The Health Economics of Automated Insulin Delivery Systems and the Potential Use of Time in Range in Diabetes Modeling: A Narrative Review.

Intensive therapy with exogenous insulin is the treatment of choice for individuals living with type 1 diabetes (T1D) and some with type 2 diabetes, alongside regular glucose monitoring. The development of systems allowing (semi-)automated insulin delivery (AID), by connecting glucose sensors with insulin pumps and algorithms, has revolutionized insulin therapy. Indeed, AID systems have demonstrated a proven impact on overall glucose control, as indicated by effects on glycated hemoglobin (HbA1c), risk of severe hypoglycemia, and quality of life measures. An alternative endpoint for glucose control that has arisen from the use of sensor-based continuous glucose monitoring is the time in range (TIR) measure, which offers an indication of overall glucose control, while adding information on the quality of control with regard to blood glucose level stability. A review of literature on the health-economic value of AID systems was conducted, with a focus placed on the growing place of TIR as an endpoint in studies involving AID systems. Results showed that the majority of economic evaluations of AID systems focused on individuals with T1D and found AID systems to be cost-effective. Most studies incorporated HbA1c, rather than TIR, as a clinical endpoint to determine treatment effects on glucose control and subsequent quality-adjusted life year (QALY) gains. Likely reasons for the choice of HbA1c as the chosen endpoint is the use of this metric in most validated and established economic models, as well as the limited publicly available evidence on appropriate methodologies for TIR data incorporation within conventional economic evaluations. Future studies could include the novel TIR metric in health-economic evaluations as an additional measure of treatment effects and subsequent QALY gains, to facilitate a holistic representation of the impact of AID systems on glycemic control. This would provide decision makers with robust evidence to inform future recommendations for health care interventions.