Diabetes Incidence in Children Saw Nonsignificant Increase During COVID-19 Pandemic

Rayzel Shulman, MD, PhD

A recent cross-sectional study in Canada observed a slightly higher, but ultimately non-significant, increase in the incidence of diabetes among children during the COVID-19 pandemic.

The rates were similar to previous findings from Germany wherein data show a 1.15-fold increase in type 1 diabetes incidence among German children during the COVID-19 pandemic. However, they reported no observed association regarding an increase in incidence in the months following infection.

“The lack of both an observable increase in overall diabetes incidence among children during the 18-month pandemic restrictions and a plausible biological mechanism calls into question an association between COVID-19 and new-onset diabetes,” wrote study author Rayzel Shulman, MD, PhD, Division of Endocrinology, Hospital for Sick Children.

Although a recent study reported an association between COVID-19 infection and new-onset diabetes among this population in the US, some experts have criticized the study’s conclusion due to no clear mechanism in how infection might cause the disease.

Due to Canada having one of the highest incidence rates of T1D worldwide, the current study examined whether diabetes incidence increased during the COVID-19 pandemic among children aged <18 years in Ontario, Canada.

Investigators used health administrative data from January 2017 - September 2021 analyzed at ICES and included all children and youth in Ontario eligible for universal health care insurance on January 1 of each study year. The province has a population of approximately 14.8 million people, with 3 million younger than 18 years.

Data show an estimated 3.3% of children had SARS Cov-2 infection between November 2020 and April 2021. Investigators used a generalized estimating equation for Poisson regression to model 3-year pre-COVID-19 rates adjusting for age group, sex, pre-COVID-19 month, and secular trend.

Based on these models, they then estimated expected post-COVID-19 monthly rates (95% confidence intervals [CI]) using 2-sided hypothesis tests. The exposure was considered the pandemic era starting in March 2020 with the outcome being a new diabetes diagnosis.

A total of 2,700,178 children were included in the 2021 cohort, with a mean age of 9.2 years and consisting of 48.7% female. Overall, data show no difference in observed versus expected relative rates (RRs) of new diabetes presentations (RR, 1.09 [95% CI, 0.91 - 1.30]).

Despite this, the RRs of new diabetes presentations were found to decrease in the first 3 months of the pandemic (15% - 32% lower in March to May 2020). A subsequent increase to higher expected rates (33% - 50% higher) was observed from February to July 2021.

Limitations of the study included the small population and lower power, according to investigators. They noted they cannot rule out a 1.3-fold increase in RRs.

They additionally mentioned the advantage of the study is the reported monthly variations in post-COVID-19 diabetes incidence showing a decline followed by an increase in rates. This suggested a potential delay in diabetes diagnosis for children early in the pandemic with a catch-up effect, they said.

“Given the variability in monthly RRs, additional population-based, longer-term data are needed to examine the direct and indirect effects of COVID-19 and diabetes risk among children,” Shulman concluded.

The research letter, “Examination of Trends in Diabetes Incidence Among Children During the COVID-19 Pandemic in Ontario, Canada, From March 2020 to September 2021,” was published in JAMA Network Open.