gms | German Medical Science

The German Agency for Health Technology Assessment (DAHTA) at the German Institute for Medical Documentation and Information (DIMDI) commissions research projects for Health Technology Assessment (HTA) based on the legal order to built up an information system for the evaluation of medical technology. The authors were engaged to conduct a rapid economic HTA to assess the cost-effectiveness of Quantitative Ultrasound (QUS) as a technique for screening the fracture risk due to osteoporosis and consecutive preventing fractures [1]. The rapid economic HTA was conducted according to the German methodological recommendations for health economic short HTA [2]. Information synthesis is based on a systematic literature search and subsequently on extracting cost data out of the identified literature.

Osteoporosis is a disease that mainly concerns postmenopausal women and may increase fracture risk [3]. The risk for osteoporosis and fractures is age dependent. Nearly 50% of women in the age over 70 years show osteoporotic alterations [4]. Fractures due to osteoporosis cause costs, reduce quality of life and may lead to death. Main fracture localisations are found at the distal forearm, spine and femoral neck. The annual incidence rate of fractures of the femoral neck for women is about 291 per 100,000 inhabitants [5]. Factors associated with a increased risk for osteoporosis are age, postmenopause, smoking, lack of physical exercise, nutrition, low body weight, positive family history, alcohol abuse, and medication. The causal interdependence of the mentioned risk factors with the development of osteoporosis are partially not validated.

Osteoporosis can be treated (among others) by calcium and Vitamin D in combination with biphosphonates and therapy with raloxifen [6], [7].

The diagnosis of osteoporosis covers clinical parameters and the measurement of bone mineral density (BMD). Several techniques for the determination of bone mineral density are described in the literature. The dual x-ray absorptiometry (further DXA) is at present the most frequently used method of measurement. Furthermore the peripheral quantitative computed tomography (pQCT) and the quantitative computed tomography (QCT) are of relevance. Increasingly the Quantitative Ultrasound (further QUS) is becoming important as it is an easily and quickly feasible technique [8], [9].

In Germany QUS examinations are not covered by the statutory health insurance fund and have to be financed out of pocket by the patients.

Three main proceedings do exist: (1) DXA might be used as a screening technology alone, (2) QUS might be used as a screening technology alone, (3) two step strategy with QUS as a first step followed by DXA as a second step for women with a pathological QUS examination.

Aim of the rapid health economic HTA was to evaluate the cost-effectiveness of different proceedings for the application of QUS compared to DXA as a technique for screening of osteoporotic fracture risk and subsequently for the prevention of osteoporotic fractures in postmenopausal women.

To assess the cost-effectiveness of QUS as a technique for screening of fracture risk due to osteoporosis a comprehensive literature search was performed in the medical and economical literature databases Medline, Embase and Econlit. Table 1 [Tab. 1] presents the used search strategy.

The HTA databases DARE (Database of Abstracts of Reviews of Effectiveness), NHS (Economic Evaluation Database), HTA (Health Technology Assessment) as well as the Cochrane library were searched for relevant HTA reports and reviews. Internet pages of 17 international HTA-institutions were visited and searched for references to relevant HTA-reports.

No restrictions were made in respect to publication language and publication date. The literature search was performed in November 2001.

Following a data extraction, qualitative and quantitative information was consolidated in form of tables.

Relevant qualitative characteristics have been author, year, country, type of evaluation and study population, quantitative characteristics have been the main medical effects and economic parameters.

In addition the authors calculated the incremental cost-effectiveness for the different proceedings: of a two-step proceeding compared to a one-step proceeding (DXA alone) as well as a comparison of the one step proceedings QUS alone vs. DXA alone.

The considered studies reported the costs of bone mineral density measurements in different currencies. A currency conversion to EURO was done according to the recommendations for german HTA reports [10] by the purchasing-power parities of the OECD.

67 references of publications and projects were identified. 62 references were excluded, because of being a duplicat (number: 22), they did not deal with the costs of QUS or osteoporosis as a target diagnosis (number: 38), the results of the studies have not been available at the time of the literature search.

Four of the remaining five publications dealt with the question of the cost-effectiveness of QUS as a screening instrument in a two-step proceeding [11], [12], [13], [14]. One publication was the summary of a HTA report concerning the effectiveness of osteodensitometry and gave some information about costs of DXA and QUS. Costs have not been set in relation to the medical effects [15]. For this reason this study was not considered in the further calculations.

All five publications had limitations in regard to methods, especially in the consideration of relevant cost parameters. Just Langton et al. (1997 [12] and 1999 [13]) gave information about cost types included, namely obsolescence of device, personnel costs and costs due to transport. Due to methodological weaknesses of the studies the results can not be used without limitation.

Table 2 [Tab. 2] presents the main quantitative parameters of the four relevant studies.

One study was conducted in Switzerland [11], the remaining three in Great Britain. Study population of the study of Langton et al. (1997) have been women in the age of 60 to 69 years [12], of the study of Langton et al. (1999) women between the age of 50 to 54 years [13]. The remaining two studies had an age range from 50 to 80 years [14] and 44 to 80 years [11]. None of the studies reported the cost-effectiveness of a one-step proceeding with QUS compared to one-step DXA. However, data of the studies of Langton et al. (1997 [12] and 1999 [13]) and Sim et al. (2000) [14] allowed to calculate the cost-effectiveness of the two techniques in a one step proceeding. Results of this calculation are presented in Table 3 [Tab. 3].

According to the studies of Langton et al. (1997) [12] and Sim et al. (2000) [14], approximately additional 1,000 EURO are necessary per additional identified case of osteoporosis diagnosed by one-step DXA compared to one-step QUS. The high costs resulting of the data of Langton et al. (1999) [13] are due to the low prevalence in younger women. It also has to be considered that QUS will rise more false positive cases that subsequently will get a costly treatment. These false positive cases will potentially be identified in a two-step proceeding with DXA as a second test.

The cost-effectiveness of a two-step proceeding compared to a one-step proceeding depends on the relation of costs of QUS to the costs of DXA. In all four studies a cut-off-point could be calculated, that tells under what level an examination with QUS in a two-step screening procedure is more cost-effective compared to a one-step proceeding with DXA. Depending on the considered study a two-step proceeding is cost-effective as long as the costs of one QUS examination lies between 31% and 51% of an examination with DXA.

Table 4 [Tab. 4] shows the calculations of a monetary cut-off point in a two-step proceeding.

A main limitation of the results is seen in the lack of explanatory power of the obtained study results in the context of the German health care system. So far screening for osteoporosis is not conducted in Germany by default of a valid screening instrument. It is to mention that the results of DXA and QUS examinations are hardly comparable. DXA measures bone density, but it is not proved which entity of bone is measured with QUS. Costs due to a QUS examination are not reimbursed by the statutory health insurance fund and have to be financed out of pocket by the patients. As no valid information was found in the literature concerning costs of a QUS or DXA examination in Germany we recommend to assess German cost data and perform adjusted calculations.

As pointed out above due to the methodological limitations the evidence of the studies is restricted decisively. Limitations are seen in the insuffiently performed description of quantities and prices for the calculation of costs. No study sufficiently discussed the own study results or potential limitations of the study. No study performed a sufficient sensitivity analysis.

The time horizon of considered studies was limited to the end point of diagnosis of osteoporosis. False positive results will lead to the application of a unnecessary and costly treatment, false negative will not reduce the fracture risk and associated costs. For this reason the long-term outcomes of a diagnosed case of osteoporosis in regard to fracture risk, avoided death and improvement of quality of life as much as the costs of a treatment of osteoporosis and treatment of fractures are of special interest. Because of the short time horizon none of the studies included these longtime outcomes. Therefore the scientific evidence of the results of all considered studies is strongly limited. Decision analytic models that analyse long-term outcomes do exist in the literature [e.g. [16], [17]]. We strongly recommend to adopt decision analytic models to calculate the cost-effectiveness of QUS considering long-term effects.

Due to the weakness of data and the limited time horizon of all considered studies no concluding judgement about the cost-effectiveness of QUS as a first step of a two-step proceeding (QUS followed by DXA in case of a pathological result) or as a one-step proceeding (QUS alone) can be given.