This population-based sibling-matched cohort study (Hong Kong, 2001–2023) assessed whether prenatal acetaminophen (paracetamol) exposure—identified from electronic dispensing records—was associated with incident ASD or ADHD diagnoses in offspring, using within-family comparison to control for unmeasured familial confounding.
In sibling-matched analyses (n=124,333 for ASD; n=97,285 for ADHD), prenatal paracetamol exposure was not associated with ASD (aHR 1.00; 95% CI, 0.91–1.11) or ADHD (aHR 1.01; 95% CI, 0.93–1.08); null findings held across all exposure timings, cumulative doses, and usage patterns. Positive associations seen in conventional (unmatched) cohort analyses were replicated by negative control analyses using *prepregnancy* exposure (ASD HR 1.12; ADHD HR 1.24), strongly implicating familial confounding as the driver of prior positive signals.
- Study is observational; sibling-matching controls familial confounding but cannot rule out all residual confounding. - Exposure ascertained from dispensing records only—actual consumption and over-the-counter use are not captured. - Findings are from a single health system in Hong Kong and may not generalize to other populations.
Clinicians can reassure pregnant patients that indicated paracetamol use does not appear to raise the risk of ASD or ADHD in their children—prior observational signals were likely driven by shared family factors, not the drug itself. The guidance to use paracetamol at the lowest effective dose for the shortest necessary duration during pregnancy remains prudent but need not be motivated by neurodevelopmental concerns.
