Acetaminophen use in pregnancy and autism: separating controversy from scientific evidence

Abstract

Acetaminophen (paracetamol) has long been recommended as the preferred first-line analgesic and antipyretic in pregnancy, supported by decades of clinical experience, and has a more favorable safety profile than nonsteroidal anti-inflammatory drugs or opioids. Recently, however, political statements and media coverage linking maternal tylenol use to autism risk have amplified public concern and created uncertainty in clinical practice. This narrative review summarizes the current epidemiological evidence and positions of major professional societies and regulatory agencies on prenatal acetaminophen exposure and neurodevelopmental outcomes. A large Swedish nationwide cohort with a siblingcontrol analysis found no significant association between acetaminophen use during pregnancy and children’s risk of autism spectrum disorder, attentiondeficit/hyperactivity disorder, or intellectual disability, suggesting that the previously reported modest risk elevations largely reflect confounding and methodological limitations. Similar null findings from other high-quality cohort and siblingcomparison studies indicate no causal relationship between therapeutic prenatal acetaminophen use and major neurodevelopmental disorders. Consistent with these data, organizations such as the World Health Organization, American College of Obstetricians and Gynecologists, and national regulators continue to endorse short-term guideline-concordant acetaminophen use as appropriate in pregnancy. In routine obstetric care, emphasis should be placed on maternal safety, fetal well-being, and evidence-based counseling rather than non-evidencebased concerns, with acetaminophen used at the lowest effective dose for the shortest necessary duration unless future high-quality data indicate otherwise.

Introduction

Maternal use of antipyretic and analgesic medications during pregnancy is often an essential component for safeguarding both maternal and fetal health. Pregnant women frequently experience headache, fever, musculoskeletal pain, and a variety of infectious or noninfectious conditions, and failure to treat significant symptoms can adversely affect the outcomes of both the mother and fetus. For decades, clinical practitioners and obstetric societies worldwide have recommended paracetamol (acetaminophen) as the safest antipyretic and analgesic option for pregnancy. Nevertheless, in 2025, the USA president publicly suggested that “Tylenol use in pregnancy increases the risk of autism”. This statement drew substantial public and professional attention to the safety of acetaminophen use during pregnancy and intensified concerns regarding the risk of medication-related autism. In response, regulatory and professional bodies, including the USA Food and Drug Administration (FDA), the Korean Ministry of Food and Drug Safety, and national medical associations, rapidly issued statements emphasizing that the current clinical evidence does not support a clear causal relationship and that use at a maximum daily dose of 4,000 mg, under professional guidance, remains appropriate and safe. Against this backdrop, obstetricians in everyday practice increasingly need to explain objective and clinically meaningful risks to patients and their families by drawing on long-term clinical experiences, large-scale cohort data, and contemporary high-quality evidence.

Clinical context of analgesic and antipyretic use in pregnancy

As pregnancy care must simultaneously account for the health and prognosis of the mother and fetus, medication choices must be made with particular caution. National and international obstetric guidelines concur that appropriate treatment of pain and fever during pregnancy can be critical for fetal well-being and maternal outcomes. Paracetamol is currently the most widely accepted antipyretic analgesic with several decades of use in pregnant populations. The World Health Organization (WHO) and multiple expert groups continue to recommend paracetamol as the preferred first-line treatment option during pregnancy. Where data were available, we highlighted the potential differences in risk according to the dose, timing of exposure during pregnancy, and duration of use. However, existing studies often lack precise information on indications, cumulative doses, and critical windows of susceptibility, and few have rigorously examined prolonged or repeated exposure. Further large, well-designed studies are needed to clarify whether and how the dose, timing, and duration, individually and in combination, influence neurodevelopmental outcomes.

Recent public controversy and policy responses

In 2025, President Trump publicly linked maternal tylenol use to autism risk in a White House press briefing, calling for “greater caution” regarding its use among pregnant women. Major regulatory agencies, including the USA FDA, the European Medicines Agency (EMA), and the Korean Ministry of Food and Drug Safety (MFDS), promptly issued guidelines stating that although some observational studies have reported modest associations, the current evidence base does not demonstrate causality and that pregnant women may continue to use acetaminophen when clinically indicated, in consultation with their clinicians, and within recommended dose limits. In parallel, professional bodies, such as national obstetrics and gynecology societies and medical associations, have stressed that public statements that are not grounded in current evidence should not be used as a basis for changing clinical practice. Similarly, the WHO reiterated that the risks of untreated high fever in pregnancy are likely to exceed any theoretical neurodevelopmental risk attributed to short-term, guideline-concordant paracetamol use.

Methodological scientific evidence and methods

A Swedish nationwide cohort study published in JAMA in 2024 analyzed more than 2.48 million children born between 1995 and 2019 and systematically assessed the association between prenatal acetaminophen exposure and subsequent diagnoses of autism spectrum disorder (ASD), attention-deficit/hyperactivity disorder (ADHD), and intellectual disability. A key methodological strength of this study was its use of siblingcontrol analyses, which allowed control for shared familial genetic and environmental factors, there-by reducing confounding that often limit traditional cohort and cross-sectional studies. In the crude and conventional adjusted models, small risk elevations were observed among acetaminophen-exposed offspring; however, these associations largely disappeared or were markedly attenuated in the siblingcomparison analyses. When family history and shared genetic and environmental factors were accounted for, the hazard ratios were 0.98 (95% confidence interval [CI], 0.94–1.02) for ASD, 0.98 (95% CI, 0.95–1.01) for ADHD, and 1.01 (95% CI, 0.96–1.07) for intellectual disability, indicating no statistically significant or clinically meaningful increase in risk attributable to prenatal paracetamol exposure. These findings strongly suggest that the modest risk elevations reported in earlier observational studies and meta-analyses are likely explained by residual confounding, secular changes in diagnostic criteria and awareness, and other methodological limitations, rather than a true causal effect. Commentaries in leading journals, such as Nature and JAMA, have high-lighted that in rigorous designs with high levels of control for confounding, no significant increase in neurodevelopmental risk is observed, and that self-reported exposure, imprecise dosing information, and recall bias in earlier work severely constrain causal inference. In addition, the marked increase in ASD prevalence over the past two decades is now thought to reflect, to a substantial extent, broadened diagnostic criteria, increased clinical recognition, and improved social awareness, rather than a true surge in underlying incidence [1].

A large Japanese birth cohort study published in 2024 in Pediatric and Perinatal Epidemiology applied a similar sibling-comparison design in more than 200,000 children and found no clear association between prenatal paracetamol use and subsequent ASD or ADHD. The replication of null findings across large, methodologically robust siblingcontrol cohorts in different countries has strengthened the conclusion that while weak associations can appear in conventional observational analyses, a causal relationship between therapeutic maternal acetaminophen use and child neurodevelopmental disorders is not supported [2].

In contrast, earlier observational reports, including a 2019 umbilical cord blood study by Johns Hopkins reported a 2–3-fold higher risk of ASD or ADHD in children with higher cord concentrations of acetaminophen metabolites [3]. However, these findings have not been corroborated by large national cohorts or siblingcomparison studies. Leading organizations such as the EMA, WHO, and major medical societies emphasize that such results should be interpreted as correlations rather than evidence of causality. Notably, the cordblood study itself acknowledged that acetaminophen has a relatively short half-life of approximately 3 hours, and that detectable cord levels most likely reflect peripartum use rather than cumulative exposure throughout pregnancy. More recent work from institutions such as Harvard and Mount Sinai similarly concluded that when indications, doses, and durations are considered and use remains within therapeutic ranges, any incremental neurodevelopmental risk associated with acetaminophen appears to be very small in absolute terms [4].

Taken together, contemporary siblingcomparison cohort studies overcome the key limitations of earlier cross-sectional and meta-analyses and provide a more reliable basis for judging the safety of tylenol use in pregnancy. Based on the strength of these data and the convergent positions of major professional and regulatory bodies, clinicians can communicate a clear message to patients and families: the current high-quality evidence does not support the claim that tylenol use in pregnancy significantly increases the risk of autism. In particular, the largest Swedish nationwide study found no significant association between maternal acetaminophen use during pregnancy and ASD in offspring when rigorous family based controls were applied [5].

Positions of professional societies and regulatory agencies

The American College of Obstetricians and Gynecologists (ACOG), in its 2025 Practice Advisory, stated that “acetaminophen remains the safest firstline analgesic and antipyretic in pregnancy, based on decades of clinical experience and the highestquality evidence”, and emphasized that, in light of the limitations of existing observational studies-including residual confounding, exposure misclassification, and heterogeneity of outcomes-current data do not support a causal relationship between therapeutic acetaminophen use for pain or fever in pregnancy and neurodevelopmental disorders such as autism. The ACOG further cautioned that clinical practice should not be swayed by social or political anxiety or by overinterpretation of methodologically weak studies and that risk-benefit assessments must prioritize maternal health, fetal protection, and the well-documented complications of untreated high fever and severe pain [6].

Similarly, the Royal College of Obstetricians and Gynecologists and the International Federation of Gynecology and Obstetrics have reaffirmed that based on current scientific evidence, recommendations for analgesic and antipyretic use in pregnancy remain unchanged and that short-term, guideline-concordant use of paracetamol continues to be considered safe worldwide [7,8]. Regulatory agencies in the United Kingdom (including the National Health Service, EMA, and Medicines and Healthcare products Regulatory Agency) have stated that paracetamol remains the recommended first-line analgesic during pregnancy and that contrary to recent public controversies, the risks of undertreated pain and fever are likely to be greater than any hypothetical neurodevelopmental harm from appropriate acetaminophen use [9].

In Korea, the Ministry of Food and Drug Safety and the Korean Medical Association have repeatedly emphasized that claims that maternal tylenol use “causes” autism are not supported by established scientific evidence and that pregnant women should use paracetamol when needed, after consultation with clinicians and taking into account gestational age and symptom severity, without exceeding a maximum daily dose of 4,000 mg [10]. In contrast, nonsteroidal anti-inflammatory drugs should generally be limited to the minimum effective dose for the shortest duration between 20 weeks and 30 weeks of gestation and avoided after 30 weeks because of the well-recognized fetal risks [11–13]. The MFDS has also warned that failing to treat significant maternal fever or pain may increase the risk of preterm birth, growth restriction, and other serious complications and has urged that decisions regarding indication, dose, and duration should be made in conjunction with specialist care [14].

Across the USA, Europe, Japan, and other high-income countries, authoritative medical organizations and public health agencies have consistently stated that, when interpreted in the context of recent large siblingcomparison cohorts, national clinical data, and the WHO commentary, paracetamol remains an appropriate first-line agent for treating pain and fever in pregnancy. Clinical guidelines are not modified based on media controversy alone; instead, academic societies and regulators stress that individualized risk-benefit assessment, adherence to official recommendations, and scientifically grounded counseling are essential for protecting both maternal and fetal health.

As this article is a narrative (non-systematic) review, it does not aim to exhaustively identify all existing studies, and is inherently more susceptible to selection and interpretive bias than a formal systematic review. These limitations should be considered when interpreting the conclusions of this study.

Conclusion

Synthesizing the latest large-scale cohort and siblingcomparison evidence together with the official positions of the WHO, ACOG, and national regulatory bodies indicates that the current data do not support a causal relationship between therapeutic tylenol (paracetamol, acetaminophen) use in pregnancy and autism or other major neurodevelopmental disorders. Non-evidence based concerns about acetaminophen should not lead pregnant women to forgo the necessary treatment of pain and fever, as uncontrolled high fever and significant pain may themselves increase the risk of adverse outcomes such as miscarriage, congenital anomalies, and preterm birth. In clinical practice, the primary focus must remain on maternal safety, fetal well-being, and practical patient counseling rather than on social controversy or public statements not supported by current evidence, and care should be guided by evidence-based medicine. Considering the currently available body of evidence, tylenol (acetaminophen) should remain the first-line option for managing pain and fever during pregnancy. Unless future high-quality data indicate otherwise, it is recommended that it be used at the lowest effective dose for the shortest necessary duration under appropriate medical supervision. A Korean-language factsheet summarizing key counseling points for pregnant women is provided in Supplementary Material 1.