Breastfeeding and Autism Spectrum Disorder in the National Survey of Children's Health

Background: Breastfeeding is thought to promote healthy cognitive development. A small number of studies have also reported a protective association between breastfeeding and autism spectrum disorder (ASD). The National Survey of Children's Health (2007, 2011), a large nationally representative survey of US children, was examined to determine whether breastfeeding of infants is associated with later development of ASD. Methods: Respondents with a child ages 2-5 years (n = 37,901) were queried about whether their child was currently diagnosed with ASD and about their child's breastfeeding history. Additional information was available about child and family health history and family demographics. Survey-weighted logistic regressions were employed to examine ASD diagnosis as associated with several breastfeeding exposure metrics including a history of partial or exclusive breastfeeding for the first 3 or 6 months, and duration of partial or exclusive breastfeeding. Results: Across models, a current diagnosis of ASD (n = 391) was unassociated with any measure of breastfeeding history. Adjusted odds ratios for categorical breastfeeding exposures ranged from 0.68 (confidence interval [CI] = 0.4, 1.3) for any partial breastfeeding to 0.74 (CI = 0.3, 1.7) for 6 months of exclusive breastfeeding. When exposure was measured continuously, adjusted odds ratios were 1.03 (CI = 0.97, 1.10) for each additional month of partial breastfeeding and 1.04 (CI = 0.96, 1.13) for each additional month of exclusive breastfeeding. Conclusion: These findings call into question the results from the small body of research that has examined this issue to date.