Large-Scale Study Finds No Association Between Acetaminophen in Pregnancy and Autism

A long-standing debate has gripped public health discussions: does acetaminophen, the active ingredient in popular pain relievers like Tylenol or paracetamol, pose a risk to fetal brain development when used during pregnancy? Concerns arose from early observational studies hinting at possible links to autism spectrum disorder (ASD), attention-deficit/hyperactivity disorder (ADHD), and intellectual disability. However, a monumental investigation led by researchers from prestigious universities has delivered compelling evidence to the contrary, alleviating fears for millions of expectant mothers worldwide.

This breakthrough comes from an unprecedented large-scale analysis involving over 2.4 million children, spearheaded by epidemiologists at Karolinska Institutet in Sweden and Drexel University in the United States. The study's rigorous design, particularly its innovative use of sibling comparisons, cuts through confounding factors like shared genetics and family environments, providing the strongest evidence yet that there is no causal connection between maternal acetaminophen use and these neurodevelopmental conditions.

🌡️ The Origins of the Controversy

Acetaminophen, known as paracetamol outside North America, is the world's most recommended analgesic and antipyretic for pregnant women due to its favorable safety profile compared to alternatives like ibuprofen or aspirin, which carry higher risks of complications such as bleeding or preterm labor. First approved for over-the-counter use decades ago, it alleviates common pregnancy ailments including headaches, back pain, fever from infections, and muscle strains.

Initial alarms sounded around 2018-2019 with studies from Johns Hopkins University and Harvard T.H. Chan School of Public Health reporting associations between prenatal exposure and elevated odds of ASD and ADHD. For instance, a 2019 analysis of umbilical cord blood samples suggested newborns with higher acetaminophen levels faced triple the risk of ADHD. These findings, echoed in meta-analyses, prompted cautionary advisories and even lawsuits alleging birth defects. Yet, critics noted these were observational, prone to biases where sicker mothers (with fevers or pain indicating underlying issues) used more medication, confounding results.

By 2025, the U.S. Food and Drug Administration (FDA) responded to mounting pressure by initiating label updates to warn of potential risks, amid political rhetoric from figures like former President Trump claiming a "very increased risk of autism." The World Health Organization (WHO) countered, stating no conclusive causal evidence existed. This tension underscored the need for definitive research from academic powerhouses.

Unpacking the Mega-Study's Design

The landmark research, published in the Journal of the American Medical Association (JAMA) in April 2024, harnessed Sweden's gold-standard national registries—a treasure trove for longitudinal epidemiology. Covering singleton births from 1995 to 2019 (follow-up to 2021), it tracked 2,480,797 children, with 185,909 (7.5%) exposed to acetaminophen via prescriptions or antenatal reports.

Exposure was meticulously defined: 'ever-use' from the Medical Birth Register (prospective data at prenatal visits) and Prescribed Drug Register (post-2005). Outcomes—ASD, ADHD, intellectual disability—drawn from ICD-coded National Patient Register diagnoses, capturing 2.8%, 5.9%, and 1.0% incidences respectively.

The game's changer: sibling control analysis on 1,773,747 full siblings. By comparing exposed and unexposed siblings within families, researchers neutralized unmeasured confounders like parental genetics, socioeconomic status, or prenatal care habits. Advanced Cox proportional hazards models adjusted for 20+ covariates, including maternal BMI, smoking, infections, migraines, and income. Dose-response curves and sensitivity tests (e.g., multiple imputation for missing data) fortified robustness.

Results That Reshape the Narrative

Population-wide, ever-use showed tiny hazard ratios (HRs): 1.05 for ASD (95% CI 1.02-1.08), 1.07 for ADHD, 1.05 for intellectual disability—barely above 1, with absolute risk differences under 0.3% at age 10. But E-values revealed modest unmeasured confounding could explain them away.

Sibling controls flipped the script: HRs plummeted to 0.98 (ASD), 0.98 (ADHD), 1.01 (intellectual disability)—all inclusive of 1.0, signaling no association. No dose-response emerged; high-dose siblings even trended protective (HR 0.88 ASD). Similar nulls held for other analgesics, implicating familial liabilities over drugs.

• Key Statistic: Among siblings discordant for exposure, autism risk difference was a negligible 0.02% (95% CI -0.14% to 0.18%).
• Follow-up: Median 13.4 years, up to 26 years—ample for diagnoses.
• Validation: High ASD diagnostic accuracy (94% in subsets).

University Luminaries Behind the Science

Lead architects hail from top-tier institutions. Viktor H. Ahlqvist, PhD, from Karolinska's Department of Global Public Health and Neuroscience, co-led with Hugo Sjöqvist. Co-senior authors Brian K. Lee, PhD (Drexel Dornsife School of Public Health and A.J. Drexel Autism Institute), and Renee M. Gardner, PhD (Karolinska), brought transatlantic expertise. Christina Dalman, MD, PhD, and others from Karolinska University Hospital rounded the team.

Lee emphasized: "Sibling comparisons provide the gold standard for causal inference in observational data." Funded by NIH's National Institute of Neurological Disorders and Stroke, this exemplifies university-driven public health impact. Drexel's press release hailed it as "the largest study to date," reassuring clinicians amid prior fears that deterred 60% of Swedish pregnant individuals from medications.

Recent corroboration: A Danish nationwide cohort in JAMA Pediatrics (April 2026), analyzing 1.5 million births (1997-2022), found no autism link (1.8% exposed vs. unexposed rates, adjusted OR near 1.0, upper CI <1.12). Details in the Danish registry study mirror Swedish rigor.

Addressing Confounders and Past Research

Prior associations stemmed from residual confounding. Mothers using acetaminophen often had migraines, fevers, or infections—known ASD/ADHD risks themselves. Genetic pleiotropy (genes predisposing pain and neurodevelopment) or indication bias (pain signaling inflammation) explained signals. Sibling designs dissect these.

Contrasting views persist: A 2025 Harvard update and Mount Sinai study suggested links, but lacked sibling controls. Systematic reviews (e.g., BMJ 2025 of nine meta-analyses) increasingly favor nulls. WHO's 2025 statement: "No consistent association established." Access the full Karolinska-Drexel JAMA paper for methodologies.

Real-World Implications for Expectant Families

For pregnant individuals, acetaminophen remains first-line: Untreated fever raises miscarriage/neural tube defect risks; pain exacerbates stress. Guidelines (ACOG, WHO) endorse shortest duration, lowest dose. Post-study, usage dipped amid hype, but evidence supports judicious use.

Stakeholders—pediatricians, obstetricians, policymakers—gain clarity. No mass label panics needed; focus shifts to holistic prenatal care.

Higher Education's Pivotal Role

Universities like Karolinska (Nobel Prize birthplace) and Drexel exemplify how academic research translates to policy. Interdisciplinary teams blending epidemiology, neuroscience, biostatistics drive discoveries. Training grounds for future experts, these institutions offer programs in public health, fueling global health security.

Amid rising neurodevelopmental diagnoses (ASD prevalence ~1 in 36 U.S. children), university labs pioneer biomarkers, interventions. This study underscores epidemiology's power in debunking myths.

• Benefits of university research: Massive cohorts via registries, ethical sibling designs, rapid peer-review.
• Challenges: OTC exposure gaps, diagnosis validations.

Future Trajectories in Prenatal Research

Prospects gleam: Mendelian randomization (genetic proxies for exposure), animal models probing mechanisms, prospective biomarkers. Emerging 6G AI at Khalifa University predicts networks; similar tech could model fetal exposures. NTU Singapore mandates AI literacy 2026—tools for tomorrow's researchers.

Global collaborations, like EU-funded cohorts, promise deeper insights. For neurodevelopment, prioritize infections, pollutants over safe meds.

In summary, this university-led odyssey dispels fears, empowering informed choices. Academic rigor safeguards maternal-fetal health, one mega-study at a time.

Stakeholder perspectives abound: Autism Speaks lauds null findings; obstetric societies reaffirm safety. Actionable: Consult providers, track symptoms, embrace evidence-based care.

Photo by Annie Spratt on Unsplash