Maternal Acetaminophen Use and Child Neurodevelopment

Importance: Whether maternal use of acetaminophen during pregnancy is associated with offspring's neurodevelopment remains debated. Objective: To evaluate the associations of maternal prenatal prescriptions of acetaminophen with attention-deficit/hyperactivity disorder (ADHD) and autism spectrum disorders (ASDs) among offspring. Design, Setting, Participants: This cohort study analyzed 2 092 926 singleton births between 2004 and 2015 in Taiwan, with 1 231 819 having at least 1 sibling. Estimated hazard ratios (HRs) and 95% CIs were examined for offspring ADHD or ASDs according to prenatal acetaminophen prescriptions, adjusting for confounding factors. Exposure: Prescriptions of acetaminophen were extracted from the National Health Insurance Research Database (NHIRD). Acetaminophen use was defined as having at least 2 dispense records during pregnancy, and the total number of prescriptions and the estimated mean daily dispensed dose were examined. Main Outcomes and Measures: The primary outcome was ascertaining diagnoses of ADHD and ASDs from the NHIRD. Results: Of the 2 092 926 singleton births between 2004 and 2015, 48.3% (n = 1 012 159) were born to mothers with at least 2 acetaminophen prescriptions during pregnancy. In the full cohort ASD dataset (N = 2 092 926), 23 557 children (1.1%) had ASD, and in the full ADHD dataset (N = 2 079 935), 116 387 children (5.6%) had ADHD. In the full cohort, offspring ADHD and ASDs were associated with prenatal prescriptions to acetaminophen, with associations noted for higher frequencies of acetaminophen use or higher mean daily dispensed doses of acetaminophen. In sibling-matched analyses, the associations between prenatal exposures to acetaminophen and ADHD and ASDs among offspring were null. However, a positive association was observed when only the older sibling was exposed (HR, 1.33 [95% CI, 1.17-1.52] for ADHD; HR, 1.75 [95% CI, 1.29-2.36] for ASDs), and a negative association was observed when only the younger sibling was exposed (HR, 0.75 [95% CI, 0.67-0.84] for ADHD; HR, 0.74 [95% CI, 0.57-0.96] for ASDs). The divergence of associations persisted in the bidirectional analyses of higher frequencies of acetaminophen use or higher mean daily doses of acetaminophen. Conclusions and Relevance: The findings of this cohort study in Taiwan suggest that positive associations were observed between maternal prenatal acetaminophen prescriptions and offspring's ADHD or ASDs in the full cohort but not in the sibling-matched analyses. A substantial divergence in associations in the sibling bidirectional analyses indicates unaddressed sources of bias and prevents firm conclusions from being drawn using the sibling design.