Science is inherently messy. Studies contradict each other. Results that look definitive turn out to be flukes. Effects that seem small end up mattering, while effects that seem huge evaporate on closer inspection. Researchers argue with each other in journal pages, conferences, and increasingly on social media. Uncertainty isn't a bug in the scientific process; it's a feature. It's how we gradually zero in on truth through replication, debate, and refinement.
But here's the thing: that messiness can be exploited. A Nature editorial calls out exactly this phenomenon, using a recent FDA statement about acetaminophen (that's Tylenol, for those keeping track at home) during pregnancy as a case study in how scientific uncertainty gets weaponized for political purposes.
The "Tylenol Causes Autism" Scare
The FDA announced plans to update labeling for acetaminophen to suggest that use during pregnancy "may be associated with an increased risk of neurological conditions such as autism and ADHD in children."
Sounds pretty alarming, right? A government agency saying a common medication might cause developmental disorders? That's the kind of thing that makes pregnant women immediately clear out their medicine cabinets.
Except the scientific picture is way more complicated than that label suggests.
The World Health Organization responded almost immediately: "There is currently no conclusive scientific evidence confirming a possible link between autism and use of acetaminophen during pregnancy." The European Medicines Agency agreed. So did regulators in Australia and the UK. So did most U.S. state medical boards. So did major professional medical societies representing obstetricians, pediatricians, and neurologists.
When that many independent regulatory bodies all say essentially the same thing, you're looking at something close to scientific consensus. And that consensus, expressed with appropriate nuance, is: yes, some studies have shown correlations. No, the evidence doesn't support a causal link. The studies have methodological problems. The effect sizes are small and inconsistent. When you look at the totality of evidence, there's no strong reason to think acetaminophen causes autism.
The Anatomy of a Scientific Distortion
So how does a statement like the FDA's happen despite this consensus? This is where things get uncomfortable, because the answer involves politics.
The problem isn't that someone mentioned the studies exist. The problem is selectively presenting data to support a predetermined conclusion. You cite the three studies that show an association. You ignore the ten that don't. You strip away the methodological caveats. You turn "we found a weak correlation in a study with significant confounding variables" into "this drug may cause autism."
When politicians and officials make definitive statements like "Don't take Tylenol" and "Fight like hell not to take it," they're presenting scientific uncertainty as if it were scientific consensus in the opposite direction. That's not caution. It's not even skepticism. It's propaganda with footnotes.
And it matters in real ways. Pregnant women need safe pain relief options. Fever and pain during pregnancy carry their own risks to both mother and fetus. Scaring people away from acetaminophen, which remains the pain reliever with the best safety profile for pregnant women, doesn't protect anyone. It just replaces one theoretical risk with other more concrete ones: untreated fever, maternal distress, or switching to medications that are actually known to be problematic during pregnancy.
The Autism Community Gets Used Again
Here's the particularly frustrating part: autism keeps getting used as a rhetorical weapon, and it never actually helps autistic people.
We saw this with vaccines. For years, a discredited study was used to scare parents away from vaccination. The result wasn't fewer kids with autism. The result was outbreaks of measles and whooping cough, while autistic individuals and their families had to deal with the stigma of their condition being treated as a catastrophe to be prevented at all costs.
Now we're seeing it with Tylenol. And as Nature Medicine noted, people with autism deserve evidence-based policy and care. They don't benefit from their condition being used as a scare tactic. They don't benefit from fear-mongering that generates more heat than light.
If you genuinely cared about autistic people, you'd fund more research into the actual neurobiological causes of autism. You'd improve access to services and support. You'd work on reducing stigma. What you wouldn't do is grab whatever study seems alarming and wave it around to score political points.
Living With Scientific Uncertainty
Science is going to continue being uncertain. New studies will come out with concerning results. Other studies will fail to replicate those results. Meta-analyses will be inconclusive. Experts will disagree. This is the normal, healthy functioning of a knowledge-generating enterprise.
The question is whether we're going to be honest about that uncertainty or exploit it. Being honest means presenting the full picture, including the parts that don't support your preferred narrative. It means explaining what we don't know and why. It means acknowledging that preliminary findings are preliminary, that correlations aren't causations, and that a single study rarely settles anything.
Exploiting uncertainty means cherry-picking. It means treating every scary study as definitive while dismissing every reassuring one as industry propaganda. It means claiming scientific authority while ignoring what most scientists actually think.
We deserve better than that. And so do pregnant women trying to make reasonable decisions about their health care.
Reference: Weaponizing uncertainty in science and in public health puts people in harm's way. (2025). Nature. doi: 10.1038/d41586-025-03167-5 | PMID: 41044274
Disclaimer: The image accompanying this article is for illustrative purposes only and does not depict actual experimental results, data, or biological mechanisms.