Cost-effectiveness of insulin detemir compared to NPH insulin for type 1 and type 2 diabetes mellitus in the Canadian payer setting: modeling analysis.

OBJECTIVE

This study was conducted to quantify the long-term cost-effectiveness of insulin detemir (Levemir) versus intermediate-acting neutral protamine Hagedorn (NPH) insulin for the treatment of type 1 diabetes mellitus (T1DM) and type 2 diabetes mellitus (T2DM) in Canada, and to assess the sensitivity of results to dis-utilities for hypoglycemic events. dagger Levemir is a trade name of Novo Nordisk, Princeton, NJ, USA RESEARCH DESIGN AND METHODS: The web-based IMS-CORE diabetes model has a menu-driven interface programmed in hypertext markup language (HTML). It was used to project lifetime (60 years for T1DM and 35 years for T2DM) clinical and economic outcomes for patients on detemir vs. NPH. Cohort characteristics, utilities, and costs were derived from published literature. For T1DM, clinical trial data for HbA(1c) improvement (detemir -0.94% +/- 1.07; NPH -0.82% +/- 1.01) from baseline, and rates of hypoglycemic events (major events: 0.20 vs. 0.80 per patient-year for detemir vs. NPH, respectively) were modeled. For T2DM, observational study data for HbA(1c) improvement (detemir -0.18%) from baseline, and reductions in hypoglycemic events (major events: 0.0995 vs. 1.33 per patient-year for detemir vs. NPH, respectively) were modeled. Base-case hypoglycemia dis-utilities were -0.0118 for major and -0.0035 for minor events. Sensitivity analyses were conducted on discount rate and hypoglycemia dis-utility.

OUTCOME MEASURES

Outcomes included costs of treatment/management and costs (and incidence) of diabetes-related complications. Incremental cost-effectiveness ratios (ICERs) were calculated from differences in total costs and quality-adjusted life-years (QALYs).

RESULTS

Average total costs for T1DM were $CAN 83 622 +/- 4585 for detemir and $CAN 72 016 +/- 4593 for NPH. QALYs increased by 0.475 years with detemir, with an ICER of $CAN 24 389/QALY. Average direct costs for T2DM were $CAN 74 919 +/- 6391 (detemir) and $CAN 69 230 +/- 6840 (NPH). QALYs increased by 0.305 years. The ICER was $CAN 18 677. Although detemir was associated with slightly lower costs for most complications, results were driven by the differences in rates and costs for hypoglycemic events, and their assumed dis-utility. Study limitations include the use of single trials for clinical assumptions and the lack of analyses for patient risk sub-groups.

CONCLUSIONS

Findings provide evidence for the cost-effectiveness of detemir vs. NPH in treating T1 and T2DM in Canada, and support the key role of assumptions regarding the impact of hypoglycemic events. Additional work is needed to determine the extent to which results are robust for different sub-groups of patients and for variation in assumptions around HbA(1c) improvements and hypoglycemic event rates.