|
Screening for osteoporosis using easily obtainable biometrical data: diagnostic accuracy of measured, self-reported and recalled BMI, and related costs of bone mineral density measurements |
van der Voort D J, Brandon S, Dinant G J, van Wersch J W |
|
|
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 Three methods to assess body mass index (BMI) for the screening of both osteoporosis and low bone mineral density (BMD; osteopenia plus osteoporosis) were examined. These were objective BMI, self-reported BMI and recalled BMI.
The objective BMI equals the measured weight divided by the square of the measured height. The self-reported BMI equals the measured weight divided by the square of the self-reported height. The recalled BMI equals the recalled weight at age of 20 to 30 years divided by the square of the self-reported height.
Economic study type Cost-effectiveness analysis.
Study population The study population comprised the general population of postmenopausal women aged between 50 and 75 years. Patients presenting abnormalities that could interfere with osteoporosis, such as scoliosis or excessive osteophytosis, were excluded from the study.
Setting The setting was primary care. The economic study was carried out in 12 practice areas in south-east Netherlands.
Dates to which data relate The effectiveness and resource use data were gathered between October 1992 and March 1994. No price year was reported.
Source of effectiveness data The effectiveness data were derived from a single study.
Link between effectiveness and cost data The costing was performed prospectively on the same patient sample as that used in the effectiveness analysis.
Study sample Power calculations to determine the sample size were not performed. Of a total population of 9,107 women in the study area, 6,246 women were invited to participate in the study and 4,725 patients agreed to do so. A sample of 4,203 women (whole population) with a mean age of 63 (+/- 7) years was actually included in the study, as data for 522 patients were not assessed appropriately. A subsample consisting of 1,155 women with a mean age of 63 (+/- 7) years, who provided data on recalled weight, was considered to be the main study sample.
Study design This was a cross-sectional cohort study carried out in 12 practice areas in south-east Netherlands. The women completed a questionnaire containing a section on recalled weight at the age of 20 to 30 years. Immediately after the questionnaire was completed, their actual weight, height and BMD were measured. The BMD was measured using computer-guided dual-energy X-ray absorptiometry (DXA). A single cohort of women was enrolled in the study. No follow-up was performed.
Analysis of effectiveness The basis of the analysis was treatment completers only. The primary health outcomes assessed in the analysis were:
the number of DXA scans performed with each test,
the positive predictive value (PPV) of the three tests,
the number of patients with osteoporosis or low BMD,
the number of scans required to detect one patient with osteoporosis or low BMD,
the percentage of missed patients with osteoporosis or low BMD, and
the overall prevalence of osteoporosis and low BMD.
The BMD cut-off was 0.88 g/cm2 for osteoporosis and 0.970 g/cm2 for low BMD. A receiver operating characteristic (ROC) analysis showed that the cut-off point was 60 years for age and 27 kg/m2 for BMI. The diagnostic accuracy of the three methods for BMI measurement was compared in the sub-group of women aged at least 60 years who had a BMI less than or equal to 27. This was because this patient group had higher predictive values for osteoporosis and low BMD than prior probabilities for these conditions.
Effectiveness results In the study sample (1,155 women), the number of scans performed for the detection of osteoporosis was 352 with objective BMI, 410 with self-reported BMI, and 637 with recalled BMI.
The PPV was 41% with objective BMI, 30% with self-reported BMI, and 31% with recalled BMI.
The number of patients with osteoporosis was 143 with objective BMI, 156 with self-reported BMI, and 196 with recalled BMI (285 in the whole population).
The number of scans required to detect one patient with osteoporosis was 2.46 with objective BMI, 2.62 with self-reported BMI, and 3.25 with recalled BMI (4.05 in the whole population).
The percentage of missed patients with osteoporosis was 50% (142 patients) with objective BMI, 45% (129 patients) with self-reported BMI, and 31% (89 patients) with recalled BMI.
The overall prevalence of osteoporosis was 25%.
In terms of low BMD, the number of scans performed was the same as for osteoporosis.
The PPV was 81% with objective BMI, 77% with self-reported BMI, and 71% with recalled BMI.
The number of patients with low BMD was 285 with objective BMI, 316 with self-reported BMI, and 452 with recalled BMI (750 in the whole population).
The number of scans required to detect one patient with low BMD was 1.24 with objective BMI, 1.30 with self-reported BMI, and 1.41 with recalled BMI (1.54 in the whole population).
The percentage of missed patients with low BMD was 62% (466 patients) with objective BMI, 58% (434 patients) with self-reported BMI, and 39% (296 patients) with recalled BMI.
The overall prevalence of low BMD was 65%.
Clinical conclusions The effectiveness analysis showed that the recalled BMI was associated with higher detection rates (the inverse of the percentage of missed patients with osteoporosis or low BMD) than self-reported BMI and objective BMI.
Measure of benefits used in the economic analysis The benefit measure used in the economic analysis was the number of patients with osteoporosis or low BMD in the study sample, as assessed in the effectiveness analysis.
Direct costs Discounting was irrelevant due to the short time horizon of the study. Only the cost of DXA measurement in a bone density unit or diagnostic centre was included in the analysis. The unit cost was reported separately from the resource quantities. The cost/resource boundary adopted in the analysis was not stated, but it might have been that of the diagnostic centre. The costs for calling up patients were not considered. The costs were estimated from two published studies. The resource quantities were measured from the effectiveness study. No price year was reported.
Statistical analysis of costs The costs were treated deterministically.
Indirect Costs The indirect costs were not included in the analysis.
Currency Euros. The authors stated that 1 euro was approximately equal to 1 US dollar.
Sensitivity analysis No sensitivity analyses were performed.
Estimated benefits used in the economic analysis The number of patients with osteoporosis was 143 for objective BMI, 156 for self-reported BMI, and 196 for recalled BMI (285 in the whole population).
The number of patients with low BMD was 285 for objective BMI, 316 for self-reported BMI, and 452 for recalled BMI (750 in the whole population).
Cost results The cost of one scan was 75 euros.
The total costs (cost of one scan multiplied by the number of scans performed) were 26,400 euros in the objective BMI group, 30,700 euros in the self-reported BMI group, and 47,775 euros in the recalled BMI group.
Synthesis of costs and benefits An average cost-effectiveness analysis was performed to combine the costs and benefits. No incremental analysis was performed.
In screening for osteoporosis, the cost per detected patient was 185 euros in the objective BMI group, 197 euros in the self-reported BMI group, 243 euros in the recalled BMI group, and 304 euros in the overall study population of women scanned.
In screening for low BMD, the cost per detected patient was 93 euros in the objective BMI group, 97 euros in the self-reported BMI group, 105 euros in the recalled BMI group, and 116 euros in the overall study population of women scanned.
Authors' conclusions In the general population of women aged 50 to 80 years, screening for osteoporosis would cost 304 euros per patient detected while screening for low bone mineral density (BMD) would cost only 116 euros per patient detected. Among the three body mass index (BMI) tests for screening, the objective BMI test offered the lowest cost-effectiveness ratio. A BMI below 27 kg/m2 only proved to be a significant predicting factor for osteoporosis and low BMD in women over 60 years of age. There is no need for an actual height measurement, as self-reported height appears to be a good approximation of the real height (although there is systematic overestimation). Age and BMI do not represent significant risk factors in screening for osteoporosis and low BMD. Finally, the authors stated that "measurements of BMD could be implemented within the current screening program for breast cancer".
CRD COMMENTARY - Selection of comparators The rationale for the choice of the comparators was clear. The three methods for BMI measurement were selected since they all represented possible options used for osteoporosis screening. You should decide whether they represent feasible options in your own setting.
Validity of estimate of measure of effectiveness The effectiveness analysis used a cross-sectional design, but the authors acknowledged that a longitudinal design would have been more appropriate for the study question. The study sample appears to have been representative of the study population, as unselected postmenopausal women were enrolled. The sub-groups of patients undergoing the three BMI tests were not compared. The main analysis was performed on 1,155 patients of the 4,203 enrolled in the study. However, it appears that there were no substantial differences between the whole study sample and the sub-group (24%) of patients used in the effectiveness analysis.
Validity of estimate of measure of benefit The detection rate was used as the benefit measure in the economic analysis. It was derived directly from the effectiveness analysis. It represents a common benefit measure in screening programmes.
Validity of estimate of costs The perspective adopted in the analysis was not stated. However, only the costs of DXA scans were included in the analysis. The costs of calling up patients were not considered which, if included, would have an impact on the cost results. The unit cost and the quantities of resource were reported separately. The cost data were derived from two published studies. However, the costs were treated deterministically and sensitivity analyses were not performed. In addition, no price year was reported. Therefore, the cost results have some limitations in terms of their generalisability, and reflation exercises to other settings would be difficult.
Other issues The authors made some comparisons of their findings with those from other studies. The issue of the generalisability of the study results to other settings was not addressed, and sensitivity analyses were not performed. Thus, the external validity of the analysis was limited. The study enrolled a sample of postmenopausal women aged 50 to 80 years, and this was reflected in the conclusions of the study. The authors reported the average cost-effectiveness ratios rather than the more informative incremental cost-effectiveness ratios.
Implications of the study The authors noted that future research should use a longitudinal study, and should assess the real role of age and BMI as predicting factors in the screening for osteoporosis and low BMD.
Bibliographic details van der Voort D J, Brandon S, Dinant G J, van Wersch J W. Screening for osteoporosis using easily obtainable biometrical data: diagnostic accuracy of measured, self-reported and recalled BMI, and related costs of bone mineral density measurements. Osteoporosis International 2000; 11(3): 233-239 Indexing Status Subject indexing assigned by NLM MeSH Absorptiometry, Photon /economics; Aged; Aged, 80 and over; Biometry; Body Mass Index; Bone Density; Bone Diseases, Metabolic /diagnosis /epidemiology /prevention & Costs and Cost Analysis; Cross-Sectional Studies; Female; Humans; Mass Screening /economics /methods /standards; Middle Aged; Netherlands /epidemiology; Osteoporosis, Postmenopausal /diagnosis /epidemiology /prevention & Predictive Value of Tests; Prevalence; Self Disclosure; Sensitivity and Specificity; control; control AccessionNumber 22000000868 Date bibliographic record published 31/03/2003 Date abstract record published 31/03/2003 |
|
|
|