Acetaminophen (paracetamol), considered the drug of choice and most commonly used by pregnant women, is at the center of an intense public health and bioethical controversy.
A recent report reviewing 46 observational studies suggested the need for “immediate precautionary measures” and to restrict its prenatal use due to a possible association with neurodevelopmental disorders (NDDs).
However, other experts cautiously caution that these conclusions “exceed the strength of the available evidence” and present “internal dissonance” by confusing statistical correlation with causality, which cannot be established with observational designs.
The main ethical concern is that this overinterpretation could generate unjustified alarm in vulnerable populations, leading pregnant women to abandon a safe medication. Or perhaps to opt for alternatives with a “clearly more unfavorable” safety profile, drawing a parallel with the risks of misinformation observed in the Andrew Wakefield case and vaccines.
Experts urge caution following a report calling for restrictions on acetaminophen use during pregnancy due to its possible link to neurodevelopmental disorders. Scientific critics warn that the conclusions go beyond the evidence and fear a similar effect to that of the MMR vaccine scandal.
Acetaminophen, universally known as paracetamol, the analgesic and antipyretic of choice and considered safe for pregnant women, is at the center of a scientific and public health controversy.
A recent research study, led by Prada et al., applied the Navigation Guide methodology to review 46 observational studies on prenatal use of this drug and its possible association with neurodevelopmental disorders (NDDs). The authors concluded that there is sufficient evidence to recommend “immediate precautionary measures” and restrict the use of acetaminophen in pregnant women.
However, this conclusion has been described by other experts as an overinterpretation that ‘exceeds the strength of the available evidence’ and presents an ‘internal dissonance’ by stating that the evidence is ‘strong and consistent’ while simultaneously admitting that causality cannot be established.
The drug of choice questioned
Critics fear that spreading messages suggesting a causal relationship without solid foundation could lead pregnant women, especially those with less access to reliable medical information, to abandon a drug considered safe and with no clear alternatives.
In fact, discontinuing acetaminophen could lead to increased use of NSAIDs or opioids, whose safety profile is “clearly more unfavorable” for pregnancy. Furthermore, discontinuing treatment for fever or pain carries proven risks for the mother and fetus.
Methodological limitations that preclude causality
The basis of the controversy lies in the methodology of the 46 studies analyzed. Experts point out that the conclusions of Prada et al. exceed the actual strength of the evidence due to crucial limitations:
- Observational nature: All studies are cohort or case-control, which introduces an inevitable risk of residual confounding. Unmeasured factors, such as fever, underlying infections, or genetic and socioeconomic variables, could be influencing the results. The article by Prada et al. itself acknowledges that it did not perform a quantitative bias analysis, making it impossible to assess the impact of these confounding factors.
- Inconsistency and heterogeneity: There is considerable variability in the way exposure was measured (self-report, biomarkers, or clinical records) and outcomes (clinical diagnoses, parental questionnaires, or symptom scales). This heterogeneity was such that a meta-analysis was precluded, leaving the resulting narrative synthesis more vulnerable to interpretation bias.
- Publication bias: It has been noted that most studies reporting positive associations were published after 2013, suggesting a possible bias in the available literature.
One point highlighted by critics is the existence of studies with sibling-controlled designs. These designs are less vulnerable to shared familial confounding and, in fact, tend to show null associations between acetaminophen and NDDs.
The commentary authors point out that the work by Prada et al. downplays these findings, even though they may reflect that the associations observed in conventional cohorts are the result of confounding.
The risk of unjustified alarm and the ghost of Wakefield
Experts advocate the need to interpret observational evidence with caution and avoid hasty extrapolations to clinical practice.
Currently, no international clinical guidelines have modified their recommendations based on these data, arguing that the precautionary principle cannot be invoked without a balanced risk-benefit analysis.
To illustrate the risks of disseminating premature or inaccurate scientific information, critics draw a direct parallel with the case of Andrew Wakefield. In 1998, Wakefield stated unequivocally at a press conference that the MMR vaccine was linked to the development of autism.
His work, published in The Lancet, was retracted 12 years later, in 2010, but during that time it fostered a false belief that has fueled anti-vaccine movements to this day, seriously endangering the population. Respect for the life, dignity, and autonomy of patients requires that they have access to rigorous, complete, and prudent information.
This type of error is frequently related to confusing a simple statistical correlation with a causal relationship. Critics cite examples of spurious correlations, such as the statistical association between stork nesting rates and the birth rate in London, or the correlation between the divorce rate in Maine and per capita margarine consumption. “Storks don’t bring babies, and margarine consumption doesn’t cause divorces,” they point out.
According to experts, the risk is that, if the lack of rigor in the design of studies is combined with recklessly drawing unsustainable conclusions, the result can be catastrophic, as was the case with Wakefield and his role as a “true agitator of the current anti-vaccine movements.”
In this context, the imprudence of certain public figures, such as Donald Trump, who has made an alarmist announcement about the dangers of paracetamol in pregnancy, could result in many mothers choosing not to seek treatment or looking for alternative medications with greater risk, especially given the foreseeable changes in the package leaflets warning of a risk.
Conclusions
In the opinion of experts who criticize overinterpretation, there is no justification for modifying current clinical recommendations or transmitting alarmist messages to the general population. The priority should be to communicate cautiously, promoting more studies with biomarkers and well-designed cohorts, and avoiding inducing unjustified fear.
The most rigorous and ethical assessment would be expressed in the following message: “Observational studies show a possible association between prenatal acetaminophen and NDDs, but heterogeneity, residual confounding, and inconsistent results prevent establishing causality. More studies with robust designs are required before making clinical recommendations.”
The ethics of scientific research and publication require the prudent interpretation of data, drawing only the conclusions that scientific rigor allows, so that science truly benefits citizens.
Marcelino Pérez, María Teresa Murillo and Julio Tudela
Bioethics Observatory – Institute of Life Sciences
Catholic University of Valencia
References:
- Prada D, Ritz B, Bauer AZ, Baccarelli AA. Evaluation of the evidence on acetaminophen use and neurodevelopmental disorders using the Navigation Guide methodology. Environ Health. 2025; 24:56.
- American College of Obstetricians and Gynecologists ACOG. Committee Opinion No. 742: Postpartum Pain Management. Obstet Gynecol. 2018;132(1):e35-e43. doi: 10.1097/AOG.0000000000002683.
- Ji Y, Azuine RE, Zhang Y, Hou W, Hong X, Wang G, et al. Association of cord plasma biomarkers of in utero acetaminophen exposure with risk of attention-deficit/hyperactivity disorder and autism spectrum disorder in childhood. JAMA Psychiatry. 2020;77(2):180–189.
- Gustavson K, Ystrom E, Ask H, Ask Torvik F, Hornig M, Susser E, Lipkin WI, Lupattelli A, Stoltenberg C, Magnus P, Mjaaland S, Askeland RB, Walle KM, Bresnahan M, Nordeng H, Reichborn-Kjennerud T. Acetaminophen use during pregnancy and offspring attention deficit hyperactivity disorder – a longitudinal sibling control study. JCPP Adv. 2021;1(2):e12020. doi: 10.1002/jcv2.12020.
- Avella-Garcia CB, Julvez J, Fortuny J, Rebordosa C, García-Esteban R, Galán IR, Tardón A, Rodríguez-Bernal CL, Iñiguez C, Andiarena A, Santa-Marina L, Sunyer J. Acetaminophen use in pregnancy and neurodevelopment: attention function and autism spectrum symptoms. Int J Epidemiol. 2016;45(6):1987-1996. doi: 10.1093/ije/dyw115.
- Baker BH, Lugo-Candelas C, Wu H, Laue HE, Boivin A, Gillet V, et al. Association of prenatal acetaminophen exposure measured in meconium with risk of attention-deficit/hyperactivity disorder mediated by frontoparietal network brain connectivity. JAMA Pediatr. 2020;174(11):1073–1081.
- Wakefield AJ, Murch SH, Anthony A, et al. Ileal-lymphoid-nodular hyperplasia, non-specific colitis, and pervasive developmental disorder in children. Lancet. 1998;351:637–41 [RETRACTED.].
- Matthews R. Storks deliver babies (p= 0.008). Teaching Statistics. 2000;22(2), 36-38.
- Banks C, Boothroyd III. P. Spurious Correlations: The Comedy and Drama of Statistics.
The World Health Organization emphasizes that there is currently no conclusive scientific evidence linking acetaminophen (also known as paracetamol) use during pregnancy to autism.
Globally, nearly 62 million people (approximately 1 in 127) are affected by autism spectrum disorder, a diverse group of conditions related to brain development. Although awareness and diagnosis have improved in recent years, the exact causes of autism remain unclear, with multiple factors likely contributing.
If a strong link between acetaminophen and autism existed, it would likely have been consistently observed across multiple studies. However, the available data do not show a consistent association. While some cohort studies report weak or modest associations, recent research comparing children prenatally exposed to acetaminophen with their siblings finds no significant relationship between maternal acetaminophen use and autism. This lack of replicability underscores the need for caution in drawing causal conclusions.
It is also important to highlight the need for the ethical and timely inclusion of pregnant women in clinical trials. Historically, this population has often been excluded from research, leading to gaps in evidence about the safety and efficacy of medications during pregnancy. Ensuring their inclusion in well-designed studies allows for evidence-based guidance that protects both maternal and fetal health while maintaining high ethical standards.