In Reply Dr Rahmoune and colleagues and Mr Uzoigwe suggest that unmeasured confounding might have influenced the results of our study. The results are consistent with findings from a previous analysis that observed a strong association between respiratory infections during the first 6 months of life and development of beta cell autoimmunity in infants with a family history of T1D.1 This association was independent of early variables such as delivery mode, age at gluten introduction, season of birth, and antibiotic use in children, and could not be explained by modulations of the gut microbiome,2 suggesting that such confounding factors may not have a major role in determining the association between infection rates and T1D risk.
Beyerlein A, Donnachie E, Ziegler A. Association of Infection in Early Life and Risk of Developing Type 1 Diabetes—Reply. JAMA. 2016;316(8):883. doi:10.1001/jama.2016.10294