Key points from the evidence

Trials comparing closed loop systems with usual care generally have small sample sizes and evaluate interventions over short time periods with a study population of people with well controlled type 1 diabetes who have lived with the condition for many years. The closed loop systems used in trials have often been superseded by more advanced versions. Meta-analyses show high levels of statistical heterogeneity based on these trials.
A network meta-analysis of 14 trials (n=1,043) in adults with type 1 diabetes found that the mean percentage time in normal glycaemic range was significantly greater with closed loop systems compared with other diabetes technologies, including continuous glucose monitoring plus insulin pump therapy. A meta-analysis of 12 trials (n=344) that compared closed loop systems with sensor-augmented pump therapy reached similar conclusions.
A meta-analysis of 19 studies (n=364) compared closed loop systems with continuous subcutaneous insulin infusion in adolescents and children with type 1 diabetes. There was a statistically significant difference in mean percentage time in normal glycaemic range that favoured closed loop systems: 11.97%, 95% confidence interval (CI) 5.54% to 18.40%, p=0.0003.
A meta-analysis of 41 trials (n=1,042) comparing closed loop systems with continuous subcutaneous insulin infusion or sensor-augmented pump therapy, in people of any age with type 1 diabetes, found a statistically significant improvement in weighted mean percentage time in normal glycaemic range in the closed loop group compared with the control group: 9.62%, 95% CI 7.54% to 11.70%, p<0.001. A subgroup analysis was consistent with the overall meta-analysis for comparisons of an artificial pancreas (a dual hormone closed loop system) with continuous subcutaneous insulin infusion or sensor augmented pump therapy.
The meta-analyses described in key points 3 and 4 found corresponding statistically significant reductions in the mean percentage time spent in hypoglycaemia and hyperglycaemia with closed loop systems.
The results of 13 randomised controlled trials (RCTs), published after the most recent meta-analysis, are consistent with the findings reported in the secondary literature. Two of the trials tested closed loop systems in people with a moderate-to-high risk of hypoglycaemia or suboptimal glycaemic control.
Clinical safety outcomes, such as severe hypoglycaemia or diabetic ketoacidosis, were rarely reported in the secondary literature. As a result there is uncertainty around the frequency of these outcomes with closed loop systems compared with other diabetes management options.
Device-associated safety concerns related to either technical difficulties affecting components within the closed loop system or human factors. The main safety issues reported with closed loop systems were about loss of connectivity between component devices often owing to the devices being too far apart.
In the published literature on patient experiences and views of closed loop systems, people with type 1 diabetes described how closed loop systems improved their glycaemic control, gave them increased flexibility around eating and exercise, and provided ‘time off’ from managing their diabetes. People also described a burden of treatment associated with this technology such as the need to respond to frequent alarms, replace sensors and deal with technical problems. Some people expressed concerns about the trustworthiness of closed loop systems or found the systems challenging to use when exercising.
An audit of services in England and Wales for children and young people with type 1 diabetes found that continuous glucose monitors and insulin pumps were significantly more likely to be used by children and young people with diabetes who live in the most affluent areas and are of white ethnicity.
The patient organisation Insulin Pump Awareness Group (iPAG) Scotland identified inequalities in access to closed loop systems in Scotland created by the current requirement for people to self-fund the use of these systems. This is a cost that many from lower income areas cannot afford.
Three patient organisations (iPAG Scotland, Juvenile Diabetes Research Foundation (JDRF), and Diabetes Scotland) outlined the substantial impact the condition has on those living with type 1 diabetes and all strongly supported access to closed loop systems in Scotland.
- Managing type 1 diabetes is a daily burden that has a major impact on the daily lives of people with diabetes, their families and carers.
- The adverse effects of managing type 1 diabetes are both physical and mental and include diabetes-related distress and reduced quality of life.
- There are currently financial barriers to accessing closed loop systems because of the requirement to self-fund and barriers to education about diabetes and the role of closed loop systems.
- Equal access to closed loop systems across NHSScotland is highly desirable for people with type 1 diabetes.
Two studies reporting interviews with healthcare professionals in NHS England identified the importance of defining priority groups and ensuring consistency of access to closed loop systems in clinical practice.
SHTG adapted an economic model comparing closed loop systems with continuous glucose monitoring plus multiple daily injections, flash glucose monitoring plus multiple daily injections, continuous subcutaneous insulin infusion pumps plus continuous glucose monitoring, and finger prick testing plus multiple daily injections. Based on the available clinical evidence in people with well controlled type 1 diabetes:
- closed loop systems were associated with the highest costs and quality adjusted life years in a Scottish adult population with type 1 diabetes (except in the comparison with continuous glucose monitoring plus continuous subcutaneous insulin infusion, where associated closed loop system costs were lower)
- base case results showed that closed loop systems are cost-effective compared with continuous subcutaneous insulin infusion plus continuous glucose monitoring (non-integrated)
- base case results showed that closed loop systems are unlikely to be cost-effective compared with flash or continuous glucose monitoring plus multiple daily injections. The results are sensitive to baseline HbA1c, technology costs and effects on hypoglycaemia. The model does not capture day-to-day quality of life improvements associated with the use of closed loop systems, and
- the costs associated with closed loop systems (device and consumables cost) should be considered in the context of the reduction in the costs of managing long-term complications of type 1 diabetes.

Key points from the evidence

Editorial Information