Evidence supporting the association between breastfeeding patterns and autism is inconsistent. This study examined sociodemographic and birth factors related to breastfeeding duration and subsequent autism spectrum disorder (autism) diagnosis according to the Diagnostic and Statistical Manual of Mental Disorders, 5 th Edition, compared to a neurotypically developing cohort, based on electronic health records. Demographics, feeding preferences, and breastfeeding duration as reported by parents during routine baby wellness visits were analyzed for a cohort of 11,766 (1.9%) children with autism spectrum disorder diagnosis and a random subsample of 12,000 (2.03%) neurotypically developing children. Autism spectrum disorder diagnosis was based on a national autism registry and assigned after electronic health records were completed. Preterm, very low birth weight, multiple births, and complex medical comorbidities were excluded. Infants subsequently diagnosed with autism were breastfed for an average of 5.0 months, 1.5 months shorter than neurotypically developing. Fewer autism spectrum disorder infants were exclusively breastfed in the first year of life. Two-way analysis of variance indicated a significant effect of socioeconomic status and autism spectrum disorder on breastfeeding duration, and a significant interaction of socioeconomic status with autism spectrum disorder. Shorter breastfeeding duration among infants with subsequent autism spectrum disorder was confirmed, calling for closer monitoring for autism traits in infants with breastfeeding difficulties. These challenges were independent of birth parameters; however, they were influenced by socioeconomic factors. Lay Abstract This study found that infants later diagnosed with autism were breastfed for a shorter duration than their typically developing peers, with differences influenced by socioeconomic status. These findings highlight the importance of monitoring breastfeeding challenges as a potential early indicator of autism, particularly in families of mid-range socioeconomic status.
Ben-Sasson et al. (Sat,) studied this question.