"Given the upper limits of the CI, a relatively higher risk of more than 12% for autism is unlikely to be associated with acetaminophen exposure," the study authors noted.
A large-scale nationwide cohort study published in JAMA Pediatrics found no significant association between prenatal acetaminophen exposure and the risk of autism in offspring. The research, which utilized Danish national registers to track over 1.5 million children, provided new evidence to address ongoing public concerns regarding the safety of acetaminophen use during pregnancy.1
Evidence regarding the link between in utero acetaminophen exposure and neurodevelopmental outcomes has historically been inconsistent. Although some previous population-level analyses reported a small increase in risk, sibling-matched studies often failed to observe the same association, suggesting that previous findings may have been influenced by residual confounding.
The topic has been particularly highlighted in the United States, where, in 2025, federal government agencies expressed concerns about associations between acetaminophen use during pregnancy. In September of 2025, the FDA initiated a process for a label change of acetaminophen to “reflect evidence suggesting that the use of acetaminophen by pregnant women may be associated with an increased risk of neurological conditions such as autism and ADHD in children,” the agency stated at the time.2
This prompted several medical societies, including the American College of Obstetricians & Gynecologists (ACOG), to release statements affirming the safe use of acetaminophen during pregnancy.3
“Suggestions that acetaminophen use in pregnancy causes autism are not only highly concerning to clinicians but also irresponsible when considering the harmful and confusing message they send to pregnant patients, including those who may need to rely on this beneficial medicine during pregnancy,” said Steven J. Fleischman, MD, MBA, FACOG, president, ACOG, in September 2025.
To evaluate this potential association further, researchers of the current study conducted a comprehensive analysis of Danish demographic and healthcare data spanning more than 25 years.1
The study source population included all children born from singleton pregnancies in Denmark between January 1, 1997, and July 31, 2022, who remained alive at age 1. Researchers excluded children with missing gestational or maternal age, diseases inherently linked to autism, or those who died or emigrated before the age of 1. This resulted in a final cohort of 1,506,155 children.
Exposure was identified through the National Prescription Register, specifically maternal fulfillment of a prescription for acetaminophen. The primary outcome was an incident diagnosis of autism based on the International Statistical Classification of Diseases and Related Health Problems, Tenth Revision (ICD-10) codes. Children were followed from age 1 until they reached an autism diagnosis, emigrated, or reached the end of the follow-up period on July 31, 2023.
Of the 1,506,155 children included in the study, 31,098 (2.1%) were exposed to prescribed acetaminophen during pregnancy. Within this exposed group, 554 (1.8%) were later diagnosed with autism. In the unexposed group, 44,667 of 1,475,057 (3.0%) children received an autism diagnosis. The data indicated that women who were prescribed acetaminophen during pregnancy generally had different clinical profiles than those who were not. Specifically, exposed women were older, with a median age of 31.2 years compared to 30.5 years in the unexposed group. They also exhibited higher parity, higher body mass index, more comorbidities, and utilized more prescription medicine.
After adjusting for these confounders and other analgesics, the research team calculated an adjusted hazard ratio (aHR) of 1.03 (95% CI, 0.95-1.12) in the population analysis. To further account for shared familial and environmental factors, a sibling analysis was conducted comparing groups with discordant exposure. This analysis yielded an aHR of 1.09 (95% CI, 0.91-1.27), which was not statistically significant.
The study found no evidence of a significant association across several stratified analyses. This included evaluations of dose-response patterns and exposure during different trimesters of pregnancy. Furthermore, the results remained consistent in analyses restricted to pregnancies occurring after 2013.
The researchers noted that given the upper limits of the confidence intervals, it is unlikely that acetaminophen exposure is associated with a higher risk of autism exceeding 12%.
Although the study lacked individual-level data for over-the-counter (OTC) medication use, previous simulations indicated that the resulting bias from low-level OTC exposure is generally negligible.
References:
