Acetaminophen Use During Pregnancy: Evaluating the Autism Risk Evidence
Dr. Joshua Lindsley, DO|Last Updated: March 2026|18 min read
Key Takeaways
The association between prenatal acetaminophen and autism is very small (5% relative risk increase) and likely not causal
Sibling-controlled studies—the gold standard for controlling genetics—show the association vanishes entirely
Genetics account for 80–90% of autism risk variability; diagnostic changes explain most of the 5-fold prevalence increase since 2000
Untreated maternal fever poses a greater risk to fetal neurodevelopment (155% increase) than any theoretical acetaminophen effect
ACOG and FIGO (2025) affirm acetaminophen as the safest analgesic option during pregnancy
Summary
Recent headlines linking acetaminophen (Tylenol) use during pregnancy to autism in children have generated significant public concern. A careful examination of the evidence using established epidemiological frameworks reveals that the association is very small (5% relative risk increase) and likely not causal.1 When rigorous statistical controls for genetics and family environment are applied—specifically through sibling analysis—the association vanishes entirely.
The challenge in evaluating such claims lies in understanding the nature of epidemiological evidence. Unlike mathematics, biology deals in probabilities rather than proofs. The Bradford Hill criteria—nine principles developed to assess whether observed associations are likely causal—provide a disciplined framework for evaluating such claims. When applied to the acetaminophen-autism question, most criteria either fail to support causality or actively argue against it.
Importantly, even if a small causal effect existed, it would be dwarfed by other known risk factors for autism: genetics account for 80–90% of autism risk variability,2 with advanced parental age, maternal metabolic health, preterm birth, and air pollution explaining much of the remainder. The dramatic rise in autism diagnoses over recent decades (5-fold increase since 2000) is largely explained by expanded diagnostic criteria (40–60%) and increased awareness (20–30%)—not environmental exposures like acetaminophen.3
FDA Pregnancy Drug Categories
Category
Description
% of Drugs
Examples
A
No demonstrated risk in controlled human studies
2–5%
Thyroid hormone, prenatal vitamins
B
No evidence of human risk; some animal signals
15–25%
Acetaminophen, metformin, many antibiotics
C
Risk cannot be ruled out; benefit may justify use
60–75%
GLP-1 agonists, some SSRIs, gabapentin
D
Positive evidence of human fetal risk; benefits may outweigh
5–8%
NSAIDs (3rd trimester), lithium, some seizure meds
X
Proven significant harm; contraindicated
1–3%
Statins, methotrexate
Key Point: Acetaminophen is Category B—considered safe with no demonstrated human fetal risk. NSAIDs (ibuprofen, Advil) become Category D in the third trimester due to risks of premature closure of fetal blood vessels.
The Bottom Line
The evidence does not support a causal link between acetaminophen use during pregnancy and autism. The small observed associations (5–6% relative risk) completely disappear when controlling for genetics and family environment through sibling analysis.1,4 Acetaminophen remains the safest option for pain and fever during pregnancy (Category B), and the risks of untreated maternal fever to fetal development are well-established and likely exceed any theoretical acetaminophen risk.5,6,7
Pregnant women should discuss medication use with their physicians, avoid unnecessary medications, but not fear appropriate use of acetaminophen when indicated—especially for fever.
Continue Reading This Evidence-Based Guide
Free access to 235+ health education guides from Highland Longevity
No spam. Unsubscribe anytime.
Key Evidence
The Bradford Hill Criteria Applied
Criterion
Assessment
Evidence
Strength
WEAK
5% relative risk (1.05); threshold for “interesting” in pharmacoepidemiology is 1.51
Consistency
MODERATE
Multiple studies show small associations, but these vanish with sibling controls1,4
Specificity
WEAK
Many variables linked to autism; acetaminophen also linked to ADHD, language issues
Temporality
MODERATE
Exposure precedes outcome, but no consensus on critical windows
Biological Gradient
MODERATE
Some dose-response data, but inconsistent across studies
Plausibility
WEAK
Mechanism unclear; prostaglandin inhibition is speculative
Analogy
AGAINST
Aspirin (similar mechanism) shows protective effect against autism
Coherence
WEAK
Animal studies inconsistent; extreme doses used
Experiment
N/A
No RCT data available
Conclusion: The Bradford Hill criteria do not support a causal relationship.
NO significant association (HR 0.98; 95% CI 0.94–1.02)
Critical Finding
When comparing siblings discordant for acetaminophen exposure, the association completely disappeared—suggesting the observed correlation in the general population was due to confounding variables (genetics, family environment), not acetaminophen itself.1
Japanese Cohort Study (Replication)
Metric
Finding
Sample size
217,602 children
Exposure rate
~40% (similar to US rates)
Full cohort result
HR 1.06 (95% CI: 0.98–1.15) for ASD
Sibling-controlled analysis
Association completely abolished (HR 0.86; 95% CI 0.52–1.44 for ADHD)
Significance: This study addresses concerns that the Swedish study had unusually low acetaminophen use rates (7.5%). Even with 40% exposure rates, the same pattern emerged: small associations vanish with proper controls.4
Clinical Applications
Framework for Evaluating Drug-Disease Associations
Step 1: Confirm Statistical Association
Does the data actually show a significant correlation?
Is the effect size meaningful (>1.5 for pharmacoepidemiology)?
Step 2: Assess Probability of Causality
Apply Bradford Hill criteria systematically
Look for sibling-controlled or twin studies
Consider confounding variables
Step 3: Evaluate Effect Size
Calculate absolute risk, not just relative risk
Compare to other known risk factors
Consider clinical significance
Why the Association Is Likely Confounding
The “Ice Cream and Drowning” Problem
Ice cream consumption correlates with drowning deaths. Both are caused by a third variable: hot weather. Similarly, genetic predisposition may cause both increased likelihood of autism in offspring and increased acetaminophen use by the mother (autism relates to sensory/pain perception).
Known Autism Risk Factors (By Magnitude)
Risk Factor
Contribution
Notes
Genetics
80–90% of variability
Monozygotic twin concordance rate 0.982
Expanded diagnosis criteria
40–60% of prevalence increase
DSM changes 1987–20133
Increased awareness/screening
20–30% of prevalence increase
Racial/socioeconomic diagnostic gaps narrowing3
Advanced parental age
5–15% of increase
Paternal age especially important
Maternal obesity/metabolic disease
Contributing factor
Obesity in pregnancy tripled in 30 years
Preterm birth
Contributing factor
Rates rising since 2010s
Air pollution (PM 2.5)
Contributing factor
38% global increase
Acetaminophen
Minimal if any
0.09% absolute risk increase (if causal)1
Pregnancy Medication Guidance
General Principle: Stop non-essential medications during pregnancy, but maternal health directly affects fetal health.
When to Consider Acetaminophen
Situation
Recommendation
Minor headache/discomfort
May skip; try rest instead
Debilitating pain affecting function
Judicious use appropriate
Fever
Strong indication for use
Risks of Untreated Fever During Pregnancy
Outcome
Risk Increase
Notes
Neural tube defects
25–200% higher
First trimester especially5,6
Oral clefts
25–200% higher
First trimester especially5
Autism risk (from fever)
34–40% higher
Second trimester peak5,6
Autism risk (untreated fever)
155% higher
OR 2.55 without antipyretics7
Critical Point
Fever-reducing medication (acetaminophen) appears to attenuate autism risk from fever (OR reduced from 2.55 to 1.30 with antipyretics).7 The risk of NOT treating fever likely exceeds any theoretical risk from acetaminophen.
Key Studies & Data
Finding
Result
Significance
Swedish Cohort acetaminophen-autism link1
HR 1.05 in full cohort; HR 0.98 (NS) with sibling control
Largest study (2.48M); authors conclude “non-causal association”
Japanese Cohort replication4
HR 1.06 (NS); sibling analysis abolishes association
Confirms Swedish findings at 40% exposure rate
CDC Autism Prevalence 20223
32.2/1000 (1 in 31) among 8-year-olds
5-fold increase since 2000 largely explained by diagnostic changes
Twin Heritability Meta-analysis2
64–91% heritability; MZ concordance 0.98
Genetics dominate autism risk
Maternal Fever Meta-analysis5
OR 1.32 (95% CI: 1.20–1.46)
12–17% of autism cases potentially preventable with fever treatment
Antipyretic attenuation of fever risk7
OR 2.55 → 1.30 with treatment
Supports treating fever during pregnancy
The Multiple Comparisons Problem
When researchers test thousands of potential associations with autism, some will appear significant by chance alone. With a p-value threshold of 0.05:
Margarine consumption correlates 98.5% with Maine divorce rates. California physicist count correlates 97.1% with Michael Schumacher’s F1 rankings. Correlation found through data mining is not evidence of causation, especially when effect sizes are small (1.05–1.10 range).
Additional Considerations
Study Limitations
Both Swedish and Japanese studies used prescription records, potentially underestimating over-the-counter acetaminophen use
Sibling comparisons have smaller sample sizes than full cohort analyses, reducing statistical power for rare outcomes
Self-reported exposure data subject to recall bias
Unable to assess dose-response relationships accurately due to variable recording practices
Conflicting Evidence
Some earlier observational studies (pre-sibling control era) reported significant associations between prenatal acetaminophen and neurodevelopmental outcomes
A 2022 meta-analysis of conventional observational studies suggested modest associations (pooled OR ~1.2), but these did not account for familial confounding
Industry-funded studies and independent academic studies have reached similar conclusions, reducing concerns about funding bias
Individual Variation
Genetic polymorphisms in acetaminophen metabolism (CYP2E1, SULT1A1) may create subpopulations with different risk profiles
Maternal conditions requiring acetaminophen use (infections, inflammatory states) may themselves affect fetal neurodevelopment
Trimester of exposure may have differential effects, though evidence is inconsistent
Co-administration with other medications could theoretically modify risk
Safety Notes
Acetaminophen overdose remains a leading cause of acute liver failure; pregnancy does not change this risk
Maximum daily dose (4g/day, or 3g/day with regular alcohol use) should be respected
Extended-release formulations may have different pharmacokinetics
Avoid combining with other acetaminophen-containing products (many cold remedies contain hidden acetaminophen)
Evidence Gaps
No randomized controlled trials exist or are ethically feasible for this question
Mechanism by which acetaminophen might theoretically affect neurodevelopment remains speculative
Long-term follow-up beyond childhood is lacking in most studies
Effects of different formulations (liquid, tablet, combination products) unstudied
Recent Developments
ACOG (2025) issued guidance affirming acetaminophen as the safest analgesic option during pregnancy8
FIGO (2025) published consensus statement that evidence does not support causal association8
NIH (2024) press release highlighted the Swedish study findings, emphasizing the importance of sibling-controlled analysis1
Ongoing litigation against acetaminophen manufacturers continues despite scientific evidence; legal standards differ from scientific standards
References
Hagberg KW, Jick SS, Habel LA, et al. Acetaminophen Use During Pregnancy and Children’s Risk of Autism, ADHD, and Intellectual Disability. JAMA. 2024;331(14):1205-1214. doi:10.1001/jama.2024.3172
Tick B, Bolton P, Happé F, Rutter M, Rijsdijk F. Heritability of autism spectrum disorders: a meta-analysis of twin studies. Journal of Child Psychology and Psychiatry. 2016;57(5):585-595. doi:10.1111/jcpp.12499
Shaw KA, Maenner MJ, Bakian AV, et al. Prevalence and Early Identification of Autism Spectrum Disorder Among Children Aged 4 and 8 Years — Autism and Developmental Disabilities Monitoring Network, 16 Sites, United States, 2022. MMWR Surveill Summ. 2024;74(SS-2):1-32. doi:10.15585/mmwr.ss7402a1
Okubo Y, Hayakawa I, Sugitate R, Nariai H. Maternal Acetaminophen Use and Offspring’s Neurodevelopmental Outcome: A Nationwide Birth Cohort Study. Paediatric and Perinatal Epidemiology. 2025;39(5):412-421. doi:10.1111/ppe.70071
Tioleco N, Silberman AE, Engel SM. Prenatal maternal infection and risk for autism in offspring: A meta-analysis. Autism Research. 2021;14(6):1296-1310. doi:10.1002/aur.2499
Perin J, Thompson J, McBratney-Owen B. Fever during pregnancy as a risk factor for neurodevelopmental disorders: results from a systematic review and meta-analysis. Molecular Autism. 2021;12:60. doi:10.1186/s13229-021-00464-4
Zerbo O, Iosif AM, Walker C, Ozonoff S, Hansen RL, Hertz-Picciotto I. Is Maternal Influenza or Fever During Pregnancy Associated with Autism or Developmental Delays? Results from the CHARGE Study. Journal of Autism and Developmental Disorders. 2013;43(1):25-33. doi:10.1007/s10803-012-1540-x
Louwen F, Filipek A, Reynolds JL. Paracetamol (acetaminophen) use during pregnancy and autism risk: Evidence does not support causal association. International Journal of Gynecology & Obstetrics. 2025;169(1):1-4. doi:10.1002/ijgo.70577
Medical Disclaimer: This educational brief is for informational purposes only and does not constitute medical advice, diagnosis, or treatment. Always consult with a qualified healthcare provider before starting any new health regimen. Individual results may vary. The information presented reflects current research as of March 2026 and may be updated as new evidence becomes available.
Have Questions About Your Health?
Dr. Lindsley can help you build a comprehensive, evidence-based health optimization plan.