|
Interest of a prescreening questionnaire to reduce the cost of bone densitometry |
Ben Sedrine W, Broers P, Devogelaer J P Depresseux G, Kaufman J M, Goemaere S, Reginster J |
|
|
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 A pre-screening questionnaire to identify patients at risk of osteoporosis, who may benefit from bone mineral density (BMD) measurement, was examined. The questionnaire comprised ten historical, clinical and behavioural risk factors:
(1) prolonged amenorrhoea;
(2) early menopause (before the age of 45 years);
(3) late puberty (after the age of 15 years);
(4) natural or surgical menopause after the age of 45 years with at least two additional risk factors (such as no hormonal substitution treatment, family history of osteoporosis, BMI less than <25 kg/m2);
(5) primary or secondary hypogonadism;
(6) history of traumatic fractures;
(7) fortuitous discovery of osteopenia by radiogram;
(8) history of long-term immobilisation;
(9) diseases affecting the skeleton; and
(10) long-term use of medications affecting the skeleton.
Factors 1 to 4 were only relevant for women, while factors 5 to 10 applied to men and women.
Economic study type Cost-effectiveness analysis.
Study population The study population consisted of the general population of men and women aged more than 20 years. Realistically, the analysis was limited to persons at or over the age of 50 years.
Setting The setting was an outpatient osteoporosis centre at a university hospital. The economic study was carried out at Liege in Belgium.
Dates to which data relate The effectiveness data were gathered between January 1996 and December 1998. No dates for the resource use data were reported. The price year appears to have been 1987.
Source of effectiveness data The effectiveness evidence came from a single study.
Link between effectiveness and cost data The costing was performed retrospectively on the same sample of patients as that used in the effectiveness study.
Study sample Power calculations do not appear to have been conducted. A sample of 3,998 consecutive patients, either having a spontaneous consultation or being referred for a BMD scan at the study centre, was included in the analysis. No comparison group was used in the study. There were 3,677 (92%) women and 321 (8%) men. The mean age was 60.1 (+/- 10.3) years (median age: 61; range: 20 - 98). The sub-sample of those aged 50 years or more numbered 3,405 individuals, whilst those aged 61 years or more numbered 2,306.
Study design This was a retrospective case series, which was carried out at the Department of Epidemiology and Public Health of the University of Liege. No follow-up was conducted. The outcome measurement was based on complete medical records. BMD measurements, using dual-energy X-ray absorptiometry (DXA) technology, were performed at three sites (total hip, femoral neck and lumbar spine). BMD values were expressed as T-scores. A T-score of at least -1 indicated a normal individual, a score of -1 to greater than -2.5 indicated an osteopenic person, while a score of less than or equal to -2.5 indicated an osteoporotic person.
Analysis of effectiveness All of the patients included in the initial study sample were accounted for in the effectiveness study. The health outcomes used in the analysis were:
the distribution of the population according to the number of risk factors present; and
the sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NVP) of being exposed to one risk factor or more for identifying patients with osteoporosis (T-score =/< -2.5) or with either osteoporosis or osteopenia (T-score =/< -1).
No weighting of the risk factors was undertaken, thus exposure to at least one of the ten factors was considered as a sign that the test was positive. As BMD represents a decreasing function of age, age was introduced into a receiver operating characteristics (ROC) analysis to determine the optimal cut-off value that best discriminated between osteoporotic and non-osteoporotic patients. This value was 61 years, thus this was considered as the basis of strategy C.
Effectiveness results The distribution of the population according to the number of risk factors present was as follows:
33.3% did not have any of the suggested risk factors and would be excluded from the BMD scan;
52.2% presented with one risk factor, 10.8% with two risk factors, 3.4% with three risk factors, and 0.2% with four risk factors.
The most common positive sign was menopause after the age of 45 years, with at least two additional risk factors (in 44% of women).
In the whole population (n=3,998), the sensitivity was 74.9%, the specificity 37%, the PPV 34.8% and the NPV 76.7% for the identification of patients with osteoporosis. The corresponding values for the detection of patients with either osteoporosis or osteopenia at any site were 70.7% (sensitivity), 43.6% (specificity), 76.2% (PPV) and 36.8% (NPV).
In the population aged 50 years or more (n=3,405), the sensitivity was 75.6%, the specificity 34.6%, the PPV 36.5% and the NPV 74% for the identification of patients with osteoporosis. The corresponding values for the detection of patients with either osteoporosis or osteopenia at any site were 72.4% (sensitivity), 41.4% (specificity), 77.7% (PPV) and 34.7% (NPV).
In the population aged 60 years or more (n=2,036), the sensitivity was 75.3%, the specificity 35%, the PPV 42.9% and the NPV 68.6% for the identification of patients with osteoporosis. The corresponding values for the detection of patients with either osteoporosis or osteopenia at any site were 72.5% (sensitivity), 43.7% (specificity), 82.9% (PPV) and 29.7% (NPV).
Clinical conclusions The effectiveness analysis showed that the pre-screening questionnaire was effective in identifying patients for whom screening with BMD could be more beneficial, but low specificity values were reported. Thus, there were a high number of false positives, meaning that a substantial number of unnecessary BMD tests would still be performed.
Measure of benefits used in the economic analysis The summary benefit measure was the number of patients detected with the three strategies. This was derived directly from the effectiveness study.
Direct costs Discounting was not relevant since the costs were incurred over a short time period. The unit cost was reported separately from the quantity of resources used. The economic evaluation included only the costs of the BMD scan. The cost/resource boundary was that of the third-party payer. Resource use was based on the hypothesis of performing BMD tests on the eligible population, which depended on the screening strategy. The source of the cost data was official reimbursement rates. The price year was 1985 and no discounting to present values was conducted.
Statistical analysis of costs The costs were treated deterministically.
Indirect Costs The indirect costs were not included in the economic analysis.
Currency Euros (Euro). The authors reported that the exchange rate from Belgian francs (Bfr) into Euros, considering the cost of a BMD scan in 1997, was Bfr 875 = Euros 21.7. Euro 1 = approximately US $1.
Sensitivity analysis Sensitivity analyses were not conducted.
Estimated benefits used in the economic analysis The number of patients with osteoporosis detected was 1,131 with strategy A, 855 with strategy B, and 603 with strategy C. Consequently, strategy A did not miss any patient with osteoporosis, while strategy B missed 276 patients (24.4%) and strategy C missed 198 patients (24.7%).
The number of patients with either osteoporosis or osteopenia detected was 2,515 with strategy A, 1,820 with strategy B, and 1,165 with strategy C. Again, strategy A did not miss any patient, while strategy B missed 694 patients (27.6%) and strategy C 443 patients (27.5%).
Cost results The total costs of the screening strategies were not reported in the article.
Synthesis of costs and benefits The costs and benefits were combined by calculating the cost-effectiveness ratios. The cost per patient detected with osteoporosis was Euro 63.55 with strategy A, Euro 59.44 with strategy B, and Euro 50.60 with strategy C. The cost per patient detected with either osteopenia or osteoporosis was Euro 29.38 with strategy A, Euro 27.92 with strategy B, and Euro 26.19 with strategy C.
Authors' conclusions The pre-screening questionnaire, especially when focusing on patients older than 60 years of age, appeared to be efficient as it was cheaper than the mass screening campaign. However, there were a substantial number of undetected patients and unnecessary tests performed. Thus, the use of the ten prognostic factors as a selective programme for the target population in the screening for osteoporosis (and osteopenia) was unsatisfactory.
CRD COMMENTARY - Selection of comparators The authors stated that mass screening was selected as the basic comparator because it represented the standard approach recommended for the detection of osteoporosis. The age of the target population was restricted to individuals older than 50 years of age because this was a realistic hypothesis. You should decide whether mass screening represents a valid comparator in your own setting.
Validity of estimate of measure of effectiveness The effectiveness evidence came from a retrospective case series at the study institution. The authors admitted that this source of data entailed some limitations. In addition, that selection bias may have affected the study results especially in the case of male patients, who were likely to have developed the disease. This casts some doubts on how representative the study sample was. Further, there was no formal control group as the effectiveness study evaluated only the results of the pre-screening test, while mass screening was supposed to detect all cases of osteoporosis. These issues may limit the internal validity of the analysis.
Validity of estimate of measure of benefit The benefit measure used in the economic analysis was derived directly from the effectiveness study. The use of detection rate appears to be the natural outcome of the intervention, although the use of a final health outcome measure would have enhanced the comparability of the study intervention with the benefits of other programmes.
Validity of estimate of costs The perspective adopted in the study was reported and only the costs strictly related to the DXA test were included in the analysis. The unit cost was reported. Resource use was based on one DXA test being performed on each patient, depending on the selected target population. The authors stated that the costs were underestimated as reimbursement rates in 1987 were used, but the focus of the analysis was on the relative difference between the screening strategies rather than on the absolute values. The costs were treated deterministically and no sensitivity analyses were conducted. The authors stated that further cost-savings could be derived by from the inclusion of further categories of costs.
Other issues The authors compared their findings with those from other studies and reported the results of other questionnaires used for the detection of osteoporosis. However, the issue of the generalisability of the study results to other settings was not addressed and sensitivity analyses were not conducted. Thus, the external validity of the analysis is limited. The study enrolled persons at risk of osteoporosis and this was reflected in the conclusions of the analysis. The authors stressed that the questionnaire should have included a question regarding body weight or BMI, in order to improve both the specificity and sensitivity of the programme. It would have been interesting had the authors performed an incremental cost-effectiveness analysis.
Implications of the study The study suggests that the pre-screening test reduces the costs of screening for osteoporosis, but is unable appropriately to identify the patients who could benefit the most from the screening. The authors noted some critical issues in the workability of the pre-screen questionnaire. First, some of the questions included in the questionnaire required the interpretation of medical experts, thus reducing the widespread use of the programme. Second, age was considered as a prognostic factor only in a sub-group analysis. These limitations should be considered when interpreting the results obtained.
Bibliographic details Ben Sedrine W, Broers P, Devogelaer J P Depresseux G, Kaufman J M, Goemaere S, Reginster J. Interest of a prescreening questionnaire to reduce the cost of bone densitometry. Osteoporosis International 2002; 13(5): 434-442 Indexing Status Subject indexing assigned by NLM MeSH Absorptiometry, Photon /economics; Adult; Aged; Aged, 80 and over; Bone Density; Cost Control; Female; Fractures, Spontaneous /prevention & Humans; Male; Mass Screening /economics /methods; Middle Aged; Osteoporosis /economics /prevention & Patient Selection; Predictive Value of Tests; Risk Factors; Sensitivity and Specificity; Surveys and Questionnaires; control; control /radiography AccessionNumber 22002001078 Date bibliographic record published 31/10/2003 Date abstract record published 31/10/2003 |
|
|
|