Tylenol and Autism, Part II: The Swedish Study That Got It Wrong
A perfect example of a poorly designed study producing results favorable to Tylenol promoters and contributing to delays in banning the use of this highly neurotoxic compound in pregnant women.
This is Part II in our series.
Part I: Breaking: Government Finally Admits Tylenol-Autism Link After Years of Corporate Cover-Up
Part III: Broken Trust: The Tylenol Cover-Up That May Have Damaged Millions of Children
Part IV: Tylenol: From Painkiller to Empathy Killer
In 2024, a massive Swedish study of 2.48 million children was published in JAMA. Its headline: acetaminophen use during pregnancy was not associated with autism or ADHD once sibling comparisons were made.¹
This conclusion was quickly seized upon as proof that concerns about Tylenol were overblown. Yet the reality is that the study’s design virtually guaranteed a false null:
Severe undercounting of exposure. Only 7.5% of Swedish mothers were recorded as Tylenol users.¹ By contrast, surveys consistently show 40–65% of pregnant women in Europe and North America use acetaminophen.² This means five out of six actual users were misclassified as “unexposed,” mathematically collapsing the signal.
Over-adjustment of causal pathways. The Swedish team controlled for more than 20 inflammation- and oxidative stress-related variables.³ But these are not neutral confounders—they are the very conditions that magnify acetaminophen’s neurotoxicity. Adjusting them away erased the biological mechanism itself.
Sibling design over-correction. Sibling analysis assumes familial traits only confound, when in fact they can enable harm. Families with impaired glutathione metabolism, for example, may be especially vulnerable. By comparing siblings, the Swedish design canceled out the very contrast that reveals how acetaminophen interacts with susceptibility.²
What emerged was an illusion of safety—not proof of absence, but absence of proof.
Parker’s Rebuttal: Why the Null Was False
Shortly after the Swedish study’s release, William Parker and colleagues published a detailed rebuttal in Life.² Using in silico modeling, they demonstrated exactly how the Swedish results could have been manufactured:
When actual acetaminophen use is ~60% but only 7.5% is recorded, a true two-fold risk collapses to an apparent hazard ratio near 1.0 (“no risk”).
When susceptibility cofactors like inflammation are wrongly treated as confounders, the signal disappears completely.
Parker’s conclusion was clear: the Swedish sibling “no association” result was a statistical artifact, not evidence of safety.
The rebuttal was not just theoretical. Parker and colleagues marshaled 22+ lines of biological and epidemiological evidence—from oxidative stress and glutathione depletion to cord-blood biomarker studies showing dose–response risk—that all converge on acetaminophen as a real neurodevelopmental hazard.²
The Preprint With the Powerful Title
In August 2025, a new preprint appeared with a title that pulled no punches:
“Evidence That Acetaminophen Triggers Autism in Susceptible Individuals Has Been Ignored and Mishandled for More than a Decade.”³
The authors reviewed the literature and found:
66% of studies mishandled data by treating interacting variables (like oxidative stress) as simple confounders.
77% of studies examined only part of the exposure window (e.g., pregnancy only, ignoring neonatal use).
And most strikingly: early high-risk signals like Schultz’s 2008 finding of a 20.9-fold increased risk of regressive autism after acetaminophen use at 12–18 months were dismissed instead of investigated.⁴
This is not just a historical footnote—it is the very signal many parents had long reported, now vindicated by careful re-analysis.
The Bigger Picture
When you step back, the convergence is undeniable:
Epidemiology: Meta-analyses consistently show ~20–30% increased risk of autism/ADHD with prenatal Tylenol.²
Biomarkers: Cord-blood studies reveal dose–response risk, with the highest acetaminophen exposure tripling autism odds.²
Mechanism: Oxidative stress, glutathione depletion, sulfation deficits, and endocrine disruption—all biologically plausible, all supported by data.²
History: A forgotten 2008 survey showing a 20.9-fold risk for regressive autism has returned as one of the most important red flags of all.⁴
Against this backdrop, the Swedish “no association” study is not reassurance—it is a distraction from the weight of evidence.
Where We Are Now
With HHS preparing to formally acknowledge a Tylenol–autism link, the narrative is shifting rapidly. The precautionary principle—limit exposure until proven safe—should have been applied years ago.
The real question now is not whether acetaminophen carries risk. The evidence says it does. The question is: how many cases of autism could have been prevented if warnings had been heeded sooner?
Learn more by reading Part I below.
References
Viktor H. Ahlqvist et al., “Acetaminophen Use During Pregnancy and Children’s Risk of Autism, ADHD, and Intellectual Disability,” JAMA 331, no. 14 (April 9, 2024): 1205–14. https://doi.org/10.1001/jama.2024.3172.
William Parker et al., “Evaluating the Role of Susceptibility Inducing Cofactors and of Acetaminophen in the Etiology of Autism Spectrum Disorder,” Life 14, no. 8 (July 23, 2024): 918. https://doi.org/10.3390/life14080918.
M. Vishnu Patel et al., “Evidence That Acetaminophen Triggers Autism in Susceptible Individuals Has Been Ignored and Mishandled for More than a Decade,” Preprints (August 1, 2025). https://doi.org/10.20944/preprints202508.0006.v1.
Stephen T. Schultz et al., “Acetaminophen (Paracetamol) Use, Measles-Mumps-Rubella Vaccination, and Autistic Disorder: The Results of a Parent Survey,” Autism 12, no. 3 (2008): 293–307. https://doi.org/10.1177/1362361308089516.
Sayer Ji, “Breaking: Government Finally Admits Tylenol–Autism Link After Years of Corporate Cover-Up,” GreenMedInfo Substack, September 5, 2025. https://greenmedinfo.substack.com.