Continuous glucose monitoring for self-management of diabetes in people living with type 2 diabetes mellitus on basal insulin therapy: A microsimulation model and cost-effectiveness analysis from a US perspective with relevance to Medicaid
Jerry Frank,Daniel Son,Kirk Szafranski,Yeesha Poon
DOI: https://doi.org/10.18553/jmcp.2024.24025
Abstract:Background: Reducing the risks of complications is a primary goal of diabetes management, with effective glycemic control a key factor. Glucose monitoring using continuous glucose monitoring (CGM) technology is an important part of diabetes self-management, helping patients reach and maintain targeted glucose and glycated hemoglobin (HbA1c) levels. Although clinical guidelines recommended CGM use, coverage by Medicaid is limited, likely because of cost concerns. Objective: To assess the cost-effectiveness of FreeStyle Libre CGM systems, compared with capillary-based self-monitoring of blood glucose (SMBG), in US individuals with type 2 diabetes mellitus using basal insulin. Methods: A patient-level microsimulation model was used to compare CGM with SMBG for a population of 10,000 patients. A 10-year horizon was used, with an annual discount rate of 3.0% for costs and utilities. Model population characteristics were based on US national epidemiology data. Patient outcomes were based on published clinical trials and real-world studies. Annual costs, reflective of 2023 values, included CGM and SMBG acquisition costs and the costs of treating diabetic ketoacidosis, severe hypoglycemia, and diabetes complications. The effect of CGM was modeled as a persistent 1.1% reduction in HbA1c relative to SMBG based on US real-world evidence. Disutilities were based on published clinical trials and other relevant literature. The primary outcome was cost per quality-adjusted life-year (QALY) gained. Sensitivity analyses were performed to test the validity of the model results when accounting for a plausible variation of inputs. Results: In the base case analysis, CGM was dominant to SMBG, providing more QALYs (6.18 vs 5.97) at a lower cost ($70,137 vs $71,809) over the 10-year time horizon. A $10,456 increase in glucose monitoring costs was offset by a $12,127 reduction in treatment costs. Cost savings reflected avoidance of acute diabetic events (savings owing to reductions in severe hypoglycemia and diabetic ketoacidosis were $271 and $2,159, respectively) and a reduced cumulative incidence of diabetes complications, particularly renal failure (saving $5,292), myocardial infarction (saving $1,996), and congestive heart failure (saving $1,061). Scenario analyses were consistent with the base case results, and the incremental cost-effectiveness ratio for CGM vs SMBG ranged from dominant to cost-effective. In probabilistic analysis, CGM was 100% likely to be cost-effective at a willingness-to-pay threshold of $50,000/QALY. Conclusions: CGM is cost-effective compared with SMBG for US patients with type 2 diabetes mellitus receiving basal insulin therapy. This suggests that state Medicaid programs could benefit from broader coverage of CGM.