Paracetamol in pregnancy: what the evidence shows about the autism claims

The best available evidence does not show that paracetamol — known as acetaminophen in the United States — taken during pregnancy causes autism or attention-deficit/hyperactivity disorder in children: the largest study, which compared siblings to remove shared family influences, found no effect [s1]. Obstetric guidance continues to describe paracetamol as the preferred option for pain and fever in pregnancy when a painkiller is needed, used at the lowest effective dose for the shortest time [s2][s3]. That combination — reassuring headline evidence, sensible caution about overuse — is the honest summary a wave of alarming claims has obscured.

Why this became a fight

Paracetamol is one of the most widely used medicines in pregnancy, in part because the safer alternatives are limited: anti-inflammatories such as ibuprofen are generally avoided, especially later in pregnancy. It is the active ingredient in more than 600 medications for pain and fever [s2]. So a claimed link to neurodevelopmental disorders is not a niche concern; it touches almost every pregnancy.

In 2025 those claims intensified in public debate, and obstetric bodies pushed back. The American College of Obstetricians and Gynecologists states that more than two decades of research has shown acetaminophen is safe to use during pregnancy in consultation with a physician, and that misinformation — including claims incorrectly tying acetaminophen to autism and neurodevelopmental disorders — is growing [s3]. Untreated high fever and severe pain in pregnancy carry their own risks, which is part of why a blanket "avoid it" message can do harm [s2][s3].

The study that changed the picture

Earlier observational studies did report associations between paracetamol exposure in the womb and later autism or ADHD, and those findings drove genuine scientific concern [s2]. The problem with such studies is confounding: the reasons a woman takes paracetamol — infections, inflammation, pain, chronic conditions, genetics — may themselves relate to a child's neurodevelopment, so an association is not proof of cause.

A 2024 nationwide study was designed to break that deadlock. It followed 2,480,797 children born in Sweden between 1995 and 2019, of whom 185,909 (7.49%) were exposed to paracetamol in pregnancy [s1]. In a conventional analysis, ever-use was associated with a marginally higher risk of autism (hazard ratio 1.05, 95% confidence interval 1.02 to 1.08) [s1]. But the researchers then compared full siblings — one exposed in pregnancy, one not — which holds constant the genetics and family environment siblings share. In that sibling-controlled analysis the association vanished: the hazard ratio for autism was 0.98 (95% CI 0.93 to 1.04), for ADHD 0.98 (95% CI 0.94 to 1.02), and for intellectual disability 1.01 (95% CI 0.92 to 1.10) [s1]. There was also no dose-response pattern once siblings were compared [s1]. The authors concluded the associations seen in weaker designs were likely due to familial confounding, not the drug [s1].

Where legitimate caution remains

None of this makes paracetamol a substance to reach for casually. A 2021 consensus statement signed by 91 scientists, clinicians and public-health professionals called for precautionary action, arguing that accumulating experimental and epidemiological research warranted a focused research effort and greater awareness [s2]. Its practical advice is measured and worth quoting in substance: use paracetamol in pregnancy only when medically indicated, consult a clinician or pharmacist if unsure or before long-term use, and minimise exposure by using the lowest effective dose for the shortest possible time [s2]. That is the same principle that applies to almost any medicine in pregnancy, and it sits comfortably alongside the reassuring sibling data [s1][s2].

The nausea-drug question, briefly

The other common medication dilemma in early pregnancy is nausea and vomiting. For the anti-sickness drug ondansetron, a large study of 1,816,414 pregnancies — 88,467 (4.9%) exposed in the first trimester — found no increased risk of cardiac malformations (adjusted relative risk 0.99, 95% CI 0.93 to 1.06) [s4]. It did find a small increased risk of oral clefts (adjusted relative risk 1.24, 95% CI 1.03 to 1.48), an absolute difference of about 2.7 additional cases per 10,000 births (95% CI 0.2 to 5.2) [s4]. That is a genuinely small signal, and it illustrates the general point: the honest answer to "is this drug safe in pregnancy?" is usually a specific, quantified balance, not a yes or a no.

For the drug outside pregnancy, our explainers on paracetamol and the liver and on paracetamol for back pain and osteoarthritis cover its safety margin and its limits as a painkiller. The broader evidence on what does and does not cause autism is in our explainer on early signs and screening.

How to read this without overreaching

The evidence supports a specific conclusion, not a slogan. The strongest study, using a design built to defeat confounding, found no link between paracetamol in pregnancy and autism, ADHD or intellectual disability [s1]; guidelines still favour it over the alternatives when a painkiller is genuinely needed [s2][s3]; and the sensible rule is the ordinary one of using the least medicine that does the job [s2].

The limits are worth stating. The sibling analysis is observational and cannot exclude every non-shared confounder, and consensus scientists still urge more research rather than declaring the question closed [s1][s2]. This is a fast-moving area of debate and regulation.

This article is informational and is not medical advice. It does not tell any pregnant person which medicine to take or avoid; those decisions should be made with a qualified clinician who knows the individual case.