|
Comparative cost-effectiveness of bazedoxifene and raloxifene in the treatment of postmenopausal osteoporosis in Europe, using the FRAX algorithm |
Kim K, Svedbom A, Luo X, Sutradhar S, Kanis JA |
|
|
Record Status This is a critical abstract of an economic evaluation that meets the criteria for inclusion on NHS EED. Each abstract contains a brief summary of the methods, the results and conclusions followed by a detailed critical assessment on the reliability of the study and the conclusions drawn. CRD summary This study evaluated the cost-effectiveness of raloxifene and bazedoxifene for postmenopausal women with osteoporosis, based on their fracture probability, in eight European countries. The authors concluded that bazedoxifene seemed cost-effective, compared with raloxifene, particularly for women with a high risk of fracture. The study appears to have been well conducted, but it was unclear how the sources were chosen, some methods were unclear, and valid comparators were omitted. The validity of the authors' conclusions is tenuous. Type of economic evaluation Study objective The aim was to evaluate the cost-effectiveness of raloxifene and bazedoxifene for postmenopausal women with osteoporosis, based on their fracture probability, in eight European countries. Interventions Two selective oestrogen-receptor modulators were compared: raloxifene and bazedoxifene. Placebo was included in the model as a comparator. Location/setting The UK, Belgium, France, Germany, Ireland, Italy, Spain, and Sweden/secondary care. Methods Analytical approach:The lifetime costs and benefits were estimated using a Markov tunnel model from a published study (see Other Publications of Related Interest). The model incorporated the annual risk of hip, vertebra, arm and other fractures, as well as breast cancer and venous thromboembolisms (VTE). The tunnel states allowed the risk of fracture to vary depending on the time since the occurrence of an event. The authors stated that a public health care perspective was adopted. Effectiveness data:The main effectiveness data were the risk of fracture and the risk of fracture conditional on the 10-year probability of fracture. These were from pivotal published studies. Women entered the model at 65 years old with a T-score of -2.5 standard deviations at the femoral neck, a body mass index of 26kg/m², and a prevalent fracture. Time to a fracture was modelled as a function of age, treatment, and baseline fracture probability, using a Cox-regression model that included an interaction term for treatment and baseline fracture risk. For most countries, the initial risk was determined by the FRAX algorithm, developed by one of the authors of this study. Hazard ratios were derived for each treatment, compared with placebo. For the analyses for Belgium and Ireland, the initial mortality and event rates were from the original model publication. Monetary benefit and utility valuations:EQ-5D utilities were obtained for each health state. For Sweden, these were age-dependent and based on the general population. For the UK, they were based on the general population, and these utilities were used for every other country. Measure of benefit:The measure of benefit was the quality-adjusted life-year (QALY). These were discounted at a rate of 3% per year, for every country except the UK (3.5%) and Ireland (4%), based on local guidelines. Cost data:The costs of fractures for each country, except Belgium and Ireland, were from the original model publication. The costs for Belgium and Ireland were from published literature. All patients were assumed to receive osteoporosis monitoring – an annual physician visit and biennial bone mineral density (BMD) testing. All costs were converted to 2008 Euros (EUR), using the health-related customer purchasing index. They were discounted at a rate of 3% per year for every country, except the UK (3.5%) and Ireland (4%), based on local guidelines. Analysis of uncertainty:Many sensitivity analyses were conducted, including varying the VTE and breast cancer effects, the time horizon, the persistence with treatment, the discount rate, and the starting year. Results In the main analysis, where the effectiveness of treatment was not varied by baseline fracture risk, bazedoxifene dominated raloxifene, as it was more effective and less costly. For net monetary benefit, a EUR 30,000 threshold was used for every country. The median net monetary benefit of bazedoxifene, compared with raloxifene, increased according to a function (monotonically) with the 10-year fracture probability, in all countries. In general, the median net monetary benefit increased above zero for women when their 10-year probability of a major osteoporotic fracture rose to between 0.05 and 0.10. Excluding VTE effects, and modelling women from the age of 55 years, changed the results significantly for Spain, but not for other countries. Modelling treatment for women from the age of 55 years made raloxifene dominant. Authors' conclusions The authors concluded that bazedoxifene appeared to be cost-effective, compared with raloxifene, for the treatment of postmenopausal women with osteoporosis, in Europe, particularly for women at a high risk of fracture. CRD commentary Interventions:The treatments were not described in detail. It was unclear how frequently the drugs were delivered, and for how long. The study focused on comparing bazedoxifene with raloxifene, so their cost-effectiveness compared with no treatment, or other relevant treatments, was not assessed. The authors mentioned several comparators that were not analysed: strontium ranelate, oral clodronate, and denosumab. Their exclusion and that of other comparators (alendronate, risedronate, and teriparatide) limits the use of this study for decision-makers. Effectiveness/benefits:Few details on the derivation of the effectiveness data were reported; the authors referred to the original publications. No systematic process for selecting evidence was reported. It was unclear how the sources of evidence were selected, which leaves potential for bias. The fit of the Cox-regression model to the data was not clear from this article. Relevant health outcomes appear to have been evaluated. The hazard ratio regression only included complete data; it was unclear how many data were missing, and their effect on the results. It was also unclear how hazard ratios from more than one source were combined. The utility scores were EQ-5D values from the general population, modified by a multiplier derived from published studies; the details of the measures and methods were not provided. Costs:Few of the costing details were reported; the authors referred to the original model publication. The main cost components appear to have been included. Analysis and results:The analysis and results were adequately reported. Many sensitivity analyses were conducted, but few results were presented. The authors gave a thorough analysis of the effect of fracture risk on cost-effectiveness. There were no probabilistic sensitivity analyses – as acknowledged by the authors, the impact of uncertainty was not fully explored, and the effects of uncertainty could be significant. The authors described several study limitations. These and others were: omitting the effects of most adverse events, pooling of effectiveness for 20mg and 40mg bazedoxifene, assuming 100% adherence, variable methods and quality of the sources, soon-to-expire patent on raloxifene (March 2014), and assumptions on the applicability of data from one country to another. Concluding remarks:This study appears to have been generally well conducted, but it was unclear how the sources were chosen, some methods were unclear for the model and analyses, and valid comparators were omitted. How useful the study results might be to a decision-maker is unclear, and the validity of the authors' conclusions is tenuous. Funding Funded by Pfizer, manufacturer of bazedoxifene. Bibliographic details Kim K, Svedbom A, Luo X, Sutradhar S, Kanis JA. Comparative cost-effectiveness of bazedoxifene and raloxifene in the treatment of postmenopausal osteoporosis in Europe, using the FRAX algorithm. Osteoporosis International 2014; 25(1): 325-337 Other publications of related interest Borgstrom F, Strom O, Kleman M, McCloskey E, Johansson H, Oden A, Kanis JA. Cost-effectiveness of bazedoxifene incorporating the FRAX algorithm in a European perspective. Osteoporosis International 2010; 22: 955-965. Indexing Status Subject indexing assigned by NLM MeSH Aged; Algorithms; Bone Density Conservation Agents /economics /therapeutic use; Cost-Benefit Analysis; Decision Support Techniques; Dose-Response Relationship, Drug; Europe /epidemiology; Female; Health Care Costs /statistics & Health Services Research /methods; Humans; Indoles /administration & Middle Aged; Osteoporosis, Postmenopausal /drug therapy /economics /epidemiology; Osteoporotic Fractures /economics /epidemiology /prevention & Quality of Life; Raloxifene Hydrochloride /economics /therapeutic use; Risk Assessment /methods; Sensitivity and Specificity; control; dosage /economics /therapeutic use; numerical data AccessionNumber 22013046055 Date bibliographic record published 31/10/2013 Date abstract record published 08/08/2014 |
|
|
|