A COMPARISON OF DIFFERENT METHODS OF ESTIMATING FRACTURE RISK AND FRACTURE RISK REDUCTION IN COST-EFFECTIVENESS ANALYSES OF THE OSTEOPOROSIS TREATMENT BAZEDOXIFENE

OBJECTIVES To compare the traditionally used approach for fracture risk assessment in cost-effectiveness analysis (CEA) compared to the use of FRAX® models based on multiple individual clinical risk factors (CRFs) using the osteoporosis treatment bazedoxifene. METHODS In CEA of osteoporosis the fracture risk has traditionally been calculated with risk-adjustments based on age, bone mineral density (BMD) and prior fracture. The treatment effect has been derived from clinical trials and the same efficacy has been assumed irrespective of the fracture risk of the population. A novel approach to fracture risk assessment considers the contribution of 8 individual CRFs on fracture risk and mortality using the FRAX®-tool. The application of FRAX® to clinical trial populations has shown that treatment efficacy increases with higher fracture risk. The cost-effectiveness was estimated in a Markov cohort model with US data using a health care perspective. The CEA compared the osteoporosis treatment bazedoxifene (BZA) to no treatment in women with a T-score for femoral neck BMD of -2.5SD and previous fracture. Using the old approach BZA was set to reduce the risk of vertebral fracture by 42% based on the overall analysis of the phase III study. Using the FRAX®-approach the vertebral fracture risk reduction varied depending on pre-hoc fracture probability in the target population and was estimated at 59%. The relative risk of a vertebral fracture compared to the general population was 3.96 and 1.82 with the old and FRAX® -approach, respectively. RESULTS The cost per QALY gained with bazedoxifene treatment was estimated at $54,712 using the old approach and $33,650 using the FRAX®-approach. This is due to differences in the assessment of fracture risk efficacy, fracture risk and mortality. CONCLUSIONS The advent of more accurate assessment of fracture risk assessment and its use as a determinant of efficacy has important consequences for CEA.