Using Real World Data to Estimate Complication Costs for a Diabetes Cost-Effectiveness Model: An Australian Public Healthcare System Perspective
Author(s)
Gabrielle Challis, BAppSc1, Asli Zeynep Ozdemir Saltik, MSc2, Liesl Strachan, BSc, PhD2, Marianne Huynh2.
1Medtronic, Macquarie Park, Australia, 2
1Medtronic, Macquarie Park, Australia, 2
OBJECTIVES: To conservatively estimate the cost of complications of diabetes to inform the cost-effectiveness of an automated insulin delivery (AID) system for managing type 1 diabetes mellitus (T1DM), from the Australian public healthcare perspective.
METHODS: The National Hospital Cost Data Collection (NHCDC), an annual collection of public hospital costs, was used to update the IQVIA Core Diabetes Model (CDM). The CDM is a validated model based on a series of inter-dependent sub-models that simulate the progression of diabetes and diabetes-related complications. Key outcomes include quality-adjusted life expectancy and incremental cost-effectiveness ratios (ICERs). Model inputs were determined for many complications including cardio- and peripheral vascular disease, haemodialysis and cataract, using the weighted average of medical-partition diagnosis related groups (DRGs) where possible. Non-admitted costs were used for peritoneal dialysis and eye laser treatment. Model inputs were deemed conservative due to: not inflating 2021-22 costs to 2024; using medical-partition only DRGs where possible; assuming one hospital admission per year; not including out-of-hospital costs. Published Australian cost data for heart failure and stroke was used. Hypoglycaemic medical episodes were costed from the Medicare Benefits Schedule (MBS) and Pharmaceutical Benefits Schedule (PBS). Published data provided weights that were used to discount first-year costs.
RESULTS: Using the NHCDC, MBS and PBS data, and published studies resulted in a conservative dataset of Australian costs for complications of diabetes that can be used as a source for diabetes cost-effectiveness model inputs. When incorporated into the CDM, outcomes replicated published studies, demonstrating that an AID system is cost-effective for managing T1DM (ICER below $50,000), even with conservative data sources.
CONCLUSIONS: Real world data in the form of annually updated NHCDC data can be readily accessed to replicate and update previously published cost analyses. Alternative Australian sources for costs, including published studies, MBS and PBS data, are also important inputs for data modelling.
METHODS: The National Hospital Cost Data Collection (NHCDC), an annual collection of public hospital costs, was used to update the IQVIA Core Diabetes Model (CDM). The CDM is a validated model based on a series of inter-dependent sub-models that simulate the progression of diabetes and diabetes-related complications. Key outcomes include quality-adjusted life expectancy and incremental cost-effectiveness ratios (ICERs). Model inputs were determined for many complications including cardio- and peripheral vascular disease, haemodialysis and cataract, using the weighted average of medical-partition diagnosis related groups (DRGs) where possible. Non-admitted costs were used for peritoneal dialysis and eye laser treatment. Model inputs were deemed conservative due to: not inflating 2021-22 costs to 2024; using medical-partition only DRGs where possible; assuming one hospital admission per year; not including out-of-hospital costs. Published Australian cost data for heart failure and stroke was used. Hypoglycaemic medical episodes were costed from the Medicare Benefits Schedule (MBS) and Pharmaceutical Benefits Schedule (PBS). Published data provided weights that were used to discount first-year costs.
RESULTS: Using the NHCDC, MBS and PBS data, and published studies resulted in a conservative dataset of Australian costs for complications of diabetes that can be used as a source for diabetes cost-effectiveness model inputs. When incorporated into the CDM, outcomes replicated published studies, demonstrating that an AID system is cost-effective for managing T1DM (ICER below $50,000), even with conservative data sources.
CONCLUSIONS: Real world data in the form of annually updated NHCDC data can be readily accessed to replicate and update previously published cost analyses. Alternative Australian sources for costs, including published studies, MBS and PBS data, are also important inputs for data modelling.
Conference/Value in Health Info
2025-09, ISPOR Real-World Evidence Summit 2025, Tokyo, Japan
Value in Health Regional, Volume 49S (September 2025)
Code
RWD309
Topic Subcategory
Reproducibility & Replicability
Disease
SDC: Diabetes/Endocrine/Metabolic Disorders (including obesity)