HYPE: “Taking paracetamol during pregnancy causes autism.”
But what does the evidence actually show?
Some observational studies report an association, but an association is not proof of causation. The strongest family-controlled evidence and recent reviews do not support the claim that appropriate paracetamol use causes autism. Pregnant people should not stop clinically appropriate treatment because of a viral headline.
Why People Believe It
Paracetamol — called acetaminophen in the US and sold under names including Tylenol — can cross the placenta. Researchers have proposed effects involving oxidative stress, hormones and brain development. Those are reasons to investigate, not demonstrations that ordinary use causes autism.
The concern also comes from real observational signals. Earlier cohorts and pooled analyses sometimes found more autism diagnoses or autistic traits among exposed children. The strongest reasonable question is whether the medicine explains that difference, or whether the circumstances surrounding its use do.
What the Evidence Shows
What changed when researchers compared siblings?
A Swedish study followed 2.48 million children. In conventional adjusted analyses, prenatal exposure was associated with a small increase in autism diagnoses: hazard ratio 1.05. The estimated difference at age ten was 0.09 percentage points — about nine additional diagnoses per 10,000 children — with an interval compatible with no difference.
Within-family comparisons changed the picture: the autism hazard ratio was 0.98 (95% confidence interval 0.93–1.04), with no association or dose–response pattern. Comparing siblings helps account for shared family background and genetics. It does not make an observational study a randomised trial.
Read the research ↗More reviews do not necessarily mean stronger evidence
A BMJ umbrella review examined nine reviews covering 40 primary studies. Seven reviews were rated critically low quality and two low quality. Studies overlapped substantially; counting the same families repeatedly does not create independent confirmation. The authors judged the evidence insufficient to establish a causal link.
Read the research ↗Better-controlled analyses remain reassuring
D’Antonio and colleagues included 43 studies, with 17 entering meta-analysis. Their sibling-comparison analysis found no increased likelihood of autism: odds ratio 0.98 (0.93–1.03). Analyses restricted to studies at low risk of bias also found no association. ADHD was examined separately and showed the same broad pattern; it is not interchangeable with autism.
This synthesis reuses important cohorts, including the Swedish study, rather than supplying a wholly new population. Reliable detail on trimester, frequency and prolonged exposure remains limited.
Read the research ↗Why positive pooled associations still need interpretation
Tayebi-Hillali and colleagues found a small autism association in unrestricted pooled reviews (relative risk 1.17). Estimates weakened substantially when restricted to better-quality evidence, sometimes approaching no association. They identified heterogeneity, overlapping studies and residual confounding, and did not conclude that a robust causal relationship had been established.
Read the research ↗UK guidance is consistent with that interpretation. The MHRA and RCOG reaffirmed in January 2026 that paracetamol remains an appropriate option for pain or fever in pregnancy when used as directed. Their guidance takes account of the reasons treatment is needed, rather than assuming that avoiding all medication is automatically safer.
The Catch
Confounding by indication means the reason a medicine is taken may help explain an association. Maternal fever, infections, pain and underlying illness can differ between exposed and unexposed pregnancies. Genetic susceptibility, parental characteristics and shared environments can also influence both medicine use and neurodevelopment.
A longer duration of use may indicate a longer illness. A statistical dose–response in a conventional cohort therefore does not settle causation. Sibling comparisons address shared factors but cannot remove every difference between pregnancies; incomplete reporting of over-the-counter use can also blur exposure.
The evidence cannot prove zero risk for every possible dose, duration or circumstance. That limitation should not be converted into a positive claim that paracetamol causes autism. Mechanistic experiments and measurements of metabolites do not establish the clinical effect of ordinary use.
Autism is a complex neurodevelopmental condition. These data do not provide a basis for blaming a parent’s medication choice.
What This Means for You
If you are pregnant and need treatment for pain or fever, follow the product instructions and advice from your midwife, pharmacist or doctor. Ask for advice about persistent symptoms, repeated use or uncertainty over dose.
Do not discontinue medically appropriate treatment or substitute another medicine on the basis of this article. The practical decision depends on your symptoms and clinical circumstances, not a social-media assertion.
An observational association has been reported. Paracetamol has not been shown to cause autism, and the best-controlled evidence does not support that claim.
Research & Sources
Ahlqvist et al. (2024)
Acetaminophen Use During Pregnancy and Children’s Risk of Autism, ADHD, and Intellectual Disability
JAMA
Original source ↗BMJ umbrella review (2025)
Maternal paracetamol use during pregnancy and risk of autism spectrum disorder and attention deficit/hyperactivity disorder in offspring
BMJ 391: e088141
Original source ↗D’Antonio et al. (2026)
Prenatal paracetamol exposure and child neurodevelopment: a systematic review and meta-analysis
The Lancet Obstetrics, Gynaecology, & Women’s Health
Original source ↗Tayebi-Hillali et al. (2026)
Prenatal exposure to paracetamol and the risk of autism and attention-deficit/hyperactivity disorder: An umbrella review and meta-meta-analysis
Developmental Medicine & Child Neurology
Original source ↗MHRA (2026)
Statement on new review of paracetamol safety during pregnancy
UK Medicines and Healthcare products Regulatory Agency
Original source ↗RCOG (2026)
Paracetamol in pregnancy
Royal College of Obstetricians and Gynaecologists
Original source ↗Evidence reviewed: October 2026. This is an editorial investigation, not an exhaustive systematic review or personal medical advice. The featured artwork is conceptual illustration, not a study result. Evidence Ratings assess support for the specific claim described above.