New meta-analysis published in The Lancet Obstetrics, Gynecology & Women’s Health reviewed roughly 60 studies and found no causal link between acetaminophen taken as directed during pregnancy and a later diagnosis of autism, attention-deficit/hyperactivity disorder (ADHD) or intellectual disability. The research prioritized sibling-comparison studies to reduce the influence of family-level genetic and environmental factors. Authors and several professional bodies said the findings support existing guidance that acetaminophen (paracetamol/Tylenol) remains the first-line option for fever or pain in pregnancy when used at the lowest effective dose for the shortest time. Some government officials and commentators criticized the review’s scope and interpretation, a dispute that the authors acknowledge as part of an ongoing scientific debate.
Key Takeaways
- The review synthesized results from about 60 studies and concentrated on sibling-comparison designs to control for familial confounders.
- In sibling analyses, prenatal acetaminophen exposure was not associated with increased risk of autism spectrum disorder, ADHD, or intellectual disability, including in studies with more than five years of follow-up.
- Authors caution that most earlier observational studies may have been confounded by the underlying reasons for medication use, such as fever or infection.
- Professional bodies including the American College of Obstetricians and Gynecologists (ACOG) align with the finding that acetaminophen is appropriate when used per label during pregnancy.
- The review’s limitations include the relatively small number of sibling-design studies and remaining uncertainty about less common patterns of long-term exposure.
- Public-health messaging must balance the risks of untreated fever and pain against potential, but unproven, medication harms.
- Manufacturers and some officials welcomed the analysis; others, including an HHS official, argued the review selectively narrowed evidence.
Background
Acetaminophen, also called paracetamol and sold under brand names such as Tylenol, is widely recommended as the preferred analgesic and antipyretic during pregnancy because alternatives like NSAIDs and opioids carry recognized risks. Use during pregnancy is common; clinicians typically advise limited, label-directed dosing only when needed. Over the past decade, multiple observational studies reported mixed results, with some indicating small associations between prenatal acetaminophen exposure and later neurodevelopmental diagnoses, prompting public concern and debate. Epidemiologists have highlighted a key methodological problem in the literature: people take acetaminophen for fever, infection or inflammation—conditions that themselves may affect fetal brain development—so distinguishing effects of the medication from effects of the underlying illness is difficult.
Sibling-comparison studies compare outcomes among siblings born to the same parent when one pregnancy involved acetaminophen use and another did not, which helps control for shared genetics and many household factors. That design does not eliminate all sources of bias—time-varying exposures and measurement differences remain possible—but it reduces confounding that can inflate associations in simple cohort analyses. Regulators and professional groups have previously issued cautious statements noting that available evidence has not established causality, and they emphasize treating fever in pregnancy because elevated maternal temperature can carry risks for both the pregnant person and the fetus.
Main Event
The analysis published Friday aggregated results from approximately 60 published studies and gave greater weight to sibling-comparison designs, which the authors say offer stronger protection against family-level confounding. In these sibling-based analyses, the team found no statistically meaningful association between recommended acetaminophen use in pregnancy and diagnoses of autism, ADHD, or other intellectual disabilities in the children studied. The conclusion persisted in studies that followed children for more than five years, a timeframe commonly used to capture developmental diagnoses.
Study co-author Dr. Asma Khalil, a professor of obstetrics and maternal-fetal medicine at St. George’s Hospital in London, summarized the primary message as: the highest-quality evidence does not support a causal link between taking paracetamol during pregnancy and autism or ADHD. She reiterated clinical advice that pregnant people should use the lowest effective acetaminophen dose for the shortest necessary period and seek medical care for severe or persistent symptoms, particularly fever. The authors and several medical societies noted that avoiding treatment for fever carries its own risks, and alternatives to acetaminophen may be less safe in pregnancy.
The paper arrived amid a public controversy: in September 2025, President Donald Trump publicly warned that acetaminophen in pregnancy was linked to a ‘very increased risk of autism’ and urged pregnant women not to take Tylenol; that remark was repeated on social media and prompted responses from public-health agencies. The Food and Drug Administration at the time issued a note indicating that there was no established evidence that acetaminophen causes autism and that the association remained under scientific investigation. HHS later criticized the new review’s methodology, saying the authors privileged one study design and excluded a majority of relevant evidence; the review’s authors and other researchers dispute that characterization.
Analysis & Implications
The sibling-comparison approach strengthens causal inference by holding constant many nonmeasured family-level variables, but it is not immune to bias. Time-varying confounders—such as changes in maternal health, behavior, or environment between pregnancies—can still influence both medication use and child outcomes. Measurement error in exposure timing, dose, or duration can also dilute detectable associations, particularly if prenatal use is under-reported or misclassified in medical records or questionnaires.
From a clinical and public-health standpoint, the review supports current guidance that acetaminophen remains an appropriate first-line agent for fever and pain in pregnancy when used per label. Importantly, untreated fever has documented associations with adverse pregnancy outcomes, meaning messaging that discourages necessary use of acetaminophen could inadvertently increase harm. Clinicians should therefore emphasize symptomatic assessment, treat significant fever or pain, and review persistent or recurrent symptoms rather than advise blanket avoidance of the drug.
Policy and research implications include a need for larger, well-measured prospective studies with precise exposure data and attention to dose-response relationships, timing in gestation, and long-term follow-up. In addition, harmonized reporting standards and pooled individual-participant data might clarify whether specific patterns of prolonged or high-dose exposure—rather than typical short-term, label-directed use—carry risks. Communication strategies must also counteract oversimplified social-media narratives that can misrepresent nuanced evidence.
Comparison & Data
| Study Design | Approx. Share among Reviewed Studies | Typical Follow-up |
|---|---|---|
| Sibling-comparison | Minority (small number) | Up to and beyond 5 years |
| Standard cohort/observational | Majority | Variable, often shorter |
| Case-control or registry analyses | Some | Variable |
The review examined roughly 60 studies overall and elevated sibling-comparison analyses for inference. The table above summarizes study types and typical follow-up windows; exact counts per design vary across the literature and many cohort studies differ in how exposure and outcomes are measured. The authors report that null findings in sibling analyses help explain why some earlier cohort studies suggested an association: confounding by indication (illness prompting acetaminophen use) may have inflated associations in simpler designs.
Reactions & Quotes
‘The clearest takeaway is that the best-quality evidence does not support a causal link between taking paracetamol during pregnancy and autism or ADHD in children.’
Dr. Asma Khalil, study co-author (St. George’s Hospital)
Dr. Khalil framed the result as reassurance for pregnant people and clinicians while reiterating typical dosing cautions.
‘Untreated fever has documented associations with serious pregnancy and fetal risks.’
Jessica B. Steier, science communicator (Unbiased Science)
Steier emphasized the risk balance, noting that alternatives to acetaminophen have their own documented pregnancy risks.
‘By excluding the vast majority of relevant evidence, privileging one study design known to bias results toward the null, the authors engineer a finding rather than evaluate causality.’
HHS official (statement relayed to ABC News)
The HHS comment represents a skeptical view of the review’s methodological choices; the authors and several professional societies contested that interpretation.
Unconfirmed
- The precise number of sibling-comparison studies and their statistical power to detect small effects remains limited and not fully enumerated in public summaries.
- Claims that acetaminophen causes a ‘very increased risk’ of autism, as stated publicly by some commentators, lack corroborating evidence in high-quality causal analyses.
- Whether rare patterns of prolonged high-dose prenatal acetaminophen exposure carry risk is still unresolved and requires targeted investigation.
Bottom Line
This comprehensive review, emphasizing sibling comparisons across roughly 60 studies, finds no support for a causal link between usual, label-directed acetaminophen use in pregnancy and later autism, ADHD or intellectual disability. The findings align with guidance from major medical organizations that recommend acetaminophen as the preferred medication for fever and pain in pregnancy when used appropriately.
Remaining uncertainties — notably about uncommon exposure patterns and limitations in available sibling-design studies — mean continued research is needed. In clinical practice, the priority is to treat significant fever or pain with the safest available option, use the lowest effective dose for the shortest necessary period, and seek medical evaluation for persistent or severe symptoms.
Sources
- ABC News (news report summarizing the review and reactions)
- The Lancet Obstetrics, Gynecology & Women’s Health (peer-reviewed journal where the meta-analysis was published)
- American College of Obstetricians and Gynecologists (ACOG) (professional medical organization guidance)
- U.S. Food and Drug Administration (FDA) (government agency statements on medication safety)
- Kenvue (Tylenol maker) (company statement on acetaminophen safety)