Cost effectiveness of ibandronate for the prevention of fractures in inflammatory bowel disease-related osteoporosis: cost-utility analysis using a Markov model.

BACKGROUND Osteoporosis is a frequent complication in patients with inflammatory bowel disease. Recent studies have shown bisphosphonates to considerably reduce fracture risk in patients with osteoporosis, and preventing fractures with bisphosphonates has been reported to be cost effective in older populations. However, no studies of the cost effectiveness of these agents in preventing fractures in patients with inflammatory bowel disease are available. OBJECTIVE To investigate the cost effectiveness of the bisphosphonate ibandronate combined with calcium/colecalciferol ('ibandronate') in patients with osteopenia or osteoporosis due to inflammatory bowel disease in Germany. Treatment strategies used for comparison were sodium fluoride combined with calcium/colecalciferol ('fluoride') and calcium/colecalciferol ('calcium') alone. STUDY DESIGN AND METHODS A cost-utility analysis was conducted using data from a randomized controlled trial (RCT). Changes in bone mineral density (BMD) were adjusted and predicted for a standardized population receiving each respective treatment. A Markov model was developed, with probabilities of transition to fracture states consisting of BMD-dependent and -independent components. The BMD-dependent component was assessed using predicted change in BMD from the RCT. The independent component captured differences in bone quality and micro-architecture resulting from prevalent fractures or treatment with anti-resorptive drugs. The analysis was conducted for a population with a mean age of the RCT patients (women aged 36 years, men aged 38 years) with osteopenia (T-score about -2.0 at baseline), a population of the same age with osteoporosis (T-score of -3.0 at baseline) and for an older population (both sexes aged 65 years) with osteoporosis (T-score of -3.0). Outcomes were measured as costs per QALY gained from a societal perspective. The treatment duration in the RCT was 42 months. A 5-year period was assumed to follow, during which the treatment effects linearly declined to 0. The simulation time was 10 years. Prices for medication and treatment were presented as year 2004 values; costs and effects were discounted at 5%. To test the robustness of the results, univariate and probabilistic sensitivity analyses (Monte Carlo simulation) were conducted. RESULTS The calcium strategy dominated the fluoride strategy. When the ibandronate strategy was compared with the calcium strategy, the base-case cost-effectiveness ratios (costs per QALY gained) were between euro 407 375 for an older female population with osteoporosis and euro 6 516 345 for a younger female population with osteopenia. Univariate sensitivity analyses resulted in variations between 4% of base-case results and dominance of calcium. In Monte Carlo simulations, conducted for the various populations, the probability of an ICER of ibandronate below euro 50 000 per QALY was never greater than 20.2%. CONCLUSION The ibandronate strategy is unlikely to be considered cost effective by decision makers in men or women with characteristics of those in the target population of the RCT, or in older populations with osteoporosis.