In a recent study published on the medRxiv* preprint server, researchers investigate whether the risk of new-onset type 1 diabetes (T1D) is elevated by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infections among children living in Denmark.
Previous reports on the potential association between coronavirus disease 2019 (COVID-19) and the risk of new-onset T1D underpins the probability of viral organisms as etiological agents for T1D. As COVID-19 positivity rates remain high in many countries worldwide, including Denmark, this correlation could increase existing global health burdens associated with SARS-CoV-2 infections.
About the study
In the present nationwide prospective study, researchers assess the association between COVID-19 and the subsequent risk of new-onset T1D among Danish children.
Data from multiple nationwide Danish health registers of individuals below 18 years of age with at least one COVID-19 positive test result between March 1, 2020, and August 25, 2022, were analyzed.
Denmark residents were identified by the civil registration system (CPR) of Denmark. In addition, COVID-19 testing data were obtained from the nationwide SARS-CoV-2 infection surveillance system, including reverse-transcription-polymerase chain reaction (RT-PCR) analysis reports.
The diagnoses of T1D and diabetic ketoacidosis were based on the international classification of diseases 10th revision (ICD-10) DE10 and DE101 codes, respectively, in the national patient register (NPR). In addition, the team linked registration data using CPR numbers that are unique to each Danish citizen.
Follow-up assessments were performed between 30 days post-initial SARS-CoV-2 testing and study termination, or until the individual turned 18 years, was deceased, or left Denmark and was designated as missing in the Danish CPR register.
The team excluded individuals who were registered as suffering from T1D or diabetic ketoacidosis before the study commenced. Cox regression modeling was used for the analysis, and hazard ratios (HR) were calculated, with data adjustments for sex, comorbidities, COVID-19 vaccination status (unvaccinated, one dose, two doses or more), parental T1D history, and the calendar year month.
A secondary analysis was also performed to assess the association between COVID-19-associated hospitalizations and subsequent T1D diagnosis among children during the study period.
A total of 613 individuals received new-onset T1D diagnoses in 1,593,937 individual years for 1,115,716 individuals below 18. Moreover, the incidence rate of T1D was 39 cases for every 100,000 individual years.
Of the 613 cases, 144 were detected in 419,260 individual years during follow-up assessments for 720,648 children infected with SARS-CoV-2.
A greater risk of new-onset T1D among the pediatric population after 30 or more days of a COVID-19 positive report, in comparison to uninfected children, was not found (HR 0.9). These findings were similar across sex, age, comorbidities, vaccination status, history of T1D among parents, and calendar month of T1D diagnosis.
Between 30 days and six months of testing positive for COVID-19, the HR value was 0.9 (102.0 events, and 0.8 (42.0 events), respectively, for greater than six months after a positive result, as compared to uninfected or SARS-CoV-2-negative individuals aged below 18 years of age. The HR values for the simultaneous diagnoses of COVID-19 and T1D, as well as COVID-19 with diabetic ketoacidosis, were 0.6 (17.0 events) and 0.9 (127.0 events), respectively.
The subgroup analysis, by dividing the periods of different SARS-CoV-2 variants’ predominance, did not show any statistically significant differences in the HR estimates. The secondary analysis findings identified 936 individuals diagnosed with T1D over 2,817,858 individual years; however, the team did not observe any T1D cases in the 30 days or more following the initial COVID-19-associated hospital admissions (939.0 individual years).
Overall, the study findings indicate that COVID-19 does not predispose children to T1D; therefore, T1D does not need to be an area of concern in SARS-CoV-2-infected children. However, the study findings contrasted the results of studies conducted by the Centers for Disease Control and Prevention (CDC) in the United States and Norway.
The CDC analysis was based on the Health Verity, and the IQVIA claims U.S. databases. In that study, prevalent T1D cases may have been misclassified as incident ones, thus resulting in an overestimation of incidence rates. Likewise, the U.S. study was based on the TriNetX LLC claims database.
Identifying exposed individuals from health claim databases and subsequent analysis of individuals with exposure to health disorders apart from COVID-19 as comparison or reference groups could make the relevant study population estimation difficult, thus limiting the study findings’ generalizability.
The studies conducted in Norway and Scotland were national health registry-based, similar to the present study. However, the distribution of confounding factors might have differed across the study populations and, as a result, led to differential risk estimations. Moreover, data on T1D etiology, other than the genetic factor, are limited and require further investigation.