THE IMPORTANCE OF HBA1C EVOLUTION IN COST-EFFECTIVENESS MODELING OF TYPE 2 DIABETES MELLITUS (T2DM)
Author(s)
Willis M1, He J2, Neslusan C2, Johansen P1, Worbes-Cerezo M31The Swedish Institute for Health Economics, Lund, Sweden, 2Janssen Global Services LLC, Raritan, NJ, USA, 3Janssen-Cilag SA, Madrid, Madrid, Spain
INTRODUCTION: In T2DM, HbA1c tends to drift up over time and the extent to which anti-hyperglycemic agents can maintain initial glucose lowering effect (or durability) varies. HbA1c evolution is an important determinant of future outcomes and costs. Currently there is no consensus on how to model upward drift in HbA1c or the durability of treatments. OBJECTIVES: Review different approaches to modeling HbA1c evolution and assess their impact on economic evaluations of T2DM interventions. METHODS: We reviewed the ways in which HbA1c evolution has been modeled. Lifetime simulations were performed that compared two hypothetical treatments: 1) initial HbA1c reduction of 1.25% and annual cost of $1,000 and 2) initial HbA1c reduction of 1% and annual cost of $200, using ECHO-T2DM, a validated micro-simulation model. . Treatment was intensified in both arms when HbA1c exceeded 7.0%, first by adding basal insulin and subsequently by adding 3x daily short-acting insulin. RESULTS: Four different approaches were identified: (1) no HbA1c evolution; (2) constant increase in HbA1c, irrespective of treatment; (3) constant treatment-specific increase in HbA1c; and (4) non-linear increase in HbA1c, irrespective of treatment. The simulations confirmed that these assumptions are critical. While the incremental life-years (LY’s) and Quality-Adjusted LYs (QALYs) were similar in the first 3 scenarios, the absolute values were highest for (1). Cost-savings and QALY gains were largest in (3), which allowed HbA1c to drift apart over time in each arm, and smallest in (2) (because treatment intensification reduced the HbA1c gap). The incremental cost-effectiveness ratio (ICER) ranged from $3,196 in (3) to $32,444 in (2). (4) could not be implemented in this version of ECHO-T2DM. CONCLUSIONS: Assumptions used to model HbA1c evolution have important consequences for estimates of cost-effectiveness, a 10-fold difference in the ICER in this hypothetical example, and should be addressed with sensitivity analysis in health economic evaluations.
Conference/Value in Health Info
2012-11, ISPOR Europe 2012, Berlin, Germany
Value in Health, Vol. 15, No. 7 (November 2012)
Code
PDB40
Topic
Economic Evaluation
Topic Subcategory
Cost-comparison, Effectiveness, Utility, Benefit Analysis
Disease
Diabetes/Endocrine/Metabolic Disorders