Is Tylenol Linked to Autism?

The question of whether acetaminophen (Tylenol) use during pregnancy increases autism risk has become one of the most debated topics in prenatal care. Recent headlines have raised concerns, but the scientific evidence tells a more nuanced story that every expectant parent deserves to understand.

Current Autism Rates in the United States

Before examining any potential connections, it’s important to understand the current landscape of autism diagnoses.

According to the CDC’s most recent data from April 2025, autism spectrum disorder (ASD) now affects 1 in 31 children aged 8 years, representing 3.2% of this population. This marks a significant increase from the previous estimate of 1 in 36 children just two years earlier.

The statistics reveal several important patterns:

• Boys are diagnosed at 4.9% compared to girls at 1.4%, making boys 3.4 times more likely to receive an autism diagnosis
• Asian/Pacific Islander and American Indian/Alaska Native children show the highest rates at 3.8%
• Black children follow at 3.7%, Hispanic children at 3.3%, and White children at 2.8%
• Geographic location matters significantly, with California showing the highest rate at 1 in 19 children, while Texas-Laredo reports the lowest at 1 in 103

Among adults, the national prevalence stands at 1 in 45, or 2.2%. The median age of diagnosis remains 47 months (nearly 4 years), with only half of children diagnosed by age 3. Nearly 40% of children with ASD also have co-occurring intellectual disability.

Since 2000, autism diagnoses have increased by 382%, from 1 in 150 to 1 in 31. This dramatic rise reflects improved awareness, better screening tools, expanded diagnostic criteria, and reduced stigma rather than a sudden environmental change.

The Largest Study Ever Conducted on Acetaminophen and Autism

In April 2024, researchers from Drexel University and Sweden’s Karolinska Institutet published what remains the most comprehensive investigation into prenatal acetaminophen exposure and neurodevelopmental outcomes.

The study followed 2,480,797 children born in Sweden between 1995 and 2019, tracking them for a median of 13.4 years, with some followed up to age 26. Of these children, 185,909 (7.5%) were exposed to acetaminophen during pregnancy, while 2,294,888 (92.5%) were not.

Initial Population Results

When researchers first compared all exposed children to unexposed children, they found small increases in risk:

• Autism: 1.05 times higher risk (95% confidence interval: 1.02 to 1.08), representing an absolute risk increase of 0.09% by age 10
• ADHD: 1.07 times higher risk (95% confidence interval: 1.05 to 1.10)
• Intellectual disability: 1.05 times higher risk (95% confidence interval: 1.00 to 1.10)

These findings seemed to confirm concerns raised by earlier, smaller studies.

The Critical Sibling Comparison That Changed Everything

The Swedish researchers then employed what epidemiologists consider the gold standard for controlling confounding factors: sibling comparison analysis. They identified 94,105 sibling pairs where one sibling was exposed to acetaminophen in utero and the other was not, allowing them to compare children who share the same parents, genetics, home environment, and socioeconomic factors.

When comparing these siblings:

• Autism: 0.98 times the risk (95% confidence interval: 0.93 to 1.04), essentially identical risk with an absolute difference of just 0.02%
• ADHD: 0.98 times the risk (95% confidence interval: 0.94 to 1.02)
• Intellectual disability: 1.01 times the risk (95% confidence interval: 0.92 to 1.10)

The associations completely disappeared.

Why Dose Response Matters

If acetaminophen caused autism, higher doses should increase risk. The sibling analysis examined this directly by comparing no use, low use (below 25th percentile), medium use (25th to 75th percentile), and high use (above 75th percentile).

The results showed no dose-response pattern:

• Low use: 0.85 times the risk of autism
• Medium use: 0.96 times the risk
• High use: 0.88 times the risk

Higher acetaminophen exposure did not increase autism risk. This pattern strongly argues against a causal relationship.

Japanese Study Confirms Swedish Findings

A 2025 Japanese population study replicated the Swedish research design with over 200,000 children and reached identical conclusions. In population-wide analyses, acetaminophen exposure appeared associated with increased risk. In sibling-controlled analyses, these associations vanished.

The Japanese researchers made an additional critical observation: acetaminophen use after pregnancy also correlated with increased ADHD risk in children. Since acetaminophen taken after birth cannot affect fetal brain development, this finding points to an alternative explanation: the propensity to use acetaminophen reflects other factors that genuinely affect neurodevelopment risk.

The Boston Birth Cohort Study Shows Different Results

The Boston Birth Cohort study, published in JAMA Psychiatry in 2020, took a different approach and reached different conclusions. Researchers measured acetaminophen metabolites directly in cord blood samples from 996 mother-infant pairs at Boston Medical Center between 1998 and 2018.

Unchanged acetaminophen was detectable in 100% of cord blood samples, confirming that acetaminophen crosses the placental barrier. Children were followed for an average of 9.8 years.

The outcomes showed:

• 257 children (25.8%) developed ADHD only
• 66 children (6.6%) developed ASD only
• 42 children (4.2%) developed both ADHD and ASD
• 304 children (30.5%) had other developmental disabilities
• 327 children (32.8%) developed typically

Dose Response in Cord Blood

Researchers divided cord blood acetaminophen levels into three groups (tertiles) and compared developmental outcomes:

ADHD Risk:

• Second tertile: 2.26 times higher (95% confidence interval: 1.40 to 3.69)
• Third tertile: 2.86 times higher (95% confidence interval: 1.77 to 4.67)

ASD Risk:

• Second tertile: 2.14 times higher (95% confidence interval: 0.93 to 5.13)
• Third tertile: 3.62 times higher (95% confidence interval: 1.62 to 8.60)

These results suggested a clear dose-response relationship, with the highest cord blood levels associated with nearly three times the ADHD risk and 3.6 times the ASD risk.

Critical Limitations of the Boston Study

The Boston study’s findings must be interpreted within its methodological constraints:

• Sample size of 996 compared to 2.48 million in the Swedish study
• No sibling controls, making it vulnerable to the same familial confounding that the Swedish and Japanese studies controlled for
• Single measurement at birth captures only a snapshot, not cumulative exposure throughout pregnancy
• Population predominantly low-income, urban, minority families, limiting generalizability
• Cannot determine whether acetaminophen exposure or the underlying reasons for acetaminophen use drive the associations

What Two Major Research Reviews Concluded

BMJ Umbrella Review (November 2025)

Researchers conducted an umbrella review. This is a systematic review of systematic reviews, which is done by examining 9 systematic reviews that collectively included 40 primary studies. These studies comprised 37 prospective cohorts, 2 case-control studies, and 1 ecological study.

Seven of the nine reviews suggested possible to strong associations between maternal acetaminophen use and autism or ADHD. However, seven of the same nine reviews urged caution due to risk of bias and confounding.

When quality was assessed using the AMSTAR 2 tool, confidence in the reviews ranged from low (2 reviews) to critically low (7 reviews). Critically, only one review included sibling-controlled studies, and those studies showed no associations when siblings were compared.

The umbrella review concluded that “quality and validity of evidence does not support causal relationship.” The associations observed in population analyses did not persist in sibling analyses, suggesting that shared family factors explain the risk rather than acetaminophen itself.

Mount Sinai and Harvard Navigation Guide Review (2025)

Using the Navigation Guide framework, which is considered the gold standard for synthesizing environmental health data, the researchers reviewed 46 studies involving more than 100,000 participants. Twenty-seven studies showed positive associations with neurodevelopmental disorders.

This review reached different conclusions, finding:

• Strong evidence of relationship between prenatal acetaminophen and ASD
• Higher-quality studies were more likely to show positive associations
• For ADHD: average bias score 0.85, average strength of evidence 0.75, expert opinion 0.70

However, this review’s scoring system downgraded the Swedish sibling study to “very weak” evidence because it lacked large effect sizes and showed negative findings. The scoring prioritized studies showing dose-response and large effects over studies designed to control for confounding.

Why Family Factors Matter More Than Initially Recognized

The difference between population studies and sibling-controlled studies reveals that what appears to be a drug effect may actually reflect family characteristics.

Genetic Confounding

Genes associated with autism and ADHD also increase the risk of maternal health conditions that require pain medication, including migraines, chronic pain, infections, and psychiatric disorders. Mothers with genetic predisposition to autism or ADHD are more likely to experience pregnancy complications and use pain medication.

A striking finding supports this: paternal acetaminophen use in the six months before pregnancy correlates with offspring ADHD to the same extent as maternal use during pregnancy. Since fathers’ medication use cannot directly affect fetal brain development, this pattern indicates genetic confounding rather than direct drug effects.

Confounding by Indication

The conditions that lead women to take acetaminophen during pregnancy: fever, infection, pain, inflammation, are independently associated with increased autism and ADHD risk in offspring. When researchers examined other pain medications, they found the same pattern: aspirin, NSAIDs, opioids, and antimigraine drugs all showed associations with increased ASD and ADHD risk in population analyses but not in sibling analyses.

This strongly suggests that underlying maternal health conditions, not the medications used to treat them, drive the risk.

Socioeconomic and Environmental Factors

Acetaminophen use during pregnancy is more common among mothers with lower socioeconomic status, higher BMI, smoking habits, and psychiatric disorders. All of these factors independently increase autism risk and are difficult to fully account for in standard statistical analyses.

Recall Bias

Studies relying on mothers to recall medication use years after birth may introduce bias. Mothers of children with autism or ADHD may be more likely to recall and report medication exposures, searching for explanations for their child’s diagnosis.

How Acetaminophen Might Affect Fetal Brain Development (Theoretical Mechanisms)

Researchers have proposed several biological pathways through which acetaminophen could theoretically affect neurodevelopment. These mechanisms remain hypothetical rather than proven in humans.

Endocrine Disruption: Acetaminophen crosses the placental barrier and may interfere with hormones involved in fetal brain development.

Prostaglandin Synthesis Inhibition: Prostaglandins play critical roles in neurodevelopment. Acetaminophen inhibits their synthesis, and animal studies show that disruption during sensitive developmental windows causes altered brain structure and decreased social interaction, particularly in males.

Oxidative Stress and Mitochondrial Function: The acetaminophen metabolite NAPQI generates oxidative stress and may impair mitochondrial function in developing brain tissue.

Epigenetic Changes: Studies have documented that acetaminophen alters DNA methylation patterns in placenta (42 CpG sites) and cord blood, affecting genes like PTGDR (prostaglandin receptor) and CYP2E1.

Endocannabinoid System Disruption: Acetaminophen acts as an indirect cannabinoid receptor agonist through its metabolite AM404, potentially disrupting the endocannabinoid system critical for brain development.

While these mechanisms are biologically plausible and supported by laboratory research, they do not constitute evidence that acetaminophen causes autism in humans at clinically relevant doses.

What the FDA, CDC, and Medical Organizations Recommend

FDA Position (September 2025)

The FDA initiated a label change process for acetaminophen to reflect a “possible association” with neurodevelopmental disorders. However, their key statement clarified: “A causal relationship has NOT been established.”

The FDA continues to consider acetaminophen the “safest over-the-counter alternative among all analgesics and antipyretics in pregnancy.” They recommended that clinicians minimize use for routine low-grade fevers but continue use when medically indicated.

CDC and NIH

The CDC continues to recommend acetaminophen for fever and pain management during pregnancy. The NIH funded the 2024 Swedish study that found no causal link in sibling analyses.

American College of Obstetricians and Gynecologists (August-October 2025)

ACOG maintains that acetaminophen remains the analgesic and antipyretic of choice during pregnancy. Their position states that current evidence does not support a causal link and that untreated fever and pain pose greater risks than judicious acetaminophen use. They have not recommended changes to clinical practice.

Understanding the Evidence Quality Hierarchy

Not all research studies provide equally reliable evidence. Understanding why sibling-controlled studies are considered superior helps interpret conflicting findings.

Population Studies compare all exposed children to all unexposed children. These studies can identify associations but cannot definitively establish causation because they cannot fully control for all factors that differ between families who use acetaminophen and those who don’t.

Sibling-Controlled Studies compare children within the same family who differ only in their prenatal acetaminophen exposure. Because siblings share parents, genes, home environment, socioeconomic status, and many other unmeasured factors, these studies automatically control for familial confounding that cannot be measured or adjusted for in standard regression models.

Biomarker Studies measure acetaminophen metabolites directly in biological samples rather than relying on maternal recall, eliminating recall bias. However, without sibling controls, they remain vulnerable to familial confounding.

The Swedish and Japanese sibling-controlled studies, with their massive sample sizes (2.48 million and over 200,000 children respectively), represent the highest quality evidence currently available on this question.

What This Means for Decision Making During Pregnancy

The evidence presents a complex picture that requires weighing multiple factors:

The strongest evidence from the largest, most methodologically rigorous studies with sibling controls shows no causal link between acetaminophen use during pregnancy and autism when familial factors are properly controlled.

Smaller studies without sibling controls show associations, but these likely reflect shared family characteristics (genetics, maternal health, socioeconomic factors) rather than direct drug effects.

The same pattern appears with all pain medications, not just acetaminophen, suggesting that maternal pain, inflammation, and infection themselves may carry risk.

Untreated fever, particularly in the first trimester, carries documented risks for neural tube defects and other developmental problems. Untreated severe pain can lead to increased stress hormones, elevated blood pressure, and other physiological changes that may affect fetal development.

Acetaminophen remains the only over-the-counter pain reliever and fever reducer considered safe during pregnancy. NSAIDs like ibuprofen carry risks of premature closure of the ductus arteriosus and other complications, particularly in the third trimester. Aspirin carries bleeding risks and is generally not recommended.

Questions to Discuss With Your Healthcare Provider

Every pregnancy and every situation is unique. When deciding whether to use acetaminophen during pregnancy, consider discussing these questions with your obstetrician or midwife:

• What are the risks of leaving my current symptoms (fever, pain, infection) untreated?
• Are there non-medication approaches that might be effective for my specific situation?
• What is the lowest effective dose and shortest duration of use for my condition?
• How do my personal and family health history factors into this decision?
• Are there specific circumstances or symptoms that warrant acetaminophen use despite theoretical concerns?
• If I have used acetaminophen during pregnancy, should this change prenatal care or early childhood screening?

Medical decisions during pregnancy involve balancing known risks against uncertain or theoretical risks. Your healthcare provider can help you navigate this balance based on your specific circumstances.

The Bottom Line on Acetaminophen and Autism Risk

The largest and most methodologically rigorous studies, tracking nearly 2.5 million children over decades and controlling for familial factors through sibling comparisons, found no evidence that acetaminophen directly causes autism. When researchers compared siblings, eliminating genetic and family environmental confounding, the associations observed in population studies completely disappeared. The lack of dose-response relationship further argues against causation.

While acetaminophen crosses the placenta and theoretical biological mechanisms exist, these do not constitute evidence of harm in humans at therapeutic doses. All major medical organizations, including the FDA, CDC, and ACOG, conclude that acetaminophen remains the safest option for treating fever and pain during pregnancy when medically indicated, though they recommend using the lowest effective dose for the shortest necessary duration.

The decision to use any medication during pregnancy should be made in consultation with your healthcare provider, weighing the documented risks of untreated symptoms against the current evidence on medication safety. The goal is always to make informed decisions that protect both maternal and fetal health.