|
At what hip fracture risk is it cost-effective to treat? International intervention thresholds for the treatment of osteoporosis |
Borgstrom F, Johnell O, Kanis J A, Jonsson B, Rehnberg C |
|
|
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. Health technology The study examined a generic 5-year treatment for osteoporotic fractures with an assumed relative risk reduction. This was compared with no treatment.
Study population The study population comprised postmenopausal women requiring treatment for osteoporosis.
Setting The setting was secondary care. The economic study was carried out in seven countries, specifically, Australia, Germany, Japan, Sweden, Spain, the UK and the USA.
Dates to which data relate The clinical data came from studies published between 1994 and 2005. No dates for resource use were clearly reported, although much of the data were derived from Swedish sources. The price year was 2004.
Source of effectiveness data The clinical data used in the analysis were:
the rates of osteoporotic fractures in different sites (vertebrae, wrist, ribs, pelvis, humerus, clavicle, scapula, sternum, other femoral fractures, tibia and fibula) and for different age groups;
the effectiveness of the intervention (relative risk reduction);
the offset time of treatment effect;
the age-differentiated hip fracture risk for women;
the age-differentiated mortality for the general female population; and
the age-differentiated excess mortality after hip fracture.
Modelling A Markov model with four health states (well, fracture, post-fracture, and death) and a lifetime horizon was constructed to estimate the 10-year probability of hip fracture at which the hypothetical intervention was cost-effective. The transitions across health states were described. Two main analyses were undertaken depending on the site of fracture. One analysis included all osteoporotic fractures, which were estimated using estimates of hip fracture equivalent. The other analysis considered hip fractures only. Annual cycles were considered.
Sources searched to identify primary studies The authors made an assumption about treatment effectiveness and the offset time of treatment effect. Life expectancy in the general female population came from country-specific published life tables. The age-differentiated hip fracture risk was taken from country-specific epidemiological studies. Excess mortality due to hip fracture was based on a Swedish study and applied to all countries. There was limited information on the sources of other estimates.
Methods used to judge relevance and validity, and for extracting data No systematic search for data was reported, therefore the primary studies might have been identified selectively. The authors tried to use country-specific sources whenever possible, while Swedish studies were used in other circumstances because of their high quality.
Measure of benefits used in the economic analysis The summary benefit measure used in the decision analysis was the expected number of quality-adjusted life-years (QALYs) associated with treatment versus no treatment. The utility values were derived from a Swedish study and other published studies (details not given), and were estimated as quality of life decrements due to hip fractures. An annual discount rate of 3% was applied.
Direct costs The analysis of the costs was carried out from a societal perspective. It included the costs of intervention, hip fractures (both short- and long-term) and increased survival. The intervention costs covered drugs and monitoring. The fracture costs were adjusted using a hip fracture cost equivalent index (HFCQ). The unit costs and resource quantities were not presented separately. All costs of fractures and drugs were derived from country-specific sources in the base-case analysis. The costs of increased survival came from a Swedish source. Discounting was relevant, as the long-term costs were evaluated, and an annual rate of 3% was used for all countries in the base-case analysis. The price year was 2004. All previous costs were inflated, when necessary, to 2004 values using country-specific consumer price indices.
Statistical analysis of costs The costs were treated deterministically.
Indirect Costs Productivity costs do not appear to have been considered.
Currency US dollars ($). Country-specific costs were converted into US dollars using annual exchange rates among currencies and purchasing power parities.
Sensitivity analysis Extensive deterministic sensitivity analyses, sub-group analyses and scenario analyses were carried out to determine the impact of individual model inputs on the cost-effectiveness ratios and ITs. For example, ITs were estimated with and without the costs of increased survival, in different age groups, for all osteoporotic fractures and for hip fractures only, with different discount rates (national rates rather than the common 3% rate), direct Swedish costs in place of country-specific costs, different values of willingness-to-pay (WTP) for a gained QALY, and changes in assumptions about mortality, fracture risk and treatment effectiveness. Alternative values were either retrieved from the literature or were based on authors' opinions.
Estimated benefits used in the economic analysis The expected QALYs were not reported.
Cost results The total costs were not reported.
Synthesis of costs and benefits The costs and benefits were combined by calculating the incremental cost per QALY gained with the hypothetical intervention in comparison with no intervention. However, most of the results for cost-utility ratios were presented graphically and cannot be reported. In general, the cost-effectiveness of the intervention improved with increasing starting age of treatment.
The differences in cost-utility ratios when accounting for all osteoporotic fractures and when looking at hip fracture alone diminished with increasing starting age. The exclusion of the costs of increased survival did not alter the main conclusions of the analysis. Overall, the cost-utility ratios were similar between countries, except for Australia and Spain which had markedly higher cost-utility ratios than the other countries.
ITs were determined with a value of WTP for a QALY gained which was set equal to the gross domestic product (GDP) per capita in each country multiplied by two (ranging from about $47,000 for Spain to about $75,000 for the USA). In general, the ITs (i.e. 10-year probability of hip fracture at which the intervention becomes cost-effective) rose for all countries with increasing starting age of treatment. The inclusion of costs in added life-years gave slightly higher ITs, owing to the pattern of higher consumption than production among the elderly.
ITs were low for the UK and Japan and high in Spain. For example in the UK, when accounting for all fractures and including future costs, the 10-year probability of hip fracture at which the intervention becomes cost-effective was 1.02 at age 50 years, 2.03 at age 55 years, 3.18 at age 60 years, 4.35 at age 65 years, 5.70 at age 70 years, 7.43 at age 75 years, 8.44 at age 80 years, 7.46 at age 85 years, and 6.48 at age 90 years. The respective values for Spain were 3.05, 5.32, 8.73, 10.83, 14.66, 18.04, 18.91, 17.49 and 15.79.
The sensitivity analysis showed the following results. When the costs and benefits were not discounted, ITs decreased by about 30% in all countries. The use of country-specific discount rates enhanced the differences in ITs among countries. The use of Swedish costs for fractures did not alter the conclusions of the analysis, although ITs were quite sensitive to variations in intervention costs. When the same WTP value was used for all countries ($60,000), differences among countries were reduced. The use of an alternative estimate of WRP (GDP multiplied by three) decreased the threshold by approximately 35%. Alterations in population mortality and fracture risk did not affect the results of the analysis. Treatment effectiveness had a strong impact on the cost-utility ratios, and variations from 25% relative risk reduction to 50% relative risk reduction (this was 35% in the base-case) led to 41% increments and decrements in cost-effectiveness ratios.
Authors' conclusions The intervention thresholds (ITs) presented in the paper should be used in treatment guidelines in order to improve the selection of patients for whom treatment is cost-effective. The analysis suggested that the most important variables for explaining differences among countries were fracture-related costs, intervention costs and effectiveness, and the willingness-to-pay (WTP) for a quality-adjusted life-year (QALY) gained.
CRD COMMENTARY - Selection of comparators The authors considered an intervention that was not defined, thus the analysis referred to any treatment for postmenopausal women with osteoporosis. This generic intervention was compared with no treatment, which represents a valid alternative given that the objective of the study was not to select the most cost-effective treatment but to decide whether to treat or not to treat. You should decide whether these are valid comparators in your own setting. Validity of estimate of measure of effectiveness The clinical data were obtained from studies that might have been identified selectively in order to choose sources representative of each country. The methods and conduct of a systematic review of the literature were not reported. Key clinical inputs were tested in the sensitivity analysis. When country-specific data were not available, Swedish estimates were used because of their high quality. Validity of estimate of measure of benefit The benefits (QALYs) were modelled using an appropriate model. Some information on the utility weights used to adjust survival was provided. However, the impact of variations in utility weights on the ITs was not tested in the sensitivity analysis. Validity of estimate of costs The analysis of the costs did not include productivity costs, which might not have been relevant given the age of the population considered. The inclusion of future costs was appropriate given the impact of the intervention on survival. The authors reported the sources of the cost data, which were derived from country-specific sources whenever possible. The information on resource consumption was less clear. Statistical analyses were not performed, but the impact of varying some key cost estimates was tested in the sensitivity analysis. The price year was reported, which will assist reflation exercises in other time periods. Other issues The authors made a few comparisons of their results with those from other studies. However, estimates of ITs have been published only for the UK and Sweden. The issue of the generalisability of the study results to other settings was clearly addressed as seven countries were considered. Further, the use of sensitivity analysis enhances the external validity of the study. The results of the analysis were selectively reported. The expected costs and QALYs were not presented, and the cost-utility ratios were only depicted graphically. The authors noted some critical issues about the estimation of hip fracture equivalents. Another critical aspect of the analysis was that Swedish data were often used for all countries, which implicitly assumes that the fracture patterns in Sweden are similar to those in other countries. Implications of the study The authors state that their decision model might be used for future studies when new information becomes available.
Source of funding Funded by the International Osteoporosis Foundation.
Bibliographic details Borgstrom F, Johnell O, Kanis J A, Jonsson B, Rehnberg C. At what hip fracture risk is it cost-effective to treat? International intervention thresholds for the treatment of osteoporosis. Osteoporosis International 2006; 17(10): 1459-1471 Other publications of related interest Because readers are likely to encounter and assess individual publications, NHS EED abstracts reflect the original publication as it is written, as a stand-alone paper. Where NHS EED abstractors are able to identify positively that a publication is significantly linked to or informed by other publications, these will be referenced in the text of the abstract and their bibliographic details recorded here for information.
Kanis JA, Johnell O, Oden A, et al. Intervention thresholds for osteoporosis. Bone 2002;31:26-31.
Kanis JA, Johnell O, Oden, A et al. Intervention thresholds for osteoporosis in men and women: a study based on data from Sweden. Osteoporos Int 2005;16:6-14.
Kanis JA, Borgstrom F, Zethraeus N, et al. Intervention thresholds for osteoporosis in the UK. Bone 2005;36:22-32.
Indexing Status Subject indexing assigned by NLM MeSH Age Factors; Aged; Aged, 80 and over; Cost-Benefit Analysis; Epidemiologic Methods; Female; Fractures, Bone /economics /etiology /prevention & Health Care Costs /statistics & Hip Fractures /economics /epidemiology /etiology /prevention & Humans; Middle Aged; Osteoporosis, Postmenopausal /complications /economics /therapy; Quality-Adjusted Life Years; control; control; numerical data AccessionNumber 22006001933 Date bibliographic record published 31/12/2007 Date abstract record published 31/12/2007 |
|
|
|