In a Sept. 22 press conference that the New York Times dubbed “rambling,” President Donald Trump’s message came through loud, clear, and often.
“Don’t take Tylenol,” he said repeatedly. “And don’t give it to your babies.”
For nearly two hours, Trump – along with Health and Humans Services Secretary Robert Kennedy Jr. and others – pointed to a causal link between acetaminophen and autism (without any new data). All of this despite the assurances of multiple health agencies from around the world and medical experts.
“Suggestions that acetaminophen use in pregnancy causes autism are not only highly concerning to clinicians but also irresponsible when considering the harmful and confusing message they send to pregnant patients, including those who may need to rely on this beneficial medicine during pregnancy,” American College of Obstetricians and Gynecologists President Steven J. Fleischman, MD, said in a statement. “Today’s announcement by HHS is not backed by the full body of scientific evidence and dangerously simplifies the many and complex causes of neurologic challenges in children. It is highly unsettling that our federal health agencies are willing to make an announcement that will affect the health and well-being of millions of people without the backing of reliable data.”
Kenvue – Tylenol’s manufacturer – also pushed back in a statement sent to multiple media outlets.
“We believe independent, sound science clearly shows that taking acetaminophen does not cause autism,” a company spokesperson said. “We strongly disagree with any suggestion otherwise and are deeply concerned with the health risk this poses for expecting mothers.”
The Administration’s Case
HHS – along with its sister agencies – relied on a recently published review of nearly four dozen eligible studies. (It’s worth noting that the National Institute of Health funded this review.) These included 20 on ADHD, 8 on ASD, and 18 on other NDDs.
Among them, 27 studies reported positive associations, nine uncovered no significant association, and four suggested protective effects.
The researchers documented reported risks that include increased diagnoses of ADHD, ASD, and related behavioral or cognitive deficits. The authors added that sensitivity analyses confirmed that methodological biases such as recall bias, confounding by indication, or exposure misclassification couldn’t fully account for the associations.
The authors, citing what they see as a convergence of evidence, conclude that prenatal acetaminophen exposure is “consistently associated” with elevated risks of NDDs.
But ACOG’s Fleischman took issue with the researchers’ claims.
“The studies that are frequently pointed to as evidence of a causal relationship, including the latest systematic review released in August, include the same methodological limitations,” he said, “for example, lack of a control for confounding factors or use of unreliable self-reported data – that are prevalent in the majority of studies on this topic.”
Testifying for the Defense…
Fleischman also countered by citing another study, which JAMA published last year, that found slight risks that disappeared once the researchers introduced confounding factors.
Using national health registries, a group of Swedish researchers followed nearly 2.5 million children born between 1995 and 2019, including more than 185,000 exposed to acetaminophen in utero. The team tracked participants for a decade.
In standard population models, the researchers found a small but statistically significant higher risk with acetaminophen use:
- 0.09% for autism.
- 0.26% for ADHD, and
- 0.08% for ID.
But when the researchers applied sibling-comparison models, those associations leveled off and appeared to be no longer significant. The sibling design, the authors contend, suggested that confounding factors, such as parental health conditions and familial traits, might explain much of the perceived increase in risk.
On the other hand, exposure to opioid and antimigraine medications during pregnancy showed stronger associations with neurodevelopmental disorders, persisting even after sibling comparisons.
In short, the authors concluded that while acetaminophen use in pregnancy appears linked to modest increases in child neurodevelopmental diagnoses in conventional analyses, the absence of association in sibling comparisons suggests no causal relationship.
Closing Arguments
But the presser didn’t end there. Trump also took the opportunity to sow doubt about the existing child and adolescent vaccination schedule that the country’s pediatricians (and most parents) have relied on for roughly 30 years.
“Don’t do it,” Trump said at one point in his argument against the evidence-based schedule. “Get them broken into four or five visits.
“There’s no downside. It can only be good,” he added.
But the American Academy of Pediatrics released a statement that suggests otherwise.
“Today’s White House event on autism was filled with dangerous claims and misleading information that sends a confusing message to parents and expecting parents and does a disservice to autistic individuals,” it read. “Studies have repeatedly found no credible link between life-saving childhood vaccines and autism. This research, in many countries, involving thousands of individuals, has spanned multiple decades. Any effort to misrepresent sound, strong science poses a threat to the health of children.
“Spacing out or delaying vaccines means children will not have immunity against these diseases at times when they are most at risk,” the pediatrics group concluded.
Further Reading
New Autism Study Uncovers Four Biological Subtypes
Siblings With Autism Share More of Father’s DNA, Not Mother’s
Prenatal Exposure to Acetaminophen and the Risk of Attention-Deficit/Hyperactivity Disorder
