Error bars indicate 95% confidence intervals. Values adjusted for sex and month of birth.
T1D indicates type 1 diabetes; JIA, juvenile idiopathic arthritis.
aDiagnoses of infection, T1D and JIA (as a control autoimmune disease) were obtained using International Classification of Diseases, Tenth Revision, codes recorded in 3-month age intervals over 2 years.
Beyerlein A, Donnachie E, Jergens S, Ziegler A. Infections in Early Life and Development of Type 1 Diabetes. JAMA. 2016;315(17):1899-1901. doi:10.1001/jama.2016.2181
Viral infections, particularly enteroviruses,1 have been hypothesized to cause type 1 diabetes (T1D).2 Recent studies suggest that respiratory tract infections are associated with increased T1D risk if they are encountered within the first 6 months.3 We explored associations between infection types during the first 2 years and between respiratory tract infections in the first 6 months and T1D in a population-based cohort.
The Kassenärztliche Vereinigung Bayern processes claims data for all statutorily insured patients in Bavaria, Germany (approximately 85% of the total Bavarian population). Infants born between 2005 and 2007 were included and observed until March 2015 or last contact with a physician. Diagnoses of infection, T1D, and juvenile idiopathic arthritis (JIA, as a control autoimmune disease) were obtained using International Classification of Diseases, Tenth Revision, codes recorded in 3-month age intervals over 2 years (eg, birth to 2.9 months). Infections were categorized by symptoms (respiratory, gastrointestinal, dermal, and eye) and causes (viral, bacterial, and mycoses). Cox proportional hazards models were used to calculate hazard ratios (HRs) and 95% confidence intervals of time to T1D diagnosis by infection event, adjusting for sex and calendar month of birth. Infections were treated as binary time-varying covariates with nonexposure to a specific infection in a quarterly interval as reference category.
Kaplan-Meier analysis was used to estimate cumulative risks of T1D and JIA by respiratory tract infections or viral respiratory tract infections in the first 6 months. Differences were assessed by the log-rank test. Statistical analyses were conducted using SAS (SAS Institute), version 9.3, and R (R Foundation), version 3.0.3. Statistical significance was determined at the 5% level (2-sided). Data release was approved by the data protection officer according to German Guidelines for Secondary Data Analysis.4
Of the 295 420 infants included (male, 55.0%), 720 were diagnosed with T1D over a median follow-up of 8.5 years (interquartile range, 7.5-9.3), for an incidence of 29 diagnoses per 100 000 children annually. At least 1 infection was reported during the first 2 years of life in 92.9% of all children, and in 96.7% of children with T1D (χ2P < .001). Most children experienced respiratory (87.1%) and viral (83.5%) infections.
Respiratory tract infections between birth and 2.9 months of age or between 3 and 5.9 months occurred in 278 children who developed T1D (38.6%) and 100 693 children who did not develop T1D (34.2%). T1D risk was increased in children who had a respiratory tract infection compared with children who had no respiratory tract infections in these age intervals (HR, 1.17 [95% CI, 1.00-1.37]) (Figure 1).
Viral infections occurred between birth and 5.9 months of age in 243 children who developed T1D (33.8%) and 86 758 children who did not develop T1D (29.4%). An increased T1D risk was observed compared with children without viral infections (HR, 1.19 [95% CI, 1.01-1.39]).
Respiratory tract infections occurring in both age intervals (birth to 2.9 months and 3 to 5.9 months) were reported for 56 children with T1D (7.8%). Having respiratory tract infections in both age intervals was associated with an increased T1D risk compared with having infections in only 1 age interval or having no infections (cumulative 5-year risk per 100 000: 206 for infections in both intervals, 142 for infection in 1 age interval, 118 for no infections, P = .01). Risk was particularly increased if the infections were caused by viruses (cumulative 5-year risk per 100 000: 270 for infections in both age intervals, 145 for infection in 1 age interval, 120 for no infections; P < .001, Figure 2). No association was found between respiratory tract infections in the first 6 months and JIA (cumulative 5-year risk per 100 000: 102 for infections in both age intervals, 99 for infection in 1 age interval, 89 for no infections; P = .78).
Recurrent viral respiratory tract infections during the first 6 months were associated with T1D development by age 8 years in a large, population-based sample. It is unknown whether the association with early infections reflects increased exposure to virus or an impairment of the immune system response, perhaps due to genetic susceptibility. A similar study from Taiwan found that enterovirus infections were associated with increased T1D risk but was unable to investigate different time windows for infection exposure.5
We were unable to adjust for potential confounding factors such as family history of T1D or delivery mode as these data were unavailable. We did not have data on specific viral infections. Furthermore, we investigated several infection types with different exposure ages, potentially introducing multiple testing errors. However, the association of respiratory tract infections in the first 6 months with T1D is consistent with smaller studies assessing autoantibody development,3,6 suggesting that the first half-year of life is crucial for the development of the immune system and autoimmunity.
Corresponding Author: Anette-G. Ziegler, MD, Institute of Diabetes Research, Helmholtz Zentrum München, Ingolstädter Landstraße 1, 85764 Neuherberg, Germany (firstname.lastname@example.org).
Author Contributions: Drs Beyerlein and Donnachie had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Beyerlein, Ziegler.
Acquisition, analysis, or interpretation of data: Beyerlein, Donnachie, Jergens, Ziegler.
Drafting of the manuscript: Beyerlein, Ziegler.
Critical revision of the manuscript for important intellectual content: Donnachie, Jergens, Ziegler.
Statistical analysis: Beyerlein, Donnachie.
Administrative, technical, or material support: Donnachie, Jergens.
Study supervision: Ziegler.
Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest and none were reported.
Funding/Support: The work was supported by grants from the Kompetenznetz Diabetes mellitus (Competence Network for Diabetes mellitus), the Federal Ministry of Education and Research (FKZ 01GI0805-07, FKZ 01GI0805), the Deutsche Forschungsgemeinschaft (DFG ZI-310/14-1 to -4), the JDRF (JDRF-No 17-2012-16), the German Federal Ministry of Education and Research, the German Center for Diabetes Research (DZD e.V.), and iMed (Helmholtz Initiative on Personalized Medicine).
Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Additional Contributions: We thank Ezio Bonifacio, PhD (Technische Universität Dresden, Germany), for critically reviewing the manuscript without receiving compensation.