Postmenopausal Osteoporotic Women at High Risk of Fracture: DISCUSSION
Health care budgets are limited and, therefore, new therapies are increasingly required to demonstrate both clinical and economic benefits. In the case of PMO, demographic trends indicate that in the coming years a larger proportion of the population will be at risk of PMO because of advancing age, thereby placing an increased burden on MCOs to provide PMO care.
To assess the cost-effectiveness of PMO therapy, we conducted an economic analysis of two leading bisphosphonates from a MCO perspective over a three-year period. A comprehensive model of PMO was used to estimate the cost-effectiveness of these therapies in terms of cost per QALY gained, cost per hip fracture averted, and the total budget impact over a three-year period.
The results obtained indicate that risedronate is more effective at fracture prevention, and less costly. As a consequence, budget savings are possible for MCO plans when rise-dronate is prescribed instead of alendronate. Compared to no treatment, risedronate therapy provides an attractive incremental cost per gain in health. The latter comparison is provided primarily for illustrative purposes, because not treating patients is not an acceptable clinical option.
The model is subject to limitations common to all decision-analytic models, in that it combines data from numerous sources, and it requires structural and data assumptions and can be subject to certain biases. The first two limitations cannot be avoided because the primary motivation for creating any decision-analytic model is to assess comparative strategies in the absence of complete data. Whenever possible, assumptions were made that reduced any potential bias toward either intervention.
The efficacy data were obtained from separate clinical trials, which included some differences in study design and patient populations. The most important differences included a lower femoral neck T-score in the risedronate group (-3.7 [or -2.8 when NHANES correction was applied] versus -2.5 in the alendronate study) and a greater number of prevalent vertebral fractures at baseline in the risedronate study. On the basis of these differences, it appears that the risedronate populations might have been at greater risk of fracture than the alendronate population, but it is unclear whether this would increase or decrease the potential impact of therapy. Adjustments to the reported efficacy data to correct for differences in the trial populations were not conducted because of the uncertainty regarding the impact of such differences on efficacy. canadian antibiotics
The estimates of RR of fracture in patients of various ages and T-scores are based on the best available data and produce lifetime risks of hip fracture within a reasonable range. However, the epidemio-logical data regarding fracture risk is still developing and some uncertainty thus exists in these parameters. The cost-effectiveness results were sensitive to RR when lowered below RR 5, but were relatively stable in the range of 5 to 7. Given the requirement for both low BMD and previous vertebral fracture in the target population, it is likely that the risk of fracture would be considerably higher than the risk in the general population, which includes patients without osteoporosis.
Because of a lack of comprehensive data on efficacy, cost, and fracture incidence data, the analysis did not include othe fracture sites in the skeleton. Their exclusion results in a bias against both interventions compared to no therapy, as averting these fractures would provide additional cost and quality-of-life benefits.
The impact of therapy side effects on cost and quality of life were not included in the analysis. Neither the risedronate nor the alen-dronate trials reported a significant difference in adverse events compared to placebo, but the drug alendronate trial excluded patients with peptic-ulcer disease or dyspepsia who required daily treatment. Some observational data indicate that alendronate might cause a greater incidence of gastrointestinal side effects, compared to os-teopenic and osteoporotic controls, but the majority of these events are assumed to be minor with little cost or quality-of-life implication. It is therefore assumed that the inclusion of side effects would not alter the study conclusions. Similarly, the disutility associated with taking the medications was not included but would probably have little impact on the results.
The base case analysis assumed no discontinuation of therapy because data are insufficient regarding continuation of osteoporosis therapies to inform the model. However, sensitivity analyses were conducted to simulate the potential impact of therapy discontinuation and revealed higher cost-effectiveness ratios owing to a decrease in clinical benefits observed over the three-year period. Although therapy discontinuation reduced pharmacy costs, it resulted in higher overall treatment costs because the possible foregone cost reductions from averted fractures outweighed the savings in drug acquisition costs. However, there are no data to suggest that continuation would be higher with either risedronate or alendronate.
The study utilized a three-year period to reflect interest in short-term budget implications among payers. To fully assess the impact of an intervention, however, cost-effectiveness analyses should utilize a period of time sufficient to capture all relevant subsequent effects and their associated costs. Prevention of osteoporotic fractures avoids not only the cost and quality-of-life effects of the acute event but also all subsequent effects related to permanent disability resulting from the fracture. Because the negative impact of fractures can continue for a patient’s lifetime, the disparity between treated and untreated patients increases each year after the initial event. The sensitivity analysis in which patients were treated for three years but were observed for their lifetimes indicated that inclusion of these eventual effects can almost completely offset the acquisition cost of therapy, resulting in near cost-neutrality over the cohort’s lifetime.
When selecting a bisphosphonate for the treatment of high-risk PMO patients, MCOs may realize cost savings and better health outcomes with the use of risedronate over alendronate (Fosamax canadian). When treatment effects on chronic disability are considered over the patient’s lifetime, three years, compared with no therapy, may produce cost-neutral results.
This study was funded by the Alliance for Better Bone Health, Procter & Gamble Pharmaceuticals, and Aventis Pharmaceuticals. The authors retained control over all data selection, methodology, and interpretation decisions.