What does the science say on both sides of Trump’s new Tylenol warning?

Key Takeaways: The Evidence at a Glance

Understanding Association vs Causation: A Critical Distinction

Before examining the evidence, it’s crucial to understand that association does not equal causation. When studies find that children exposed to acetaminophen in utero have higher rates of autism or ADHD, this is an association – two things occurring together. Causation means one directly causes the other, which is much harder to prove.

Consider this analogy: Ice cream sales and drowning deaths both increase in summer. They’re associated, but ice cream doesn’t cause drowning – warm weather is the common factor. Similarly, women who take acetaminophen during pregnancy may have underlying conditions (pain, infections, stress) that could independently affect fetal development. This is called “confounding by indication.”

The gold standard for proving causation would be randomized controlled trials, but these are unethical in pregnant women. Instead, scientists use observational studies and look for consistent patterns, dose-response relationships, biological plausibility, and what happens when confounding factors are controlled – which is exactly where the debate stands today.

Research Finding Associations (Not Proven Causation)

The Numbers Show Consistent Associations

The most comprehensive analysis to date, published in Environmental Health (2025), examined 46 studies and found that 27 studies showed positive associations with neurodevelopmental disorders (not specifically autism/ADHD), while 9 showed no effect, 4 suggested protective effects, and 6 had mixed results.² Importantly, the researchers noted these were associations, not proven causal relationships, but concluded that “higher-quality studies were more likely to show positive associations.”²

Biomarker Studies: Stronger Evidence but Still Observational

Perhaps the most compelling associative evidence comes from a 2020 Boston Birth Cohort study published in JAMA Psychiatry that measured acetaminophen directly in umbilical cord blood:⁵

• Babies with the highest acetaminophen exposure had 3.62 times higher autism risk⁵
• These same babies had 2.86 times higher ADHD risk⁵
• 100% of cord blood samples contained acetaminophen, even though only 45% of mothers reported using it⁵
• The researchers found a clear dose-response pattern: more acetaminophen = higher risk⁵

Why this matters but isn’t definitive: While biomarker studies eliminate recall bias and show dose-response relationships (strengthening the case for causation), they remain observational. The mothers who had higher acetaminophen levels may have had more severe illnesses, pain, or stress – factors that could independently affect neurodevelopment.

Multiple Meta-Analyses Find Remarkably Consistent Associations

Three independent meta-analyses found nearly identical results:

• Masarwa et al. (2018) analyzed 132,738 mother-child pairs and found a 19% increased autism risk⁷
• Alemany et al. (2021) studied 73,881 European children and found 19% higher autism symptoms with prenatal exposure⁸
• Ricci et al. (2023) reviewed 22 studies and confirmed the 19% increased risk⁹

The association vs causation caveat: All three research teams explicitly noted that their findings showed associations only and could not establish causation due to potential unmeasured confounding factors.^{7 8 9}

Proposed Biological Mechanisms: Plausible but Unproven

Scientists have identified several potential mechanisms, as documented in the Journal of Clinical Endocrinology & Metabolism (2022):¹⁰

• Acetaminophen crosses the placenta and may act as an endocrine disruptor¹⁰
• It depletes glutathione, the brain’s main antioxidant defense³
• It potentially disrupts the endocannabinoid system critical for brain development³
• It may interfere with hormone production necessary for normal fetal development¹⁰

Critical context: While these mechanisms are biologically plausible, they haven’t been definitively proven to cause autism or ADHD in humans at typical therapeutic doses.

Scientists Calling for Precautionary Action

91 international scientists signed a consensus statement in Nature Reviews Endocrinology (2021) stating: “Based on experimental and epidemiological literature, we believe the potential for harm from continued inaction exceeds the harm that might arise from precautionary action.”¹¹

However, they acknowledged the evidence shows associations, not proven causation, and recommended a precautionary approach rather than definitive conclusions.¹¹

Research Challenging the Causal Link

The Largest Study: Associations Disappear with Better Controls

The most powerful challenge to causation comes from a 2024 Swedish study in JAMA that followed 2.48 million children:⁴

• Initial population analyses found small but statistically significant associations⁴
• Crucially: When comparing siblings with different exposures, all associations completely disappeared⁴
• Autism risk with acetaminophen in sibling analysis: 0.98 (meaning no increased risk)⁴
• ADHD risk with acetaminophen in sibling analysis: 0.98 (no increased risk)⁴
• The researchers concluded: “The lack of association in the sibling-control analysis suggests confounding by unmeasured familial factors”⁴

Why sibling studies matter: They control for genetic factors, family environment, and socioeconomic status – major confounders that other studies can’t adequately address.

The Confounding Problem: Association Doesn’t Equal Causation

A 2024 clinical review in Obstetrics & Gynecology identified critical reasons why associations may not indicate causation:¹²

• Genetic confounding: Parents with ADHD genes are more likely to need pain medication AND have children with ADHD¹²
• Confounding by indication: Women who take acetaminophen often have infections, fever, or chronic pain – all independently linked to autism¹²
• The review concluded the associations are “unlikely to confer clinically important increased risk”¹²

Known Risks of NOT Treating Pain and Fever

Multiple studies document that untreated conditions during pregnancy cause established harms:

• Untreated fever: Proven to increase risk of neural tube defects, heart defects, and facial abnormalities^{6 13}
• Untreated pain: Associated with maternal depression, anxiety, and hypertension³
• Alternative medications: NSAIDs like ibuprofen have FDA warnings for pregnancy complications¹⁴

Medical Organizations: Associations Don’t Warrant Policy Changes

American College of Obstetricians and Gynecologists (ACOG): States that despite associations found in some studies, “not a single reputable study has successfully concluded that the use of acetaminophen causes neurodevelopmental disorders.”¹⁵ They emphasize the distinction between correlation and causation.¹⁵

Society for Maternal-Fetal Medicine: Maintains “the weight of evidence is inconclusive regarding a possible causal relationship” and notes that associations in observational studies cannot prove causation.¹⁶

European Medicines Agency: Reviewed all available evidence and concluded acetaminophen “can be used during pregnancy if clinically needed” while acknowledging some studies show associations.¹⁷

The Core Debate: Can We Move from Association to Causation?

Arguments that associations may indicate causation:

• Consistency across multiple studies and populations²
• Dose-response relationships in biomarker studies⁵
• Biologically plausible mechanisms¹⁰
• Temporal relationship (exposure precedes outcome)

Arguments against causation:

• Associations disappear in sibling-controlled analyses⁴
• Strong potential for confounding by indication¹²
• No randomized controlled trial evidence
• Alternative explanations for all observed associations¹²
• Lack of specificity (acetaminophen associated with multiple different outcomes)²

TLDR

References

1. HHS Press Release – Trump administration on autism
2. Environmental Health – Systematic review of acetaminophen and neurodevelopmental disorders (2025)
3. Nature Reviews Endocrinology – Prenatal acetaminophen exposure and neurodevelopmental outcomes
4. JAMA – Swedish sibling-controlled study of 2.48 million children (2024)
5. JAMA Psychiatry – Boston Birth Cohort cord blood biomarker study
6. Duke Health – Fever during pregnancy and birth defects
7. American Journal of Epidemiology – Meta-analysis of prenatal acetaminophen exposure (2018)
8. European Journal of Epidemiology – European birth cohort meta-analysis (2021)
9. Italian Journal of Pediatrics – Systematic review and meta-analysis (2023)
10. Journal of Clinical Endocrinology & Metabolism – Endocrine disruption mechanisms (2022)
11. Nature Reviews Endocrinology – Consensus statement from 91 scientists (2021)
12. Obstetrics & Gynecology – Clinical review of confounding factors (2024)
13. Birth Defects Research – Maternal fever and congenital anomalies
14. FDA – NSAIDs pregnancy warning
15. ACOG – Official statement on acetaminophen safety in pregnancy
16. Society for Maternal-Fetal Medicine – Position statement
17. European Medicines Agency – Safety review recommendations