A decade of improved glycemic control in young children with type 1 diabetes: A population-based cohort study

Frida SundbergJonatan NåtmanStefan FranzenKarin ÅkessonStefan Särnblad

First published: 10 April 2021


Read the article and abstract pdf free




Early-onset type 1 diabetes (T1D) is associated with high risk of early cardiovascular complications and premature death. The strongest modifiable risk factor is HbA1c. Other modifiable factors, such as overweight, also increase the risk of complications. During the last decade, the introduction of continuous glucose monitoring (CGM) has offered new options in the treatment of T1D.


To compare treatment outcomes in children younger than 7 years with T1D in Sweden in two separate cohorts: one in 2008 and one in 2018.


All children in the national pediatric diabetes registry (SWEDIABKIDS) younger than 7 years with T1D were included. Data from 2008 and 2018 were analyzed.


Data were available on 666 children (45% girls) in 2008 and 779 children (45% girls) in 2018. Mean age was 5.6 (1.4) versus 5.5 (1.4) years and mean diabetes duration 2.3 (1.4) versus 2.2 (1.4) years. The use of CGM increased from 0% to 98% and the use of an insulin pump from 40% in 2008 to 82% (p < 0.01)in 2018.Mean HbA1c was 58 mmol/mol (7.4%) in 2008 and 50 mmol/mol (6.7%) in 2018 (p < 0.01). The frequency of overweight and obesity was the same in 2008 and 2018(26% vs. 29%).


During this decade, usage of CGM and insulin pump increased and HbA1c decreased. However, HbA1c remained higher than the physiological level and thus continued to represent a cardiovascular risk, especially in combination with overweight or obesity. The frequency of overweight and obesity remained unchanged.

From the article


Early-onset type 1 diabetes (T1D) is associated with a high risk of early cardiovascular complications and premature death.1 The strongest modifiable risk factor associated with diabetes-related mortality and morbidity due to microvascularand macrovascular complications is HbA1c.2-4 HbA1c target setting has been shown to affect glycemic outcome in insulin treatment.56

The treatment of T1D has developed considerably over the last decade. The use of continuous glucose monitoring (CGM) has become an integral part of routine care of young children with T1D in many parts of the world. During the same period, the HbA1c target has internationally been lowered to 52 mmol/mol (7.0%) by International Society for Pediatric and Adolescent Diabetes (ISPAD)78 and American Diabetes Association (ADA)9 then to 48 mmol/mol (6.5%) by the National Institute for Health and Care Excellence (NICE) in the UK,10and it was also set to 48 mmol/mol (6.5%) in Sweden in 2017. In 2008 target HbA1c in Sweden was 57 mmol/mol (7.5%). Other factors, such as overweight and a centralized adiposity, also contribute to increased risk of cardiovascular disease via abnormal blood lipid profile and hypertension.11 Thus, lowering HbA1c and avoiding overweight and obesity are both important to reduce the risk of diabetes complications and to delay their onset.

Previously, there was a clear association between a lower HbA1c and a higher frequency of severe hypoglycemia12 but this association has now disappeared1314 and severe hypoglycemia, defined as hypoglycemia with unconsciousness with or without seizures has become less frequent in recent decades.15 Nevertheless, fear of hypoglycemia is still a common barrier for optimal glycemic control and an active lifestyle.16

Treatment of T1D in preschool children is especially challenging and the cognitive, motor, and social immaturity of preschool children, as well as their small body size and their growth pattern, need to be considered. These young children are dependent on others for decision-making and implementation of all aspects of insulin treatment but also for establishing health-promoting lifestyle habits such as food choices and physical activity. Early childhood is a window of opportunity for the establishment of habits that have a high probability of following the child throughout itslife.17 In particular, concerns regarding high glucose variability and a high frequency of undetected hypoglycemia episodes have been expressed.1819 Our hypothesis is that treatment outcomes have improved over the past decade in preschool children in parallel with the increased usage of CGM and insulin pumps.

The aim of this study was to compare treatment outcomes in children younger than 7 years with T1D in Sweden in two separate cohorts: one in 2008 and one in 2018. The first representing a one-year cohort before the introduction of CGM and a lower HbA1c target, and the second a more recent one-year cohort after these changes were established.


The use of CGM in the everyday treatment of children younger than 7 years with T1D in Sweden increased from none in 2008 (when CGM was unavailable) to almost total (98% of patients) in 2018. During the same decade, mean HbA1c decreased from 58 mmol/mol (7.4%) to 50 mmol/mol (6.7%), and the proportion of children reaching the current ISPAD target of 52 mmol/mol (7%) increased from 22% to 61%. In contrast to these improvements, the frequency of overweight and obesity remained higher than in the background population.

While age at diagnosis is currently a risk factor that is impossible to influence, it is well documented that HbA1c is the strongest modifiable risk factor for microvascular and macrovascular complications.34 By lowering HbA1c in this patient group, their prognosis has improved significantly.

There have been no major changes in social support to families of children with diabetes between 2008 and 2018. All necessary equipment for insulin treatment, such as insulin, pumps, infusion sets, CGM, and glucometers with strips are available free on prescription for all children with diabetes living in Sweden. In 2009, new legislation22 secured the right for children with diabetes to get treatment support in school, but the children in the present study were all of preschool age and thus did not benefit from this legislation. Therefore, the explanation for the improved glycemic control we found in young children with T1D is most likely to be attributable to the technical advances in treatment and improvements in the healthcare approach.

Our results contrast with the report by Foster et al.,23 which showed an increasing HbA1c in children despite the introduction of new treatment technologies and the sense of urgency for lowering HbA1c based on the findings from DCCT/EDIC.12 This suggests that not only access to technology but also access to guidelines, targets, and experience might affect the outcome of treatment. An Australian single-center study24 showed results well in line with ours, indicating that improved glycemic control can be achieved in other social contexts than the Swedish welfare system. During the period 2008 to 2018, both international and national recommendations8-10 have lowered the HbA1c target in this age group. In Sweden, HbA1c in all pediatric age groups was lowered to 48 mmol/mol (6.5%) in 2017.

Few studies are currently available for comparison of time in target for young children with T1D. In a recent multicenter study from USA (published in 2020) on 143 children aged 2–8 years with T1D, diMeglio et al. report that children with MDI or pump treatment, with access to SMBG but without CGM, spend 40% of the time within a broader range, 3.9–10 mmol/L (70–180 mg/dl), and 4.1% below target. The DiMeglio cohort had a mean HbA1c of 66 mmol/mol (8.2%),19 representing a significantly higher risk of complications than our 2008 cohort, which at that time had access to treatment technology comparable with the DiMeglio cohort.

It has been feared that lowering mean HbA1c could lead to an unacceptably high frequency of severe hypoglycemia. On the contrary, we observed a low frequency of this kind of episode. The finding that HbA1c can be lowered without an increased rate of severe hypoglycemia,14 especially when CGM is used,25 has been previously described.

Monitoring of time spent below target (< 4.0 mmol/L, < 70 mg/dl) was available for the 2018 cohort only. The Swedish target set in 2017 was that <10% of time should be below target. In later recommendations,26 this target has been adjusted to <4%, which equals less than 1 h per day. In an observational CGM study on healthy children aged 2–8 years, time below target was 9%.27 In older groups (aged 8–65 years) of healthy individuals, time below target was 1.7%28 and 1.1%.29 Whether this observed difference in time below target in the youngest age group is constant and clinically significant remains to be further investigated.

In this study we found that one-third of the children were overweight or obese. This is in line with a recent multicenter study from the SWEET registry.30 This high proportion of overweight and obesity in children with T1D warrants attention. First, it seems that young children with T1D are more susceptible to developing overweight compared to healthy children. Data on the prevalence of overweight and obesity in the general population in Sweden is limited, but in a 2014 study from Jönköping county in Sweden, the prevalence of overweight and obesity was found to be 14.2% in girls and 10.2% in boys.31 Furthermore, national data from 96,316 children measured during 2012 and 2013 at the age of 4 years, found a prevalence of 11.7% of overweight and obesity (personal communication Charlotte Nylander). Second, previous publications on healthy children have found that the overweight tends to track from the age of 4 years to 10 years32 and even into adolescence.33To our knowledge, there are no studies that have examined tracking of BMI among young children with T1D, but we know that overweight is common in older children with T1D, and it is most probable that the increased weight gain begins in earlier childhood. Third, subjects with T1D have an increased risk of cardiovascular events; BMI can be regarded as a marker of increased insulin resistance and is also a separate cardiovascular risk factor. It is not possible from this study to determine the cause of the high proportion of overweight. However, it has previously been shown that Swedish children younger than 7 years with T1D are less physically active34 but have the same caloric intake as healthy children of the same age.35

The 2008 cohort of children had access to MDI, CSII, and SMBG. From 2015, CSII with addition of CGM (sensor augmented pump, SAP) was recommended as the treatment of choice from diagnosis in children younger than 5 years. The 2018 cohort had access to SAP but not to more advanced automated insulin delivery.

Upcoming automated insulin delivery systems need to prove better treatment results than current open-loop SAP systems, and therefore our report can serve for comparison in young children with T1D.

A strength of our study is the use of two unselected entire national cohorts. Since this is a cross-sectional study, no conclusions can be drawn on causality. We can only observe different treatment outcomes before and after the introduction of new technology and guidelines, while most other factors (such as health care system resources and social welfare system) were unaltered. Given the design of the study (all children in the study were younger than 7 years, and 10 years elapsed between the observations), it is obvious that no child could have contributed data at both observations.

In a situation where 98% of the studied children used CGM, it is no longer ethically possible to do a randomized study to explore the effect of introducing CGM. Nevertheless, our study contributes a description of treatment outcomes in a situation with a population of families of children with diabetes with experience of using CGM in collaboration with diabetes teams.

A weakness in the study is that international comparisons need to take into consideration that SWEDIABKIDS collects data on time in target using the range 4–8 mmol/L (72–144 mg/dl). The reason for this is historical: the national Swedish consensus for data collection was established before any international consensus on reporting time in target. In practical patient education much focus is on strategies to maintain glucose levels within this target rather than the broader metric Time In Range (4–10 mmol/L, 70–180 mg/dl). One can speculate if the use of this narrower target and focus on strict normoglycaemia might contribute to the relatively low mean HbA1c in Swedish children with T1D, but it is not possible to draw any conclusions on causality from a cross sectional observation study.

To conclude, we show that young children with T1D in Sweden had lower HbA1c after the introduction of, and experience with, new treatment tools such as CGM and SAP. However, the high frequency of overweight and obesity found in both cohorts warrants attention, since this contributes to a high cardiovascular risk.

Read the article pdf free



www red DiabetologNytt