Abstract
Objectives
Low bone mineral density (BMD) is a major global health concern due to fracture-related morbidity and mortality. BMD is currently assessed by dual-energy X-ray absorptiometry (DXA) against a US reference population. The aim of the study was to establish an ethnicity- and sex-specific reference for calculating BMD at different femoral sites including femoral trochanter, Ward's triangle and femoral neck in Qatar.
Methods
This cross-sectional study analyzed BMD in 4727 (2277 females and 2450 males) healthy Qatari aged between 20 and 82 years participating in the Qatar Biobank (QBB) study. Standard T-scores provided by the densitometer (TDXA) were compared with ethnicity- and sex-specific T-scores for the Qatari population (TQA) calculated using data from the 20-29-year age group for different femoral sites as the reference. The concordance between TDXA and TQA was analyzed using kappa statistics.
Results
Males consistently exhibited higher BMD values for the femoral trochanter, Ward's triangle, and the femoral neck across all age groups compared with females. Using TQA rather than TDXA as a reference at any site classified fewer individuals as having osteoporosis and osteopenia, especially for females. Agreement between TDXA and TQA was higher in males than in females.
Conclusions
These findings underscore the need for local ethnicity- and sex-specific BMD reference values. The use of TQA significantly reduced overdiagnosis of osteopenia and osteoporosis compared with TDXA, thereby decreasing overtreatment and impacting participant recruitment decisions into clinical studies.
Keywords: Bone mineral density, Osteoporosis, Reference values, Qatar Biobank, Population
1. Introduction
Osteoporosis is a systemic skeletal disease characterized by low bone mineral density (BMD) and disruption of the bone microarchitecture through trabecular thinning, decreased cortical thickness, and a deterioration of bone tissues. These processes ultimately compromise bone strength and increase the risk of fracture [1]. Osteoporosis is underpinned by imbalanced bone homeostasis caused by decreased bone formation or increased bone resorption [2]. It is a global and increasing health problem, with the International Osteoporosis Foundation (IOF) estimating that osteoporosis causes over 8.9 million fractures annually worldwide and that one-third of females and one-fifth of males over 50 years will experience an osteoporotic fracture in their lifetime [3].
Therefore, the early identification and detection of bone pathology are critical for the prevention of bone loss and the effective treatment of conditions like osteoporosis. Pathological changes in bone structure and density are assessed using a variety of clinical and laboratory techniques that allow a diagnosis of osteoporosis, including single or dual photon absorptiometry, quantitative compound tomography, neutron and proton activation analysis, and quantitative magnetic resonance [4]. Of these, dual X-ray absorptiometry (DXA), which is accurate, precise, and has low radiation exposure, is the gold standard for measuring peripheral and central BMD and for diagnosing and monitoring osteoporosis [[5], [6], [7]]. Using DXA, BMD is typically measured at specific sites, including the lumbar vertebrae, femoral neck, and hip, although this can lead to variability in peak BMD values according to the non-uniform measurement locations specified in the WHO's diagnostic criteria for osteoporosis [2].
An individual's BMD is commonly expressed as a T-score [8], calculated by subtracting the mean BMD of the reference population from the patient's BMD and dividing this difference by the standard deviation (SD) of the reference population [9]. In many countries, T-score levels defined by the WHO serves as both diagnostic and intervention thresholds. The WHO reference standard is based on females in the US participating in the third National Health and Nutrition Evaluation Survey (NHANES III), regardless of race or sex/gender, and the reference is based on the assumption that peak femoral BMD is typically reached at around 25 years of age [10]. The International Society for Clinical Densitometry (ISCD) report that fractures occur at comparable BMDs in males and females, so they suggest using this uniform White, non-ethnicity-adjusted female reference for both sexes [10,11]. Most research on osteoporosis has focused on females due to their greater risk of osteoporosis and subsequent fractures [12]. However, given that the genetic, nutritional, cultural, and lifestyle factors that affect bone health vary significantly between populations, using a young White reference in an ethnically quite different population such as that seen in Qatar could lead to overdiagnosis or underdiagnosis of osteoporosis.
Therefore, the objectives of this study were to: 1) describe the BMD distribution at various femoral sites in males and females; 2) identify sex- and age-specific reference values for calculating T-scores for the Qatari population; and 3) analyze the prevalence of osteopenia and osteoporosis in the Qatari population based on the newly established T-scores.
2. Methods
2.1. Ethical approval
The Qatar Biobank Institutional Review Board (E−2021-QF-QBB-RES-ACC-00050-0172) and Qatar University Institutional Review Board (1648-E/22) granted ethical approval for this study. All participants provided informed consent to share their data for research governed by an information security management system implemented by QBB to guarantee the confidentiality and integrity of all participant information [13].
2.2. Study population
Data from Qatar Biobank (QBB) was used for this cross-sectional study. The inclusion criteria were healthy Qatari, without a history of major chronic conditions, aged between 20 and 82 years, the wide age range chosen to track BMD progression across different life stages within the Qatari population. Participants with chronic illnesses known to impact BMD such as diabetes mellitus, all cancers, thyroid diseases, inflammatory bowel diseases such as Crohn's disease and ulcerative colitis, prior fractures, autoimmune diseases, arthritis, and gastrointestinal surgeries were excluded. Participants underwent comprehensive health assessments for a wide range of factors including lifestyle and sociodemographic characteristics [13,14].
2.3. BMD measurements
BMD was assessed by certified professionals at the QBB using the Lunar Prodigy (GE Healthcare, Madison, WI, USA) system with Encore Software. BMD measurements were taken from various femoral sites prone to fractures including the femoral trochanter, Ward's triangle, and the femoral neck, which are crucial for diagnosing and evaluating fracture risk. DXA measurements from male and female participants, segmented into specific age categories, were analyzed to determine T-scores. Osteoporosis was diagnosed based on WHO criteria, where T-scores less than or equal to −1 were normal, between −1 and −2.5 indicative of low bone mass (osteopenia), and −2.5 or lower as osteoporosis [1,10,15]. In this study, T-scores were compared based on two reference values: (i) the standard reference, denoted (TDXA), was based on the established standard from the NHANES III data from young White female adults, which serves as a benchmark for international comparison as per the manufacturer's recommendation and current practice within the QBB; and (ii) the local Qatar reference, denoted (TQA), derived from the mean and standard deviation (SD) of BMD from males and females separately in the 20–29 years age group to calculate sex-specific T-scores. This age group was selected as the reference population in accordance with WHO and ISCD guidelines, which recommend defining T-scores based on BMD in young adults. This standardization allowed for consistency across studies and facilitated comparison with internationally accepted thresholds.
2.4. Data analysis
For comparison with NHANES, we combined BMD data into standard age groups similar to other studies (20–29, 30–39, 40–49, 50–59 and ≥ 60) for the three femoral sites (femoral neck, trochanter, and Ward's triangle), which show distinct BMD patterns according to sex. Participants ≥ 60 years were grouped separately.
Data were analyzed using Stata/SE v18 software. Data were preprocessed and cleaned with management of missing values, comprehensive review and removal of outliers, and assessment of skew and normality to ensure the quality of the data prior to further analysis. Then, descriptive statistics were computed for normalized variables, expressing data as means and standard deviations (SD) for 4727 participants stratified by sex and age group. Using age group, femoral site, sex, and BMD values, T-scores were calculated and classified according to WHO criteria into normal, osteopenia, or osteoporosis. Statistical agreement between TDXA and TQA was analyzed using kappa statistics with 95% confidence intervals (CI).
3. Results
3.1. Demographic characteristics of the study population
BMD data from 4727 participants were analyzed, and their demographic characteristics are summarized in Table 1. The cohort included 2277 females (48.2%) and 2450 males (51.8%) with a mean (SD) age of 34.6 (10.2) years for females and 35.2 (10.0) years for males.
Table 1.
Baseline characteristics of study participants (N = 4727). Values are presented as mean (SD).
| Variable | Female N = 2277 (48.2%) |
Male N = 2450 (51.8%) |
P-value |
|---|---|---|---|
| Age, yrs | 34.63 (10.17) | 35.16 (9.93) | 0.05 |
| BMI, kg/m2 | 28.30 (6.27) | 28.15 (5.36) | 0.88 |
| Height, cm | 158.40 (5.79) | 173.01 (6.48) | <0.05 |
| Weight, kg | 70.97 (16.13) | 84.41 (17.52) | <0.05 |
| Ward's triangle BMD, g/cm2 | 0.804 (0.146) | 0.885 (0.174) | <0.05 |
| Femoral trochanter BMD, g/cm2 | 0.767 (0.118) | 0.877 (0.135) | <0.05 |
| Femoral neck BMD, g/cm2 | 0.942 (0.124) | 1.035 (0.152) | <0.05 |
3.2. BMD in Qatar
Males consistently exhibited higher BMD values across all age groups compared with females (Table 2), with the highest BMD values observed in males aged 20–29 at the femoral neck (BMD 1.072 (0.163) g/cm2). The highest BMD in females was within the 40–49 age group at the femoral neck (0.959 (0.126) g/cm2). BMD measured at the femoral trochanter was higher in males than in females, peaking at 0.914 (0.127) g/cm2 in the 50–59 age group and in females at 0.806 (0.119) g/cm2 in the 40–49 age group. The highest mean BMD at Ward's triangle was in the 20–29 age group for both sexes, with a BMD of 0.952 (0.182) g/cm2 in males and 0.835 (0.142) g/cm2 in females.
Table 2.
BMD measurements across three different femoral sites in males and females. Values are presented as mean (SD). BMD values are in g/cm2.
| Age group | Count | Femoral trochanter | Ward's triangle | Femoral neck |
|---|---|---|---|---|
| Female | ||||
| 20–29 | 839 | 0.744 (0.117) | 0.835 (0.142) | 0.936 (0.125) |
| 30–39 | 739 | 0.764 (0.113) | 0.800 (0.137) | 0.948 (0.121) |
| 40–49 | 475 | 0.806 (0.119) | 0.795 (0.153) | 0.959 (0.126) |
| 50–59 | 193 | 0.785 (0.112) | 0.733 (0.133) | 0.915 (0.120) |
| ≥60 | 31 | 0.748 (0.100) | 0.644 (0.116) | 0.863 (0.114) |
| Male | ||||
| 20–29 | 845 | 0.878 (0.144) | 0.952 (0.182) | 1.072 (0.163) |
| 30–39 | 873 | 0.867 (0.127) | 0.875 (0.158) | 1.030 (0.141) |
| 40–49 | 525 | 0.880 (0.132) | 0.830 (0.154) | 1.007 (0.143) |
| 50–59 | 157 | 0.914 (0.127) | 0.815 (0.141) | 1.006 (0.133) |
| ≥60 | 50 | 0.861 (0.155) | 0.705 (0.158) | 0.910 (0.156) |
3.3. T-score calculations for the Qatari population
The local Qatari population aged 20–29 years was used as a reference group to calculate ethnicity- and sex-specific T-scores, denoted "TQA", which were compared with T-scores provided by the densitometer (TDXA). The number of individuals with a low femoral BMD calculated using TQA was lower than that determined by TDXA. At the femoral trochanter, more males and a significantly larger proportion of females were identified as having a normal bone density using TQA than with TDXA. Furthermore, 0.9% of males were defined as having osteoporosis using TDXA compared to only 0.1% using TQA (P < 0.05; Fig. 1A). Similarly, 3.2% of females had osteoporosis at the femoral trochanter using TDXA compared with only 0.2% using TQA (P < 0.05; Fig. 1B). At Ward's triangle, 3.10% of males had osteoporosis using TDXA compared with only 0.30% using TQA (P < 0.05; Fig. 1C). Similarly, 4.90% of females had osteoporosis at the femoral trochanter using TDXA compared with 0.40% using TQA (P < 0.05; Fig. 1D). At the femoral neck, TDXA defined 80.3% of males as having a normal bone density compared with 79.6% by TQA (P = 0.5; Fig. 1E). By contrast, 62.7% of females had a normal BMD by TDXA compared with 85.2% by TQA (P < 0.05; Fig. 1F). The prevalence of diagnoses for the entire study population divided into age groups is presented in Table 3, Table 4 for females and males, respectively.
Fig. 1.
Comparative prevalence of osteopenia and osteoporosis measured at the femoral trochanter (A, males; B females), Ward's triangle (C, males; D females), and femoral neck (E, males; F females) using different standards.
TDXA, T-scores provided by the densitometer; TQA, T-scores for the Qatari population.
Table 3.
Prevalence of osteopenia and osteoporosis in the female Qatari population using both the TQA and TDXA according to different age groups.
| Age Group | Condition | TQA |
TDXA |
||||
|---|---|---|---|---|---|---|---|
| Femoral trochanter N (%) | Ward's triangle N (%) | Femoral neck N (%) | Femoral trochanter N (%) | Ward's triangle N (%) | Femoral neck N (%) | ||
| 20–29 | Normal | 711 (84.7) | 707 (84.3) | 709 (84.5) | 444 (52.9) | 538 (64.1) | 510 (60.8) |
| Osteopenia | 123 (14.7) | 130 (15.5) | 128 (15.3) | 354 (42.2) | 279 (33.3) | 312 (37.2) | |
| Osteoporosis | 5 (0.6) | 2 (0.2) | 2 (0.2) | 41 (4.9) | 22 (2.6) | 17 (2.0) | |
| 30–39 | Normal | 660 (89.3) | 581 (78.6) | 639 (86.5) | 436 (59.0) | 388 (52.5) | 478 (64.7) |
| Osteopenia | 79 (10.7) | 158 (21.4) | 100 (13.5) | 282 (38.2) | 322 (43.6) | 259 (35.0) | |
| Osteoporosis | 0 (0.0) | 0 (0) | 0 (0) | 21 (2.8) | 29 (3.9) | 2 (0.3) | |
| 40–49 | Normal | 455 (95.8) | 356 (74.9) | 425 (89.5) | 347 (73.1) | 241 (50.7) | 318 (66.9) |
| Osteopenia | 20 (4.2) | 119 (25.1) | 50 (10.5) | 123 (25.9) | 211 (44.4) | 154 (32.4) | |
| Osteoporosis | 0 (0.0) | 0 (0) | 0 (0) | 5 (1.1) | 23 (4.8) | 3 (0.6) | |
| 50–59 | Normal | 181 (93.8) | 117 (60.6) | 149 (77.2) | 132 (68.4) | 74 (38.3) | 111 (57.5) |
| Osteopenia | 12 (6.2) | 71 (36.8) | 43 (22.3) | 56 (29.0) | 93 (48.2) | 76 (39.4) | |
| Osteoporosis | 0 (0.0) | 5 (2.6) | 1 (0.5) | 5 (2.6) | 26 (13.5) | 6 (3.1) | |
| ≥60 | Normal | 28 (90.3) | 10 (32.3) | 19 (61.3) | 16 (51.6) | 3 (9.7) | 10 (32.3) |
| Osteopenia | 3 (9.7) | 19 (61.3) | 12 (38.7) | 14 (45.2) | 17 (54.8) | 20 (64.5) | |
| Osteoporosis | 0 (0.0) | 2 (6.5) | 0 (0) | 1 (3.2) | 11 (35.5) | 1 (3.2) | |
Table 4.
Prevalence of osteopenia and osteoporosis in the male Qatari population using both the TQA and TDXA according to different age groups.
| Age Group | Condition | TQA |
TDXA |
||||
|---|---|---|---|---|---|---|---|
| Femoral trochanter N (%) | Ward's triangle N (%) | Femoral neck N (%) | Femoral trochanter N (%) | Ward's triangle N (%) | Femoral neck N (%) | ||
| 20–29 | Normal | 719 (85.1) | 713 (84.4) | 709 (83.9) | 690 (81.7) | 692 (81.9) | 716 (84.7) |
| Osteopenia | 125 (14.8) | 131 (15.5) | 135 (16.0) | 148 (17.5) | 146 (17.3) | 128 (15.1) | |
| Osteoporosis | 1 (0.1) | 1 (0.1) | 1 (0.1) | 7 (0.8) | 7 (0.8) | 1 (0.1) | |
| 30–39 | Normal | 741 (84.9) | 659 (75.5) | 703 (80.5) | 706 (80.9) | 620 (71.0) | 708 (81.1) |
| Osteopenia | 132 (15.1) | 214 (24.5) | 170 (19.5) | 157 (18.0) | 227 (26.0) | 160 (18.3) | |
| Osteoporosis | 0 (0) | 0 (0) | 0 (0) | 10 (1.1) | 26 (3.0) | 5 (0.6) | |
| 40–49 | Normal | 460 (87.6) | 318 (60.6) | 395 (75.2) | 435 (82.9) | 294 (56.0) | 396 (75.4) |
| Osteopenia | 65 (12.4) | 207 (39.4) | 130 (24.8) | 87 (16.6) | 207 (39.4) | 125 (23.8) | |
| Osteoporosis | 0 (0) | 0 (0) | 0 (0) | 3 (0.6) | 24 (4.6) | 4 (0.8) | |
| 50–59 | Normal | 144 (91.7) | 95 (60.5) | 122 (77.7) | 140 (89.2) | 90 (57.3) | 124 (79.0) |
| Osteopenia | 13 (8.3) | 61 (38.9) | 35 (22.3) | 17 (10.8) | 61 (38.9) | 33 (21.0) | |
| Osteoporosis | 0 (0) | 1 (0.6) | 0 (0) | 0 (0.0) | 6 (3.8) | 0 (0) | |
| ≥60 | Normal | 38 (76.0) | 13 (26.0) | 22 (44.0) | 37 (74.0) | 12 (24.0) | 24 (48.0) |
| Osteopenia | 11 (22.0) | 31 (62.0) | 27 (54.0) | 11 (22.0) | 25 (50.0) | 22 (44.0) | |
| Osteoporosis | 1 (2.0) | 6 (12.0) | 1 (2.0) | 2 (4.0) | 13 (26.0) | 4 (8.0) | |
3.4. Agreement of diagnosis
Kappa (κω) statistics with 95% CI were used to evaluate the diagnostic agreement between TDXA and TQA at the three femoral sites (Table 5). There were varying levels of agreement across different age groups and sexes according to standard definitions [16]. In females, there was slight to moderate agreement across all femoral sites. For individuals aged 20–29, the κω was 0.324 for the femoral trochanter, 0.431 at the femoral neck, and 0.478 at Ward's triangle, and the agreement was similar for the other age groups. The lowest κω of 0.192 (95% CI: 0.000–0.446) was in the ≥ 60 age group at the femoral trochanter, and the highest was at the femoral neck for the 50–59 age group, with a κω of 0.550 (95% CI: 0.446–0.656).
Table 5.
Diagnostic agreement for osteoporosis at different femoral sites for TDXA and TQA.
| Femoral trochanter TQA | Femoral neck TQA | Ward's triangle TQA | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Gender | Age group | Normal | Osteopenia | Osteoporosis | Kappa (BC 95% CI) | Normal | Osteopenia | Osteoporosis | Kappa (BC 95% CI) | Normal | Osteopenia | Osteoporosis | Kappa (BC 95% CI) | ||
| TDXA | Female | 20–29 | Normal | 444 | 0 | 0 | 0.324 (0.279–0.371) | 510 | 0 | 0 | 0.431 (0.379–0.485) | 538 | 0 | 0 | 0.478 (0.425–0.530) |
| Osteopenia | 267 | 87 | 0 | 199 | 113 | 0 | 169 | 110 | 0 | ||||||
| Osteoporosis | 0 | 36 | 5 | 0 | 15 | 2 | 0 | 20 | 2 | ||||||
| 30–39 | Normal | 436 | 0 | 0 | 0.276 (0.229–0.327) | 478 | 0 | 0 | 0.442 (0.382–0.503) | 388 | 0 | 0 | 0.428 (0.377–0.479) | ||
| Osteopenia | 224 | 58 | 0 | 161 | 98 | 0 | 193 | 129 | 0 | ||||||
| Osteoporosis | 0 | 21 | 0 | 0 | 2 | 0 | 0 | 29 | 0 | ||||||
| 40–49 | Normal | 347 | 0 | 0 | 0.205 (0.133–0.283) | 318 | 0 | 0 | 0.378 (0.301–0.459) | 241 | 0 | 0 | 0.467 (0.406–0.525) | ||
| Osteopenia | 108 | 15 | 0 | 107 | 47 | 0 | 115 | 96 | 0 | ||||||
| Osteoporosis | 0 | 5 | 0 | 0 | 3 | 0 | 0 | 23 | 0 | ||||||
| 50–59 | Normal | 132 | 0 | 0 | 0.233 (0.132–0.350) | 111 | 0 | 0 | 0.550 (0.446–0.656) | 74 | 0 | 0 | 0.511 (0.428–0.599) | ||
| Osteopenia | 49 | 7 | 0 | 38 | 38 | 0 | 43 | 50 | 0 | ||||||
| Osteoporosis | 0 | 5 | 0 | 0 | 5 | 1 | 0 | 21 | 5 | ||||||
| ≥60 | Normal | 16 | 0 | 0 | 0.192 (0.000–0.446) | 10 | 0 | 0 | 0.436 (0.236–0.701) | 3 | 0 | 0 | 0.293 (0.104–0.526) | ||
| Osteopenia | 12 | 2 | 0 | 9 | 11 | 0 | 7 | 10 | 0 | ||||||
| Osteoporosis | 0 | 1 | 0 | 0 | 1 | 0 | 0 | 9 | 2 | ||||||
| Male | 20–29 | Normal | 690 | 0 | 0 | 0.856 (0.810–0.897) | 705 | 11 | 0 | 0.933 (0.898–0.963) | 692 | 0 | 0 | 0.890 (0.847–0.925) | |
| Osteopenia | 29 | 119 | 0 | 4 | 124 | 0 | 21 | 125 | 0 | ||||||
| Osteoporosis | 0 | 6 | 1 | 0 | 0 | 1 | 0 | 6 | 1 | ||||||
| 30–39 | Normal | 706 | 0 | 0 | 0.826 (0.778–0.868) | 692 | 16 | 0 | 0.884 (0.842–0.920) | 620 | 0 | 0 | 0.824 (0.787–0.860) | ||
| Osteopenia | 35 | 122 | 0 | 11 | 149 | 0 | 39 | 188 | 0 | ||||||
| Osteoporosis | 0 | 10 | 0 | 0 | 5 | 0 | 0 | 26 | 0 | ||||||
| 40–49 | Normal | 435 | 0 | 0 | 0.794 (0.717–0.862) | 385 | 11 | 0 | 0.874 (0.824–0.919) | 294 | 0 | 0 | 0.828 (0.785–0.868) | ||
| Osteopenia | 25 | 62 | 0 | 10 | 115 | 0 | 24 | 183 | 0 | ||||||
| Osteoporosis | 0 | 3 | 0 | 0 | 4 | 0 | 0 | 24 | 0 | ||||||
| 50–59 | Normal | 140 | 0 | 0 | 0.853 (0.712–0.994) | 120 | 4 | 0 | 0.887 (0.799–0.976) | 90 | 0 | 0 | 0.880 (0.808–0.940) | ||
| Osteopenia | 4 | 13 | 0 | 2 | 31 | 0 | 5 | 56 | 0 | ||||||
| Osteoporosis | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 5 | 1 | ||||||
| ≥60 | Normal | 37 | 0 | 0 | 0.908 (0.749–1.000) | 22 | 2 | 0 | 0.832 (0.693–0.963) | 12 | 0 | 0 | 0.769 (0.609–0.906) | ||
| Osteopenia | 1 | 10 | 0 | 0 | 22 | 0 | 1 | 24 | 0 | ||||||
| Osteoporosis | 0 | 1 | 1 | 0 | 3 | 1 | 0 | 7 | 6 | ||||||
By contrast, males exhibited higher kappa values and thus better agreement between the two references for all femoral sites, ranging from substantial to several instances of almost perfect agreement. For instance, the lowest κω of 0.769 (95% CI: 0.609–0.906) was observed in the ≥ 60 age group at Ward's triangle, and the highest κω of 0.933 (95% CI: 0.898–0.963) was in the 20–29 age group at the femoral neck. Therefore, stronger agreement between TDXA and TQA was observed in males, suggesting closer alignment between the classification compared to females.
4. Discussion
Bone mineral density testing is an essential component of the diagnosis and treatment of bone disorders, especially osteoporosis. The standardized reference values used worldwide for T-score calculation are generated from a young healthy White population aged 20–29 years [1]. These values cannot adequately reflect bone health state across diverse ethnic and geographic populations [17]. Numerus studies have suggested that BMD varies according to geographical area, age, ethnicity, and sex/gender groups [18]. For instance, T-scores were found to be lower in Gulf countries including Oman, United Arab Emirates (UAE), and Qatar than the reference White female population used for DXA measurements, with differences observed even between these countries [19,20]. Recognizing this gap, many countries have set up their own reference standards based on their populations and that reflect their citizens' distinct bone health profiles [[21], [22], [23]]. By taking a similar strategy, Qatar could enhance the precision of bone health evaluations, resulting in more targeted healthcare treatments and reducing the risk of over- or underdiagnosis of osteoporosis and osteopenia.
We detected a significant discrepancy in BMD values obtained from the TDXA reference and our estimated TQA values, with further differences according to demographics. For instance, the mean TQA BMD at the femoral neck was 0.936 (0.125) for females and 1.072 (0.163) for males, considerably lower than the TDXA-derived reference values of 1.038 (0.139) for females but the same [1.070 (0.130)] in males [24]. This finding is consistent with similar studies performed in Indian, Lebanese, and other Asian populations, further emphasizing the need for local BMD reference values [17,21,25]. These differences also had clinical consequences. Using the international TDXA reference resulted in a significantly higher prevalence of osteoporosis, particularly among females, compared to the local TQA reference. This suggests that the standard reference may not be appropriately calibrated to the Qatari population and could lead to overdiagnosis. This was also observed in a Vietnamese study, where the application of NHANES reference values resulted in 35% overdiagnosis of osteoporosis and osteopenia compared with using a local reference [8]. Overdiagnosis could result in unnecessary management and patient anxiety as well as erroneously increasing the prevalence of osteoporosis in public health data [12].
Males in our cohort generally had higher mean BMD values than females, and our TQA-referenced BMD values revealed sex-specific patterns that are important for understanding bone health throughout life stages (Table 3, Table 4). In females, we observed that BMD values increased during early adulthood, peaking in the 40–49 age group at the femoral neck and trochanter, and in the 30–39 age group at Ward's triangle, before declining in older age groups. Indeed, this disparity was emphasized in the 50–59 age group, with females experiencing a more dramatic decrease in BMD together with a higher prevalence of osteopenia. In individuals aged ≥ 60 years, we observed high rates of normal BMD, although this result was influenced by the small sample size within this subgroup. This small cohort of individuals aged ≥ 60 years with relatively good BMD may be also explained by changes in lifestyles in Qatar, who are not so exposed to unhealthy behaviors such as smoking, caffeine, and carbonated beverage consumption, which are endemic in young adults in Qatar [28,29]. Our observations are also consistent with studies showing that males typically have higher peak bone mass and experience bone loss at a slower rate as they age compared with females, although BMD in males peaked later, at 50–59 years, at the femoral trochanter [30]. Sex-related decreases in bone density are well recognized and are largely attributable to hormonal changes, particularly those occurring before and after menopause, which place women at greater risk of reduced BMD and bone fractures [20,26,27,31].
The observed sex-related differences became even more obvious when comparing bone health between the two references across the entire study cohort. For example, when comparing the values derived from the femoral trochanter, there was a significant 29% difference in normal BMD in females versus 3.8% in males for the two sets of measurements, equating to a 26% and 3% decrease in osteopenia prevalence when using TQA for females and males, respectively, and a 3% and 0.8% decrease in osteoporosis prevalence for females and males, respectively. This trend towards overestimation was also observed for Ward's triangle and femoral neck measurements. Over-diagnosis is a widespread and well-recognized problem in non-Western cultures when standard reference values are utilized [21,25,32]. By contrast, a study from Ireland reported that the standard NHANES III reference values are suitable for other populations, arguing that this standard reflects the pattern seen in a healthy population [24]. Nevertheless, this approach assumes that the standard reference is sufficiently similar to the characteristics of the local population.
Assessment of agreement between values derived using the two references revealed fair to moderate agreement for females, suggesting the potential for overestimation of the diagnosis, and even greater agreement for males. Nevertheless, by adopting the TQA standard, Qatar can more accurately reflect its population's bone health characteristics, increasing the accuracy of diagnosis and the appropriateness of subsequent treatments.
This study has some limitations. This study did not take into consideration other confounding variables that could affect BMD, such as BMI, lifestyle, vitamin and drug use, dietary habits, and genetic predisposition. In particular, given that the study population exhibited a relatively high BMI and considering the known correlation between BMI and BMD, future studies should explore this relationship, including across age groups. Another limitation was the relatively small number of our participants over the age of 60, which reduces the generalizability of our findings to older populations, who are at particularly high risk for osteoporosis. Therefore, future studies should include a broader representation of older individuals and a wider range of variables to gain more accurate insights into bone health in the Qatari population. Future studies should also incorporate additional clinical indicators such as fracture incidence to validate the diagnostic performance and clinical applicability of the TQA reference values. Finally, in our analysis we did not specifically assess the age of peak BMD within the Qatari population; therefore, our use of the 20–29 age group as the reference, while consistent with WHO/ISCD recommendations, assumes that this age range represents peak BMD locally, which may not be fully validated.
5. Conclusions
This study highlights the importance of developing localized BMD reference values to accurately assess the risk of osteoporosis and bone health in the Qatari population. Using a large cohort of healthy participants in the QBB, we compared BMD values derived using the standard reference with a new Qatari reference derived from the same age category. Implementation of the TQA reference values in Qatar could improve the accuracy of diagnosis and hence the management of patients with osteopenia and osteoporosis in Qatar. However, more studies are needed to analyze the new reference values in patients with osteopenia, osteoporosis and bone fractions.
CRediT author statement
Fawziya Al-Baker: Statistical analysis, Writing – original draft, Writing – review & editing. Mujahed Shraim: Methodology, Statistical analysis, Writing – review & editing. Amal Al-Haidose: Supervision, Funding acquisition, Writing – original draft, Writing – review & editing. Atiyeh M. Abdallah: Conceptualization, Methodology, Supervision, Funding acquisition, Writing – review & editing.
Conflicts of interest
The authors declare no competing interests.
Acknowledgments
This publication was supported by Qatar University, internal grants No. QUCG-CHS-25/26-716, QUST-1-CHS-2025-212 and QUST-1-CHS-2025-245. The findings presented herein are solely the responsibility of the authors. ORCID Fawziya Al-Baker: 0009-0003-0454-9711. Mohammed Al-Hamdani: 0000-0002-2527-4492. Mujahed Shraim: 0000-0001-7972-8210. Atiyeh Abdallah: 0000-0003-4812-956X.
References
- 1.Sözen T, Özışık L, Başaran NÇ. An overview and management of osteoporosis. Eur J Rheumatol. 2017;4:46–56. doi: 10.5152/eurjrheum.2016.048. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Zhao Y, Chen H, Qiu F, He J, Chen J. Cognitive impairment and risks of osteoporosis: a systematic review and meta-analysis. Arch Gerontol Geriatr. 2023;106 doi: 10.1016/j.archger.2022.104879. [DOI] [PubMed] [Google Scholar]
- 3.Chotiyarnwong P, McCloskey EV, Harvey NC, Lorentzon M, Prieto-Alhambra D, Abrahamsen B, et al. Is it time to consider population screening for fracture risk in postmenopausal women? A position paper from the International Osteoporosis Foundation Epidemiology/Quality of Life Working Group. Arch Osteoporosis. 2022;17:87. doi: 10.1007/s11657-022-01117-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Manisali M., Ozaksoy D., Yilmaz E., Senocak O., Tatari H., Baran O., et al. Bone mineral density reference values in the normal female and male population of Izmir, Turkey. Eur Radiol. 2003;13:157–162. doi: 10.1007/s00330-002-1371-9. [DOI] [PubMed] [Google Scholar]
- 5.Al-Bogami MM, Akanle OA, Aldawood S, Alkhorayef M, Sulieman A, Jawad AS, et al. Comparison of bone mineral density changes between male and female osteoporosis patients using dual energy X-ray absorptiometry scan. Appl Radiat Isot Data Instrum Methods Use Agric Ind Med. 2023;194 doi: 10.1016/j.apradiso.2022.110624. [DOI] [PubMed] [Google Scholar]
- 6.Golding PH. Dual-energy X-ray absorptiometry (DXA) to measure bone mineral density (BMD) for diagnosis of osteoporosis - experimental data from artificial vertebrae confirms significant dependence on bone size. BoneKEy Rep. 2022;17 doi: 10.1016/j.bonr.2022.101607. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Carey JJ, Delaney MF. Utility of DXA for monitoring, technical aspects of DXA BMD measurement and precision testing. Bone. 2017;104:44–53. doi: 10.1016/j.bone.2017.05.021. [DOI] [PubMed] [Google Scholar]
- 8.Ho-Pham LT, T Nguyen UD, Pham HN, Nguyen ND, Nguyen TV. Reference ranges for bone mineral density and prevalence of osteoporosis in Vietnamese men and women. BMC Muscoskelet Disord. 2011;12:182. doi: 10.1186/1471-2474-12-182. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Lundstam K, Godang K, Pretorius M, Markwardt P, Hellström M, Bollerslev J, et al. The influence of DXA hardware, software, reference population and software analysis settings on the bone mineral density and T-score relationship. J Clin Densitom Off J Int Soc Clin Densitom. 2022;25:24–33. doi: 10.1016/j.jocd.2021.07.002. [DOI] [PubMed] [Google Scholar]
- 10.Xue S, Zhang Y, Qiao W, Zhao Q, Guo D, Li B, et al. An updated reference for calculating bone mineral density T-scores. J Clin Endocrinol Metab. 2021;106:e2613–e2621. doi: 10.1210/clinem/dgab180. [DOI] [PubMed] [Google Scholar]
- 11.Shuhart CR, Yeap SS, Anderson PA, Jankowski LG, Lewiecki EM, Morse LR, et al. Executive summary of the 2019 ISCD position development conference on monitoring treatment, DXA cross-calibration and least significant change, spinal cord injury, peri-prosthetic and orthopedic bone health, transgender medicine, and pediatrics. J Clin Densitom Off J Int Soc Clin Densitom. 2019;22:453–471. doi: 10.1016/j.jocd.2019.07.001. [DOI] [PubMed] [Google Scholar]
- 12.Salari N, Ghasemi H, Mohammadi L, Behzadi M.H, Rabieenia E, Shohaimi S, et al. The global prevalence of osteoporosis in the world: a comprehensive systematic review and meta-analysis. J Orthop Surg. 2021;16:609. doi: 10.1186/s13018-021-02772-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Al Kuwari H, Al Thani A, Al Marri A, Al Kaabi A, Abderrahim H, Afifi N, et al. The Qatar Biobank: background and methods. BMC Public Health. 2015;15:1208. doi: 10.1186/s12889-015-2522-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Younes N, Syed N, Yadav SK, Haris M, Abdallah AM, Abu-Madi M. A whole-genome sequencing association study of low bone mineral density identifies new susceptibility loci in the phase I Qatar Biobank cohort. J Personalized Med. 2021;11:34. doi: 10.3390/jpm11010034. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Professor Kanis JA, Melton LJ, Christiansen C, Johnston CC, Khaltaev N. The diagnosis of osteoporosis. J Bone Miner Res. 1994;9:1137–1141. doi: 10.1002/jbmr.5650090802. [DOI] [PubMed] [Google Scholar]
- 16.Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics. 1977;33:159–174. [PubMed] [Google Scholar]
- 17.Wáng YXJ, Blake GM, Xiao B-H, Guglielmi G, Su Y, Jiang Y, et al. East Asians’ T-scores for the diagnosis of osteoporosis should be calculated using ethnicity- and gender-specific BMD reference ranges: justifications. Skelet Radiol. 2024;53:409–417. doi: 10.1007/s00256-023-04423-3. [DOI] [PubMed] [Google Scholar]
- 18.Gerber LM, Bener A, Al-Ali HM, Hammoudeh M, Liu LQ, Verjee M. Bone mineral density in midlife women: the Study of Women’s Health in Qatar. Climacteric J Int Menopause Soc. 2015;18:316–322. doi: 10.3109/13697137.2014.944495. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Bener A, Hammoudeh M, Zirie M. Prevalence and predictors of osteoporosis and the impact of life style factors on bone mineral density. APLAR J Rheumatol. 2007;10:227–233. [Google Scholar]
- 20.Yeap SS, Thambiah SC, Samsudin IN, Appannah G, Zainuddin N, Mohamad-Ismuddin S, et al. Different reference ranges affect the prevalence of osteoporosis and osteopenia in an urban adult Malaysian population. Osteoporos Sarcopenia. 2020;6:168–172. doi: 10.1016/j.afos.2020.11.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Aggarwal A, Pal R, Bhadada SK, Ram S, Garg A, Bhansali A, et al. Bone mineral density in healthy adult Indian population: the chandigarh urban bone epidemiological study (CUBES) Arch Osteoporosis. 2021;16:17. doi: 10.1007/s11657-020-00868-4. [DOI] [PubMed] [Google Scholar]
- 22.Lee S, Choi M-G, Yu J, Ryu O-H, Yoo HJ, Ihm S-H, et al. The effects of the Korean reference value on the prevalence of osteoporosis and the prediction of fracture risk. BMC Muscoskelet Disord. 2015;16:69. doi: 10.1186/s12891-015-0523-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Lynn HS, Lau EMC, Au B, Leung PC. Bone mineral density reference norms for Hong Kong Chinese. Osteoporos Int. 2005;16:1663–1668. doi: 10.1007/s00198-005-1899-z. [DOI] [PubMed] [Google Scholar]
- 24.E E, Wang T, Yang L, Dempsey M, Brennan A, Yu M, et al. How does proximal femur BMD of healthy Irish adults compare to NHANES III? Results of the DXA-HIP Project. Arch Osteoporosis. 2021;16:170. doi: 10.1007/s11657-021-01034-0. [DOI] [PubMed] [Google Scholar]
- 25.Maalouf G, Salem S, Sandid M, Attallah P, Eid J, Saliba N, et al. Bone mineral density of the Lebanese reference population. Osteoporos Int. 2000;11:756–764. doi: 10.1007/s001980070054. [DOI] [PubMed] [Google Scholar]
- 26.Choi YJ, Oh HJ, Kim DJ, Lee Y, Chung Y-S. The prevalence of osteoporosis in Korean adults aged 50 years or older and the higher diagnosis rates in women who were beneficiaries of a national screening program: the Korea National Health and Nutrition Examination Survey 2008–2009. J Bone Miner Res. 2012;27:1879–1886. doi: 10.1002/jbmr.1635. [DOI] [PubMed] [Google Scholar]
- 27.Lo JC, Kim S, Chandra M, Ettinger B. Applying ethnic-specific bone mineral density T-scores to Chinese women in the USA. Osteoporos Int. 2016;27:3477–3484. doi: 10.1007/s00198-016-3673-9. [DOI] [PubMed] [Google Scholar]
- 28.Hamid A, Shi Z, Thalib L. Association between soft drink consumption and bone mineral density among Qatari women- analysis of Qatar Biobank data. Univ. Future Re-Imagining Res. High. Educ. 2020:208. [Google Scholar]
- 29.AlMulla A, Mamtani R, Cheema S, Maisonneuve P, Abdullah BaSuhai J, Mahmoud G, et al. Epidemiology of tobacco use in Qatar: prevalence and its associated factors. PLoS One. 2021;16 doi: 10.1371/journal.pone.0250065. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Zeng Q, Li N, Wang Q, Feng J, Sun D, Zhang Q, et al. The prevalence of osteoporosis in China, a nationwide, multicenter DXA Survey. J Bone Miner Res. 2019;34:1789–1797. doi: 10.1002/jbmr.3757. [DOI] [PubMed] [Google Scholar]
- 31.Fahimfar N, Noorali S, Yousefi S, Gharibzadeh S, Shafiee G, Panahi N, et al. Prevalence of osteoporosis among the elderly population of Iran. Arch Osteoporosis. 2021;16:16. doi: 10.1007/s11657-020-00872-8. [DOI] [PubMed] [Google Scholar]
- 32.Marwaha RK, Tandon N, Kaur P, Sastry A, Bhadra K, Narang A, et al. Establishment of age-specified bone mineral density reference range for Indian females using dual-energy X-ray absorptiometry. J Clin Densitom. 2012;15:241–249. doi: 10.1016/j.jocd.2011.09.006. [DOI] [PubMed] [Google Scholar]