Comparing the Cohort and Micro-Simulation Modeling Approaches in Cost-Effectiveness Modeling of Type 2 Diabetes Mellitus: A Case Study of the IHE Diabetes Cohort Model and the Economics and Health Outcomes Model of T2DM
Willis M, Fridhammar A, Gundgaard J, Nilsson A, Johansen P
Economic modeling is widely used in estimating cost-effectiveness in type 2 diabetes mellitus. Because type 2 diabetes is complex and patients are heterogenous, the cohort modeling approach may generate biased estimates of cost-effectiveness. The IHE Diabetes Cohort Model (IHE-DCM) was constructed using the cohort approach as an alternative for stakeholders with limited resources, some of whom have voiced reasonable concerns about a lack of transparency with type 2 diabetes micro-simulation models and long run times.
The objective of this study was to inform decision makers by investigating the direction and magnitude of bias of IHE-DCM cost-effectiveness estimates that can be attributed to the cohort modeling approach.
Simulation scenarios inspired by the 9th Mount Hood Diabetes Challenge were simulated with IHE-DCM and with a micro-simulation model, the Economic and Health Outcomes Model of T2DM (ECHO-T2DM), and key metrics (absolute and incremental costs and quality-adjusted life-years, event rates, and cost-effectiveness) were compared for evidence of systematic differences. The models were harmonized to the extent possible to ensure that differences were driven primarily by the unit of observation and not by other model differences.
IHE-DCM run times were faster and IHE-DCM produced uniformly larger estimates of absolute life-years, quality-adjusted life-years, and costs than ECHO-T2DM but smaller between-arm (incremental) differences. Estimated incremental cost-effectiveness ratios and net monetary benefits varied similarly and predictably across the scenarios. On average, IHE-DCM estimates of incremental cost-effectiveness ratios and net monetary benefits were CAN$269 (3%) and CAN$2935 (10%) smaller, respectively, than ECHO-T2DM.
There was little evidence that estimated cost-effectiveness metrics, the outcomes that matter most to stakeholders, differed systematically.