Prenatal paracetamol exposure cleared of ASD, ADHD risk in offspring

23 hours ago
Prenatal paracetamol exposure cleared of ASD, ADHD risk in offspring

Children born to mothers who used paracetamol during pregnancy are unlikely to develop autism spectrum disorder (ASD) or attention-deficit/hyperactivity disorder (ADHD), according to a study from Hong Kong.

Researchers used de-identified electronic health records within a triangulation framework and identified 708,020 mother-child pairs, with approximately 43.3 percent having prenatal paracetamol exposure. Then, they constructed sibling-matched cohorts of children from families with discordant prenatal paracetamol exposure and sufficient follow-up (≥2 years for ASD and ≥5 years for ADHD).

The final cohorts consisted of 124,333 children for ASD analysis (mean age 9.3 years, 50.3 percent male) and 97,285 for ADHD analysis (mean age 7.6 years, 50.2 percent male).

In sibling-matched analyses, prenatal paracetamol exposure had a null association with the risk of ASD (hazard ratio [HR], 1.00, 95 percent confidence interval [CI], 0.91–1.11) or ADHD (HR, 1.01, 95 percent CI, 0.93–1.08) in offspring. These null associations were consistent across exposure timing, cumulative dose, and usage patterns (sporadic, intermittent, persistent).

Conversely, conventional analysis in the full cohort (singleton live births) showed positive association between prenatal paracetamol exposure and ASD (HR, 1.17, 95 percent CI, 1.13–1.20) and ADHD (HR, 1.23, 95 percent CI, 1.20–1.27) in offspring.

Positive associations also emerged following negative control analyses. Prepregnancy paracetamol exposure was associated with increased risk of ASD (HR, 1.12, 95 percent CI, 1.08–1.17) and ADHD (HR, 1.24, 95 percent CI, 1.20–1.28) in offspring. Similar associations were observed for post-pregnancy paracetamol exposure.

Overall, the findings provide reassurance regarding the safety of indicated paracetamol use during pregnancy, the researchers said. The positive signals observed in conventional analyses may be attributed to residual familial confounding.

JAMA Intern Med 2026;doi:10.1001/jamainternmed.2026.2215