|
Development and validation of a new clinical risk index for prediction of osteoporosis in Thai women |
Pongchaiyakul C, Nguyen N D, Pongchaiyakul C, Nguyen T V |
|
|
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 new Thai-specific osteoporosis score, the Khon Kaen Osteoporosis Study (KKOS) score, was examined.
Economic study type Cost-effectiveness analysis.
Study population The study population comprised the general population of women who might be at risk of osteoporosis. Women were excluded if they had a history of metabolic bone disorders (other than postmenopausal bone loss), a presence of cancer with known metastasis to bone, or menopause before the age of 40 years. Women who had at least one ovary removed were also excluded, as were those with a history of taking medications affecting calcium and bone metabolism (e.g. steroids, thyroid hormone, bisphosphonates, fluoride or calcitonin).
Setting The setting was secondary care. The economic study was carried out inn Thailand.
Dates to which data relate The dates to which the effectiveness and resource use data referred were not reported. The price year was not stated.
Source of effectiveness data The effectiveness evidence was derived from a single study.
Link between effectiveness and cost data The costing was carried out in a hypothetical cohort of women eligible for the diagnosis of osteoporosis.
Study sample Initially, 332 letters were sent out to identify eligible women. The response rate was 100% (no women refused to join the study). However, 10 women were excluded because they did not meet the inclusion criteria. Thus, the final study sample comprised 322 women. The whole sample was randomly divided into two cohorts in a ratio of 2:3. The development cohort (n=130) was used to develop the new diagnostic tool, while the validation cohort (n=192) was used to validate it. The mean age of the participants was 59.6 (+/- 9.1) years in the development group and 60.5 (+/- 9.8) years in the validation group. Women in the development group weighed 57.0 (+/- 10.3) kg and were 152.9 (+/- 5.6) cm in height, while those in the validation group were of weight 55.7 (+/- 9.8) kg and height 152.5 (+/- 5.5) cm. Power calculations were not reported.
Study design This was a cross-sectional study that was carried out in the Muang district, Khon Kaen province, Thailand. The participants were randomly selected by an administrator of a sub-district. In both groups the BMD was measured by both the KKOS score and DXA. No follow-up was performed.
Analysis of effectiveness All of the women included in the study sample were accounted for in the analysis of effectiveness. The primary outcome measure was the prevalence of osteoporosis. This was calculated in the whole cohort at the lumbar spine and femoral neck by DXA ('gold' standard). A woman was classified as "osteoporosis" if her BMD T-score was -2.5 or less, otherwise she was classified as "non-osteoporosis".
In the development cohort, the KKOS was developed using logistic regression to evaluate the association between age and weight and BMD-based osteoporosis diagnosis. This relationship was expressed as an odds ratio (the KKOS score was derived as the sum of the odds ratios). Further, receiver operating characteristic (ROC) curves were constructed for the KKOS score system. A cut-off score with the highest discriminatory power was derived from the ROC curves.
In the validation cohort, the KKOS score was then calculated and classified into two groups (high and low risk) based on the cut-off score. The concordance between the KKOS classification and the actual BMD-based classification based on DXA (the 'gold' standard) led to the estimation of sensitivity, specificity and positive predictive value (PPV) of the KKOS score, which were the other primary outcome measures.
The two cohorts were comparable at baseline in terms of their age, body weight, height, body mass index and BMD.
Effectiveness results The prevalence of osteoporosis in the whole sample was 11% by femoral neck BMD and 32% by lumbar spine BMD. When the two measures were considered, the prevalence was 33% and increased with advancing age. There was no significant difference in the prevalence of osteoporosis between the development and validation cohorts.
In the development cohort, each 5-year increase in age was associated with a 1.6-fold (95% confidence interval, CI: 1.2 - 2.0) in the risk of osteoporosis. Each 5-kg decrease in weight was associated with a 2.0-fold (95% CI: 1.5 - 2.8) increase in the risk of osteoporosis. The range of KKOS score was between 19.5 and -21.5. The cut-off score of -1 had the highest discriminatory power. Thus, individuals with KKOS scores less than -1 were defined as "high risk" and individuals with KKOS scores greater than -1 were defined as "low risk".
In the development group, the sensitivity of the KKOS score was 75% and the specificity was 80%. The KKOS score system yielded an area under the curve of 0.85.
When applied to the validation cohort, the sensitivity of KKOS was 70% and the specificity was 73%. However, there was a significant variation in the diagnostic measures with age. For example, the sensitivity and specificity were 50% and 89%, respectively, in women aged 45 to 60 years, and 44% and 78% among women aged 60+ years.
The application of the KKOS led to 41% of women in the cohort being considered at high-risk (KKOS </= -1). The PPV of the KKOS score was 59% in the high-risk population and 15% in low-risk individuals.
Clinical conclusions The effectiveness analysis showed that the KKOS score was relatively accurate as a diagnostic tool for osteoporosis in Thai women in terms of its sensitivity and specificity. However, it was associated with a low PPV.
Measure of benefits used in the economic analysis The benefit measure was the PPV of the KKOS score. This was derived directly from the effectiveness analysis.
Direct costs The perspective used in the analysis of the costs was not stated. The economic analysis included only the cost of treatment (hormone replacement therapy and calcium supplementation) and the cost of BMD measurement. The unit costs, which were based on authors' assumptions, were reported. The resource use data were based on the hypothesis that each woman would undergo a single diagnostic evaluation. The cost analysis was based on the assumption that treatment would reduce the fracture incidence by 50%, as reported in the literature. Discounting was not relevant as the costs were incurred during a short timeframe. The price year was not reported.
Statistical analysis of costs The costs were treated deterministically.
Indirect Costs The indirect costs were not considered in the economic evaluation.
Sensitivity analysis Several scenarios were considered according to the prevalence and PPV of KKOS scores, the incidence of fracture of individuals with osteoporosis, and the cost of drug treatment.
Estimated benefits used in the economic analysis See the 'Effectiveness Results' section.
Cost results The daily cost of treatment was THB 10.00 (hormone replacement therapy and calcium supplementation) or THB 50.00 per day (anti-resorptive agent and calcium supplementation). The cost of BMD measurement was THB 600.00 per patient.
Synthesis of costs and benefits An incremental cost-effectiveness ratio was calculated to combine the costs and benefits of the KKOS score over the standard diagnostic tool.
Assuming that the annual incidence of fractures among osteoporotic and non-osteoporotic Thai women were 2% and 1%, respectively, and that all high-risk individuals (KKOS < -1) who have osteoporosis (T-scores < -2.5) were treated, then the cost of preventing one fracture was estimated to be THB 466,695 per year. Most of the cost resulted from treatment (80%); the cost of BMD measurement was modest (20%).
The cost of preventing one fracture decreased as the PPV increased, although the effect was quite modest. For example, under the same assumption as above for a PPV of 80%, the cost was THB 440,000 per year. This represented a reduction of THB 26,695 in comparison with the cost associated with a PPV of 59%.
Increasing the cost of the drug treatment to THB 50.00 per day had a higher impact on the cost-effectiveness ratios, which ranged from 1,284,464 to 1,958,333 per fracture prevented (depending on the incidence of fractures in osteoporosis patients).
Authors' conclusions A simple tool based on two factors (age and body weight) accurately identified women at an increased risk of osteoporosis (low bone mineral density). The Khon Kaen Osteoporosis Study (KKOS) score had a high sensitivity (70%) and specificity (73%) for identifying individuals with a high risk of osteoporosis at the femoral neck or lumbar spine. However, the KKOS score had a limited positive predictive value (PPV). Thus, its implementation in the general population might not be efficient on account of its high false-positive rate.
CRD COMMENTARY - Selection of comparators The selection of the comparator (i.e. diagnosis by DXA scanning) was appropriate as it represented the standard diagnostic tool for the detection of osteoporosis. You should decide whether this is a valid comparator in your own setting.
Validity of estimate of measure of effectiveness The effectiveness evidence came from a cross-sectional study. Women were included in the study to provide demographic and clinical information for the development and validation of the new diagnostic tool. Cross-sectional studies usually have a limited internal validity. However, the random selection of patients and the high response rate will have increased the robustness of the clinical data. The two cohorts of women were comparable. The analysis focused on clinical characteristics of Thai women, thus caution is required when extrapolating the results of the analysis to other patient populations.
Validity of estimate of measure of benefit The summary benefit measure was specific to the study setting. It would not be comparable with the benefits of other health care interventions. The impact of the intervention on quality of life was not investigated.
Validity of estimate of costs The analysis of the costs was restricted to costs for the diagnosis and treatment of osteoporosis. Further treatment of fractures was not considered. The unit costs were reported and were specific to the study setting. The cost analysis was based on several assumptions about the unit costs and resource use. Some of these assumptions, such as the unit costs and rate of incidence fracture, were varied in the sensitivity analysis. The price year was not reported, which would hinder reflation exercises in other time periods. The unit cost of treatment was varied in a univariate sensitivity analysis.
Other issues The authors did not compare their findings with those from other studies. In terms of the issue of the generalisability of the study results to other settings, the authors noted that caution will be necessary when extrapolating their findings to other settings. Further, both the costs and efficacy data were specific to the Thai context. In general, the external validity of the results was very low and few sensitivity analyses were performed. The results of the analysis were clearly presented.
Implications of the study The study results suggested that the KKOS score is accurate in detecting women at risk of osteoporosis. However, the authors noted that its application to the general population requires further research to arrive at the optimal cost-benefit for the community.
Source of funding Funded in party by a grant from the Faculty of Medicine, Khon Kaen University.
Bibliographic details Pongchaiyakul C, Nguyen N D, Pongchaiyakul C, Nguyen T V. Development and validation of a new clinical risk index for prediction of osteoporosis in Thai women. Journal of the Medical Association of Thailand 2004; 87(8): 910-916 Other publications of related interest Black DM, Steinbuch M, Palermo L, et al. An assessment tool for predicting fracture risk in postmenopausal women. Osteoporos Int 2001;12:519-28.
Felson DT, Zhang Y, Hannan MT, Anderson JJ. Effects of weight and body mass index on bone mineral density in men and women: the Framingham study. J Bone Miner Res 1993;8:567-73.
Cadarette SM, Jaglal SB, Murray TM. Validation of the simple calculated osteoporosis risk estimation (SCORE) for patient selection for bone densitometry. Osteoporos Int 1999;10:85-90.
Indexing Status Subject indexing assigned by NLM MeSH Absorptiometry, Photon; Aged; Bone Density; Cross-Sectional Studies; Female; Femur Neck /physiology; Fractures, Bone /economics /etiology /prevention & Health Care Costs; Humans; Middle Aged; Osteoporosis /complications /diagnosis; Preventive Medicine /economics; Risk Factors; Severity of Illness Index; control AccessionNumber 22004009110 Date bibliographic record published 31/03/2006 Date abstract record published 31/03/2006 |
|
|
|