Glycated hemoglobin at diagnosis of type 1 diabetes and at follow-up in children and adolescents during the COVID-19 pandemic in Germany.

BACKGROUND: This study investigated the diagnostic delay and the subsequent quality of care during the Covid-19 pandemic among children with new-onset type 1 diabetes.
METHODS: We compared the HbA1c levels of 3111 children at diagnosis of type 1 diabetes and of 2825 children at a median follow-up of 4.7 months (interquartile range, 4.1-5.4) together with their daily insulin requirement during the Covid-19 pandemic with the two previous years via multivariable linear regression, using data from the German Diabetes Registry DPV.
RESULTS: During the Covid-19 pandemic, HbA1c levels were higher at diagnosis of type 1 diabetes (mean estimated difference, 0.33% [95% confidence interval, 0.23-0.43], p < 0.001), but not at follow-up (mean estimated difference, 0.02% [-0.02-0.07]). Children with diabetes onset during the Covid-19 pandemic had a significantly higher daily insulin requirement after initiation of therapy (mean estimated difference, 0.08 U/kg [0.06-0.10], p < 0.001). Both the increase in HbA1c and daily insulin requirement were evident only after the first wave of the pandemic.
CONCLUSIONS: This increase in HbA1c at diagnosis of type 1 diabetes during the Covid-19 pandemic may indicate a delay in seeking medical care due to the pandemic. However, this did not affect short-term glycemic control. The increased insulin requirement at follow-up could suggest a more rapid autoimmune progression during the pandemic.

INTRODUCTION

During the coronavirus disease 2019 (Covid‐19) pandemic, contacts with the health care system have markedly declined and diagnoses were delayed, leading to more advanced stages of diseases. , For children and adolescents with new‐onset type 1 diabetes, this resulted in an increased rate of diabetic ketoacidosis. , , , However, the impact of delayed diagnosis and impaired acute care on the continued care of children and adolescents with new‐onset type 1 diabetes during the Covid‐19 pandemic is unknown.

The aim of this study was to quantify the diagnostic delay and its impact on subsequent quality of care during the Covid‐19 pandemic in Germany by comparing the levels of glycated hemoglobin (HbA1c) at diagnosis of type 1 diabetes and after initiation of therapy during the Covid‐19 pandemic with those of the two previous years.

METHODS

This study compared HbA1c levels of children and adolescents from the German Diabetes Prospective Follow‐up Registry (DPV) at diagnosis of type 1 diabetes in the year 2020, at follow‐up 2–8 months later, and daily insulin requirement (units per kilogram body weight), with data from 2018 and 2019 via multivariable linear regression, adjusted for age group (<6, 6–<12, and 12–<18 years), sex, and immigrant background (patient or at least one parent born outside Germany). Adjusted differences with the corresponding 95% confidence interval (CI) were presented for the whole year, as well as for four different periods related to the Covid‐19 pandemic: the pre‐pandemic period (January and February 2020), the first wave of the pandemic from March to May 2020, the period from June to September 2020 with a relatively low rate of new infections, and the 2nd wave starting October 2020.

The DPV registry has a nationwide coverage of more than 90% of pediatric patients with type 1 diabetes in Germany and comprises 257 pediatric diabetes centers (hospitals and practices) as of March 2021. Twice a year, locally collected longitudinal data are pseudonymized and transmitted for central plausibility checks and analyses to Ulm University, Ulm, Germany. Inconsistent data are reported back to participating centers for validation and/or correction. Data are then completely anonymized for analysis. Verbal or written informed consent for participation in the DPV registry was obtained from patients or their guardians. The ethics committee of Ulm University approved the analysis of anonymized data from the DPV registry.

Local HbA1c values were mathematically standardized to the DCCT reference range (4.05–6.05%) using the “multiple of the mean” transformation method. HbA1c at diagnosis was aggregated 10 days around the date of diagnosis. BMI values (calculated as weight in kilograms divided by height in meters squared) were transformed to standard deviation scores (SDSLMS) based on German reference values (KiGGS [German Health Interview and Examination Survey for Children and Adolescents]) by applying the Box‐Cox‐transformation method. Confidence intervals for estimated period‐specific values were adjusted according to the Bonferroni method, and corresponding p‐values according to the Holm method.

A two‐sided p‐value <0.05 was considered statistically significant. All analyses were performed with SAS version 9.4 (SAS Institute Inc., NC, USA).

RESULTS

We obtained HbA1c values at diabetes diagnosis from 3111 children and adolescents (55.5% males; median age 9.8 years [interquartile range, 5.9–12.9]) with new‐onset type 1 diabetes in 2020 from 186 diabetes centers in Germany, of whom we obtained data of 2825 patients (90.8%) at a median follow‐up of 4.7 months (interquartile range, 4.1–5.4). The median HbA1c was 11.4% (interquartile range, 9.9–13.1 [101 mmol/mol (85–120)]) at diagnosis of type 1 diabetes and 6.7% (interquartile range, 6.1–7.3 [50 mmol/mol (43–56)]) at follow‐up. The median daily insulin dose at follow‐up was 0.61 IU/kg (interquartile range, 0.44–0.83) and the median BMI‐SDS at follow‐up was −0.27 (interquartile range, −1.09–0.54). Data from the 2020 cohort were compared with data from 5256 children and adolescents (54.7% males; median age 9.8 years [interquartile range, 6.0–13.1]) with new‐onset type 1 diabetes in 2019 and 2018, and their follow‐up data (N = 4789) after a median of 4.7 months (interquartile range, 4.1–5.4). The median HbA1c of the 2019/2018 cohort was 11.1% (interquartile range, 9.6–12.81 [98 mmol/mol (82–117)]) at diagnosis and 6.6% (interquartile range, 6.1–7.3 [49 mmol/mol (43–56)]) at follow‐up. The median daily insulin dose at follow‐up was 0.57 IU/kg (interquartile range, 0.42–0.76) and the median BMI‐SDS at follow‐up was −0.29 (interquartile range, −1.10–0.55). Table 1 provides a descriptive overview of the 2020 compared to the 2018/2019 cohort.

TABLE 1

Characteristics of patients with new‐onset type 1 diabetes from 2020 and from 2018/2019

Variable	2020	2018/2019	p‐value
At diabetes onset—number of participants	3111	5256
Median age at diabetes diagnosis—years (interquartile range)	9.8 (5.9–12.9)	9.8 (6.0–13.1)	>0.99
Sex (male)—%	55.5	54.7	>0.99
Immigrant background—%	26.6	26.9	>0.99
Diabetic ketoacidosis—%	35.3	27.2	<0.001
Median HbA1c—% (interquartile range)	11.4 (9.9–13.1)	11.1 (9.6–12.8)	<0.001
At follow‐up—number of participants	2825	4789
Median time after diabetes diagnosis at follow‐up—months (interquartile range)	4.7 (4.1–5.4)	4.7 (4.1–5.4)	>0.99
Median age at diabetes diagnosis—years (interquartile range)	9.7 (6.0–12.8)	9.6 (5.8–12.8)	>0.99
Sex (male)—%	55.3	54.5	>0.99
Immigrant background—%	26.8	27.0	>0.99
Median HbA1c—% (interquartile range)	6.7 (6.1–7.3)	6.6 (6.1–7.3)	>0.99
Median daily insulin dose—IU/kg (interquartile range)	0.61 (0.44–0.83)	0.57 (0.42–0.76)	<0.001
Median BMI—SDS (interquartile range)	−0.27 (−1.09–0.54)	−0.29 (−1.10–0.55)	>0.99

Note: For demographic and clinical data, the cohort of 2020 was compared to children and adolescents with a diagnosis of type 1 diabetes in the two previous years 2019 and 2018 in Germany. All children and adolescents were between 6 months and less than 18 years of age at the time of diabetes diagnosis. Unadjusted values were compared via Wilcoxon's rank sum test for continuous variables and χ2‐test for dichotomous variables.

The significant of bold values as results with the bold p‐values.

Children and adolescents with new‐onset type 1 diabetes in 2020 had higher adjusted mean HbA1c at diagnosis compared to 2019 and 2018 (mean estimated difference, 0.33% [95% CI, 0.23–0.43], p < 0.001; Table 2A). The difference in adjusted HbA1c at diagnosis between 2020 and 2019/2018 was significant in all age groups and both sexes (Table 2A). Analysis by periods showed that the difference in HbA1c was significant after the first wave of the pandemic (Table 2B). In contrast, the HbA1c during follow‐up no longer differed (mean estimated difference, 0.02% [95% CI, −0.02–0.07]; Table 2C), not even in the patients who developed diabetes after the first Covid‐19 wave (Table 2D). However, children with diabetes onset during the Covid‐19 pandemic had a significantly higher daily insulin dose after initiation of therapy (mean estimated difference, 0.08 U/kg [95% CI, 0.06–0.10], p < 0.001; Table 2E). Compared to the 2 years before the pandemic, the increased insulin requirements affected those children who developed type 1 diabetes after the first wave of the Covid‐19 pandemic. (Table 2F).

TABLE 2

Adjusted mean HbA1c at diagnosis and at follow‐up, and adjusted daily insulin dose at follow‐up of children and adolescents with new‐onset type 1 diabetes in 2020 versus 2018/2019

	In 2020—adjusted mean (95% CI)	In 2018/2019—adjusted mean (95% CI)	Absolute difference 2020 versus 2018/2019—adjusted mean (95% CI)	p‐value
A. HbA1c (in %) at diabetes diagnosis—the whole year
All patients	11.55 (11.47–11.63)	11.22 (11.16–11.29)	0.33 (0.23–0.43)	<0.001
Females	11.76 (11.63–11.88)	11.49 (11.39–11.58)	0.27 (0.11–0.43)	<0.001
Males	11.39 (11.28–11.49)	11.01 (10.93–11.09)	0.38 (0.25–0.51)	<0.001
<6 years	10.43 (10.29–10.57)	10.22 (10.12–10.33)	0.21 (0.03–0.39)	0.020
6–11.9 years	11.76 (11.63–11.88)	11.36 (11.26–11.45)	0.40 (0.24–0.56)	<0.001
12–17.9 years	12.16 (12.01–12.32)	11.83 (11.71–11.94)	0.34 (0.14–0.53)	<0.001
B. HbA1c (in %) at diabetes diagnosis—the four pandemic‐related periods
January–February	11.27 (11.01–11.53)	11.13 (10.93–11.32)	0.15 (−0.23–0.52)	>0.99
March–May	11.51 (11.28–11.74)	11.24 (11.06–11.41)	0.27 (−0.06–0.60)	0.17
June–September	11.76 (11.57–11.96)	11.40 (11.24–11.56)	0.36 (0.08–0.65)	0.001
October–December	11.52 (11.29–11.75)	11.08 (10.91–11.25)	0.44 (0.11–0.77)	<0.001
C. HbA1c (in %) at follow‐up a —the whole year
All patients	6.77 (6.73–6.80)	6.75 (6.72–6.75)	0.02 (−0.02–0.07)	0.38
Females	6.84 (6.78–6.89)	6.80 (6.76–6.84)	0.03 (−0.03–0.10)	0.31
Males	6.71 (6.66–6.76)	6.70 (6.66–6.74)	0.01 (−0.05–0.07)	0.79
<6 years	7.13 (7.07–7.20)	7.10 (7.05–7.15)	0.03 (−0.05–0.11)	0.49
6–11.9 years	6.71 (6.66–6.76)	6.67 (6.63–6.71)	0.04 (−0.02–0.10)	0.19
12–17.9 years	6.54 (6.46–6.61)	6.55 (6.50–6.61)	−0.02 (−0.11–0.07)	0.69
D. HbA1c (in %) at follow‐up a —the four pandemic‐related periods (depending on time of diabetes diagnosis)
January–February	6.67 (6.55–6.79)	6.68 (6.59–6.76)	−0.01 (−0.17–0.16)	>0.99
March–May	6.80 (6.70–6.90)	6.69 (6.61–6.76)	0.11 (−0.03–0.26)	0.35
June–September	6.79 (6.71–6.88)	6.82 (6.75–6.89)	−0.03 (−0.15–0.10)	>0.99
October–December	6.77 (6.66–6.87)	6.77 (6.69–6.85)	0.00 (−0.15–0.14)	>0.99
E. Daily insulin dose (in IU/kg) at follow‐up a —the whole year
All patients	0.70 (0.68–0.71)	0.62 (0.61–0.63)	0.08 (0.06–0.10)	<0.001
Females	0.73 (0.71–0.75)	0.64 (0.63–0.66)	0.09 (0.06–0.12)	<0.001
Males	0.67 (0.65–0.69)	0.60 (0.58–0.61)	0.07 (0.05–0.09)	<0.001
<6 years	0.67 (0.64–0.69)	0.59 (0.57–0.60)	0.08 (0.05–0.11)	<0.001
6–11 years	0.70 (0.67–0.72)	0.61 (0.59–0.63)	0.09 (0.06–0.12)	<0.001
12–17 years	0.72 (0.70–0.74)	0.65 (0.64–0.67)	0.07 (0.04–0.10)	<0.001
F. Daily insulin dose (in IU/kg) at follow‐up a —the four pandemic‐related periods (depending on time of diabetes diagnosis)
January–February	0.65 (0.60–0.69)	0.61 (0.57–0.64)	0.04 (−0.02–0.11)	0.47
March–May	0.64 (0.61–0.68)	0.60 (0.57–0.63)	0.05 (−0.01–0.10)	0.17
June–September	0.74 (0.70–0.77)	0.63 (0.61–0.66)	0.10 (0.06–0.15)	<0.001
October–December	0.73 (0.69–0.77)	0.63 (0.60–0.65)	0.10 (0.05–0.16)	<0.001

Note: Multivariable linear regression analysis, adjusted for age group (<6, 6–<12, and 12–<18 years), sex, and immigrant background (patient or at least one parent born outside Germany). Confidence intervals for estimated period‐specific values were adjusted according to the Bonferroni method, and corresponding p‐values according to the Holm method.

Mean (standard deviation) time after diabetes diagnosis was 4.7 (1.0) months in both cohorts.

The significant of bold values as results with the bold p‐values.

DISCUSSION

This study found an increase in HbA1c at diagnosis of type 1 diabetes after the first wave of the Covid‐19 pandemic in Germany, which may indicate a delay in seeking medical care due to the pandemic. This delay is probably the main reason for the increased frequency of diabetic ketoacidosis in children with new‐onset type 1 diabetes during the pandemic. , , ,

However, this did not affect short‐term treatment response, as the identical HbA1c at follow‐up excludes major limitations in the care of chronically ill children. This also corresponds to reports on the metabolic control of children with type 1 diabetes during the Covid‐19 lockdown, which did not noticeably worsen. ,

It has been demonstrated that the incidence of type 1 diabetes in children increased only a few months after the Covid‐19 waves. , With this background, it is important to note that we found an increased insulin requirement in children with onset of type 1 diabetes after the first wave of the Covid‐19 pandemic. This may indicate more rapid autoimmune destruction of beta cells during the pandemic. Importantly, the difference in the required daily weight‐adjusted insulin amount was not due to differences in BMI and consequent differences in insulin sensitivity.

Prolonged follow‐up of patients who developed new type 1 diabetes during the pandemic is needed to further analyze this phenomenon as the pandemic continues to progress and to capture longer‐term potential adverse metabolic effects of more rapid progression of type 1 diabetes.

CONFLICT OF INTEREST

The authors declare no potential conflict of interest.

AUTHOR CONTRIBUTIONS

Clemens Kamrath conceptualized the study, interpreted the analyses, wrote the initial manuscript, and revised the manuscript. Joachim Rosenbauer analyzed the data, designed and supervised the statistical analysis, and critically reviewed and revised the manuscript. Alexander J. Eckert analyzed the data and designed the analyses, contributed to the interpretation of results, and reviewed and revised the manuscript. Reinhard W. Holl conceptualized the study, coordinated and supervised data collection, acquired funding for the study, and critically reviewed the manuscript for important intellectual content. Ute Ohlenschläger, Carmen Sydlik, and Nicole Nellen‐Hellmuth collected data, contributed intellectually to the research topics of the DPV initiative, and critically reviewed the scientific content of the manuscript. All authors approved the final manuscript as submitted and agreed to be accountable for all aspects of the work.

ETHICS APPROVAL STATEMENT

The ethics committee of Ulm University approved the analysis of anonymized data from the DPV registry.

PEER REVIEW

The peer review history for this article is available at https://publons.com/publon/10.1111/pedi.13338.