To Örebro University

Objectives: To investigate the risk of future diabetes mellitus type 1 (T1D) in children who suffered from infection at time of gluten introduction.

Study design: Population-based prospective study. Parents filled out a diary at home. We hereby obtained data on date of gluten introduction, breastfeeding duration, and infections in 9414 children born in the southeast of Sweden from October 1, 1997, through October 1, 1999 (the All Babies in Southeast Sweden cohort). The Cox proportional hazards model was used to investigate the risk of future T1D until February 1, 2012, among children with infection at time of gluten introduction.

Results: Forty-six children (0.5%) developed T1D and were compared with 9368 reference children from the general population. Some 10 of 46 children with later T1D had an infection at time of gluten introduction (22%) compared with 2520 reference children (27%, P = .43). Later T1D was not associated with age at end of breastfeeding, age at any infection, or age at gluten introduction. Breastfeeding at time of gluten introduction was not protective against future T1D (hazard ratio 1.2; 95% CI, 0.5-2.7). In our final model, when we adjusted for age at gluten introduction, age at infection, and breastfeeding duration, infection at time of gluten introduction did not influence the risk of future T1D (hazard ratio 0.8; 95% CI, 0.3-1.6).

Conclusion: Infection at time of gluten introduction is not a major risk factor for future T1D in nonselected children.