|
The Osteoporosis Self-Assessment Tool versus alternative tests for selecting postmenopausal women for bone mineral density assessment: a comparative systematic review of accuracy |
Rud B, Hilden J, Hyldstrup L, Hrobjartsson A |
|
|
CRD summary This review concluded that in post-menopausal white women, the Osteoporosis Self-Assessment Tool (OST) had similar accuracy to alternative triage tools, but the Stiffness Index was more accurate than OST. Most included studies were methodologically flawed, there were no searches for recent research, and duplication during study selection was not reported. Consequently, the review results should be viewed with caution. Authors' objectives To evaluate the accuracy of the Osteoporosis Self-Assessment Tool compared with other tests used to select post-menopausal women for bone mineral density assessment. Searching PubMed and EMBASE were searched without language restrictions from inception to June 2005. Search terms were reported in the paper and the full search strategy was available in an online appendix (see URL for additional data; accessed 18 November 2009). A Web of Science citation search was also undertaken. Bibliographies of included studies were scanned, as were conference abstracts from six relevant societies from 2000, and the European Congresses on Clinical Economic Aspects of Osteoporosis and Osteoarthritis. Study selection Studies that evaluated the Osteoporosis Self-Assessment Tool and other triage tests (used to detect low bone mineral density in the same post-menopausal women) compared with dual-energy X-ray absorptiometry measured at the femoral neck, total hip or lumbar spine, were eligible for inclusion.
In the included studies, the mean age ranged from 50.6 to 70.5 years, and the majority of participants were Caucasian. A variety of clinical decision rules were evaluated; Osteoporosis Risk Assessment Instrument and Simple Calculated Osteoporosis Risk Estimation were the most common. For the dual-energy X-ray absorptiometry, T-score targets were at least -2.0 or -2.5; a range of dual-energy X-ray absorptiometry devices were used. Prevalence of osteoporosis ranged from 0.3 to 47.3%, depending on dual-energy X-ray absorptiometry target and site of scan.
The authors did not state how many reviewers performed the study selection. Assessment of study quality Study quality was assessed using QUADAS (Quality Assessment of Diagnostic Accuracy Studies); four of the criteria were not assessed (appropriateness of reference standard, progression bias, incorporation bias and clinical review bias), and two criteria were added (use of pairing and sequencing of tests).
[A: Study quality was assessed by one reviewer and checked by a second.] Data extraction Data were extracted to construct 2x2 tables of test performance, from which sensitivity, specificity and diagnostic odds ratio (DOR) with 95% confidence intervals (CI) were calculated. Authors were contacted for missing data. Sufficient data were available to construct a 2x2 table for 15 studies. Where data for different thresholds was available, data were extracted for the threshold that provided sensitivity closest to 90%.
Data were extracted by one reviewer and checked by a second; discrepancies were resolved by consensus. Methods of synthesis Summary receiver operating characteristic (sROC) curves were produced using the Moses and Littenberg model when three or more studies were available; where fewer studies were available, results were plotted in receiver operating characteristic (ROC) space and synthesised in a narrative. Analyses were stratified according to ethnicity. Covariates were added to the regression model to test the equality of diagnostic odds ratios between the Osteoporosis Self-Assessment Tool and comparator triage test and relative summary diagnostic odds ratios (rsDOR) were reported. The Higgins permutation test was used to calculate p-values for regression co-efficient. Heterogeneity was assessed using the I2 statistic. Results of the review Fifteen studies met the inclusion criteria and provided data (n=48,148; range 208 to 23,833). All included studies had methodological flaws. Most studies did not: report the recruitment of a representative patient spectrum; report uninterpretable results; report blinding of assessors of the triage tests or dual-energy X-ray absorptiometry; or account for withdrawals.
White women: The Osteoporosis Self-Assessment Tool (OST) was more accurate than: Osteoporosis Risk Assessment Instrument (ORAI), Simple Calculated Osteoporosis Risk Estimation (SCORE) and Study of Osteoporotic Fractures Simple Useful Risk Factors (SOFSURF) when dual-energy X-ray absorptiometry was measured at the femoral neck or more than one site for any target (rsDOR ranged from 0.57 to 0.97; number of studies unclear); SOFSURF when dual-energy X-ray absorptiometry was measured at the femoral neck with a target of below or equal to -2.5 (rsDOR 0.74; one study); and case finding criteria based on risk factors for fracture across eight bone mineral density targets (one study). ORAI was more accurate than OST when dual-energy X-ray absorptiometry was measured at the lumbar spine (rsDOR ranged from 1.02 to 1.17; three studies). Significant heterogeneity was observed for comparisons made where the dual-energy X-ray absorptiometry was not conducted at the femoral neck. Stiffness Index was significantly more accurate than OST where the dual-energy X-ray absorptiometry was at the lumbar spine with a target of below or equal to -2.5 (rsDOR 1.9; four studies). OST was no significantly more accurate than weight (three studies).
Black and Asian women: Comparisons of OST, ORAI, SCORE and SOFSURF in Asian women had differences in specificity of less than 0.1, irrespective of bone mineral density target (two studies). Little difference was seen in diagnostic odds ratios between OST and Quantitative Ultrasound Index T score at the calcaneus (one study) or cortical width and shape derived from dental panoramic radiograms (one study). In African-American women, estimates of diagnostic odds ratio were 9.2 for OST, 7.1 for ORAI, 6.0 for SCORE, and 4.0 for age, body size, no oestrogen (ABONE), when the bone mineral density target was below or equal to -2.0 at the femoral neck (one study). Authors' conclusions In white women, the accuracy of the Osteoporosis Self-Assessment Tool and alternative clinical decision rules were similar, whereas the Stiffness Index was more accurate than the Osteoporosis Self-Assessment Tool. CRD commentary The authors addressed a clear research question supported by appropriate inclusion criteria. Several relevant sources were searched, with attempts were made to reduce language and publication bias. However, the search data up to 2005 was that of a previously published review (see Other Publications of Related Interest); no update searches were performed, so more recent studies may have been missed. Data extraction and the assessment of study quality were carried out in duplicate, but it was unclear whether similar methods to reduce error and bias were employed during study selection.
Most included studied had methodological flaws in terms of patient spectrum, blinding and reporting of uninterpretable results and withdrawals, which impacted on the reliability of the results of these studies, and hence the review.
Given the limitations of the included studies and review, the results of the review should be viewed with caution. Implications of the review for practice and research Practice: The authors stated that the generalisability of the results of the review to clinical practice is uncertain, and other than Stiffness Index, the low specificity of triage tests may mean their use would result in high numbers of referrals for dual-energy X-ray absorptiometry that would not have therapeutic consequences.
Research: The authors recommend future evaluations of triage tests for dual-energy X-ray absorptiometry referral are conducted in a consecutive sample of postmenopausal women attending primary care settings, and studies are reported in accordance with the STARD statement. Bibliographic details Rud B, Hilden J, Hyldstrup L, Hrobjartsson A. The Osteoporosis Self-Assessment Tool versus alternative tests for selecting postmenopausal women for bone mineral density assessment: a comparative systematic review of accuracy. Osteoporosis International 2009; 20(4): 599-607 Other publications of related interest Rud B, Hilden J, Hyldstrup L, Hrobjartsson A. Performance of the Osteoporosis Self-Assessment Tool in ruling out low bone mineral density in postmenopausal women: a systematic review. Osteoporosis International 2007;18(9):1177-1187 Indexing Status Subject indexing assigned by NLM MeSH Absorptiometry, Photon; Aged; Bone Density; Calcaneus /ultrastructure; Female; Humans; Middle Aged; Osteoporosis, Postmenopausal /diagnosis /physiopathology; Patient Selection; Research Design /standards; Risk Assessment /methods; Triage /methods AccessionNumber 12009104197 Date bibliographic record published 18/11/2009 Date abstract record published 24/03/2010 Record Status This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn. |
|
|
|