Our Professional Obligation in Counseling About Acetaminophen and Autism
It is our ethical duty not to present certainty, where none exists, nor to downplay conflicting evidence, but to honestly explain the balance of findings.
Whether you agree or disagree with the present party in power, pointing out that studies sometimes disagree is not unscientific. What is unscientic is to not pont oiut divergent studies. On the contrary, pointing out differences reflects the essence of science, which advances through questioning, comparison, and careful interpretation of evidence. Divergent findings are expected when research is conducted in different populations or with different methods, and acknowledging this variation is the first step toward building trust with patients. Ignoring uncertainty or presenting only one side is what undermines science, not the recognition that results can conflict.
Notwithstanding reassurances by some professional organizations, there are contradictory studies about acetaminophen use in pregnancy and the risk of autism. Some large cohort and biomarker-based studies have suggested associations, reporting higher rates of autism spectrum disorder (ASD) among children with the highest prenatal exposure. At the same time, other equally or even more robust investigations, especially those using sibling comparisons and more rigorous control of confounding, have found no increased risk and argue that familial, genetic, or illness-related factors are the true drivers of the observed signals.
This tension has left both clinicians and pregnant patients navigating a confusing evidence base. On one hand, the studies that raise concern cannot and should not be ignored, because they represent published peer-reviewed work. On the other hand, the methodological weaknesses—recall bias, misclassification of exposure, inadequate adjustment for maternal illness—mean these findings do not necessarily establish causality. Patients who see only the alarming headlines may believe that acetaminophen is dangerous, when the scientific picture is far more nuanced.
It is all about Informed Consent
When a patient takes a medication on the recommendation of a physician, informed consent is always required. Even for commonly used drugs, every treatment involves weighing benefits against risks, and patients have the right to understand those trade-offs before agreeing. Consent is not just about giving permission but about making sure patients truly understand what they are choosing. This responsibility becomes even more important when scientific studies give conflicting results.
Informed consent is both a process and a principle. It is not necessarily a signature on a piece of paper. At its core, it means that no medical test, treatment, or procedure should be carried out unless the patient has been given enough information to understand what is being proposed and has freely agreed to it. It is not just a legal formality but an ethical commitment to respect each patient’s right to make decisions about their own body and health. For consent to be truly informed, the information must be accurate, complete, and explained in language the patient can understand. This includes outlining the expected benefits, the known risks, any reasonable alternatives, and the option of declining treatment altogether.
Informed consent is required because medicine always involves uncertainty and trade-offs. Patients have different values, tolerances for risk, and personal circumstances, so what seems like the best medical choice to one person may not be the right choice for another. By sharing information openly, physicians help patients make choices that align with their own preferences and priorities rather than being guided only by the clinician’s judgment.
When the medical evidence is clear and consistent, the process of informed consent can be relatively straightforward. However, when findings diverge—as in the ongoing debate about acetaminophen use in pregnancy and autism—professionalism requires a deeper level of transparency. Physicians must explain that while some studies suggest a possible association, others with stronger methods do not support it. They should also note how expert organizations such as ACOG or the FDA interpret the data, and why recommendations may differ between groups. Most importantly, the discussion should acknowledge uncertainty without either minimizing concerns or creating unnecessary fear.
Done well, informed consent is not simply a signature on a document but a respectful dialogue. The physician’s role is to create a space where patients can ask questions, express concerns, and weigh the options in light of their own values and circumstances. In this way, informed consent ensures that even when science is unsettled, decisions are made with honesty, trust, and partnership between patient and physician.
Positions by Professional Organizations
Professional organizations such as the American College of Obstetricians and Gynecologists (ACOG), along with other national and international groups, have taken a clear position: acetaminophen remains the preferred analgesic and antipyretic in pregnancy. They stress that untreated maternal fever, severe pain, or preeclampsia symptoms pose proven and immediate risks to maternal and fetal health, while the hypothesized link to autism is inconsistent and likely confounded. In doing so, these bodies often dismiss the positive association studies as not sufficiently rigorous. That stance provides clarity for practice but can also create mistrust if patients perceive that unfavorable evidence is being brushed aside.
Our Ethical Obligations
From an ethical perspective, the physician must carefully balance three obligations.
1.Respect for autonomy requires telling patients that there are contradictory studies, but that the weight of the evidence—including the best-controlled analyses—does not support a causal link.
2.Beneficence require highlighting that withholding acetaminophen when it is clinically indicated, such as for fever in early pregnancy, could result in real harm, including increased risk of neural tube defects.
3.Justice requires delivering this information equitably and without paternalism, making sure that all patients—regardless of education level or background—have access to a clear explanation of what the evidence shows and how experts interpret it.
The ethically sound way to counsel is through truthful nuance. Patients should hear, in plain terms:
Some studies have found associations between acetaminophen use and autism, but these studies have important limitations.
The most rigorous and largest studies, including those comparing siblings, find no evidence of a causal link.
ACOG and other professional groups continue to recommend acetaminophen when clinically indicated because the risks of untreated maternal conditions are proven and often more serious.
Decisions should be individualized, with room for patients to ask questions and share their concerns.
The obligation of physicians is not to present certainty where none exists, nor to downplay conflicting evidence, but to honestly explain the balance of findings and the rationale behind current recommendations. Only by acknowledging both sides, while emphasizing the weight and quality of the data, can clinicians maintain patient trust and ensure that care is both evidence-based and ethically grounded.
Below is a set of peer-reviewed studies with different findings concerning a possible link between acetaminophen and autism. .
Not linked: Ahlqvist VH, Magnusson C, Dalman C, et al. Acetaminophen use during pregnancy and children’s risk of autism, ADHD, and intellectual disability. JAMA. 2024;331(14):1214-1227. JAMA Network+1
Nationwide Swedish cohort of ~2.5 million births with sibling-control analyses. Associations seen in conventional models disappeared in sibling comparisons, with HR for ASD ~0.98, indicating no increased ASD risk after accounting for shared familial and genetic factors.
Not linked: Liew Z, Ritz B, Rebordosa C, Lee P-C, Olsen J. Maternal use of acetaminophen during pregnancy and risk of autism spectrum disorders in childhood. Autism Res. 2016;9(9):951-958. Wiley Online Library
Danish National Birth Cohort. No association for ASD overall; an elevated risk was limited to the subgroup ASD+hyperkinetic symptoms, suggesting indication and behavioral comorbidity may drive signals rather than ASD broadly.
Not linked: (Contextual appraisal of above JAMA study) NIH. Study reveals no causal link between neurodevelopmental disorders and acetaminophen exposure before birth. NIH News Release summarizing Ahlqvist et al., 2024. Apr 11, 2024. National Institutes of Health (NIH)
Peer-reviewed JAMA paper summarized by NIH: when rigorous familial controls are applied, no association with ASD is observed, highlighting the role of confounding in earlier reports. (Note: summary of a peer-reviewed study; original JAMA article cited above.)
Not linked: (Meta-analysis perspective highlighting mixed/attenuated ASD findings) Prada D, Gennings C, Drury BE, et al. Evaluation of the evidence on acetaminophen use and neurodevelopmental disorders using the Navigation Guide methodology. Environ Health. 2025;24:62. BioMed Central+1
Systematic review using Navigation Guide methods. Authors note heterogeneity and potential confounding; while several studies report associations, others do not, and ASD-specific evidence is inconsistent, especially in higher-quality, confounder-resistant designs.
Not linked: (Field synthesis noting inconsistency under rigorous control) Live Science editorially interviews study authors discussing the JAMA analysis: Schweitzer K. Acetaminophen Use in Pregnancy—Study Author Explains the Data. JAMA. 2025; Published online Sept 29, 2025. JAMA Network
JAMA Medical News piece contextualizing the 2024 JAMA cohort: familial confounding likely explains earlier associations; best-controlled analyses do not show increased ASD risk. (Note: news/interpretation in a peer-reviewed journal; primary data in Ahlqvist et al.)
Positive association: Ji Y, Azuine RE, Zhang Y, et al. Association of cord plasma biomarkers of in utero acetaminophen exposure with risk of ADHD and ASD in childhood. JAMA Psychiatry. 2020;77(2):180-189. JAMA Network+1
Prospective Boston Birth Cohort using cord-blood acetaminophen metabolites. Dose-response relationship reported; children in the highest exposure tertile had significantly higher ASD and ADHD risks vs the lowest.
Positive association: Liew Z, Ritz B, Virk J, Olsen J. Maternal acetaminophen use during pregnancy and ASD risk. Autism Res. 2016;9(9):951-958. Wiley Online Library
Within the same Danish cohort, ASD with hyperkinetic symptoms showed increased risk with long-duration use (>20 weeks), whereas ASD overall did not. Authors interpret this as a signal confined to specific phenotypes or confounded by indication.
Positive association: Alemany S, Avella-García C, Liew Z, et al. Prenatal and postnatal exposure to acetaminophen in relation to autistic spectrum and ADHD symptoms in childhood. Eur J Epidemiol. 2021;36(10):993-1004. PubMed
Meta-analysis of six European cohorts (n≈73,881). Prenatal exposure associated with higher odds of borderline/clinical autistic-spectrum symptoms (OR≈1.19) and ADHD symptoms, with modest effect sizes.
Positive association: Masarwa R, Levine H, Gorelik E, et al. Prenatal exposure to acetaminophen and risk for ADHD and ASD: systematic review and meta-analysis of cohort studies. Am J Epidemiol. 2018;187(8):1817-1827. Figo
Meta-analysis of longitudinal cohorts. Reported elevated risks for ADHD and a smaller but significant increase for ASD, though authors caution about residual confounding and exposure misclassification.
Positive association: Prada D, Crespo-Amorós M, Gennings C, et al. Evaluation of the evidence on acetaminophen and neurodevelopmental disorders using the Navigation Guide methodology. Environ Health. 2025;24:62. BioMed Central
Applying GRADE-like criteria, authors conclude evidence signals increased NDD risks including ASD in several higher-quality studies, while acknowledging heterogeneity and need for mechanistic and replication work.
Notes to interpret
The largest and most confounder-resistant analysis to date is Ahlqvist et al., JAMA 2024, which found no association with ASD in sibling-control models, implying that earlier signals may reflect familial or indication-related confounding. JAMA Network
Several studies showing positive associations use maternal self-report for exposure or assess autistic traits rather than confirmed ASD diagnoses, and many acknowledge potential residual confounding by fever, infection, genetics, or co-medications. PubMed+1
Public statements / wording from professional organizations and government bodies regarding acetaminophen (paracetamol) use in pregnancy, especially in relation to neurodevelopment / autism concerns.
ACOG — “Acetaminophen Use in Pregnancy and Neurodevelopmental Outcomes” (2025 advisory)
“ACOG reaffirms that acetaminophen remains the safest first-line analgesic and antipyretic in pregnancy. Clinicians should continue to recommend it when needed.” ACOG
Firm reassurance: emphasizes base case recommendation, affirms safety, does not deny all uncertainty.
ACOG — News Release / Public statement (2025)
“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.” ACOG
Strong in tone, pushing back on alarmist messaging; frames contrary claims as “harmful” and “confusing.”
SMFM (Society for Maternal-Fetal Medicine)
“To be clear, SMFM stands behind our recommendation that acetaminophen use during pregnancy has not been shown to cause or increase the risk of autism or other neurobehavioral problems in children.” SMFM
Reinforces lack of evidence for causation, expresses confidence in their position; acknowledges “suggesting” literature but rejects causality.
SMFM (earlier statement on acetaminophen & pregnancy and autism)
“The Society for Maternal-Fetal Medicine … continues to advise physicians and patients that acetaminophen is an appropriate medication to treat pain and fever during pregnancy.” SMFM
Emphasis on appropriateness for common clinical use, implicitly reassuring.
ACOG / Response to consensus statement (2021)
“ACOG and obstetrician-gynecologists across the country have always identified acetaminophen as one of the only safe pain relievers for pregnant [people].” ACOG
Historical grounding of the position; reminds clinicians and public of long-standing view.
U.S. FDA — “FDA Responds to Evidence of Possible Association…”
“It is important to note that while an association between acetaminophen and neurological conditions has been described in many studies, a causal relationship has not been established … acetaminophen is the only over-the-counter drug approved for use to treat fevers during pregnancy … high fevers in pregnant women can pose a risk to their children.” U.S. Food and Drug Administration
More cautious / balanced tone. Acknowledges possible associations, emphasizes nonestablished causality, and defends the role of acetaminophen as unique option for fever.
U.S. Department of Health and Human Services / White House “Fact: Evidence Suggests Link”
“There is mounting evidence finding a connection between acetaminophen use during pregnancy and autism … we are issuing this new health guidance.” The White House
More assertive / precautionary language, taking a stance toward risk, emphasizing “mounting evidence” though without acknowledging limits explicitly.
HHS / “Autism Announcement Fact Sheet”
“Chronic acetaminophen use in pregnant women, especially late in pregnancy, may cause long-term neurological effects in their children.” HHS.gov
Uses “may cause” phrasing, signals worry, emphasizes “chronic use,” which is framing intended to limit overgeneralization.
Commentary and Ethical Considerations
Spectrum of caution vs reassurance
Professional bodies (ACOG, SMFM) tend toward reassurance, emphasizing current standard of care and rejecting alarmist interpretations of association studies.
Regulatory/government bodies (e.g. FDA, HHS) often adopt precautionary language (“may,” “association,” “possible risk”), partly to signal they are taking new evidence seriously without committing to causation.
Balancing transparency and avoiding panic
Reassuring statements are ethically defensible if grounded in a rigorous evaluation of evidence. But oversimplified reassurance risks losing credibility if later data challenge the position.
More cautious wording (“may cause,” “not yet established”) respects uncertainty but can generate anxiety, especially if patients hear only the risk language without context.
Framing of “chronic use” and qualifiers
Many of the more cautious statements focus on chronic or prolonged use, implicitly distinguishing occasional, short-term use from sustained exposure.
This language softens the implication for routine use, but patients may interpret “may cause” as a more general prohibition.
Disagreement among authorities
When government messages begin to emphasize risk more heavily than professional societies, clinicians face tension: do they side with specialty consensus or evolving regulatory caution?
In such cases, the clinician should explain both the consensus views and the regulatory concerns, noting that regulatory decisions often incorporate precaution even in uncertainty.
Ethical responsibility in quoting guidance
When quoting professional or regulatory guidance to patients, clinicians should also explain the strengths, limitations, and context of the statements (e.g. “your specialty society says X based on existing evidence,” “FDA is proposing changes but has not confirmed causation”).
This preserves transparency and supports truly informed patient decision-making.