can we all calm down about Tylenol?

Since President Trump blamed Tylenol - the trade name for acetaminophen - for rising autism diagnoses and told pregnant women not to use it, the drug has become the latest flashpoint in America’s cultural and political wars.

First things first. Tylenol is a lousy drug. It’s sold over-the-counter and heavily advertised, so people assume it’s very safe. It’s not. In overdose, it is toxic to the liver. And overdosing is easy. Just four grams of Tylenol — roughly six extra-strength 650 milligram tablets — in 24 hours can cause liver damage, and acetaminophen overdoses cause half of liver failures in the United States, according to a 2023 paper.

—

(But overdosing on the truth is impossible. Please help me find it.)

—

Still, regulators allow people to buy acetaminophen over-the-counter because all painkillers have risks. Aspirin and ibuprofen (Advil) can cause stomach bleeds and are linked to some birth defects. Opioids carry a huge risk of addiction and overdose.

But the question isn’t if people should be careful with acetaminophen. Yes, they should. Very careful. The question is if acetaminophen causes autism.

The evidence that it does is weak at best. (And yes, I am aware of the documents the Daily Caller published this week that show officials at Johnson & Johnson, which used to own Tylenol, discussed a potential link.)

—

(Maximum strength liver failure)

—

The Tylenol-causes-autism theory has two big problems.

The first is that it is not clear how much rates of autism have risen in the United States. Many children who have what is now called “profound autism” — that is, who are unable to communicate meaningfully and are severely developmentally disabled — may previously have been diagnosed as mentally retarded.

Children who have less severe forms of autism, such as patterns of repetitive behavior or difficulty understanding emotions, might not have been diagnosed at all in the past.

So the epidemic may well come mostly from increased awareness and diagnoses, as well as increased mainstreaming of children with developmental difficulties, making them more visible in schools and elsewhere. To the extent it is real, rising parental age — a known risk factor for autism — probably accounts for some of it.

I realize family members of people with autism, including many Unreported Truths readers, want a more definitive answer than this one. The truth is not always satisfying.

—

The second problem with the acetaminophen-causes-autism theory is that… the studies that the proponents of the link rely on show a connecton that barely qualifies as provisional. The big, key study, which is based on decades of data from Sweden, showed an extra risk of 5 percent for autism for children whose mothers had ever taken acetaminophen (and a 7 percent higher risk of ADHD).

That risk disappeared entirely when researchers performed a more sophisticated analysis called sibling control, but let’s assume for the sake of argument it’s real. A 5 percent extra risk obviously cannot explain the massive increases in diagnosed autism in the last 25 years — especially since Tylenol has been around the entire time.

When environmental risks are real and substantial, they tend to be obvious to casual observers and have VERY high rate ratios when they’re studied. Smoking increases lung cancer risk up to 30 times. That’s 3000 percent¹. Not 5 percent. Heavy cannabis use increases the risk for schizophrenia two- to six-fold — 100 to 500 percent.

Those are real risks, and they are visible to users and physicians even without complex database analysis. Lung surgeons were aware by the 1950s that lung cancer was rising and that most cases were in smokers. Even so, proving the connection scientifically took decades.

—

The Food and Drug Administration clearly understands the complexities of the issue. In its letter to physicians last week, it wrote:

[W]hile an association between acetaminophen and autism has been described in many studies, a causal relationship has not been established and there are contrary studies in the scientific literature. The association is an ongoing area of scientific debate…

In the spirit of patient safety and prudent medicine, clinicians should consider minimizing the use of acetaminophen during pregnancy for routine low-grade fevers.

Hardly a call for a ban.

—

The FDA knows:

—

Unfortunately, President Trump was not in the mood for subtlety last week.

“Taking Tylenol is not good — I’ll say it: It’s not good,” he said at the press conference, telling pregnant women to “tough it out.”

Trump’s advice is troubling in at least three ways.

First, other over-the-counter painkillers have known risks for developing fetuses, which is why Tylenol is recommended at all. Second, untreated fevers can have risk, and Tylenol is a fever reducer as well as a painkiller. There will be times when women need to take something for fever and pain. What should it be?

That’s the worst part of Trump’s words: they risk making women who do have autistic children blame themselves if they took Tylenol, when in reality the causes of autism are unknown and likely mostly genetic.

Of course, Trump’s comments led to the usual stupidity on the left, as pregnant women filmed themselves taking (or pretending to take Tylenol). If the Orange Man says arsenic is bad, gimme a hit!

Then yesterday on X, Heather Heying — whose husband, Bret Weinstein, bizarrely insists that the mRNA Covid jabs killed 17 million people (there is simply no evidence of this, none) — took the anti-acetaminophen hysteria to another, dangerous level.

Heying responded to a woman who said she had treated her baby’s fever with Tylenol by telling parents not to treat fevers with any medicine. Giving kids fever-reducing medicine is “poisoning” them, Heying wrote.

This advice is not merely stupid or wrong, it’s dangerous.

Yes, small children regularly spike fevers, and a 101- or 102-degree fever is less meaningful in a kid than than an adult. Most pediatricians will encourage parents to wait before treating it aggressively. But a 103-degree fever merits close watching and is likely to make a child very uncomfortable. A fever of 104 degrees or more can cause a seizure that in the worst case can kill.

So just how hot does Heying think kids need to get before parents try to break their fever with medicine? In response to a follow-up question, she wouldn’t say.

—

(Not medical advice, just the truth.)

—

If you read Unreported Truths regularly, you know I believe Americans are overmedicated, particularly for chronic conditions. You know I think drug companies overprice, overpromise, and hide side effects and overpromise. You know I prefer the lowest possible dose for the shortest possible time.

But it’s 2025, not 1025.

We have drugs that work, particularly for simple and acute conditions like fever. Refusing to use them is stupid. And telling mothers that they’ll be responsible if their kids develop autism if they pop a couple of Tylenol tablets is cruel, wrong — and untrue.

1. Technically 2900 percent increased risk.