Journal of Clinical Densitometry: Assessment & Management of Musculoskeletal Health, vol. -, no. -, 1e7, 2015 Ó Copyright 2015 by The International Society for Clinical Densitometry 1094-6950/-:1e7/$36.00 http://dx.doi.org/10.1016/j.jocd.2015.01.001
Original Article
Performance of the Osteoporosis Self-Assessment Tool for Asians (OSTA) in Screening Osteoporosis Among Middle-Aged and Old Women in the Chengdu Region of China Ji-Yuan Huang,* Wen-Zhong Song, Hui-Rong Zeng, Mei Huang, and Qun-Fang Wen Department of Nuclear Medicine, Sichuan Provincial People’s Hospital, Sichuan Academy of Medical Sciences, Chengdu City, Sichuan Province, China
Abstract The purpose of this study was to assess the performance of an osteoporosis self-assessment tool for Asians (OSTA) for screening middle-aged and old healthy women in the Chengdu region of China. OSTA scores were used to evaluate the effect of age on the development of osteoporosis. A total of 15,752 healthy women older than 40 yr, who reside in the Chengdu region, were assigned to 9 age groups divided by 5-yr intervals. Bone mineral density of the lumbar vertebrae, L1eL4, and the left hip was measured with a GE Lunar Prodigy advance dual-energy X-ray absorptiometry system (GE Healthcare, Cincinnati, OH). All subjects were classified into 3 categories according to OSTA cutoff values. The OSTA values used to define the categories were 1, O 4 and ! 1, and 4. The accuracy of the OSTA index was compared with T-scores measured by dual-energy X-ray absorptiometry at various skeletal sites. All statistical analyses were performed using SPSS, version 17.0 (SPSS, Inc., Chicago, IL). The detection rate of osteoporosis and osteopenia at the lumbar vertebrae and the femur was found to increase with age. Furthermore, the sensitivity of OSTA value for detection of osteoporosis also gradually enhanced with subject age. In contrast, the specificity gradually decreased with age. The sensitivity and negative predictive value of the OSTA index for detecting osteoporosis gradually increased with higher OSTA cutoff values, and the specificity and positive predictive value appeared to gradually decline. It was concluded that the OSTA is a useful screening tool to detect osteoporosis in middle-aged and old women in the Chengdu region of China. Furthermore, specific OSTA cutoff values should be selected for screening individuals who belong to different age groups. Key Words: Chengdu region; middle-aged and old women; osteoporosis; osteoporosis self-assessment tool for Asians; screening.
Dual-energy X-ray absorptiometry (DXA) is found to be of great value in the diagnosis, prevention, and control of osteoporosis by being able to identify high-risk populations and assessing the risk of bone fracture (2). However, the relatively few DXA scanners are available, and the high cost of each measurement limit its wide application. Development of rapid, accurate, and simple screening tools to identify populations at high risk for osteoporosis is, therefore, of great practical value for the early detection, diagnosis, and treatment of the disease. Several tools have been used for screening for human osteoporosis (3e5). Among these currently used screening tools, the osteoporosis self-assessment tool for Asians (OSTA) is the simplest and most effective one (6,7).
Introduction With the acceleration of population aging, the prevalence of osteoporosis appears to be gradually increasing, and this greatly affects human health and quality of life. As a highrisk population for osteoporosis and bone fracture, middleaged and old women should receive more attention (1). Received 11/26/14; Revised 12/30/14; Accepted 01/15/15. *Address correspondence to: Ji-Yuan Huang, MD, Department of Nuclear Medicine, Sichuan Provincial People’s Hospital, Sichuan Academy of Medical Sciences, No. 32 West Second Section First Ring Road, Chengdu City, Sichuan Province 610072, China. E-mail: [email protected]
1
2 The OSTA has been shown to be highly effective for identifying the risk of osteoporosis. However, various inadequacies have been reported. The effectiveness of the OSTA risk index was shown to be unsatisfactory for identifying the risk of osteoporosis in Han Chinese, postmenopausal women of the Sichuan Province (8). However, the OSTA index was considered as a simple and effective tool for identifying the risk of osteoporosis in a large selected Han Chinese population in Beijing (9). Many factors are reported to account for the difference in these results, and these include age, ethnicity, subject selection, sample size, and region. The purpose of this study was to assess the effectiveness of the OSTA for screening osteoporosis among middle-aged and old healthy women in the Chengdu region, Sichuan province of China. OSTA scores in various age groups were correlated to subjects with osteoporosis, so as to provide a reference for using this screening tool to diagnose osteoporosis in women residing in the Chengdu region.
Subjects and Methods Subjects A total of 15,752 women receiving health examinations in our hospital during the period from June 2009 through February 2014 were enrolled in this study. The subject age range was from 40 to 96 yr. Participants were excluded for the following disorders revealed by a diagnostic examination and laboratory tests: (1) history of bone fracture or vertebral body fracture detected by X-ray scan; (2) diseases known to potentially induce secondary osteoporosis, including diabetes, thyroid disease, hypothyroidism, thyrotoxicosis, primary or secondary hyperparathyroidism, and other chronic disorders; (3) administration of drugs that may affect bone metabolism; (4) severe liver and renal dysfunction; and (5) surgical resection of the ovary or uterus.
Measurement of BMD and Quality Control Bone mineral density (BMD) was measured with a Lunar Prodigy DXA system (GE Healthcare, Madison, WI), and routine quality control of the equipment was performed each day to make sure the coefficient of variation (CV) of the measurements was less than 1%. A spine phantom was scanned by DXA each week to evaluate long-term precision. Short-term precision was determined through estimation of CV for duplicate BMD measurements at 1 site in each of the 30 selected subjects (10). Areal BMD of lumbar vertebrae, L1eL4, left femoral neck, trochanter and Ward triangle, femoral shaft, and the total hip was measured using a DXA system, with CVs of 0.87%, 0.93%, 0.89%, 1.6%, 0.97%, and 0.99%, respectively. BMD at lumbar spine, total hip, and femoral neck was used for diagnosis as per International Society for Clinical Densitometry recommendations.
Huang et al.
Fig. 1. Age-specific prevalence of osteoporosis in lumbar vertebrae detected by dual-energy X-ray absorptiometry. OP, osteoporosis. score of 4 was defined as high risk; OSTA scores between O 4 and ! 1, as moderate risk (osteopenia); and an OSTA score of 1 as low risk (normal BMD). A BMD T-score of 2.5 was classified as osteoporosis; a T-score of O 2.5 and ! 1 as osteopenia; and T-score of 1 as normal BMD. All subjects were assigned to 9 age groups that were divided by 5-yr intervals. The age groups included 40e44, 45e49, 50e54, 55e59, 60e64, 65e69, 70e74, 75e79, and 80 or greater years.
Statistical Analysis All statistical analyses were performed using SPSS, version 17.0, statistical software (SPSS, Inc., Chicago, IL). BMD T-score, which served as a gold standard for osteoporosis diagnosis, was used to estimate the sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and the area under the receiver operating characteristic curve for the OSTA diagnosis of osteoporosis. Correlation analyses were done with the Pearson correlation analysis. Area under the curve (AUC) was estimated using the nonparametric test by Hanley and McNeil. A p value of !0.05 was considered statistically significant.
OSTA Calculation The OSTA score was calculated using the following formula (6): OSTA 5 0.2 (body weight age). An OSTA
Fig. 2. Age-specific prevalence of femoral osteoporosis detected by dual-energy X-ray absorptiometry. OP, osteoporosis.
Journal of Clinical Densitometry: Assessment & Management of Musculoskeletal Health
Volume
-,
2015
Performance of the OSTA in Screening Osteoporosis
3
Table 1 Comparison of Effectiveness of OSTA Index and T-Score for Screening OP at Lumbar Vertebrae and the Femur Lumbar vertebra T-score
Normal, n (%)
Low risk 5430 Moderate risk 605 High risk 160 Total 6195
Osteopenia, n (%)
Femur
OP, n (%)
Normal, n (%)
n (%)
(56.9) 3214 (33.6) 904 (9.5) 9548 (14.6) 1822 (43.9) 1726 (41.6) 4153 (7.8) 536 (26.1) 1355 (66.1) 2051 (39.3) 5572 (35.4) 3985 (25.3) 15,752
Osteopenia, n (%)
(60.6) 3957 (51.2) 3501 (26.4) 400 (11.3) 2244 (13.0) 48 (2.7) 608 (100.0) 4405 (33.8) 6353
OP, n (%)
n (%)
(45.3) 270 (3.5) 7728 (63.4) 898 (25.3) 3542 (34.7) 1095 (55.5) 1751 (48.8) 2263 (17.4) 13,021
(59.4) (27.2) (13.4) (100.0)
Abbr: OP, osteoporosis; OSTA, osteoporosis self-assessment tool for Asians.
Results The prevalence of osteoporosis and detection of osteopenia at lumbar vertebrae and femurs gradually increased with age. The detection rates were 25.3% and 17.4% for osteoporosis at lumbar vertebrae and femurs, respectively. The detection rate for osteopenia at lumbar vertebrae and femur sites were 35.4% and 48.8%, respectively (Figs. 1 and 2). Age was found to negatively correlate with height, body weight, OSTA score, and BMD T-score. The OSTA index had a significantly positive correlation with T-score, height, and body weight ( p ! 0.001) (Tables 1 and 2). The OSTA index also showed a gradual increased sensitivity and gradual reduced specificity for screening for osteoporosis with increasing age. The sensitivity, specificity, PPV, NPV, and accuracy of OSTA score cutoff values 1 and 1 were 0.3%e99.2%, 2.2%e99.8%, 25%e91.7%, 27.1% e81.5%, and 49.8%e84.3%, respectively, for screening osteoporosis at lumbar vertebra, and 0.7%e99%, 4%e99.9%, 75%e96.5%, 16.7%e69.2%, and 52.1%e94.6%, respectively, for screening femoral osteoporosis (Figs. 3 and 4). The sensitivity and NPV for T-score cutoff values of ! 1 or 2.5 gradually increased with the OSTA index for detecting osteoporosis at lumbar vertebra and femur sites. In contrast, the specificity and PPV gradually decreased. However, the sensitivity and NPV for T-scores 2.5 were higher than those for T-scores ! 1. Lower specificity and PPV for
T-scores 2.5 were observed relative to those for T-scores ! 1 (Tables 3 and 4). AUC used for detecting osteoporosis ranged from 0.653 to 0.789 in patients belonging to different age groups, and the AUC for screening osteoporosis at lumbar vertebra and femurs (T-score cutoff values ! 1 or 2.5) ranged from 0.812 to 0.875 (Table 5).
Discussion BMD, measured by DXA, is currently accepted as the gold standard for the diagnosis of osteoporosis. The International Society for Clinical Densitometry proposes indications for DXA scan and recommends a DXA scan for women who are at or older than 65 yr and women who are perimenopausal/postmenopausal with osteoporosis or risk factors for bone fractures (10). In women older than 40 yr, a decline in estrogen level results in rapid loss of BMD. Our findings show that age has a significantly negative correlation with both the OSTA scores and T-scores used for detecting osteoporosis at lumbar vertebra and femurs (r 5 0.812 and 0.809). These findings indicate that age is an important factor and may affect BMD loss. The detection rates for osteoporosis and low BMD at lumbar vertebrae and femurs were 18.5% and 31%, respectively, in women in the 40e44-yr age group. The incidence of osteoporosis gradually increased with age. This observation suggests that much more attention
Table 2 Pearson Correlation Analysis of Variables Lumbar Vertebrae and the Femur Site of BMD OP at lumbar vertebra
Femoral OP
Variable Age OSTA T-score Age OSTA index T-score
Height 0.374* 0.516* 0.383* 0.360* 0.484* 0.362*
Weight 0.105* 0.665* 0.335* 0.100* 0.665* 0.354*
Age
OSTA
1.000* 0.812* 0.555* 1.000 0.809* 0.604*
0.812* 1.000* 0.614* 0.809* 1.000* 0.662*
T-score 0.555* 0.614* 1.000* 0.604* 0.662* 1.000*
Abbr: BMD, bone mineral density; OP, osteoporosis; OSTA, osteoporosis self-assessment tool for Asians. *p ! 0.001. Journal of Clinical Densitometry: Assessment & Management of Musculoskeletal Health
Volume
-,
2015
4
Fig. 3. Age-specific changes in the parameters for diagnosis of OP in lumbar vertebrae by dual-energy X-ray absorptiometry. The changes in parameters for diagnosis of osteoporosis of lumbar vertebrae by dual-energy X-ray absorptiometry when OSTA values 1 and the T-score ! 1 serve as the cutoff values. accu, accuracy; NPV, negative predictive value; OP, osteoporosis; PPV, positive predictive value; sen, sensitivity; spe, specificity. should be paid to the prevention and control of osteoporosis in women who are at or older than 40 yr. Because of the limited quantity, high equipment cost, and high detection cost, the large-scale use of DXA to screen for osteoporosis seems unrealistic. It appears more feasible to identify subjects at high risk for developing osteoporosis using simple and effective approaches that reflect the BMD indicators detected by DXA. After comparing the effectiveness of 6 commonly used screening tools for osteoporosis, including the Simple Calculated Osteoporosis Risk Estimation, the Osteoporosis Risk Assessment Instrument, the Osteoporosis SelfAssessment Tool (OST), the body weight criterion, the Osteoporosis Index of Risk, and Age, Body Size, No Estrogen, the OSTA was recognized as the simplest approach for
Fig. 4. Age-specific changes in the parameters for diagnosis of femoral OP by dual-energy X-ray absorptiometry. The changes in parameters for diagnosis of hip osteoporosis by dual-energy X-ray absorptiometry when an osteoporosis selfassessment tool for Asians value 1 and T-score ! 1 serve as the cutoff values. accu, accuracy; NPV, negative predictive value; OP, osteoporosis; PPV, positive predictive value; sen, sensitivity; spe, specificity.
Huang et al. identifying individuals at risk for osteoporosis and who should be recommended for additional DXA screening at a clinical or community practice (11). All the aforementioned screening tools were considered useful in identifying postmenopausal Caucasian women at risk for osteoporosis and requiring further DXA screening and possible intervention for osteoporosis. A systematic study to evaluate the performance of the OST for ruling out low BMD in postmenopausal women revealed that the OST performed moderately well in ruling out femoral neck T-scores 2.5 in whites (12). The OST index, which is calculated based on age and body weight, is a method used to screen populations at high risk for developing osteoporosis (6) and has similar diagnostic accuracy with other relatively complicated tools (13,14). In the present study, the sensitivity for OSTA scores with cutoff values 1 and T-score cutoff values ! 1 was 56.9% and 51.2%, respectively, for detecting the presence of osteoporosis at lumbar vertebrae and femurs. The sensitivity of the OSTA increased to 83.6% and 80.7% for screening osteoporosis at lumbar vertebrae and femurs, respectively, in patients older than 65 yr. However, in this older age group, specificity was reduced. This would indicate that OSTA values 1 may be of greater value for screening osteoporosis in women who are older than 60 yr. Furthermore, adjustment of the OSTA index cutoff values may achieve higher effectiveness for diagnosing osteoporosis in women who are younger than 60 yr. To date, there have not been any studies, with a large number of participants, that report the differences in the efficiency of osteoporosis screening tests between various age groups. It has been demonstrated that OST, body weight, and body mass index show a high efficacy to predict femoral neck osteoporosis in women, aged 40e59 yr (15). Among Taiwanese women, aged 30e49 yr, OSTA, body mass index, and body weight can still serve as good indicators for predicting osteoporosis in subjects with an OSTA cutoff value of 2.0 (16). In addition, the OST was found to have greater AUC than Osteoporosis Risk Assessment Instrument and Simple Calculated Osteoporosis Risk Estimation in women who were 67 yr and older and was easier to perform (17). In individuals who are 70 yr and older, the OST also showed some value in screening osteoporosis, with AUC between 0.76 and 0.82 (18). In the present study, low sensitivity, specificity, and AUC for an OSTA cutoff value of 1 were detected for screening osteoporosis at lumbar vertebra and femurs among patients in the 40e59-yr age group. This suggests that an OSTA cutoff value of 1 has a low efficiency to screen osteoporosis in individuals who are younger than 60 yr, and diagnosis of these women requires adjustment of the OSTA cutoff value. The sensitivity of the OSTA index for screening for osteoporosis gradually increased with age. In contrast, the specificity was gradually reduced with age. These findings indicate that there is a significant effect of age on the ability to screen for osteoporosis with the OSTA. Our findings also show that the AUC for the OSTA cutoff value of 1 ranged from 0.702 to 0.789 in patients who were 60 yr and older and demonstrates the value of an OSTA cutoff value of 1 in screening for osteoporosis.
Journal of Clinical Densitometry: Assessment & Management of Musculoskeletal Health
Volume
-,
2015
Performance of the OSTA in Screening Osteoporosis
5
Table 3 Performance of Various OSTA Cutoff Values for Screening for OP at Lumbar Vertebrae T-score cutoff value of ! 1
T-score cutoff value of 2.5
OSTA index Sensitivity (%) Specificity (%) PPV (%) NPV (%) Sensitivity (%) Specificity (%) PPV (%) NPV (%) 3 2 1 0 1 2 3
30.0 42.8 56.9 71.0 83.3 91.5 96.2
95.8 92.9 87.7 78.3 62.7 41.9 22.7
91.6 90.3 87.7 83.5 77.5 70.8 65.8
47.0 51.3 56.9 63.6 70.9 76.1 79.7
47.8 63.3 77.3 88.2 94.8 97.7 99.4
89.6 83.0 73.5 60.8 44.8 28.2 14.8
61.0 55.7 50.0 43.2 36.8 31.6 28.3
83.5 87.0 90.5 93.9 96.2 97.3 98.6
Abbr: NPV, negative predictive value; OP, osteoporosis; OSTA, osteoporosis self-assessment tool for Asians; PPV, positive predictive value.
The OSTA index and T-score vary in their efficiency for screening for osteoporosis. In 4343 postmenopausal women from Mendoza, Argentina, 17.7%e91.6% sensitivity and 30.2%e95.7% specificity were reported for a T-score cutoff value of 2.5 and OST cutoff value between 3 and 3 (19). These previously reported results are significantly different from the results of our present study. In addition, Kamondetdecha et al (20) detected 68.8% sensitivity and 52.2% specificity for OSTA score cutoff values 1 for detecting osteoporosis in the lumbar spine as well as 82.4% sensitivity and 57.2% specificity for detecting osteoporosis at the femoral neck. In the present study, the sensitivity and specificity of OSTA cutoff values of 1 and T-score cutoff values of 2.5 were 77.3% and 73.5%, respectively, for detecting osteoporosis at lumbar vertebrae and 88.1% and 69.3%, respectively, for screening for osteoporosis in the total hip. These sensitivities and specificities are higher than those reported in the previously published studies and may be because of differences in sample size and subject selection. The enhanced sensitivity for detecting osteoporosis with increased OSTA cutoff values, although specificity is reduced, is consistent with a previous study (19). It is generally considered that an AUC of !0.7 indicates a low diagnostic
accuracy, an AUC of 0.7e0.9 indicates moderate diagnostic accuracy, and an AUC of O0.9 indicates a high degree of diagnostic accuracy (21). Our findings show that the AUCs for screening osteoporosis at lumbar vertebrae and femurs were 0.812 and 0.822, respectively and thus, indicate that the OSTA is useful for identifying osteoporosis in middleaged and old women who reside in the Chengdu region. The following causes may be responsible for the differences in the results of the presented osteoporosis screening study with the OSTA and previously published studies. (1) In the study conducted by Koh et al (6), the calculation of T-score was based on femoral neck BMD. However, BMDs of lumbar vertebra, femoral neck, and total hip were used in various studies, and there were significant differences in the detection rate of osteoporosis at these various sites in women. (2) BMD varies with ethnicity, region, and age. Therefore, the selection of study subjects may contribute to the differences found in the screening results. (3) Various sample sizes were used. (4) The selection of T-score and OSTA cutoff values may enhance sensitivity but reduce specificity. Therefore, cutoff values should be determined by multiple factors, which include disease severity, subject populations screened, demographic characteristics, screening efficiency, cost and benefit.
Table 4 Performance of Various OSTA Cutoff Values for Screening for Femoral OP T-score cutoff value of ! 1
T-score cutoff value of 2.5
OSTA index Sensitivity (%) Specificity (%) PPV (%) NPV (%) Sensitivity (%) Specificity (%) PPV (%) NPV (%) 3 2 1 0 1 2 3
29.8 42.4 56.2 69.8 81.8 90.5 96.1
95.9 95.1 89.8 80.0 64.7 44.3 25.2
93.5 94.4 91.5 87.2 82.0 76.1 71.5
41.1 45.8 51.2 57.5 64.6 70.4 76.6
63.9 78.2 88.1 94.7 98.1 99.4 99.8
88.7 80.5 69.3 55.8 40.7 25.6 13.4
54.3 45.8 37.7 31.1 25.8 22.0 19.5
92.1 94.6 96.5 98.1 99.1 99.5 99.7
Abbr: NPV, negative predictive value; OP, osteoporosis; OSTA, osteoporosis self-assessment tool for Asians; PPV, positive predictive value. Journal of Clinical Densitometry: Assessment & Management of Musculoskeletal Health
Volume
-,
2015
6
Huang et al. Table 5 AUC of Lumbar Vertebrae and Femurs in Various Age Groups and Subjects With Different BMD T-Scores
Age group (yr)
No. of patients
AUC of lumbar vertebra
No. of patients
AUC of femur
40e 45e 50e 55e 60e 65e 70e 75e 80e Total
1844 2557 2151 2478 1966 1694 1522 942 598 15,752
0.618* 0.616* 0.672* 0.675* 0.691* 0.715* 0.706* 0.728* 0.727* 0.812*
1483 2000 1681 2067 1662 1444 1309 840 535 13,021
0.654* 0.653* 0.686* 0.669* 0.707* 0.702* 0.707* 0.789* 0.750* 0.822*
AUC 0.812 0.827 0.822 0.875
Significance 0.000 0.000 0.000 0.000
0.805 0.820 0.815 0.868
T-score T-score T-score T-score T-score
! !
1 for OP at lumbar vertebra 2.5 for OP at lumbar vertebra 1 for femoral OP 2.5 for femoral OP
95% CI 0.819 0.834 0.830 0.882
Note: T-score cutoff value of ! 1 was selected for various age groups; the lowest T-score at the femoral neck and total hip was used. Abbr: AUC, area under the curve; BMD, bone mineral density; CI, confidence interval; OP, osteoporosis. *p ! 0.001.
In the present study, healthy subjects were selected to evaluate the performance of the OSTA index for screening osteoporosis among middle-aged and old women of the Chengdu region. There may be selection bias in the present study. However, the sample size is large, and subjects are representative of the population residing in this area. The OSTA is useful to predict women who have low BMD or osteoporosis. Furthermore, age has a significant effect on the ability to detect osteoporosis using the OSTA. In addition, appropriate OSTA cutoff values should be selected for subjects who are optimized for age range. Integration of the OSTA, other screening tools, and risk factors for osteoporosis may enhance the accuracy of screening for osteoporosis.
5.
6.
7. 8. 9.
References 1. Kanis JA, McCloskey EV, Johansson H, et al. Scientific Advisory Board of the European Society for Clinical and Economic Aspects of Osteoporosis and Osteoarthritis (ESCEO) and the Committee of Scientific Advisors of the International Osteoporosis Foundation (IOsF). 2012 European guidance for the diagnosis and management of osteoporosis in postmenopausal women. Osteoporos Int 24:23e57. 2. Blake GM, Fogelman I. 2007 Role of dual-energy X-ray absorptiometry in the diagnosis and treatment of osteoporosis. J Clin Densitom 10:102e110. 3. Lydick E, Cook K, Turpin J, et al. 1998 Development and validation of a simple questionnaire to facilitate identification of women likely to have low bone density. Am J Manag Care 4: 37e48. 4. Cadarette SM, Jaglal SB, Kreiger N, et al. 2000 Development and validation of the Osteoporosis Risk Assessment Instrument
10. 11. 12.
13.
to facilitate selection of women for bone densitometry. CMAJ 162:1289e1294. Sedrine WB, Chevallier T, Zegels B, et al. 2002 Development and assessment of the Osteoporosis Index of Risk (OSIRIS) to facilitate selection of women for bone densitometry. Gynecol Endocrinol 16:245e250. Koh LK, Sedrine WB, Torralba TP, et al. Osteoporosis SelfAssessment Tool for Asians (OSTA) Research Group. 2001 A simple tool to identify Asian women at increased risk of osteoporosis. Osteoporos Int 12:699e705. Chan SP, Teo CC, Ng SA, et al. 2006 Validation of various osteoporosis risk indices in elderly Chinese females in Singapore. Osteoporos Int 17:1182e1188. Lu C, Chen D, Cai Y, Wei S. 2006 Concordane of OSTA and lumbar spine BMD by DXA in identifying risk of osteoporosis. J Orthop Surg 1:14. Yang Y, Wang B, Fei Q, et al. 2013 Validation of an osteoporosis self-assessment tool to identify primary osteoporosis and new osteoporotic vertebral fractures in postmenopausal Chinese women in Beijing. BMC Musculoskelet Disord 14:271. 2013 Official Positions of the International Society for Clinical Densitometry. Available at: http://www.iscd.org/official-positions/ 2013-iscd-official-positions-adult/. Accessed March 10, 2015. McLeod KM, Johnson CS. 2009 Identifying women with low bone mass: a systematic review of screening tools. Geriatr Nurs 30:164e173. Rud B, Hilden J, Hyldstrup L, Hr objartsson A. 2007 Performance of the Osteoporosis Self-Assessment Tool in ruling out low bone mineral density in postmenopausal women: a systematic review. Osteoporos Int 18:1177e1187. Rud B, Hilden J, Hyldstrup L, Hr objartsson A. 2009 The osteoporosis self-assessment tool versus alternative tests for selecting post-menopausal women for bone mineral density assessment: a comparative systematic review of accuracy. Osteoporos Int 20: 599e607.
Journal of Clinical Densitometry: Assessment & Management of Musculoskeletal Health
Volume
-,
2015
Performance of the OSTA in Screening Osteoporosis
7
14. Martinez-Aguila D, Gomez-Vaquero C, Rozadilla A, et al. 2007 Decision rules for selecting women for bone mineral density testing: application in postmenopausal women referred to a bone densitometry unit. J Rheumatol 34:1307e1312. 15. Morin S, Tsang JF, Leslie WD. 2009 Weight and body mass index predict bone mineral density and fractures in women aged 40 to 59 years. Osteoporos Int 20:363e370. 16. Chang SF, Yang RS. 2014 Determining the cut-off point of osteoporosis based on the osteoporosis self-assessment tool, body mass index and weight in Taiwanese young adult women. J Clin Nurs 23:2628e2635. 17. Gourlay ML, Powers JM, Lui LY, Ensrud KE. 2008 Clinical performance of osteoporosis risk assessment tools in women aged 67 years and older. Osteoporos Int 19:1175e1183.
18. Pang WY, Inderjeeth CA. 2014 FRAX without bone mineral density versus osteoporosis self-assessment screening tool as predictors of osteoporosis in primary screening of individuals aged 70 and older. J Am Geriatr Soc 62:442e446. 19. Saravı FD. 2013 Osteoporosis self-assessment tool performance in a large sample of postmenopausal women of Mendoza, Argentina. J Osteoporos 2013:150154. 20. Kamondetdecha R, Panyakhamlerd K, Chaikittisilpa S, et al. 2013 Value of Osteoporosis Self-assessment Tools for Asians (OSTA) with or without Brown’s clinical risk factors in detection of postmenopausal osteoporosis. Climacteric 16:127e132. 21. Hanley JA, McNeil BJ. 1982 The meaning and use of the area under a receiver operating characteristic (ROC) curve. Radiology 143:29e36.
Journal of Clinical Densitometry: Assessment & Management of Musculoskeletal Health
Volume
-,
2015
Original Article
Performance of the Osteoporosis Self-Assessment Tool for Asians (OSTA) in Screening Osteoporosis Among Middle-Aged and Old Women in the Chengdu Region of China Ji-Yuan Huang,* Wen-Zhong Song, Hui-Rong Zeng, Mei Huang, and Qun-Fang Wen Department of Nuclear Medicine, Sichuan Provincial People’s Hospital, Sichuan Academy of Medical Sciences, Chengdu City, Sichuan Province, China
Abstract The purpose of this study was to assess the performance of an osteoporosis self-assessment tool for Asians (OSTA) for screening middle-aged and old healthy women in the Chengdu region of China. OSTA scores were used to evaluate the effect of age on the development of osteoporosis. A total of 15,752 healthy women older than 40 yr, who reside in the Chengdu region, were assigned to 9 age groups divided by 5-yr intervals. Bone mineral density of the lumbar vertebrae, L1eL4, and the left hip was measured with a GE Lunar Prodigy advance dual-energy X-ray absorptiometry system (GE Healthcare, Cincinnati, OH). All subjects were classified into 3 categories according to OSTA cutoff values. The OSTA values used to define the categories were 1, O 4 and ! 1, and 4. The accuracy of the OSTA index was compared with T-scores measured by dual-energy X-ray absorptiometry at various skeletal sites. All statistical analyses were performed using SPSS, version 17.0 (SPSS, Inc., Chicago, IL). The detection rate of osteoporosis and osteopenia at the lumbar vertebrae and the femur was found to increase with age. Furthermore, the sensitivity of OSTA value for detection of osteoporosis also gradually enhanced with subject age. In contrast, the specificity gradually decreased with age. The sensitivity and negative predictive value of the OSTA index for detecting osteoporosis gradually increased with higher OSTA cutoff values, and the specificity and positive predictive value appeared to gradually decline. It was concluded that the OSTA is a useful screening tool to detect osteoporosis in middle-aged and old women in the Chengdu region of China. Furthermore, specific OSTA cutoff values should be selected for screening individuals who belong to different age groups. Key Words: Chengdu region; middle-aged and old women; osteoporosis; osteoporosis self-assessment tool for Asians; screening.
Dual-energy X-ray absorptiometry (DXA) is found to be of great value in the diagnosis, prevention, and control of osteoporosis by being able to identify high-risk populations and assessing the risk of bone fracture (2). However, the relatively few DXA scanners are available, and the high cost of each measurement limit its wide application. Development of rapid, accurate, and simple screening tools to identify populations at high risk for osteoporosis is, therefore, of great practical value for the early detection, diagnosis, and treatment of the disease. Several tools have been used for screening for human osteoporosis (3e5). Among these currently used screening tools, the osteoporosis self-assessment tool for Asians (OSTA) is the simplest and most effective one (6,7).
Introduction With the acceleration of population aging, the prevalence of osteoporosis appears to be gradually increasing, and this greatly affects human health and quality of life. As a highrisk population for osteoporosis and bone fracture, middleaged and old women should receive more attention (1). Received 11/26/14; Revised 12/30/14; Accepted 01/15/15. *Address correspondence to: Ji-Yuan Huang, MD, Department of Nuclear Medicine, Sichuan Provincial People’s Hospital, Sichuan Academy of Medical Sciences, No. 32 West Second Section First Ring Road, Chengdu City, Sichuan Province 610072, China. E-mail: [email protected]
1
2 The OSTA has been shown to be highly effective for identifying the risk of osteoporosis. However, various inadequacies have been reported. The effectiveness of the OSTA risk index was shown to be unsatisfactory for identifying the risk of osteoporosis in Han Chinese, postmenopausal women of the Sichuan Province (8). However, the OSTA index was considered as a simple and effective tool for identifying the risk of osteoporosis in a large selected Han Chinese population in Beijing (9). Many factors are reported to account for the difference in these results, and these include age, ethnicity, subject selection, sample size, and region. The purpose of this study was to assess the effectiveness of the OSTA for screening osteoporosis among middle-aged and old healthy women in the Chengdu region, Sichuan province of China. OSTA scores in various age groups were correlated to subjects with osteoporosis, so as to provide a reference for using this screening tool to diagnose osteoporosis in women residing in the Chengdu region.
Subjects and Methods Subjects A total of 15,752 women receiving health examinations in our hospital during the period from June 2009 through February 2014 were enrolled in this study. The subject age range was from 40 to 96 yr. Participants were excluded for the following disorders revealed by a diagnostic examination and laboratory tests: (1) history of bone fracture or vertebral body fracture detected by X-ray scan; (2) diseases known to potentially induce secondary osteoporosis, including diabetes, thyroid disease, hypothyroidism, thyrotoxicosis, primary or secondary hyperparathyroidism, and other chronic disorders; (3) administration of drugs that may affect bone metabolism; (4) severe liver and renal dysfunction; and (5) surgical resection of the ovary or uterus.
Measurement of BMD and Quality Control Bone mineral density (BMD) was measured with a Lunar Prodigy DXA system (GE Healthcare, Madison, WI), and routine quality control of the equipment was performed each day to make sure the coefficient of variation (CV) of the measurements was less than 1%. A spine phantom was scanned by DXA each week to evaluate long-term precision. Short-term precision was determined through estimation of CV for duplicate BMD measurements at 1 site in each of the 30 selected subjects (10). Areal BMD of lumbar vertebrae, L1eL4, left femoral neck, trochanter and Ward triangle, femoral shaft, and the total hip was measured using a DXA system, with CVs of 0.87%, 0.93%, 0.89%, 1.6%, 0.97%, and 0.99%, respectively. BMD at lumbar spine, total hip, and femoral neck was used for diagnosis as per International Society for Clinical Densitometry recommendations.
Huang et al.
Fig. 1. Age-specific prevalence of osteoporosis in lumbar vertebrae detected by dual-energy X-ray absorptiometry. OP, osteoporosis. score of 4 was defined as high risk; OSTA scores between O 4 and ! 1, as moderate risk (osteopenia); and an OSTA score of 1 as low risk (normal BMD). A BMD T-score of 2.5 was classified as osteoporosis; a T-score of O 2.5 and ! 1 as osteopenia; and T-score of 1 as normal BMD. All subjects were assigned to 9 age groups that were divided by 5-yr intervals. The age groups included 40e44, 45e49, 50e54, 55e59, 60e64, 65e69, 70e74, 75e79, and 80 or greater years.
Statistical Analysis All statistical analyses were performed using SPSS, version 17.0, statistical software (SPSS, Inc., Chicago, IL). BMD T-score, which served as a gold standard for osteoporosis diagnosis, was used to estimate the sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and the area under the receiver operating characteristic curve for the OSTA diagnosis of osteoporosis. Correlation analyses were done with the Pearson correlation analysis. Area under the curve (AUC) was estimated using the nonparametric test by Hanley and McNeil. A p value of !0.05 was considered statistically significant.
OSTA Calculation The OSTA score was calculated using the following formula (6): OSTA 5 0.2 (body weight age). An OSTA
Fig. 2. Age-specific prevalence of femoral osteoporosis detected by dual-energy X-ray absorptiometry. OP, osteoporosis.
Journal of Clinical Densitometry: Assessment & Management of Musculoskeletal Health
Volume
-,
2015
Performance of the OSTA in Screening Osteoporosis
3
Table 1 Comparison of Effectiveness of OSTA Index and T-Score for Screening OP at Lumbar Vertebrae and the Femur Lumbar vertebra T-score
Normal, n (%)
Low risk 5430 Moderate risk 605 High risk 160 Total 6195
Osteopenia, n (%)
Femur
OP, n (%)
Normal, n (%)
n (%)
(56.9) 3214 (33.6) 904 (9.5) 9548 (14.6) 1822 (43.9) 1726 (41.6) 4153 (7.8) 536 (26.1) 1355 (66.1) 2051 (39.3) 5572 (35.4) 3985 (25.3) 15,752
Osteopenia, n (%)
(60.6) 3957 (51.2) 3501 (26.4) 400 (11.3) 2244 (13.0) 48 (2.7) 608 (100.0) 4405 (33.8) 6353
OP, n (%)
n (%)
(45.3) 270 (3.5) 7728 (63.4) 898 (25.3) 3542 (34.7) 1095 (55.5) 1751 (48.8) 2263 (17.4) 13,021
(59.4) (27.2) (13.4) (100.0)
Abbr: OP, osteoporosis; OSTA, osteoporosis self-assessment tool for Asians.
Results The prevalence of osteoporosis and detection of osteopenia at lumbar vertebrae and femurs gradually increased with age. The detection rates were 25.3% and 17.4% for osteoporosis at lumbar vertebrae and femurs, respectively. The detection rate for osteopenia at lumbar vertebrae and femur sites were 35.4% and 48.8%, respectively (Figs. 1 and 2). Age was found to negatively correlate with height, body weight, OSTA score, and BMD T-score. The OSTA index had a significantly positive correlation with T-score, height, and body weight ( p ! 0.001) (Tables 1 and 2). The OSTA index also showed a gradual increased sensitivity and gradual reduced specificity for screening for osteoporosis with increasing age. The sensitivity, specificity, PPV, NPV, and accuracy of OSTA score cutoff values 1 and 1 were 0.3%e99.2%, 2.2%e99.8%, 25%e91.7%, 27.1% e81.5%, and 49.8%e84.3%, respectively, for screening osteoporosis at lumbar vertebra, and 0.7%e99%, 4%e99.9%, 75%e96.5%, 16.7%e69.2%, and 52.1%e94.6%, respectively, for screening femoral osteoporosis (Figs. 3 and 4). The sensitivity and NPV for T-score cutoff values of ! 1 or 2.5 gradually increased with the OSTA index for detecting osteoporosis at lumbar vertebra and femur sites. In contrast, the specificity and PPV gradually decreased. However, the sensitivity and NPV for T-scores 2.5 were higher than those for T-scores ! 1. Lower specificity and PPV for
T-scores 2.5 were observed relative to those for T-scores ! 1 (Tables 3 and 4). AUC used for detecting osteoporosis ranged from 0.653 to 0.789 in patients belonging to different age groups, and the AUC for screening osteoporosis at lumbar vertebra and femurs (T-score cutoff values ! 1 or 2.5) ranged from 0.812 to 0.875 (Table 5).
Discussion BMD, measured by DXA, is currently accepted as the gold standard for the diagnosis of osteoporosis. The International Society for Clinical Densitometry proposes indications for DXA scan and recommends a DXA scan for women who are at or older than 65 yr and women who are perimenopausal/postmenopausal with osteoporosis or risk factors for bone fractures (10). In women older than 40 yr, a decline in estrogen level results in rapid loss of BMD. Our findings show that age has a significantly negative correlation with both the OSTA scores and T-scores used for detecting osteoporosis at lumbar vertebra and femurs (r 5 0.812 and 0.809). These findings indicate that age is an important factor and may affect BMD loss. The detection rates for osteoporosis and low BMD at lumbar vertebrae and femurs were 18.5% and 31%, respectively, in women in the 40e44-yr age group. The incidence of osteoporosis gradually increased with age. This observation suggests that much more attention
Table 2 Pearson Correlation Analysis of Variables Lumbar Vertebrae and the Femur Site of BMD OP at lumbar vertebra
Femoral OP
Variable Age OSTA T-score Age OSTA index T-score
Height 0.374* 0.516* 0.383* 0.360* 0.484* 0.362*
Weight 0.105* 0.665* 0.335* 0.100* 0.665* 0.354*
Age
OSTA
1.000* 0.812* 0.555* 1.000 0.809* 0.604*
0.812* 1.000* 0.614* 0.809* 1.000* 0.662*
T-score 0.555* 0.614* 1.000* 0.604* 0.662* 1.000*
Abbr: BMD, bone mineral density; OP, osteoporosis; OSTA, osteoporosis self-assessment tool for Asians. *p ! 0.001. Journal of Clinical Densitometry: Assessment & Management of Musculoskeletal Health
Volume
-,
2015
4
Fig. 3. Age-specific changes in the parameters for diagnosis of OP in lumbar vertebrae by dual-energy X-ray absorptiometry. The changes in parameters for diagnosis of osteoporosis of lumbar vertebrae by dual-energy X-ray absorptiometry when OSTA values 1 and the T-score ! 1 serve as the cutoff values. accu, accuracy; NPV, negative predictive value; OP, osteoporosis; PPV, positive predictive value; sen, sensitivity; spe, specificity. should be paid to the prevention and control of osteoporosis in women who are at or older than 40 yr. Because of the limited quantity, high equipment cost, and high detection cost, the large-scale use of DXA to screen for osteoporosis seems unrealistic. It appears more feasible to identify subjects at high risk for developing osteoporosis using simple and effective approaches that reflect the BMD indicators detected by DXA. After comparing the effectiveness of 6 commonly used screening tools for osteoporosis, including the Simple Calculated Osteoporosis Risk Estimation, the Osteoporosis Risk Assessment Instrument, the Osteoporosis SelfAssessment Tool (OST), the body weight criterion, the Osteoporosis Index of Risk, and Age, Body Size, No Estrogen, the OSTA was recognized as the simplest approach for
Fig. 4. Age-specific changes in the parameters for diagnosis of femoral OP by dual-energy X-ray absorptiometry. The changes in parameters for diagnosis of hip osteoporosis by dual-energy X-ray absorptiometry when an osteoporosis selfassessment tool for Asians value 1 and T-score ! 1 serve as the cutoff values. accu, accuracy; NPV, negative predictive value; OP, osteoporosis; PPV, positive predictive value; sen, sensitivity; spe, specificity.
Huang et al. identifying individuals at risk for osteoporosis and who should be recommended for additional DXA screening at a clinical or community practice (11). All the aforementioned screening tools were considered useful in identifying postmenopausal Caucasian women at risk for osteoporosis and requiring further DXA screening and possible intervention for osteoporosis. A systematic study to evaluate the performance of the OST for ruling out low BMD in postmenopausal women revealed that the OST performed moderately well in ruling out femoral neck T-scores 2.5 in whites (12). The OST index, which is calculated based on age and body weight, is a method used to screen populations at high risk for developing osteoporosis (6) and has similar diagnostic accuracy with other relatively complicated tools (13,14). In the present study, the sensitivity for OSTA scores with cutoff values 1 and T-score cutoff values ! 1 was 56.9% and 51.2%, respectively, for detecting the presence of osteoporosis at lumbar vertebrae and femurs. The sensitivity of the OSTA increased to 83.6% and 80.7% for screening osteoporosis at lumbar vertebrae and femurs, respectively, in patients older than 65 yr. However, in this older age group, specificity was reduced. This would indicate that OSTA values 1 may be of greater value for screening osteoporosis in women who are older than 60 yr. Furthermore, adjustment of the OSTA index cutoff values may achieve higher effectiveness for diagnosing osteoporosis in women who are younger than 60 yr. To date, there have not been any studies, with a large number of participants, that report the differences in the efficiency of osteoporosis screening tests between various age groups. It has been demonstrated that OST, body weight, and body mass index show a high efficacy to predict femoral neck osteoporosis in women, aged 40e59 yr (15). Among Taiwanese women, aged 30e49 yr, OSTA, body mass index, and body weight can still serve as good indicators for predicting osteoporosis in subjects with an OSTA cutoff value of 2.0 (16). In addition, the OST was found to have greater AUC than Osteoporosis Risk Assessment Instrument and Simple Calculated Osteoporosis Risk Estimation in women who were 67 yr and older and was easier to perform (17). In individuals who are 70 yr and older, the OST also showed some value in screening osteoporosis, with AUC between 0.76 and 0.82 (18). In the present study, low sensitivity, specificity, and AUC for an OSTA cutoff value of 1 were detected for screening osteoporosis at lumbar vertebra and femurs among patients in the 40e59-yr age group. This suggests that an OSTA cutoff value of 1 has a low efficiency to screen osteoporosis in individuals who are younger than 60 yr, and diagnosis of these women requires adjustment of the OSTA cutoff value. The sensitivity of the OSTA index for screening for osteoporosis gradually increased with age. In contrast, the specificity was gradually reduced with age. These findings indicate that there is a significant effect of age on the ability to screen for osteoporosis with the OSTA. Our findings also show that the AUC for the OSTA cutoff value of 1 ranged from 0.702 to 0.789 in patients who were 60 yr and older and demonstrates the value of an OSTA cutoff value of 1 in screening for osteoporosis.
Journal of Clinical Densitometry: Assessment & Management of Musculoskeletal Health
Volume
-,
2015
Performance of the OSTA in Screening Osteoporosis
5
Table 3 Performance of Various OSTA Cutoff Values for Screening for OP at Lumbar Vertebrae T-score cutoff value of ! 1
T-score cutoff value of 2.5
OSTA index Sensitivity (%) Specificity (%) PPV (%) NPV (%) Sensitivity (%) Specificity (%) PPV (%) NPV (%) 3 2 1 0 1 2 3
30.0 42.8 56.9 71.0 83.3 91.5 96.2
95.8 92.9 87.7 78.3 62.7 41.9 22.7
91.6 90.3 87.7 83.5 77.5 70.8 65.8
47.0 51.3 56.9 63.6 70.9 76.1 79.7
47.8 63.3 77.3 88.2 94.8 97.7 99.4
89.6 83.0 73.5 60.8 44.8 28.2 14.8
61.0 55.7 50.0 43.2 36.8 31.6 28.3
83.5 87.0 90.5 93.9 96.2 97.3 98.6
Abbr: NPV, negative predictive value; OP, osteoporosis; OSTA, osteoporosis self-assessment tool for Asians; PPV, positive predictive value.
The OSTA index and T-score vary in their efficiency for screening for osteoporosis. In 4343 postmenopausal women from Mendoza, Argentina, 17.7%e91.6% sensitivity and 30.2%e95.7% specificity were reported for a T-score cutoff value of 2.5 and OST cutoff value between 3 and 3 (19). These previously reported results are significantly different from the results of our present study. In addition, Kamondetdecha et al (20) detected 68.8% sensitivity and 52.2% specificity for OSTA score cutoff values 1 for detecting osteoporosis in the lumbar spine as well as 82.4% sensitivity and 57.2% specificity for detecting osteoporosis at the femoral neck. In the present study, the sensitivity and specificity of OSTA cutoff values of 1 and T-score cutoff values of 2.5 were 77.3% and 73.5%, respectively, for detecting osteoporosis at lumbar vertebrae and 88.1% and 69.3%, respectively, for screening for osteoporosis in the total hip. These sensitivities and specificities are higher than those reported in the previously published studies and may be because of differences in sample size and subject selection. The enhanced sensitivity for detecting osteoporosis with increased OSTA cutoff values, although specificity is reduced, is consistent with a previous study (19). It is generally considered that an AUC of !0.7 indicates a low diagnostic
accuracy, an AUC of 0.7e0.9 indicates moderate diagnostic accuracy, and an AUC of O0.9 indicates a high degree of diagnostic accuracy (21). Our findings show that the AUCs for screening osteoporosis at lumbar vertebrae and femurs were 0.812 and 0.822, respectively and thus, indicate that the OSTA is useful for identifying osteoporosis in middleaged and old women who reside in the Chengdu region. The following causes may be responsible for the differences in the results of the presented osteoporosis screening study with the OSTA and previously published studies. (1) In the study conducted by Koh et al (6), the calculation of T-score was based on femoral neck BMD. However, BMDs of lumbar vertebra, femoral neck, and total hip were used in various studies, and there were significant differences in the detection rate of osteoporosis at these various sites in women. (2) BMD varies with ethnicity, region, and age. Therefore, the selection of study subjects may contribute to the differences found in the screening results. (3) Various sample sizes were used. (4) The selection of T-score and OSTA cutoff values may enhance sensitivity but reduce specificity. Therefore, cutoff values should be determined by multiple factors, which include disease severity, subject populations screened, demographic characteristics, screening efficiency, cost and benefit.
Table 4 Performance of Various OSTA Cutoff Values for Screening for Femoral OP T-score cutoff value of ! 1
T-score cutoff value of 2.5
OSTA index Sensitivity (%) Specificity (%) PPV (%) NPV (%) Sensitivity (%) Specificity (%) PPV (%) NPV (%) 3 2 1 0 1 2 3
29.8 42.4 56.2 69.8 81.8 90.5 96.1
95.9 95.1 89.8 80.0 64.7 44.3 25.2
93.5 94.4 91.5 87.2 82.0 76.1 71.5
41.1 45.8 51.2 57.5 64.6 70.4 76.6
63.9 78.2 88.1 94.7 98.1 99.4 99.8
88.7 80.5 69.3 55.8 40.7 25.6 13.4
54.3 45.8 37.7 31.1 25.8 22.0 19.5
92.1 94.6 96.5 98.1 99.1 99.5 99.7
Abbr: NPV, negative predictive value; OP, osteoporosis; OSTA, osteoporosis self-assessment tool for Asians; PPV, positive predictive value. Journal of Clinical Densitometry: Assessment & Management of Musculoskeletal Health
Volume
-,
2015
6
Huang et al. Table 5 AUC of Lumbar Vertebrae and Femurs in Various Age Groups and Subjects With Different BMD T-Scores
Age group (yr)
No. of patients
AUC of lumbar vertebra
No. of patients
AUC of femur
40e 45e 50e 55e 60e 65e 70e 75e 80e Total
1844 2557 2151 2478 1966 1694 1522 942 598 15,752
0.618* 0.616* 0.672* 0.675* 0.691* 0.715* 0.706* 0.728* 0.727* 0.812*
1483 2000 1681 2067 1662 1444 1309 840 535 13,021
0.654* 0.653* 0.686* 0.669* 0.707* 0.702* 0.707* 0.789* 0.750* 0.822*
AUC 0.812 0.827 0.822 0.875
Significance 0.000 0.000 0.000 0.000
0.805 0.820 0.815 0.868
T-score T-score T-score T-score T-score
! !
1 for OP at lumbar vertebra 2.5 for OP at lumbar vertebra 1 for femoral OP 2.5 for femoral OP
95% CI 0.819 0.834 0.830 0.882
Note: T-score cutoff value of ! 1 was selected for various age groups; the lowest T-score at the femoral neck and total hip was used. Abbr: AUC, area under the curve; BMD, bone mineral density; CI, confidence interval; OP, osteoporosis. *p ! 0.001.
In the present study, healthy subjects were selected to evaluate the performance of the OSTA index for screening osteoporosis among middle-aged and old women of the Chengdu region. There may be selection bias in the present study. However, the sample size is large, and subjects are representative of the population residing in this area. The OSTA is useful to predict women who have low BMD or osteoporosis. Furthermore, age has a significant effect on the ability to detect osteoporosis using the OSTA. In addition, appropriate OSTA cutoff values should be selected for subjects who are optimized for age range. Integration of the OSTA, other screening tools, and risk factors for osteoporosis may enhance the accuracy of screening for osteoporosis.
5.
6.
7. 8. 9.
References 1. Kanis JA, McCloskey EV, Johansson H, et al. Scientific Advisory Board of the European Society for Clinical and Economic Aspects of Osteoporosis and Osteoarthritis (ESCEO) and the Committee of Scientific Advisors of the International Osteoporosis Foundation (IOsF). 2012 European guidance for the diagnosis and management of osteoporosis in postmenopausal women. Osteoporos Int 24:23e57. 2. Blake GM, Fogelman I. 2007 Role of dual-energy X-ray absorptiometry in the diagnosis and treatment of osteoporosis. J Clin Densitom 10:102e110. 3. Lydick E, Cook K, Turpin J, et al. 1998 Development and validation of a simple questionnaire to facilitate identification of women likely to have low bone density. Am J Manag Care 4: 37e48. 4. Cadarette SM, Jaglal SB, Kreiger N, et al. 2000 Development and validation of the Osteoporosis Risk Assessment Instrument
10. 11. 12.
13.
to facilitate selection of women for bone densitometry. CMAJ 162:1289e1294. Sedrine WB, Chevallier T, Zegels B, et al. 2002 Development and assessment of the Osteoporosis Index of Risk (OSIRIS) to facilitate selection of women for bone densitometry. Gynecol Endocrinol 16:245e250. Koh LK, Sedrine WB, Torralba TP, et al. Osteoporosis SelfAssessment Tool for Asians (OSTA) Research Group. 2001 A simple tool to identify Asian women at increased risk of osteoporosis. Osteoporos Int 12:699e705. Chan SP, Teo CC, Ng SA, et al. 2006 Validation of various osteoporosis risk indices in elderly Chinese females in Singapore. Osteoporos Int 17:1182e1188. Lu C, Chen D, Cai Y, Wei S. 2006 Concordane of OSTA and lumbar spine BMD by DXA in identifying risk of osteoporosis. J Orthop Surg 1:14. Yang Y, Wang B, Fei Q, et al. 2013 Validation of an osteoporosis self-assessment tool to identify primary osteoporosis and new osteoporotic vertebral fractures in postmenopausal Chinese women in Beijing. BMC Musculoskelet Disord 14:271. 2013 Official Positions of the International Society for Clinical Densitometry. Available at: http://www.iscd.org/official-positions/ 2013-iscd-official-positions-adult/. Accessed March 10, 2015. McLeod KM, Johnson CS. 2009 Identifying women with low bone mass: a systematic review of screening tools. Geriatr Nurs 30:164e173. Rud B, Hilden J, Hyldstrup L, Hr objartsson A. 2007 Performance of the Osteoporosis Self-Assessment Tool in ruling out low bone mineral density in postmenopausal women: a systematic review. Osteoporos Int 18:1177e1187. Rud B, Hilden J, Hyldstrup L, Hr objartsson A. 2009 The osteoporosis self-assessment tool versus alternative tests for selecting post-menopausal women for bone mineral density assessment: a comparative systematic review of accuracy. Osteoporos Int 20: 599e607.
Journal of Clinical Densitometry: Assessment & Management of Musculoskeletal Health
Volume
-,
2015
Performance of the OSTA in Screening Osteoporosis
7
14. Martinez-Aguila D, Gomez-Vaquero C, Rozadilla A, et al. 2007 Decision rules for selecting women for bone mineral density testing: application in postmenopausal women referred to a bone densitometry unit. J Rheumatol 34:1307e1312. 15. Morin S, Tsang JF, Leslie WD. 2009 Weight and body mass index predict bone mineral density and fractures in women aged 40 to 59 years. Osteoporos Int 20:363e370. 16. Chang SF, Yang RS. 2014 Determining the cut-off point of osteoporosis based on the osteoporosis self-assessment tool, body mass index and weight in Taiwanese young adult women. J Clin Nurs 23:2628e2635. 17. Gourlay ML, Powers JM, Lui LY, Ensrud KE. 2008 Clinical performance of osteoporosis risk assessment tools in women aged 67 years and older. Osteoporos Int 19:1175e1183.
18. Pang WY, Inderjeeth CA. 2014 FRAX without bone mineral density versus osteoporosis self-assessment screening tool as predictors of osteoporosis in primary screening of individuals aged 70 and older. J Am Geriatr Soc 62:442e446. 19. Saravı FD. 2013 Osteoporosis self-assessment tool performance in a large sample of postmenopausal women of Mendoza, Argentina. J Osteoporos 2013:150154. 20. Kamondetdecha R, Panyakhamlerd K, Chaikittisilpa S, et al. 2013 Value of Osteoporosis Self-assessment Tools for Asians (OSTA) with or without Brown’s clinical risk factors in detection of postmenopausal osteoporosis. Climacteric 16:127e132. 21. Hanley JA, McNeil BJ. 1982 The meaning and use of the area under a receiver operating characteristic (ROC) curve. Radiology 143:29e36.
Journal of Clinical Densitometry: Assessment & Management of Musculoskeletal Health
Volume
-,
2015
