Some people with diabetes need to take insulin and this has an impact on the National Health Service (NHS) in terms of its healthcare resource use (HCRU) and cost. There are different types of basal insulin: first-generation (Gen 1) and second-generation (Gen 2, insulin glargine 300U/ml [Toujeo] and insulin degludec [Tresiba]). The latter lasts longer, i.e. it needs to be administered only once daily rather than multiple times; and also reduces the risk of short-term complications of diabetes (e.g., hypoglycaemia, diabetic ketoacidosis, and hyperosmolar hyperglycaemic state). We want to test the hypothesis that people who take Gen 2 as part of their routine care treatment use less healthcare resources, with lower costs for the NHS, than people who take Gen 1 basal insulin. For this, we will compare the rates of diabetes-related short-term complications, and associated HCRU and costs between (1) patients switched from a Gen 1 to another Gen 1 basal insulin vs. patients switched from a Gen 1 to Toujeo; and (2) patients who switched from a Gen 1 basal insulin to Toujeo vs. patients who switched from a Gen 1 basal insulin to Tresiba.
It is of public health interest to understand burden on the NHS of different types of insulin treatments for people with diabetes in terms of HCRU and cost, and also to evaluate the effectiveness of these treatments in the real-world.
This retrospective cohort study will estimate the primary and secondary healthcare resource utilisation (HCRU) and associated costs in insulin-treated adult patients with type 1 or type 2 diabetes in England. Data from the Clinical Practice Research Datalink linked to Hospital Episode Statistics will be used to conduct analyses between two comparison groups: (1) patients who switched from a first-generation (Gen 1) basal insulin to an alternate Gen 1 basal insulin vs. patients who switched from a Gen 1 basal insulin to the second-generation (Gen 2) insulin glargine (Toujeo); and (2) patients who switched from a Gen 1 basal insulin to Toujeo vs. patients who switched from a Gen 1 basal insulin to the Gen 2 insulin degludec (Tresiba). In addition to HCRU and costs, the rates of diabetes-related short-term complications, including medically attended hypoglycaemia, diabetic ketoacidosis, and hyperosmolar hyperglycaemia, and HbA1c level will also be compared between the groups. To minimize selection bias, propensity-score-based inverse probability of treatment weighting will be conducted to balance baseline characteristics. Generalized linear models will be used to compare the outcomes in the weighted sample.
The study results will provide a better understanding of the impact of the different long-acting insulin analogues on clinical outcomes, and HCRU and associated costs in patients with diabetes following basal insulin initiation in England. The knowledge is of public health interest and will allow understanding of real-world value of insulin treatment in terms of costs and effectiveness for patients with diabetes.
HCRU Outcomes (All-cause, related to diabetes, and related to hypoglycaemia / diabetic ketoacidosis [DKA] / hyperosmolar hyperglycaemic state [HHS]):
Number of hospital admissions per patient-year during follow-up
Length of stay per patient-year during follow-up
Number of physician visits per patient-year during follow-up
Number of A&E visits per patient-year during follow-up
Number of primary care visits per patient-year during follow-up
Cost Outcomes (All-cause, related to diabetes, and related to hypoglycaemia / DKA / HHS):
Hospital admission costs per patient-year during follow-up
Physician visit costs per patient-year during follow-up
Accident and emergency (A&E) attendance costs per patient-year during follow-up
Costs associated with primary care prescription drugs per patient-year during follow-up
Total direct costs per patient-year during follow-up
Clinical Outcomes (Captured in primary care [Clinical Practice Research Datalink, CPRD] only, secondary care [Hospital Episode Statistics, HES] only, overall [CPRD or HES]):
Medically attended hypoglycaemic events
DKA events (as primary diagnosis, secondary diagnosis, any diagnosis position)
HHS events
HbA1c level
Craig Currie - Chief Investigator - Pharmatelligence Limited t/a Human Data Sciences
Nick Denholm - Corresponding Applicant - Harvey Walsh Ltd
Iskandar Idris - Collaborator - University of Nottingham
Karen Palmer - Collaborator - Sanofi Aventis UK Holdings Limited (UK)
Leah Fisher - Collaborator - Pharmatelligence Limited t/a Human Data Sciences
Manoharan Andiappan - Collaborator - Harvey Walsh Ltd
Neil Holden - Collaborator - Sanofi Aventis UK Holdings Limited (UK)
Onyinye Diribe - Collaborator - Sanofi Aventis UK Holdings Limited (UK)
Sue Beecroft - Collaborator - Harvey Walsh Ltd
Sushmitha Inguva - Collaborator - OPEN Health Group
Xiaocong Marston - Collaborator - Harvey Walsh Ltd
Karen Palmer - Collaborator - Sanofi Aventis UK Holdings Limited (UK)
Manoharan Andiappan - Collaborator - Harvey Walsh Ltd
Mark Evans - Collaborator - OPEN Health Group
HES Accident and Emergency;HES Admitted Patient Care;HES Outpatient;ONS Death Registration Data;Patient Level Index of Multiple Deprivation