On Monday afternoon, President Trump took the White House podium, flanked by Robert F. Kennedy Jr. and Dr. Mehmet Oz, and declared:

I’m sorry… what?

The subject wasn’t pain management, or even public health. It was autism. Specifically, Trump and his administration announced that acetaminophen — the active ingredient in Tylenol — could be a cause.

But you know all this already. So I’ll cut to the chase: this is junk science. Dangerous junk science. And I’m not going to force-feed you another 1,500 words fact-checking it. The Wall Street Journal, The New York Times, CNN, and every major outlet have already done that.

What I want to do today instead is something more useful: show you why flimsy claims like this gain traction, how studies get twisted into scary headlines, and how you can learn to tell the difference between evidence and noise.

Because if the President of the United States is now running the same playbook as supplement companies selling “clinically studied” gummies, then scientific literacy isn’t optional anymore. It’s survival.

For decades, supplements and wellness products have gotten away with a simple trick: confuse “studied” with “proven.” Stick “clinically studied” on the label, and suddenly your probiotic looks like penicillin. I wrote about this exact grift a few weeks ago in my piece, “The Asterisk Economy.”

Now that same trick is being played on the national stage. When RFK Jr. says acetaminophen is an “environmental factor” behind autism, or when Trump shouts that pregnant women should “tough it out” through fevers, they’re not citing new science. They’re recycling weak associations from observational studies and presenting them as fact.

You should be offended that they believe you are dumb enough to buy into it.

The consequences aren’t abstract. Doctors are already reporting that they may stop recommending Tylenol to pregnant women, not because it’s unsafe, but because they fear being sued later if a child is diagnosed with autism.

Imagine being told to ride out a 102-degree fever in your first trimester — when untreated fever itself can cause birth defects. That’s the reality of politicized BS science.

Let’s start with the most important concept in all of science literacy: correlation is not causation.

Here’s a classic internet example: Between 1999 and 2009, the number of Nicolas Cage movies released in theaters strongly correlated with the number of people who drowned in swimming pools. The charts line up beautifully. Which is to say: Nicolas Cage is apparently a menace to pool safety.

Anyone with a brain can see that this is merely a coincidence — a correlation. Neither is a reason for the other. But the coincidental data allows for someone to use semantics to make a compelling argument to those who may not know who Nicolas Cage is.

To make it more personal: If correlation equaled causation, the New York Knicks’ fifty-year championship drought would explain Elon Musk’s growing fortune. The more the Knicks lose, the richer Musk gets. As a lifelong Knicks fan, I’d believe almost anything to explain the misery — but even I know causation doesn’t work that way.

The same principle applies to Tylenol and autism. Yes, some studies have found “associations.” But associations are not proof. They’re just patterns in data, and those patterns can be driven by dozens of other factors.

Here’s what you all need to understand: if you design a study poorly enough — or interpret it aggressively enough — you can “prove” almost anything. That’s how we get headlines about coffee both preventing and causing cancer, or wine being simultaneously good for your heart and a straight shot to liver failure.

Let’s break down the playbook:

1. Observational vs. RCTs
The gold standard in medicine is the randomized controlled trial (RCT). You take two groups, randomize them, and control for variables. That’s how we know statins lower cholesterol or Ozempic helps with weight loss.

But with pregnant women, RCTs are off the table. You can’t introduce risk to test for it. No well-intentioned researcher would ask a bunch of pregnant women to take a drug when they don’t know the consequences. So researchers rely on observational studies: track who took what, then look at outcomes years later. These studies can find associations, but they can’t prove causation.

2. Small Sample Size = Big Claims
One of the drugs Trump’s team touted at the same event, leucovorin, has been studied in — wait for it — a few dozen children. Some parents reported improvements. That’s encouraging, but it’s not a breakthrough therapy. Small samples make noise look like signal.

3. Confounding Variables
Women don’t take Tylenol in pregnancy for fun. They take it because they’re sick, in pain, or running a fever. And fever itself is a known risk factor for neurodevelopmental issues, including autism. So is infection. So is chronic pain. Tylenol may just be a bystander in a more complicated chain of events.

4. Relative vs. Absolute Risk
A favorite trick: frame risk in relative terms. “Tylenol doubles your risk of autism” sounds terrifying. But 0.1% → 0.2% is meaningless. The absolute numbers matter, but they rarely make headlines.

5. Cherry-Picking
In August, a Harvard/Mt. Sinai review looked at 46 studies. More than half showed an association. But the authors themselves emphasized: no causation could be claimed.
Guess which part made it into the Trump press conference?

6. The Sibling Study Reality Check
The most important study looked at 2.5 million Swedish children. When researchers compared siblings — one exposed to Tylenol in utero, the other not — the link to autism disappeared. Genetics and environment, not acetaminophen, explained the difference.

7. Confirmation Bias
RFK Jr. has long argued that genetics play little role in autism. So when an observational study produces a suggestive correlation, he treats it as proof of his worldview. That’s not science — it’s advocacy dressed up in data.

If this is junk science, the real question is why push it? Who actually gains?

Who doesn’t benefit?

Patients, parents, and doctors — the people who actually have to navigate pregnancy and make decisions under stress.

This playbook should feel familiar. It’s the same one supplement brands use when they slap “clinically studied” on a bottle of green powder. It’s the same one peptide clinics use to sell injections with no FDA approval. And now it’s the one being used by the government itself.

The danger isn’t just bad science. It’s erosion of trust. When every claim is framed as fact, when correlation is sold as causation, when headlines replace evidence — people stop knowing what to believe at all. And that vacuum gets filled by whoever shouts the loudest.

Here’s the bottom line: acetaminophen has been studied for decades. It is not an autism trigger.

What Trump and RFK Jr. announced this week was not science. It was politics, lawsuits, and culture war packaged as health advice.

But the bigger danger is what it represents. If we now live in a world where the White House is willing to say whatever benefits them without proof or fear of repercussions, then understanding how to read studies is no longer a nice skill for science nerds. It’s a civic responsibility.

I hope this gave you an idea of the manipulation tactics to look out for as we enter the Age of Disinformation in Medicine.

Whether it’s gummies at the drugstore or pronouncements from the White House podium, the tactics are the same. The only defense left is knowing how to read the study for yourself.

The next drug, supplement, or therapy you’re asked to believe in may not come with a disclaimer at the bottom.

Alec Wade Ginsberg, PharmD, RPh
4th-Gen Pharmacist | Owner & COO, C.O. Bigelow
Founder, Drugstore Cowboy