Meta-analysis found that studies may have overestimated Caesarean section risks for attention-deficit hyperactivity disorder by ignoring confounding factors

Abstract

Aim: Epidemiological studies on associations between Caesarean sections (C-sections) and attention-deficit hyperactivity disorder (ADHD) have been inconsistent, and we performed a meta-analysis.

Methods: We systematically searched PubMed and Embase to December 2018 and included nine hospital-based and population registry studies published in 2011-2018. These covered a total study cohort of more than 2.5 million people in eight countries: Australia, Brazil, Denmark, Finland, Germany, Sweden, Turkey and the UK. The analysis provided summary odds ratios (ORs) and 95% confidence intervals (CI) while taking heterogeneity into account.

Results: We found that that C-sections were associated with a small increase in the risk of ADHD (OR 1.14, 95% CI 1.11, 1.17, I² 0%) in offspring. In subgroup analyses, the association remained for both infants born after elective C-sections (OR, 1.15, 1.11, 1.19, I² 0%) and emergency C-sections (OR, 1.13, 1.1, 1.17, I² 45.4%). However, these were only marginally significant when we pooled data from siblings from other pregnancies (OR, 1.06, 1.00-1.13, I² 0%), implying that the association was due to confounding.

Conclusion: The statistically significant association between C-sections and ADHD in children can be partially explained by unmeasured confounding. Further research controlling for important confounders is required before firm conclusions can be drawn.

Keywords: Caesarean section; attention-deficit hyperactivity disorder; cofounders; meta-analysis; neurodevelopment.