Application of a Standing Quad-Plate Bioelectrical Impedance Analyzer for Whole-Body Bone Mineral Density Measurement in Osteoporosis Screening

Application of a Standing Quad-Plate Bioelectrical Impedance Analyzer for Whole-Body Bone Mineral Density Measurement in Osteoporosis Screening | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Application of a Standing Quad-Plate Bioelectrical Impedance Analyzer for Whole-Body Bone Mineral Density Measurement in Osteoporosis Screening Hsueh-Kuan Lu, Ai-Chun Huang, Chien-Wei Liang, Tzu-Jung Huang, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9314239/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 10 You are reading this latest preprint version Abstract Background Dual-energy X-ray absorptiometry (DXA) is the gold standard for diagnosing osteoporosis but is limited by high cost, bulky size, and lack of portability. The SA201 is a novel bioelectrical impedance analysis (BIA) device designed to estimate whole-body bone mineral density (BMD). This study aimed to evaluate the accuracy of SA201 measurements and their agreement with DXA. Methods A total of 178 participants (56 men and 122 women; mean age 52.6 ± 14.8 years) underwent whole-body BMD assessments using both SA201 and DXA. Statistical analyses included descriptive statistics, Pearson correlation, intraclass correlation coefficient (ICC), Passing–Bablok regression, Bland–Altman analysis, paired t-tests, Wilcoxon signed-rank tests, coefficient of variation (CV%), mean absolute percentage error (MAPE), root mean square percentage error (RMSPE), and receiver operating characteristic (ROC) curve analysis. Results BIA-derived BMD was strongly correlated with DXA (r = 0.819, p < 0.001; ICC (2,1) = 0.896, indicating good reliability). Paired t-tests (p = 0.603) and Wilcoxon tests (p = 0.690) showed no significant mean differences. Bland–Altman analysis demonstrated a mean difference of − 0.003 g/cm² with 95% limits of agreement between − 0.166 and 0.160 g/cm². Passing–Bablok regression produced the equation: SA201 = 0.361 + 0.661 × DXA, with wide confidence intervals. Error metrics indicated CV% ≈ 7.7%, MAPE = 6.5%, and RMSPE = 8.3%. For osteoporosis diagnosis, ROC analysis yielded an AUC of 0.906, with an optimal cutoff of 0.969 g/cm² (sensitivity 88.2%, specificity 87.0%). Conclusions SA201 demonstrated strong correlation and good agreement with DXA for whole-body BMD measurement, as well as good discriminatory ability in osteoporosis screening (AUC > 0.9). However, individual-level errors of 6–8% and wide regression confidence intervals suggest limited interchangeability. BIA may serve as a practical tool for community screening and epidemiological research, while DXA remains indispensable for definitive clinical diagnosis of osteoporosis. Health sciences/Biomarkers Health sciences/Diseases Health sciences/Endocrinology Health sciences/Health care Health sciences/Medical research Z-score Dual-energy X-ray absorptiometry (DXA) Whole-body bone mineral density (BMD) Bioelectrical impedance analysis (BIA) Agreement Figures Figure 1 Figure 2 Figure 3 Introduction Osteoporosis is an age-related metabolic bone disease characterized by reduced bone mass and deterioration of the microarchitecture, leading to increased bone fragility and a higher risk of fractures [ 1 , 2 ]. According to the World Health Organization (WHO), more than 200 million people worldwide are affected by osteoporosis, and the associated medical and social costs are steadily increasing. Therefore, early diagnosis and screening have significant clinical and public health importance [ 3 ]. Bone mineral density (BMD) is widely recognized as the core indicator for diagnosing osteoporosis and predicting fracture risk [ 4 , 5 ]. According to the WHO definition, a Z-score ≤ − 2.5 indicates osteoporosis [ 6 ]. Numerous prospective studies have demonstrated a strong association between reduced BMD and fragility fracture risk, with each standard deviation (SD) decrease in BMD corresponding to an approximately 1.5–3-fold increase in fracture risk [ 7 , 8 ]. Consequently, BMD measurement has become one of the most critical clinical tools in strategies for osteoporosis prevention and treatment. BMD reflects the mineral content of bone (mainly hydroxyapatite), and its reduction is related to excessive bone resorption and insufficient bone formation [ 9 ]. The core mechanisms of bone metabolism involve an imbalance between increased osteoclast activity and decreased osteoblast activity, leading to trabecular thinning and microarchitectural deterioration [ 10 ]. Both clinical and basic studies have shown that estrogen deficiency, aging, chronic inflammation, and hormonal dysregulation (e.g., parathyroid hormone imbalance, vitamin D deficiency) accelerate bone loss [ 11 – 13 ]. Accurate BMD measurement therefore enables early detection of osteoporosis and serves as a key indicator for treatment monitoring and drug efficacy evaluation in clinical trials [ 14 , 15 ]. FRAX, QFracture, and the Garvan calculator integrate BMD with multiple clinical risk factors—including age, history of fragility fractures, glucocorticoid use, smoking, alcohol consumption, and secondary causes of osteoporosis. Therefore, even if a device is capable of estimating whole-body BMD, its ability to predict fracture risk cannot be inferred directly unless such estimates are combined with the relevant clinical data. Currently, several modalities are available for the assessment of bone mineral density (BMD), each with distinct advantages and limitations. Dual-energy X-ray absorptiometry (DXA) remains the clinical gold standard for BMD evaluation, providing highly accurate measurements of both whole-body and regional sites (e.g., lumbar spine, femoral neck) \(\:\left[\text{4,16}\right]\) . Despite its diagnostic reliability, DXA is constrained by its relatively high equipment cost, large footprint, the need for trained radiology personnel, and unavoidable exposure to low-dose ionizing radiation, which collectively hinder its implementation in primary care and community-based settings \(\:\left[17\right]\) . Quantitative computed tomography (QCT) offers the unique capability of separately quantifying cortical and trabecular bone with three-dimensional precision \(\:\left[18\right]\) . However, its clinical adoption is limited by substantially higher radiation exposure and operational costs compared with DXA \(\:\left[19\right]\) . Peripheral quantitative computed tomography (pQCT) provides volumetric assessment of peripheral skeletal sites such as the forearm and lower leg using relatively simple instrumentation \(\:\left[20\right]\) . Nevertheless, because pQCT is restricted to peripheral regions, it cannot yield whole-body or axial skeletal information \(\:\left[21\right]\) .Quantitative ultrasound (QUS) is a non-invasive, radiation-free, portable, and cost-efficient technique that makes it particularly suitable for primary care and large-scale population screening \(\:\left[22\right]\) . Its diagnostic utility, however, is influenced by the selected measurement site—typically the calcaneus—and its correlation with DXA-derived BMD remains inconsistent and subject to ongoing debate \(\:\left[\text{23,24}\right]\) . Bioelectrical impedance analysis (BIA) is another radiation-free, rapid, portable, and inexpensive alternative, rendering it attractive for community-based screening and epidemiological research \(\:\left[\text{25,26}\right]\) . Although increasingly explored for BMD estimation, its accuracy and clinical interchangeability with DXA remain to be fully established. Previous studies employing BIA to estimate BMD have predominantly used single-frequency foot-to-foot devices, such as those in the Tanita system, and some have utilized multi-frequency InBody analyzers; however, none of these instruments were originally designed for bone density assessment. Most of these investigations derived body composition parameters through regression-based prediction equations, yielding correlations that ranged from moderate to high (r ≈ 0.40–0.90) depending on the variable examined (e.g., body fat percentage, muscle mass, or bone mineral content) [ 25 ]. Nevertheless, the magnitude of error varied substantially across studies, and these levels of accuracy are insufficient for clinical diagnostic purposes. The estimation of whole-body BMD using bioelectrical impedance is based on electrical characteristics such as current pathways, limb geometry, tissue conductivity, the distribution of bone and soft tissues, and inferred mineral content. The predictive models for this technique were developed using databases of Taiwanese and broader Asian adults aged 20–80 years. Model calibration and training were performed with whole-body BMD obtained from DXA (GE Lunar Prodigy) as the reference standard. The initial model was constructed using multiple regression analyses supplemented by cross-validation. Several studies have reported that BIA-derived parameters—such as phase angle and impedance index—are correlated with BMD. For example, Chuang et al. [ 27 ] reported moderate to high correlations between whole-body BMD measured by DXA and values estimated using the prototype BIA device. Nonetheless, its application in BMD measurement remains relatively novel and requires further validation. Compared with DXA, BIA offers advantages of non-invasiveness, ease of operation, and lower cost. However, its measurement accuracy and consistency with DXA require rigorous scientific validation. Therefore, the present study aimed to systematically evaluate the differences and agreement between BIA and DXA in whole-body BMD measurement using multiple statistical methods—including correlation analysis, Bland–Altman plots, Passing–Bablok regression, intraclass correlation coefficients (ICC), agreement testing, and ROC analysis—and to explore the feasibility of BIA as a tool for osteoporosis screening. Methods Study Design and Participants This cross-sectional study was conducted between January 2023 and September 2023 and was approved by the Institutional Review Board (IRB) of Nantou Hospital, Ministry of Health and Welfare, Taiwan (Approval No. IRB-111047 and IRB-113002). All procedures were performed in accordance with relevant ethical guidelines, and written informed consent was obtained from all participants prior to enrollment. A total of 178 adults (56 men and 122 women), aged 25–86 years, who were able to stand and walk independently, were recruited at Puzi Hospital, Ministry of Health and Welfare, through poster advertisements and oral invitations. Exclusion criteria included individuals who had undergone surgeries that could potentially alter body composition (e.g., bariatric surgery). Pre-Assessment Protocol Pre-Assessment Protocol To minimize measurement variability, participants were required to adhere to the following pre-test instructions: Avoid strenuous exercise within 48 hours prior to testing while maintaining regular dietary habits. Refrain from consuming alcohol or high-caffeine beverages within 48 hours before testing. Avoid moderate-to-vigorous physical activity within 12 hours prior to testing. Empty the bladder immediately before measurement. Discontinue the use of diuretics for at least 7 days prior to testing. Avoid nuclear medicine examinations or contrast-enhanced imaging within 5 days before testing. Female participants undergoing menstruation were rescheduled; pregnant or potentially pregnant women were excluded. During the assessment, participants wore cotton gowns and underwear and removed all metallic objects (e.g., rings, earrings, zippers, buttons) that could interfere with X-ray scanning. All measurements were performed between 9:00 a.m. and 12:00 noon to minimize the influence of diurnal variation. All BIA and DXA measurements were completed during the same study visit. The DXA and BIA assessments were performed independently by different operators in separate rooms, and the measurement outputs were processed separately before being merged for analysis. Anthropometric Measurements Body height was measured to the nearest 0.1 cm using a digital stadiometer (Jenix DS-102, Dong Sang Jen Ix Co., Ltd., South Korea). Body weight was measured using the built-in scale of the BIA device. Bioelectrical Impedance Analysis (BIA) Whole-body BMD was measured using the SA201 foot-to-foot bioelectrical impedance analyzer (StarBIA MediTek Co., Taichung, Taiwan). The device employs a four-electrode system and dual-frequency alternating currents (5 and 50 kHz). Measurements were performed according to the manufacturer’s instructions. Participants stood barefoot on the electrode plates for approximately 3 minutes, and data acquisition (including body weight and impedance) was completed within one minute. A total of 30 participants underwent repeated BIA measurements. All retests were performed within the same testing session, with a 5-minute interval between measurements, during which participants were required to step off the device and reposition themselves. The SA201 device operated using the reference database established by the manufacturer. Dual-Energy X-ray Absorptiometry (DXA) DXA scans were performed using the GE Lunar Prodigy Advance system (GE Medical Systems Lunar, Madison, WI, USA) with enCORE software version 13.50.0. During the scan, participants lay supine with arms positioned alongside the body and palms facing downward. All scans and calibrations were conducted by trained technicians according to the standards of the International Society for Clinical Densitometry (ISCD). Precision analysis demonstrated that the root mean square standard deviation (RMS-SD) of whole-body BMD was 0.006 g/cm², corresponding to a coefficient of variation (CV) of 0.63%, with a least significant change (LSC) of 0.022 g/cm² (1.73%). DXA repeat measurements were not performed in this study, as the DXA had already established precision values—least significant change (LSC) and root-mean-square standard deviation (RMS-SD)—in accordance with ISCD guidelines. The GE Lunar Prodigy Advance scanner employed the GE Lunar reference database, which is primarily based on White male and female data from the National Health and Nutrition Examination Survey (NHANES). This reference standard is used in the GE enCORE software for calculating Z-scores and complies with ISCD recommendations. Statistical Analysis All data were presented as mean ± standard deviation, minimum, and maximum values. The Kolmogorov–Smirnov test was used to evaluate the normality of data distribution. Differences in whole-body BMD between BIA and DXA were compared using paired t-tests and Wilcoxon signed-rank tests. Linear associations were assessed using Pearson correlation coefficients, while reliability was evaluated by calculating intraclass correlation coefficients (ICC, model 2,1). Agreement between methods was further examined using Bland–Altman analysis (mean difference ± 1.96 SD) [ 28 ] and Passing–Bablok regression to assess fixed and proportional errors [ 29 , 30 ]. The sample size of 178 exceeded the minimum threshold of 30–50 cases recommended in the literature, as well as the 100 cases required for equivalence testing, thereby ensuring sufficient precision for regression coefficient estimates [ 31 ]. Error indices included the coefficient of variation (CV%), mean absolute percentage error (MAPE), and root mean square percentage error (RMSPE). Diagnostic validity for osteoporosis was evaluated using receiver operating characteristic (ROC) curve analysis. All statistical analyses were performed using SPSS version 20.0 (IBM Corp., Armonk, NY, USA), with statistical significance set at p < 0.05. Results A total of 178 participants (56 men and 122 women) were included in this study. The descriptive characteristics of the participants are presented in Table 1 . The mean age was 52.6 years (SD = 14.8, range = 25–86 years). For the primary continuous variables (DXA-derived BMD and BIA-derived BMD), both BMD DXA and BMD BIA showed no violation of normality, as indicated by the Kolmogorov–Smirnov test (p > 0.05). Table 1 Baseline Characteristics and Body Composition of Participants Age (yrs) All (n = 178) Female (n = 122) Male (n = 56) P 52.63 ± 14.83 25.0-86.1 53.71 ± 13.41 25.0-86.1 50.27 ± 17.4 25.0–82.0 < 0.05 Height (cm) 160.7 ± 8.0 143.0-185.5 156.9 ± 5.3 143.0-175.0 168.9 ± 6.6 152.0-185.0 < 0.001 Weight (kg) 61.5 ± 12.8 33.1-107.8 56.5 ± 9.3 33.1–88.5 72.5 ± 12.8 48.2-107.8 < 0.001 BMI (kg/m 2 ) 23.7 ± 3.8 13.3–39.6 22.9 ± 3.7 13.3–31.3 25.3 ± 3.8 17.3–39.6 < 0.01 BPF(%) 32.7 ± 7.5 7.3–48.7 33.6 ± 6.7 7.3–46.8 30.7 ± 8.6 7.4–48.7 < 0.01 BMC(kg) 2.12(0.55 1.19–3.82 2.1 ± 0.5 1.2–3.8 2.3 ± 0.7 1.36–3.6 < 0.01 BMD BIA (g/cm 2 ) 1.07 ± 0.12 0.78–1.37 1.03 ± 0.09 0.78–1.25 1.16 ± 0.12 0.87–1.37 < 0.01 BMD DXA (g/cm 2 ) 1.08 ± 0.14 0.74–1.49 1.03 ± 0.13 0.74–1.31 1.17 ± 0.13 0.94–1.49 < 0.01 Z-score BIA -1.64 ± 1.53 -4.13-1.74 -1.67 ± 1.36 -4.13-1.58 -1.58 ± 1.87 -4.10-1.74 < 0.05 Z-score DXA -0.79 ± 1.23 -3.67-2.44 -0.89 ± 1.24 -3.67-1.78 -0.56 ± 1.19 -2.75-2.44 < 0.001 Notes All values are mean ± SDs; • BMI = Body Mass Index • BFP = Body Fat Percentage • BMC = Bone Mineral Content • BMD = Bone Mineral Density • Subscripts BIA and DXA refer to measurements obtained using the SA201 bioelectrical impedance analyzer and dual-energy X-ray absorptiometry , respectively. As shown in Table 2 , the classification of osteoporosis (Z-score ≤ − 2.5) differed between DXA and BIA. In the total sample (n = 178), DXA identified 17 participants (9.6%) with osteoporosis, whereas BIA classified 55 participants (30.9%) as osteoporotic. Among women, DXA identified 14 cases (11.5%) compared with 35 cases (28.7%) by BIA; among men, DXA identified 3 cases (5.4%) compared with 20 cases (35.7%) by BIA. Overall, BIA demonstrated a higher detection rate of osteoporosis than DXA, with a particularly pronounced difference observed in men. Supplementary Figure S1 shows the scatter plot and regression line of whole-body BMD measured by SA201 and DXA, indicating a strong linear relationship (regression equation: BIA = 0.354 + 0.668 × DXA, R² = 0.671). The Passing–Bablok regression (BIA = 0.361 + 0.661 × DXA) revealed that the 95% confidence intervals for slope and intercept included 1 and 0, respectively, indicating no significant proportional or fixed bias, although the intervals were wide (Fig. 1 ). The intraclass correlation coefficient (ICC) between BIA and DXA was 0.896, and the Pearson correlation coefficient was r = 0.819 (p < 0.001), indicating a high degree of correlation. Supplementary Figure S2 illustrates the distribution of percentage differences between BIA and DXA, showing that the mean percentage difference was close to zero, although individual differences reached ± 10%. Bland–Altman analysis (Fig. 2 ) demonstrated a mean difference of − 0.003 g/cm², with 95% limits of agreement ranging from − 0.166 to 0.160 g/cm², and most observations fell within ± 0.16 g/cm². According to the criteria of Koo and Li, the ICC value of 0.896 represents “good” agreement, though insufficient for full interchangeability. Paired t-test results (t = − 0.521, p = 0.603) and Wilcoxon signed-rank test results (W = 7691.000, p = 0.690) indicated no statistically significant differences in mean values between BIA and DXA, confirming consistency at the group level. Error indices showed a coefficient of variation (CV%) of 7.72%, suggesting approximately 7–8% inter-individual variability. The mean absolute percentage error (MAPE) was 6.39%, indicating an average deviation of about 6.5% between BIA and DXA. The root mean square percentage error (RMSPE) was 7.83%, reflecting an error magnitude of approximately 8% at the individual level, with larger deviations exerting greater influence. Receiver operating characteristic (ROC) curve analysis (Fig. 3 ) demonstrated that BIA achieved an area under the curve (AUC) of 0.906, while DXA achieved an AUC of 0.979. Although DXA performed nearly perfectly (AUC ≈ 0.98), BIA also demonstrated good diagnostic validity (AUC ≈ 0.91). The optimal cutoff point for BIA was 0.969 g/cm², with a sensitivity of 88.2% and a specificity of 87.0%. Table 2. Osteoporosis Classification of Participants by DXA and BIA (Z-score ≤ −2.5) Z-score of -2.5 Group Method (osteoporosis) (no osteoporosis) All DXA 17 161 BIA 55 123 Female DXA 14 108 BIA 35 98 Male DXA 3 53 BIA 20 36 Notes Osteoporosis was defined as Z-score ≤ −2.5 according to the WHO diagnostic criteria [3,6]. DXA = Dual-energy X-ray absorptiometry (gold standard). BIA = SA201 bioelectrical impedance analyzer (test method). Discussion The present study demonstrated that BIA exhibited a high correlation and good agreement with DXA in whole-body BMD measurement, and showed favorable discriminatory ability for diagnosing osteoporosis. Both the paired t-test and Wilcoxon signed-rank test revealed no significant mean differences between the two methods, while Bland–Altman analysis indicated that most data points fell within the ± 0.16 g/cm² limits of agreement. These findings support the interchangeability of BIA and DXA at the population level. Such results are consistent with prior studies reporting that BIA-based devices could provide comparable trends to DXA in estimating bone mineral content and skeletal muscle mass, thereby supporting their utility in osteoporosis screening applications. Recent studies have expanded the potential applications of BIA in bone health assessment. Martins et al. [ 32 ] investigated 167 university athletes and reported that the bioelectrical impedance-derived phase angle (PhA) was positively correlated with both whole-body and regional BMD (lumbar spine and femoral neck) measured by DXA in female athletes. Their findings suggested that PhA may serve as a potential marker of bone health in female athletes. Similarly, Ngai et al. [ 33 ] examined 735 Southern Chinese adults (345 postmenopausal women and 390 men) and found that BIA was significantly associated with lumbar spine, femoral neck, and total hip BMD measured by DXA. In men, BIA was an independent predictor of BMD beyond age and body weight. ROC analyses indicated that BIA predicted osteoporosis with an AUC of 0.65–0.75, with only modest improvements when combined with age and weight. This study emphasized the potential of BIA as a clinical risk assessment tool rather than a standalone diagnostic modality. In postmenopausal women, Ono et al. [ 34 ] demonstrated that PhA derived from BIA was positively correlated with both physical function and health-related quality of life (SF-36 scores), as well as with lumbar spine and femoral neck BMD. Even after adjusting for confounders such as age, BMI, appendicular skeletal muscle mass index (ASMI), and BMD, PhA remained a significant predictor of the physical component summary score, suggesting that BIA-derived PhA reflects not only cellular health but also quality of life among patients with osteoporosis. Öztürk et al. [ 35 ] conducted bioelectrical impedance spectroscopy (BIS) in 48 postmenopausal women and reported significant negative correlations between BIS characteristic frequency and DXA-derived hip BMD (r = − 0.53, p < 0.001) and lumbar spine BMD (r = − 0.37, p < 0.05). ROC analysis demonstrated that BIS frequency effectively distinguished between normal and osteoporotic participants, particularly in the hip subgroup (AUC = 0.91). Chuang et al. [ 27 ] also compared whole-body BMD measurements between BIA and DXA in 318 participants and reported moderate-to-strong correlations but limited clinical interchangeability. Compared with their study, the present work extended methodological rigor by incorporating Passing–Bablok regression, ICC, Wilcoxon tests, MAPE, RMSPE, and ROC analysis. These additional statistical evaluations provided more comprehensive evidence regarding diagnostic validity and clinical classification accuracy. Importantly, while Chuang et al. focused on a younger sample (mean age ≈ 37 years), the present study included a broader age range, encompassing clinically relevant populations at higher risk of osteoporosis. Our ROC analysis (AUC = 0.906) further demonstrated that SA201 could serve as a reliable screening tool, with sensitivity (88.2%) and specificity (87.0%) at the optimal cutoff. Moreover, error quantification (CV ≈ 7.7%, MAPE = 6.5%, RMSPE = 8.3%) highlighted the acceptable consistency of BIA at the population level but emphasized caution in individual-level interpretation. However, the wide confidence intervals observed in the Passing–Bablok regression indicate substantial heterogeneity in individual impedance responses, particularly across sex and age groups. This suggests that fixed or proportional bias cannot be fully excluded. Although the average trend appears acceptable, such uncertainty implies that the two methods cannot be considered clinically interchangeable at the individual level. The magnitude of individual-level error (MAPE ≈ 6.5% and RMSPE ≈ 8%) is relatively high for clinical decision-making and insufficient to support diagnostic use. Errors of this magnitude may lead to misclassification for individual patients. The wide age range of participants (25–86 years) substantially increased inter-individual variability in BMD, given that age is a major determinant of bone mineral density. When both DXA and BIA capture this age-related decline, the shared variance may artificially inflate Pearson correlation coefficients, even when systematic or individual-level discrepancies remain between the two methods. Therefore, a high correlation should not be interpreted as evidence of clinical interchangeability. Correlation reflects only relative ranking rather than absolute agreement. Accordingly, method comparison should rely primarily on metrics that are less influenced by age-driven group variability—such as ICC, Passing–Bablok regression, Bland–Altman analysis, and MAPE/RMSPE—rather than correlation coefficients alone. In Supplementary Figure S1 , the regression line shows a slope similar to that of the Passing–Bablok regression in Fig. 1 . Although the 95% confidence intervals for the slope (− 25.84 to 30.20) and intercept (− 31.19 to 28.75) include the values corresponding to perfect agreement (i.e., a slope of 1 and an intercept of 0), the plotted confidence bands do not fully encompass the line of perfect concordance. This indicates that, within this study population, BIA-derived estimates of BMD may exhibit systematic underestimation or overestimation. Passing–Bablok regression was selected in this study because it is a non-parametric method that is robust to outliers and does not require assumptions of normally distributed errors or homoscedasticity. This makes it particularly suitable for BMD analyses in populations with substantial biological variability. The method also allows simultaneous estimation of both fixed bias (intercept) and proportional bias (slope). In contrast, ordinary least squares (OLS) regression assumes normally distributed errors with constant variance, which may not be appropriate under these conditions. The percentage bias displayed in Supplementary Figure S2 shows an approximately symmetric distribution; however, it does not fully conform to normality because the values are scaled by the denominator, which typically produces asymmetric and non-normally distributed data. This observation further supports the use of the non-parametric Passing–Bablok regression method in this study. Because the absolute range of whole-body BMD values is narrow (e.g., 1.0–1.2 g/cm²), even small absolute errors of 0.01–0.02 g/cm² can produce apparently high coefficients of variation (CV%). In addition, greater physiological variability within the measured population—such as fluctuations in hydration distribution, plantar contact pressure, and limb temperature—can alter impedance within short periods of time, thereby increasing retest variability. The elevated CV% further indicates that BIA and DXA are not interchangeable at the individual level. Compared with the approximately 1% precision typically achieved by DXA, the CV observed for BIA in this study reflects a repeatability that is at least an order of magnitude poorer, even under strictly standardized pre-measurement conditions. The wide limits of agreement (LOA) indicate substantial individual-level discrepancies between BIA and DXA measurements. When the BMD differences are converted to Z-scores, the magnitude corresponds to approximately ± 1.5 SD (depending on the SD of the DXA reference database). Such a range far exceeds the level of error acceptable for clinical diagnosis or longitudinal monitoring. Because these discrepancies exceed the least significant change (LSC) of DXA, BIA-derived whole-body BMD is not suitable for patient follow-up and should not be used for any diagnostic or therapeutic decision-making. Errors of this magnitude are sufficient to cause clinically meaningful misclassification, including misidentifying normal individuals as osteoporotic, classifying osteoporotic patients as normal, or incorrectly categorizing individuals within the osteopenic range. Even if the mean bias appears small, the large individual-level variability considerably limits the reliability of BIA-based BMD for diagnostic classification. The primary comparison metric in this study was the absolute difference in BMD between DXA and BIA. Absolute differences were used for the main analyses, including Bland–Altman plots, ICC, and MAPE. Percentage differences, which exhibit different distributional properties due to their dependence on the denominator, were not used as a basis for the primary inferences. Therefore, percentage differences were presented only in the supplementary materials for exploratory purposes. According to the WHO and major clinical guidelines (e.g., ISCD, NOF), osteoporosis diagnosis should primarily rely on Z-scores derived from the lumbar spine (L1–L4), femoral neck, or total hip [ 3 , 27 , 36 ]. Among these, femoral neck BMD is most predictive of hip fracture risk [ 4 ], while lumbar spine BMD is particularly sensitive to vertebral fracture risk [ 6 ]. Lumbar BMD can detect early bone loss at the onset of aging but may be falsely elevated in older individuals due to degenerative changes such as disc calcification, osteoarthritis, or scoliosis, potentially leading to underestimation of osteoporosis risk [ 37 , 38 ]. Whole-body BMD, by averaging across skeletal regions, can mitigate localized biases and is particularly useful in epidemiological and body composition studies (e.g., sarcopenia, obesity, and bone health) [ 39 ]. However, whole-body Z-scores are not currently recommended as a diagnostic criterion by WHO or clinical practice [ 3 , 36 ]. Because averaging may dilute localized bone loss, whole-body BMD has lower predictive power for fracture risk compared with site-specific measures [ 40 , 41 ]. According to WHO and ISCD guidelines, whole-body BMD is not an accepted diagnostic criterion for osteoporosis. Whole-body BMD values are generally higher than those of the lumbar spine and hip and are less sensitive to localized bone loss. Although BIA shows a high correlation with whole-body DXA BMD, this does not imply suitability for clinical diagnosis. Lumbar spine and femoral neck BMD remain the required anatomical sites for diagnostic classification. Given its portability and radiation-free nature, BIA may serve as a preliminary screening tool to identify high-risk individuals who can then be referred for confirmatory DXA assessment. The lumbar spine and femoral neck are metabolically active regions with high fracture relevance, whereas whole-body BMD represents an average across multiple skeletal sites, thereby reducing sensitivity to localized bone loss. Consequently, whole-body BMD cannot be used for clinical interpretation or diagnostic decision-making。 According to WHO and ISCD guidelines, the diagnosis of osteoporosis must be based on BMD measured at the lumbar spine, femoral neck, or total hip, rather than whole-body BMD. Whole-body BMD values are typically higher and less sensitive to localized bone loss. Consequently, applying whole-body Z-score thresholds to BIA-derived measurements will inherently yield higher rates of abnormal classification, largely because whole-body averages dilute focal reductions in bone density. In addition, foot-to-foot BIA estimation of whole-body BMD may differ substantially from clinically relevant skeletal sites such as the hip and lumbar spine. At present, no validated clinical diagnostic thresholds exist for whole-body BMD—whether measured by BIA or DXA. Therefore, even if whole-body BMD can be obtained, the absence of standardized cut-offs precludes its use for clinical diagnosis. Because this study assessed only whole-body BMD, rather than lumbar spine or hip BMD, the findings should not be applied to osteoporosis diagnosis nor to clinical fracture-risk stratification. Age and sex are major determinants of BMD; thus, the wide age range and mixed-sex composition of the sample (25–86 years) naturally produced substantial inter-individual variability. Under such conditions, even a method that merely approximates the age-related decline in BMD may appear to differentiate individuals with lower BMD. Therefore, the identification of 17 low-BMD cases in this study may largely reflect the structure of the sample itself rather than the precision of the estimation method. Several factors may explain why the device used in this study could appear more accurate: the prediction model was directly calibrated against whole-body BMD measured by DXA, and the use of multiple impedance frequencies may capture richer information on tissue properties. However, several factors may also reduce accuracy. The current pathway does not encompass clinically relevant skeletal sites such as the hip and lumbar spine. Age-related changes in hydration, muscle mass, and limb geometry may violate key model assumptions. In addition, the algorithm has not been validated in older adults or osteoporotic populations, which may introduce bias. Although this study employed stringent pre-measurement standardization procedures—such as fixed measurement time, bladder emptying, avoidance of exercise and food intake, hydration control, and standardized posture and electrode contact—these conditions are difficult to replicate consistently in real-world clinical settings. Therefore, in practical applications, the error and repeatability limitations of BIA-derived whole-body BMD may be more pronounced than those observed in this controlled research environment. The SA201 utilizes a four-electrode, foot-to-foot configuration with dual-frequency measurements and a dedicated algorithm specifically developed for estimating whole-body BMD using DXA-calibrated data. Unlike conventional BIA systems designed for general body composition assessment, the SA201 algorithm incorporates the relationship between lower-limb impedance and trunk composition. However, because the electrical current primarily travels through the lower limbs, it cannot directly capture characteristics of axial skeletal sites such as the spine and hip. In summary, BIA technology is gaining recognition not only for estimating BMD but also for providing additional insights into cellular health, quality of life, and functional outcomes. The present findings reinforce the potential of BIA as a practical tool for osteoporosis screening in both clinical and community settings, complementing but not replacing DXA. Limitations and Future Directions This study has several limitations. First, the sample size was relatively modest, which may have limited the statistical power and generalizability of the findings. Second, the study population was drawn from a single geographic region, resulting in limited heterogeneity, which may restrict extrapolation to other populations. Third, the analysis focused solely on whole-body BMD, without comparing site-specific measurements (e.g., lumbar spine and femoral neck), which are of greater clinical importance in the diagnosis of osteoporosis. Finally, as BIA is a novel device, further validation across multiple centers and diverse populations is necessary to confirm its accuracy and reproducibility. Future studies should address these limitations by: (1) recruiting larger and more diverse cohorts encompassing different age groups, sexes, and clinical conditions to improve the representativeness of findings; (2) evaluating the validity and agreement of BIA with DXA at clinically critical skeletal sites (e.g., femoral neck, lumbar spine); and (3) investigating the utility of BIA in community screening, longitudinal monitoring, and epidemiological research. Such studies will be essential to establish the broader clinical applicability of BIA in osteoporosis prevention and management. The absence of fracture outcomes represents a major limitation of this study. Clinically meaningful validation should assess whether BIA-derived measures can improve the accuracy of FRAX or other fracture-risk models. Future research should include prospective follow-up and comparisons with site-specific DXA measurements to determine the predictive value and clinical utility of this approach. Conclusion BIA demonstrated a strong correlation and good agreement with DXA in whole-body BMD measurement, as well as excellent discriminatory ability in the diagnosis of osteoporosis (AUC > 0.9). However, the results also revealed considerable individual-level error (approximately 6–8%), very wide limits of agreement, and limited repeatability. In addition, accurate estimation requires strict pre-measurement standardization, and the Passing–Bablok regression indicated substantial estimation uncertainty. BIA is therefore not interchangeable with DXA, with the overly wide LOA being the primary limiting factor. Consequently, BIA is suitable as an initial clinical screening tool and for epidemiological research, whereas DXA remains the essential standard for the diagnostic confirmation of osteoporosis. Declarations Data availability statement The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation. Ethics statement The studies involving humans were approved by Institutional Review Board (IRB) of the Ministry of Health and Welfare Nantou Hospital (IRB-111047 and IRB-113002). The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study. Conflict of interest K-CH was employed by a commercial company, StarBIA Meditek Co. Ltd., during this study. Funding The author(s) declare that financial support was received for the research, authorship, and/or publication of this article. Funding for this work has been provided by Ministry of Health and Welfare Hospital Research and Development plan. Author Contribution H.-K.L. conceived the study, designed the methodology. H.-K.L. and T.-J.H. conducted the investigations. C.-L.L. and W.-C.T. performed data curation, while C.-L.L. was responsible for funding acquisition. C.-W.L. and A.-C.H. performed the formal analysis. K.-C.H. provided the software. H.-K.L. and C.-L.L. carried out the validation of the results. A.-C.H. prepared the visualizations. H.-K.L. and C.-L.L. supervised the project. H.-K.L. wrote the original draft of the manuscript, and H.-K.L. and K.-C.H. reviewed and edited the final text. All authors reviewed the manuscript. Data Availability The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation. References NIH Consensus Development Panel on Osteoporosis Prevention. Diagnosis, and Therapy. Osteoporosis prevention, diagnosis, and therapy. JAMA 285 , 785–795 (2001). Rachner, T. D., Khosla, S. & Hofbauer, L. C. Osteoporosis: now and the future. Lancet 377 , 1276–1287 (2011). World Health Organization. Assessment of fracture risk and its application to screening for postmenopausal osteoporosis. WHO Technical Report Series, No. 843. WHO, (1994). Cummings, S. R. & Melton, L. J. Epidemiology and outcomes of osteoporotic fractures. Lancet 359 , 1761–1767 (2002). Kanis, J. A. et al. European guidance for the diagnosis and management of osteoporosis in postmenopausal women. Osteoporos. Int. 30 , 3–44 (2019). Kanis, J. A. et al. The diagnosis of osteoporosis. J. Bone Min. Res. 9 , 1137–1141 (1994). Marshall, D., Johnell, O. & Wedel, H. Meta-analysis of how well measures of bone mineral density predict occurrence of osteoporotic fractures. BMJ 312 , 1254–1259 (1996). Johnell, O. & Kanis, J. A. An estimate of the worldwide prevalence and disability associated with osteoporotic fractures. Osteoporos. Int. 17 , 1726–1733 (2006). Seeman, E. & Delmas, P. D. Bone quality—the material and structural basis of bone strength and fragility. N Engl. J. Med. 354 , 2250–2261 (2006). Raisz, L. G. Pathogenesis of osteoporosis: concepts, conflicts, and prospects. J. Clin. Invest. 115 , 3318–3325 (2005). Riggs, B. L., Khosla, S. & Melton, L. J. 3rd. Sex steroids and the construction and conservation of the adult skeleton. Endocr. Rev. 23 , 279–302 (2002). Compston, J. E. Sex steroids and bone. Physiol. Rev. 81 , 419–447 (2001). Holick, M. F. Vitamin D deficiency. N Engl. J. Med. 357 , 266–281 (2007). Eastell, R. et al. Modern Spectrum Bone turnover markers: are they clinically useful? Eur. J. Endocrinol. 178 , R19–R31 (2018). Miller, P. D. et al. Clinical utility of bone mass measurements in adults: Consensus of an international panel. Semin Arthritis Rheum. 25 , 361–372 (1996). Kanis, J. A. et al. The diagnosis of osteoporosis. J. Bone Min. Res. 9 , 1137–1141 (1994). Blake, G. M. & Fogelman, I. The role of DXA bone density scans in the diagnosis and treatment of osteoporosis. Postgrad. Med. J. 83 , 509–517 (2007). Engelke, K. et al. Clinical use of quantitative computed tomography (QCT) of the hip in the management of osteoporosis in Adults: the 2015 ISCD official Positions – Part I. J. Clin. Densitom . 18 , 338–358 (2015). Link, T. M. Osteoporosis imaging: state of the art and advanced imaging. Radiology 263 , 3–17 (2012). Bousson, V. et al. Volumetric quantitative computed tomography of the proximal femur: relationships linking geometric and densitometric variables to bone strength. Role for compact bone. Osteoporos. Int. 17 , 855–864 (2006). MacNeil, J. A. & Boyd, S. K. Bone strength at the distal radius can be estimated from high-resolution peripheral quantitative computed tomography and finite element analysis. Bone 42 , 1203–1213 (2008). Krieg, M. A. et al. Quantitative ultrasound in the management of osteoporosis: the 2007 ISCD Official Positions. J. Clin. Densitom . 11 , 163–187 (2008). Moayyeri, A. et al. Quantitative ultrasound of the heel and fracture risk assessment: an updated meta-analysis. Osteoporos. Int. 23 , 143–153 (2012). Nayak, S. et al. Meta-analysis: accuracy of quantitative ultrasound for identifying patients with osteoporosis. Osteoporos. Int. 17 , 801–810 (2006). Kyle, U. G. et al. Bioelectrical impedance analysis—part I: review of principles and methods. Clin. Nutr. 23 , 1226–1243 (2004). Kushner, R. F. & Schoeller, D. A. Estimation of total body water by bioelectrical impedance analysis. Am. J. Clin. Nutr. 44 , 417–424 (1986). Chuang, C. L. et al. Comparison of whole body bone mineral density measurement between dual energy X ray absorptiometry and novel bioelectrical impedance analysis. Sci. Rep. 14, (待補頁碼或文章編號) (2024). Bland, J. M. & Altman, D. G. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1 , 307–310 (1986). Passing, H. & Bablok, W. A new biometrical procedure for testing the equality of measurements from two different analytical methods. J. Clin. Chem. Clin. Biochem. 21 , 709–720 (1983). Bilić-Zulle, L. Comparison of methods: Passing–Bablok regression. Biochem. Med. 21 , 49–52 (2011). Koo, T. K. & Li, M. Y. A guideline of selecting and reporting intraclass correlation coefficients for reliability research. J. Chiropr. Med. 15 , 155–163 (2016). Martins, P. C., Moraes, M. S. & Silva, D. A. How is the phase angle associated with total and regional bone mineral density in university athletes? Physiol. Meas. 42 , 085001 (2021). Ngai, H. H. Y., Cheung, C. L., Yao, T. J. & Kung, A. W. C. Bioimpedance: can its addition to simple clinical criteria enhance the diagnosis of osteoporosis? J. Bone Min. Metab. 27 , 372–378 (2009). Ono, Y., Kasukawa, Y., Sasaki, K. & Miyakoshi, N. Association of the Bioimpedance Phase Angle and Quality of Life in Postmenopausal Osteoporosis. Med. Princ Pract. 32 , 71–76 (2023). Öztürk, N., Ozturk-Isik, E. & Ülgen, Y. Screening post-menopausal women for bone mineral level by bioelectrical impedance spectroscopy of dominant arm. J. Electr. Bioimp . 9 , 39–47 (2018). Lewiecki, E. M. et al. Official positions of the International Society for Clinical Densitometry. J. Clin. Endocrinol. Metab. 89 , 3651–3655 (2004). Cosman, F. et al. Clinician’s Guide to Prevention and Treatment of Osteoporosis. Osteoporos. Int. 25 , 2359–2381 (2014). Riggs, B. L. & Melton, L. J. The prevention and treatment of osteoporosis. N Engl. J. Med. 327 , 620–627 (1992). Yu, W. et al. Spinal bone mineral assessment in postmenopausal women: comparison of dual-energy X-ray absorptiometry and quantitative computed tomography. Osteoporos. Int. 5 , 433–439 (1995). Shepherd, J. A., Ng, B. K., Sommer, M. J. & Heymsfield, S. B. Body composition by DXA. Bone 104 , 101–105 (2017). De Laet, C. et al. Hip fracture prediction in elderly men and women: validation in the Rotterdam Study. J. Bone Min. Res. 13 , 1587–1593 (1998). Additional Declarations No competing interests reported. Supplementary Files floatimage4.jpeg Figure S1. Scatter plot and regression line of whole-body BMD measured by SA201 and DXA, showing a strong linear relationship: regression equation: BIA = 0.354 + 0.668 × DXA, R² = 0.671 Cite Share Download PDF Status: Under Review Version 1 posted Reviewers agreed at journal 07 May, 2026 Reviews received at journal 06 May, 2026 Reviewers agreed at journal 06 May, 2026 Reviewers agreed at journal 05 May, 2026 Reviewers agreed at journal 05 May, 2026 Reviewers invited by journal 05 May, 2026 Editor invited by journal 13 Apr, 2026 Editor assigned by journal 07 Apr, 2026 Submission checks completed at journal 07 Apr, 2026 First submitted to journal 03 Apr, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9314239","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":636338608,"identity":"afa08028-2097-48cf-b2e4-d21d8ca292fa","order_by":0,"name":"Hsueh-Kuan Lu","email":"","orcid":"","institution":"National Taiwan University of Sport","correspondingAuthor":false,"prefix":"","firstName":"Hsueh-Kuan","middleName":"","lastName":"Lu","suffix":""},{"id":636338611,"identity":"8feb94be-45db-47ba-9c37-e5766eb2961f","order_by":1,"name":"Ai-Chun Huang","email":"","orcid":"","institution":"National Kaohsiung University of Hospitality and Tourism","correspondingAuthor":false,"prefix":"","firstName":"Ai-Chun","middleName":"","lastName":"Huang","suffix":""},{"id":636338613,"identity":"946b0b9e-8f9c-4d3e-8ff3-9abe114be661","order_by":2,"name":"Chien-Wei Liang","email":"","orcid":"","institution":"National Chung Hsing University","correspondingAuthor":false,"prefix":"","firstName":"Chien-Wei","middleName":"","lastName":"Liang","suffix":""},{"id":636338614,"identity":"2cb3e41e-f9ed-497d-a590-2afbdf13bf3d","order_by":3,"name":"Tzu-Jung Huang","email":"","orcid":"","institution":"National Taiwan University of Sport","correspondingAuthor":false,"prefix":"","firstName":"Tzu-Jung","middleName":"","lastName":"Huang","suffix":""},{"id":636338617,"identity":"ae19797f-1706-4e97-8067-eec90126236d","order_by":4,"name":"Kuen-Chang Hsieh","email":"","orcid":"","institution":"Starbia Meditek Co., Ltd.","correspondingAuthor":false,"prefix":"","firstName":"Kuen-Chang","middleName":"","lastName":"Hsieh","suffix":""},{"id":636338623,"identity":"4b20aa15-cb06-4877-a5b7-e466a581ba79","order_by":5,"name":"Chung-Liang Lai","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA+klEQVRIiWNgGAWjYBACPgbGBjDDfv7jAzA2fsAGVpbAwGDAkJaApCUBnxaoAgOGHAMitbAfbnvw8YdNvjnDmW8SP3fYyDGwHz66gfHHYdxaeBLbDWckpFnubOzdJtl7Js2YgSct7QZDAh4tDIlt0jwJhw0YDvNuk+BtO5zYIMFjBtRyG7cW/odt0n8S/hswHON5JvmXKC0SQFsYEg4YGJzhYZMmzhaJh+2GPWnJBpIz2IytZdvSjNlAfklI+49TCz9/+rMHP2zsDPglmB/efNtmI8fPfvjYjQ82aTi1MMCiBghYJODcBHwakLQwf8CvcBSMglEwCkYqAACBWFIODyUTMQAAAABJRU5ErkJggg==","orcid":"","institution":"Changhua Hospital, Ministry of Health and Welfare","correspondingAuthor":true,"prefix":"","firstName":"Chung-Liang","middleName":"","lastName":"Lai","suffix":""}],"badges":[],"createdAt":"2026-04-03 15:08:45","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9314239/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9314239/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":109215902,"identity":"ed18cf46-7f99-4abf-8edf-4117951b88cc","added_by":"auto","created_at":"2026-05-13 17:57:21","extension":"jpeg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":476539,"visible":true,"origin":"","legend":"\u003cp\u003ePassing–Bablok regression analysis. Slope 95% CI: [−25.84, 30.20]; intercept 95% CI: [−31.19, 28.75]. Interpretation: the CI includes 1 (slope) and 0 (intercept), indicating no significant proportional or fixed bias, although the intervals are wide.\u003c/p\u003e","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-9314239/v1/7b6f62a799548c5bec12aa22.jpeg"},{"id":109249146,"identity":"9ff1189d-9372-4a61-ad60-457be10fc2c4","added_by":"auto","created_at":"2026-05-14 08:42:51","extension":"jpeg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":551970,"visible":true,"origin":"","legend":"\u003cp\u003eBland–Altman analysis. Mean difference = −0.003 g/cm²; 95% limits of agreement = [−0.166, 0.160] g/cm². Most observations fell within ±0.16 g/cm².\u003c/p\u003e","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-9314239/v1/41bb1741ae58bf988d46cfca.jpeg"},{"id":109249449,"identity":"de66a552-85e9-42f2-a7ba-8a31a8ce4f65","added_by":"auto","created_at":"2026-05-14 08:52:54","extension":"jpeg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":192922,"visible":true,"origin":"","legend":"\u003cp\u003eReceiver operating characteristic (ROC) curve analysis. BIA achieved an AUC of 0.906, indicating good diagnostic validity. The optimal cutoff was 0.969 g/cm², with sensitivity = 88.2% and specificity = 87.0%.\u003c/p\u003e","description":"","filename":"floatimage3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-9314239/v1/ce4667d682c42a17741fadce.jpeg"},{"id":109249143,"identity":"86978337-583a-4a3f-a91f-1958ff15d80b","added_by":"auto","created_at":"2026-05-14 08:42:49","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":796246,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9314239/v1/0341f275-3ead-4b7f-9765-5a0c02596dc3.pdf"},{"id":109215903,"identity":"46e68c1b-367e-4066-a027-f9614f981736","added_by":"auto","created_at":"2026-05-13 17:57:22","extension":"jpeg","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":374265,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eFigure S1.\u003c/strong\u003e Scatter plot and regression line of whole-body BMD measured by SA201 and DXA, showing a strong linear relationship: regression equation: BIA = 0.354 + 0.668 × DXA, R² = 0.671\u003c/p\u003e","description":"","filename":"floatimage4.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-9314239/v1/a91683b2a5896249139a0603.jpeg"}],"financialInterests":"No competing interests reported.","formattedTitle":"Application of a Standing Quad-Plate Bioelectrical Impedance Analyzer for Whole-Body Bone Mineral Density Measurement in Osteoporosis Screening","fulltext":[{"header":"Introduction","content":"\u003cp\u003eOsteoporosis is an age-related metabolic bone disease characterized by reduced bone mass and deterioration of the microarchitecture, leading to increased bone fragility and a higher risk of fractures [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. According to the World Health Organization (WHO), more than 200\u0026nbsp;million people worldwide are affected by osteoporosis, and the associated medical and social costs are steadily increasing. Therefore, early diagnosis and screening have significant clinical and public health importance [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eBone mineral density (BMD) is widely recognized as the core indicator for diagnosing osteoporosis and predicting fracture risk [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. According to the WHO definition, a Z-score\u0026thinsp;\u0026le;\u0026thinsp;\u0026minus;\u0026thinsp;2.5 indicates osteoporosis [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]. Numerous prospective studies have demonstrated a strong association between reduced BMD and fragility fracture risk, with each standard deviation (SD) decrease in BMD corresponding to an approximately 1.5\u0026ndash;3-fold increase in fracture risk [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e, \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. Consequently, BMD measurement has become one of the most critical clinical tools in strategies for osteoporosis prevention and treatment.\u003c/p\u003e \u003cp\u003eBMD reflects the mineral content of bone (mainly hydroxyapatite), and its reduction is related to excessive bone resorption and insufficient bone formation [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. The core mechanisms of bone metabolism involve an imbalance between increased osteoclast activity and decreased osteoblast activity, leading to trabecular thinning and microarchitectural deterioration [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. Both clinical and basic studies have shown that estrogen deficiency, aging, chronic inflammation, and hormonal dysregulation (e.g., parathyroid hormone imbalance, vitamin D deficiency) accelerate bone loss [\u003cspan additionalcitationids=\"CR12\" citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]. Accurate BMD measurement therefore enables early detection of osteoporosis and serves as a key indicator for treatment monitoring and drug efficacy evaluation in clinical trials [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. FRAX, QFracture, and the Garvan calculator integrate BMD with multiple clinical risk factors\u0026mdash;including age, history of fragility fractures, glucocorticoid use, smoking, alcohol consumption, and secondary causes of osteoporosis. Therefore, even if a device is capable of estimating whole-body BMD, its ability to predict fracture risk cannot be inferred directly unless such estimates are combined with the relevant clinical data.\u003c/p\u003e \u003cp\u003eCurrently, several modalities are available for the assessment of bone mineral density (BMD), each with distinct advantages and limitations. Dual-energy X-ray absorptiometry (DXA) remains the clinical gold standard for BMD evaluation, providing highly accurate measurements of both whole-body and regional sites (e.g., lumbar spine, femoral neck) \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\left[\\text{4,16}\\right]\\)\u003c/span\u003e\u003c/span\u003e. Despite its diagnostic reliability, DXA is constrained by its relatively high equipment cost, large footprint, the need for trained radiology personnel, and unavoidable exposure to low-dose ionizing radiation, which collectively hinder its implementation in primary care and community-based settings \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\left[17\\right]\\)\u003c/span\u003e\u003c/span\u003e. Quantitative computed tomography (QCT) offers the unique capability of separately quantifying cortical and trabecular bone with three-dimensional precision \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\left[18\\right]\\)\u003c/span\u003e\u003c/span\u003e. However, its clinical adoption is limited by substantially higher radiation exposure and operational costs compared with DXA \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\left[19\\right]\\)\u003c/span\u003e\u003c/span\u003e. Peripheral quantitative computed tomography (pQCT) provides volumetric assessment of peripheral skeletal sites such as the forearm and lower leg using relatively simple instrumentation \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\left[20\\right]\\)\u003c/span\u003e\u003c/span\u003e. Nevertheless, because pQCT is restricted to peripheral regions, it cannot yield whole-body or axial skeletal information \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\left[21\\right]\\)\u003c/span\u003e\u003c/span\u003e.Quantitative ultrasound (QUS) is a non-invasive, radiation-free, portable, and cost-efficient technique that makes it particularly suitable for primary care and large-scale population screening \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\left[22\\right]\\)\u003c/span\u003e\u003c/span\u003e. Its diagnostic utility, however, is influenced by the selected measurement site\u0026mdash;typically the calcaneus\u0026mdash;and its correlation with DXA-derived BMD remains inconsistent and subject to ongoing debate \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\left[\\text{23,24}\\right]\\)\u003c/span\u003e\u003c/span\u003e. Bioelectrical impedance analysis (BIA) is another radiation-free, rapid, portable, and inexpensive alternative, rendering it attractive for community-based screening and epidemiological research \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\left[\\text{25,26}\\right]\\)\u003c/span\u003e\u003c/span\u003e. Although increasingly explored for BMD estimation, its accuracy and clinical interchangeability with DXA remain to be fully established. Previous studies employing BIA to estimate BMD have predominantly used single-frequency foot-to-foot devices, such as those in the Tanita system, and some have utilized multi-frequency InBody analyzers; however, none of these instruments were originally designed for bone density assessment. Most of these investigations derived body composition parameters through regression-based prediction equations, yielding correlations that ranged from moderate to high (r\u0026thinsp;\u0026asymp;\u0026thinsp;0.40\u0026ndash;0.90) depending on the variable examined (e.g., body fat percentage, muscle mass, or bone mineral content) [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]. Nevertheless, the magnitude of error varied substantially across studies, and these levels of accuracy are insufficient for clinical diagnostic purposes. The estimation of whole-body BMD using bioelectrical impedance is based on electrical characteristics such as current pathways, limb geometry, tissue conductivity, the distribution of bone and soft tissues, and inferred mineral content. The predictive models for this technique were developed using databases of Taiwanese and broader Asian adults aged 20\u0026ndash;80 years. Model calibration and training were performed with whole-body BMD obtained from DXA (GE Lunar Prodigy) as the reference standard. The initial model was constructed using multiple regression analyses supplemented by cross-validation. Several studies have reported that BIA-derived parameters\u0026mdash;such as phase angle and impedance index\u0026mdash;are correlated with BMD. For example, Chuang et al. [\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e] reported moderate to high correlations between whole-body BMD measured by DXA and values estimated using the prototype BIA device. Nonetheless, its application in BMD measurement remains relatively novel and requires further validation.\u003c/p\u003e \u003cp\u003eCompared with DXA, BIA offers advantages of non-invasiveness, ease of operation, and lower cost. However, its measurement accuracy and consistency with DXA require rigorous scientific validation. Therefore, the present study aimed to systematically evaluate the differences and agreement between BIA and DXA in whole-body BMD measurement using multiple statistical methods\u0026mdash;including correlation analysis, Bland\u0026ndash;Altman plots, Passing\u0026ndash;Bablok regression, intraclass correlation coefficients (ICC), agreement testing, and ROC analysis\u0026mdash;and to explore the feasibility of BIA as a tool for osteoporosis screening.\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eStudy Design and Participants\u003c/h2\u003e \u003cp\u003e This cross-sectional study was conducted between January 2023 and September 2023 and was approved by the Institutional Review Board (IRB) of Nantou Hospital, Ministry of Health and Welfare, Taiwan (Approval No. IRB-111047 and IRB-113002). All procedures were performed in accordance with relevant ethical guidelines, and written informed consent was obtained from all participants prior to enrollment. A total of 178 adults (56 men and 122 women), aged 25\u0026ndash;86 years, who were able to stand and walk independently, were recruited at Puzi Hospital, Ministry of Health and Welfare, through poster advertisements and oral invitations. Exclusion criteria included individuals who had undergone surgeries that could potentially alter body composition (e.g., bariatric surgery).\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003ePre-Assessment Protocol\u003c/h3\u003e\n\u003cdiv class=\"Heading\"\u003ePre-Assessment Protocol\u003c/div\u003e \u003cp\u003eTo minimize measurement variability, participants were required to adhere to the following pre-test instructions: Avoid strenuous exercise within 48 hours prior to testing while maintaining regular dietary habits. Refrain from consuming alcohol or high-caffeine beverages within 48 hours before testing. Avoid moderate-to-vigorous physical activity within 12 hours prior to testing. Empty the bladder immediately before measurement. Discontinue the use of diuretics for at least 7 days prior to testing. Avoid nuclear medicine examinations or contrast-enhanced imaging within 5 days before testing. Female participants undergoing menstruation were rescheduled; pregnant or potentially pregnant women were excluded.\u003c/p\u003e \u003cp\u003eDuring the assessment, participants wore cotton gowns and underwear and removed all metallic objects (e.g., rings, earrings, zippers, buttons) that could interfere with X-ray scanning. All measurements were performed between 9:00 a.m. and 12:00 noon to minimize the influence of diurnal variation. All BIA and DXA measurements were completed during the same study visit. The DXA and BIA assessments were performed independently by different operators in separate rooms, and the measurement outputs were processed separately before being merged for analysis.\u003c/p\u003e\n\u003ch3\u003eAnthropometric Measurements\u003c/h3\u003e\n\u003cp\u003eBody height was measured to the nearest 0.1 cm using a digital stadiometer (Jenix DS-102, Dong Sang Jen Ix Co., Ltd., South Korea). Body weight was measured using the built-in scale of the BIA device.\u003c/p\u003e\n\u003ch3\u003eBioelectrical Impedance Analysis (BIA)\u003c/h3\u003e\n\u003cp\u003eWhole-body BMD was measured using the SA201 foot-to-foot bioelectrical impedance analyzer (StarBIA MediTek Co., Taichung, Taiwan). The device employs a four-electrode system and dual-frequency alternating currents (5 and 50 kHz). Measurements were performed according to the manufacturer\u0026rsquo;s instructions. Participants stood barefoot on the electrode plates for approximately 3 minutes, and data acquisition (including body weight and impedance) was completed within one minute. A total of 30 participants underwent repeated BIA measurements. All retests were performed within the same testing session, with a 5-minute interval between measurements, during which participants were required to step off the device and reposition themselves. The SA201 device operated using the reference database established by the manufacturer.\u003c/p\u003e\n\u003ch3\u003eDual-Energy X-ray Absorptiometry (DXA)\u003c/h3\u003e\n\u003cp\u003eDXA scans were performed using the GE Lunar Prodigy Advance system (GE Medical Systems Lunar, Madison, WI, USA) with enCORE software version 13.50.0. During the scan, participants lay supine with arms positioned alongside the body and palms facing downward. All scans and calibrations were conducted by trained technicians according to the standards of the International Society for Clinical Densitometry (ISCD). Precision analysis demonstrated that the root mean square standard deviation (RMS-SD) of whole-body BMD was 0.006 g/cm\u0026sup2;, corresponding to a coefficient of variation (CV) of 0.63%, with a least significant change (LSC) of 0.022 g/cm\u0026sup2; (1.73%). DXA repeat measurements were not performed in this study, as the DXA had already established precision values\u0026mdash;least significant change (LSC) and root-mean-square standard deviation (RMS-SD)\u0026mdash;in accordance with ISCD guidelines. The GE Lunar Prodigy Advance scanner employed the GE Lunar reference database, which is primarily based on White male and female data from the National Health and Nutrition Examination Survey (NHANES). This reference standard is used in the GE enCORE software for calculating Z-scores and complies with ISCD recommendations.\u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eStatistical Analysis\u003c/h2\u003e \u003cp\u003eAll data were presented as mean\u0026thinsp;\u0026plusmn;\u0026thinsp;standard deviation, minimum, and maximum values. The Kolmogorov\u0026ndash;Smirnov test was used to evaluate the normality of data distribution. Differences in whole-body BMD between BIA and DXA were compared using paired t-tests and Wilcoxon signed-rank tests. Linear associations were assessed using Pearson correlation coefficients, while reliability was evaluated by calculating intraclass correlation coefficients (ICC, model 2,1). Agreement between methods was further examined using Bland\u0026ndash;Altman analysis (mean difference\u0026thinsp;\u0026plusmn;\u0026thinsp;1.96 SD) [\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e] and Passing\u0026ndash;Bablok regression to assess fixed and proportional errors [\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eThe sample size of 178 exceeded the minimum threshold of 30\u0026ndash;50 cases recommended in the literature, as well as the 100 cases required for equivalence testing, thereby ensuring sufficient precision for regression coefficient estimates [\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e]. Error indices included the coefficient of variation (CV%), mean absolute percentage error (MAPE), and root mean square percentage error (RMSPE). Diagnostic validity for osteoporosis was evaluated using receiver operating characteristic (ROC) curve analysis. All statistical analyses were performed using SPSS version 20.0 (IBM Corp., Armonk, NY, USA), with statistical significance set at p\u0026thinsp;\u0026lt;\u0026thinsp;0.05.\u003c/p\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003cp\u003eA total of 178 participants (56 men and 122 women) were included in this study. The descriptive characteristics of the participants are presented in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. The mean age was 52.6 years (SD\u0026thinsp;=\u0026thinsp;14.8, range\u0026thinsp;=\u0026thinsp;25\u0026ndash;86 years). For the primary continuous variables (DXA-derived BMD and BIA-derived BMD), both BMD\u003csub\u003eDXA\u003c/sub\u003e and BMD\u003csub\u003eBIA\u003c/sub\u003e showed no violation of normality, as indicated by the Kolmogorov\u0026ndash;Smirnov test (p\u0026thinsp;\u0026gt;\u0026thinsp;0.05).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eBaseline Characteristics and Body Composition of Participants\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"15\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\"\u0026plusmn;\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c10\" colnum=\"10\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\"\u0026plusmn;\" class=\"colspec\" colname=\"c11\" colnum=\"11\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c12\" colnum=\"12\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c13\" colnum=\"13\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c14\" colnum=\"14\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c15\" colnum=\"15\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eAge (yrs)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colspan=\"3\" nameend=\"c5\" namest=\"c3\"\u003e \u003cp\u003eAll (n\u0026thinsp;=\u0026thinsp;178)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colspan=\"3\" nameend=\"c9\" namest=\"c7\"\u003e \u003cp\u003eFemale (n\u0026thinsp;=\u0026thinsp;122)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colspan=\"3\" nameend=\"c13\" namest=\"c11\"\u003e \u003cp\u003eMale (n\u0026thinsp;=\u0026thinsp;56)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c14\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c15\"\u003e \u003cp\u003eP\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003e52.63\u0026thinsp;\u0026plusmn;\u0026thinsp;14.83\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003e25.0-86.1\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003e53.71\u0026thinsp;\u0026plusmn;\u0026thinsp;13.41\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c9\"\u003e \u003cp\u003e25.0-86.1\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c11\"\u003e \u003cp\u003e50.27\u0026thinsp;\u0026plusmn;\u0026thinsp;17.4\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c12\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c13\"\u003e \u003cp\u003e25.0\u0026ndash;82.0\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c14\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c15\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.05\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHeight (cm)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e160.7\u0026thinsp;\u0026plusmn;\u0026thinsp;8.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e143.0-185.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c7\"\u003e \u003cp\u003e156.9\u0026thinsp;\u0026plusmn;\u0026thinsp;5.3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e143.0-175.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c11\"\u003e \u003cp\u003e168.9\u0026thinsp;\u0026plusmn;\u0026thinsp;6.6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c12\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c13\"\u003e \u003cp\u003e152.0-185.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c14\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c15\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWeight (kg)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e61.5\u0026thinsp;\u0026plusmn;\u0026thinsp;12.8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e33.1-107.8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c7\"\u003e \u003cp\u003e56.5\u0026thinsp;\u0026plusmn;\u0026thinsp;9.3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e33.1\u0026ndash;88.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c11\"\u003e \u003cp\u003e72.5\u0026thinsp;\u0026plusmn;\u0026thinsp;12.8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c12\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c13\"\u003e \u003cp\u003e48.2-107.8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c14\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c15\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBMI (kg/m\u003csup\u003e2\u003c/sup\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e23.7\u0026thinsp;\u0026plusmn;\u0026thinsp;3.8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e13.3\u0026ndash;39.6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c7\"\u003e \u003cp\u003e22.9\u0026thinsp;\u0026plusmn;\u0026thinsp;3.7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e13.3\u0026ndash;31.3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c11\"\u003e \u003cp\u003e25.3\u0026thinsp;\u0026plusmn;\u0026thinsp;3.8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c12\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c13\"\u003e \u003cp\u003e17.3\u0026ndash;39.6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c14\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c15\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.01\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBPF(%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e32.7\u0026thinsp;\u0026plusmn;\u0026thinsp;7.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e7.3\u0026ndash;48.7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c7\"\u003e \u003cp\u003e33.6\u0026thinsp;\u0026plusmn;\u0026thinsp;6.7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e7.3\u0026ndash;46.8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c11\"\u003e \u003cp\u003e30.7\u0026thinsp;\u0026plusmn;\u0026thinsp;8.6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c12\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c13\"\u003e \u003cp\u003e7.4\u0026ndash;48.7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c14\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c15\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.01\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBMC(kg)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2.12(0.55\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1.19\u0026ndash;3.82\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c7\"\u003e \u003cp\u003e2.1\u0026thinsp;\u0026plusmn;\u0026thinsp;0.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e1.2\u0026ndash;3.8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c11\"\u003e \u003cp\u003e2.3\u0026thinsp;\u0026plusmn;\u0026thinsp;0.7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c12\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c13\"\u003e \u003cp\u003e1.36\u0026ndash;3.6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c14\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c15\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.01\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBMD\u003csub\u003eBIA\u003c/sub\u003e(g/cm\u003csup\u003e2\u003c/sup\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.07\u0026thinsp;\u0026plusmn;\u0026thinsp;0.12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.78\u0026ndash;1.37\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c7\"\u003e \u003cp\u003e1.03\u0026thinsp;\u0026plusmn;\u0026thinsp;0.09\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e0.78\u0026ndash;1.25\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c11\"\u003e \u003cp\u003e1.16\u0026thinsp;\u0026plusmn;\u0026thinsp;0.12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c12\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c13\"\u003e \u003cp\u003e0.87\u0026ndash;1.37\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c14\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c15\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.01\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBMD\u003csub\u003eDXA\u003c/sub\u003e(g/cm\u003csup\u003e2\u003c/sup\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.08\u0026thinsp;\u0026plusmn;\u0026thinsp;0.14\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.74\u0026ndash;1.49\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c7\"\u003e \u003cp\u003e1.03\u0026thinsp;\u0026plusmn;\u0026thinsp;0.13\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e0.74\u0026ndash;1.31\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c11\"\u003e \u003cp\u003e1.17\u0026thinsp;\u0026plusmn;\u0026thinsp;0.13\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c12\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c13\"\u003e \u003cp\u003e0.94\u0026ndash;1.49\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c14\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c15\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.01\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eZ-score\u003csub\u003eBIA\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-1.64\u0026thinsp;\u0026plusmn;\u0026thinsp;1.53\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-4.13-1.74\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c7\"\u003e \u003cp\u003e-1.67\u0026thinsp;\u0026plusmn;\u0026thinsp;1.36\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e-4.13-1.58\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c11\"\u003e \u003cp\u003e-1.58\u0026thinsp;\u0026plusmn;\u0026thinsp;1.87\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c12\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c13\"\u003e \u003cp\u003e-4.10-1.74\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c14\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c15\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.05\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eZ-score\u003csub\u003eDXA\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-0.79\u0026thinsp;\u0026plusmn;\u0026thinsp;1.23\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-3.67-2.44\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c7\"\u003e \u003cp\u003e-0.89\u0026thinsp;\u0026plusmn;\u0026thinsp;1.24\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e-3.67-1.78\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c11\"\u003e \u003cp\u003e-0.56\u0026thinsp;\u0026plusmn;\u0026thinsp;1.19\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c12\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c13\"\u003e \u003cp\u003e-2.75-2.44\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c14\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c15\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"15\"\u003e\u003cb\u003eNotes\u003c/b\u003e\u003c/td\u003e\u003c/tr\u003e \u003ctr\u003e\u003ctd colspan=\"15\"\u003eAll values are mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SDs;\u003c/td\u003e\u003c/tr\u003e \u003ctr\u003e\u003ctd colspan=\"15\"\u003e\u0026bull; \u003cb\u003eBMI\u003c/b\u003e\u0026thinsp;=\u0026thinsp;Body Mass Index\u003c/td\u003e\u003c/tr\u003e \u003ctr\u003e\u003ctd colspan=\"15\"\u003e\u0026bull; \u003cb\u003eBFP\u003c/b\u003e\u0026thinsp;=\u0026thinsp;Body Fat Percentage\u003c/td\u003e\u003c/tr\u003e \u003ctr\u003e\u003ctd colspan=\"15\"\u003e\u0026bull; \u003cb\u003eBMC\u003c/b\u003e\u0026thinsp;=\u0026thinsp;Bone Mineral Content\u003c/td\u003e\u003c/tr\u003e \u003ctr\u003e\u003ctd colspan=\"15\"\u003e\u0026bull; \u003cb\u003eBMD\u003c/b\u003e\u0026thinsp;=\u0026thinsp;Bone Mineral Density\u003c/td\u003e\u003c/tr\u003e \u003ctr\u003e\u003ctd colspan=\"15\"\u003e\u0026bull; Subscripts \u003cb\u003eBIA\u003c/b\u003e and \u003cb\u003eDXA\u003c/b\u003e refer to measurements obtained using the \u003cb\u003eSA201 bioelectrical impedance analyzer\u003c/b\u003e and \u003cb\u003edual-energy X-ray absorptiometry\u003c/b\u003e, respectively.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eAs shown in \u003cb\u003eTable\u0026nbsp;2\u003c/b\u003e, the classification of osteoporosis (Z-score\u0026thinsp;\u0026le;\u0026thinsp;\u0026minus;\u0026thinsp;2.5) differed between DXA and BIA. In the total sample (n\u0026thinsp;=\u0026thinsp;178), DXA identified 17 participants (9.6%) with osteoporosis, whereas BIA classified 55 participants (30.9%) as osteoporotic. Among women, DXA identified 14 cases (11.5%) compared with 35 cases (28.7%) by BIA; among men, DXA identified 3 cases (5.4%) compared with 20 cases (35.7%) by BIA. Overall, BIA demonstrated a higher detection rate of osteoporosis than DXA, with a particularly pronounced difference observed in men.\u003c/p\u003e \u003cp\u003e \u003cb\u003eSupplementary Figure S1\u003c/b\u003e shows the scatter plot and regression line of whole-body BMD measured by SA201 and DXA, indicating a strong linear relationship (regression equation: BIA\u0026thinsp;=\u0026thinsp;0.354\u0026thinsp;+\u0026thinsp;0.668 \u0026times; DXA, R\u0026sup2; = 0.671). The Passing\u0026ndash;Bablok regression (BIA\u0026thinsp;=\u0026thinsp;0.361\u0026thinsp;+\u0026thinsp;0.661 \u0026times; DXA) revealed that the 95% confidence intervals for slope and intercept included 1 and 0, respectively, indicating no significant proportional or fixed bias, although the intervals were wide (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe intraclass correlation coefficient (ICC) between BIA and DXA was 0.896, and the Pearson correlation coefficient was r\u0026thinsp;=\u0026thinsp;0.819 (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), indicating a high degree of correlation.\u003c/p\u003e \u003cp\u003e \u003cb\u003eSupplementary Figure S2\u003c/b\u003e illustrates the distribution of percentage differences between BIA and DXA, showing that the mean percentage difference was close to zero, although individual differences reached\u0026thinsp;\u0026plusmn;\u0026thinsp;10%. Bland\u0026ndash;Altman analysis (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e) demonstrated a mean difference of \u0026minus;\u0026thinsp;0.003 g/cm\u0026sup2;, with 95% limits of agreement ranging from \u0026minus;\u0026thinsp;0.166 to 0.160 g/cm\u0026sup2;, and most observations fell within \u0026plusmn;\u0026thinsp;0.16 g/cm\u0026sup2;. According to the criteria of Koo and Li, the ICC value of 0.896 represents \u0026ldquo;good\u0026rdquo; agreement, though insufficient for full interchangeability.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003ePaired t-test results (t\u0026thinsp;=\u0026thinsp;\u0026minus;\u0026thinsp;0.521, p\u0026thinsp;=\u0026thinsp;0.603) and Wilcoxon signed-rank test results (W\u0026thinsp;=\u0026thinsp;7691.000, p\u0026thinsp;=\u0026thinsp;0.690) indicated no statistically significant differences in mean values between BIA and DXA, confirming consistency at the group level.\u003c/p\u003e \u003cp\u003eError indices showed a coefficient of variation (CV%) of 7.72%, suggesting approximately 7\u0026ndash;8% inter-individual variability. The mean absolute percentage error (MAPE) was 6.39%, indicating an average deviation of about 6.5% between BIA and DXA. The root mean square percentage error (RMSPE) was 7.83%, reflecting an error magnitude of approximately 8% at the individual level, with larger deviations exerting greater influence.\u003c/p\u003e \u003cp\u003eReceiver operating characteristic (ROC) curve analysis (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e) demonstrated that BIA achieved an area under the curve (AUC) of 0.906, while DXA achieved an AUC of 0.979. Although DXA performed nearly perfectly (AUC\u0026thinsp;\u0026asymp;\u0026thinsp;0.98), BIA also demonstrated good diagnostic validity (AUC\u0026thinsp;\u0026asymp;\u0026thinsp;0.91). The optimal cutoff point for BIA was 0.969 g/cm\u0026sup2;, with a sensitivity of 88.2% and a specificity of 87.0%.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e\u003cp\u003e\u003cstrong\u003eTable 2.\u003c/strong\u003e Osteoporosis Classification of Participants by DXA and BIA (Z-score \u0026le; \u0026minus;2.5)\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" \u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 79px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 79px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003eZ-score of \u0026lt; -2.5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003eZ-score of \u0026gt; -2.5\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 79px;\"\u003e\n \u003cp\u003eGroup\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 79px;\"\u003e\n \u003cp\u003eMethod\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003e(osteoporosis)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003e(no osteoporosis)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 79px;\"\u003e\n \u003cp\u003eAll\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 79px;\"\u003e\n \u003cp\u003eDXA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003e17\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003e161\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 79px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 79px;\"\u003e\n \u003cp\u003eBIA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003e55\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003e123\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 79px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 79px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 79px;\"\u003e\n \u003cp\u003eFemale\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 79px;\"\u003e\n \u003cp\u003eDXA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003e14\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003e108\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 79px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 79px;\"\u003e\n \u003cp\u003eBIA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003e35\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003e98\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 79px;\"\u003e\n \u003cp\u003eMale\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 79px;\"\u003e\n \u003cp\u003eDXA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003e53\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 79px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 79px;\"\u003e\n \u003cp\u003eBIA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003e20\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003e36\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cstrong\u003eNotes\u003c/strong\u003e\u003c/p\u003e\n\u003cul type=\"disc\"\u003e\n \u003cli\u003eOsteoporosis was defined as \u003cstrong\u003eZ-score \u0026le; \u0026minus;2.5\u003c/strong\u003e according to the WHO diagnostic criteria [3,6].\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eDXA\u003c/strong\u003e = Dual-energy X-ray absorptiometry (gold standard).\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eBIA\u003c/strong\u003e = SA201 bioelectrical impedance analyzer (test method).\u003c/li\u003e\n\u003c/ul\u003e"},{"header":"Discussion","content":"\u003cp\u003eThe present study demonstrated that BIA exhibited a high correlation and good agreement with DXA in whole-body BMD measurement, and showed favorable discriminatory ability for diagnosing osteoporosis. Both the paired t-test and Wilcoxon signed-rank test revealed no significant mean differences between the two methods, while Bland\u0026ndash;Altman analysis indicated that most data points fell within the \u0026plusmn;\u0026thinsp;0.16 g/cm\u0026sup2; limits of agreement. These findings support the interchangeability of BIA and DXA at the population level. Such results are consistent with prior studies reporting that BIA-based devices could provide comparable trends to DXA in estimating bone mineral content and skeletal muscle mass, thereby supporting their utility in osteoporosis screening applications.\u003c/p\u003e \u003cp\u003eRecent studies have expanded the potential applications of BIA in bone health assessment. Martins et al. [\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e] investigated 167 university athletes and reported that the bioelectrical impedance-derived phase angle (PhA) was positively correlated with both whole-body and regional BMD (lumbar spine and femoral neck) measured by DXA in female athletes. Their findings suggested that PhA may serve as a potential marker of bone health in female athletes. Similarly, Ngai et al. [\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e] examined 735 Southern Chinese adults (345 postmenopausal women and 390 men) and found that BIA was significantly associated with lumbar spine, femoral neck, and total hip BMD measured by DXA. In men, BIA was an independent predictor of BMD beyond age and body weight. ROC analyses indicated that BIA predicted osteoporosis with an AUC of 0.65\u0026ndash;0.75, with only modest improvements when combined with age and weight. This study emphasized the potential of BIA as a clinical risk assessment tool rather than a standalone diagnostic modality.\u003c/p\u003e \u003cp\u003eIn postmenopausal women, Ono et al. [\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e] demonstrated that PhA derived from BIA was positively correlated with both physical function and health-related quality of life (SF-36 scores), as well as with lumbar spine and femoral neck BMD. Even after adjusting for confounders such as age, BMI, appendicular skeletal muscle mass index (ASMI), and BMD, PhA remained a significant predictor of the physical component summary score, suggesting that BIA-derived PhA reflects not only cellular health but also quality of life among patients with osteoporosis. \u0026Ouml;zt\u0026uuml;rk et al. [\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e] conducted bioelectrical impedance spectroscopy (BIS) in 48 postmenopausal women and reported significant negative correlations between BIS characteristic frequency and DXA-derived hip BMD (r\u0026thinsp;=\u0026thinsp;\u0026minus;\u0026thinsp;0.53, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) and lumbar spine BMD (r\u0026thinsp;=\u0026thinsp;\u0026minus;\u0026thinsp;0.37, p\u0026thinsp;\u0026lt;\u0026thinsp;0.05). ROC analysis demonstrated that BIS frequency effectively distinguished between normal and osteoporotic participants, particularly in the hip subgroup (AUC\u0026thinsp;=\u0026thinsp;0.91).\u003c/p\u003e \u003cp\u003eChuang et al. [\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e] also compared whole-body BMD measurements between BIA and DXA in 318 participants and reported moderate-to-strong correlations but limited clinical interchangeability. Compared with their study, the present work extended methodological rigor by incorporating Passing\u0026ndash;Bablok regression, ICC, Wilcoxon tests, MAPE, RMSPE, and ROC analysis. These additional statistical evaluations provided more comprehensive evidence regarding diagnostic validity and clinical classification accuracy. Importantly, while Chuang et al. focused on a younger sample (mean age\u0026thinsp;\u0026asymp;\u0026thinsp;37 years), the present study included a broader age range, encompassing clinically relevant populations at higher risk of osteoporosis. Our ROC analysis (AUC\u0026thinsp;=\u0026thinsp;0.906) further demonstrated that SA201 could serve as a reliable screening tool, with sensitivity (88.2%) and specificity (87.0%) at the optimal cutoff. Moreover, error quantification (CV\u0026thinsp;\u0026asymp;\u0026thinsp;7.7%, MAPE\u0026thinsp;=\u0026thinsp;6.5%, RMSPE\u0026thinsp;=\u0026thinsp;8.3%) highlighted the acceptable consistency of BIA at the population level but emphasized caution in individual-level interpretation.\u003c/p\u003e \u003cp\u003eHowever, the wide confidence intervals observed in the Passing\u0026ndash;Bablok regression indicate substantial heterogeneity in individual impedance responses, particularly across sex and age groups. This suggests that fixed or proportional bias cannot be fully excluded. Although the average trend appears acceptable, such uncertainty implies that the two methods cannot be considered clinically interchangeable at the individual level. The magnitude of individual-level error (MAPE\u0026thinsp;\u0026asymp;\u0026thinsp;6.5% and RMSPE\u0026thinsp;\u0026asymp;\u0026thinsp;8%) is relatively high for clinical decision-making and insufficient to support diagnostic use. Errors of this magnitude may lead to misclassification for individual patients.\u003c/p\u003e \u003cp\u003eThe wide age range of participants (25\u0026ndash;86 years) substantially increased inter-individual variability in BMD, given that age is a major determinant of bone mineral density. When both DXA and BIA capture this age-related decline, the shared variance may artificially inflate Pearson correlation coefficients, even when systematic or individual-level discrepancies remain between the two methods. Therefore, a high correlation should not be interpreted as evidence of clinical interchangeability. Correlation reflects only relative ranking rather than absolute agreement. Accordingly, method comparison should rely primarily on metrics that are less influenced by age-driven group variability\u0026mdash;such as ICC, Passing\u0026ndash;Bablok regression, Bland\u0026ndash;Altman analysis, and MAPE/RMSPE\u0026mdash;rather than correlation coefficients alone.\u003c/p\u003e \u003cp\u003eIn \u003cb\u003eSupplementary Figure S1\u003c/b\u003e, the regression line shows a slope similar to that of the Passing\u0026ndash;Bablok regression in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. Although the 95% confidence intervals for the slope (\u0026minus;\u0026thinsp;25.84 to 30.20) and intercept (\u0026minus;\u0026thinsp;31.19 to 28.75) include the values corresponding to perfect agreement (i.e., a slope of 1 and an intercept of 0), the plotted confidence bands do not fully encompass the line of perfect concordance. This indicates that, within this study population, BIA-derived estimates of BMD may exhibit systematic underestimation or overestimation.\u003c/p\u003e \u003cp\u003ePassing\u0026ndash;Bablok regression was selected in this study because it is a non-parametric method that is robust to outliers and does not require assumptions of normally distributed errors or homoscedasticity. This makes it particularly suitable for BMD analyses in populations with substantial biological variability. The method also allows simultaneous estimation of both fixed bias (intercept) and proportional bias (slope). In contrast, ordinary least squares (OLS) regression assumes normally distributed errors with constant variance, which may not be appropriate under these conditions.\u003c/p\u003e \u003cp\u003eThe percentage bias displayed in \u003cb\u003eSupplementary Figure S2\u003c/b\u003e shows an approximately symmetric distribution; however, it does not fully conform to normality because the values are scaled by the denominator, which typically produces asymmetric and non-normally distributed data. This observation further supports the use of the non-parametric Passing\u0026ndash;Bablok regression method in this study.\u003c/p\u003e \u003cp\u003eBecause the absolute range of whole-body BMD values is narrow (e.g., 1.0\u0026ndash;1.2 g/cm\u0026sup2;), even small absolute errors of 0.01\u0026ndash;0.02 g/cm\u0026sup2; can produce apparently high coefficients of variation (CV%). In addition, greater physiological variability within the measured population\u0026mdash;such as fluctuations in hydration distribution, plantar contact pressure, and limb temperature\u0026mdash;can alter impedance within short periods of time, thereby increasing retest variability. The elevated CV% further indicates that BIA and DXA are not interchangeable at the individual level. Compared with the approximately 1% precision typically achieved by DXA, the CV observed for BIA in this study reflects a repeatability that is at least an order of magnitude poorer, even under strictly standardized pre-measurement conditions.\u003c/p\u003e \u003cp\u003eThe wide limits of agreement (LOA) indicate substantial individual-level discrepancies between BIA and DXA measurements. When the BMD differences are converted to Z-scores, the magnitude corresponds to approximately\u0026thinsp;\u0026plusmn;\u0026thinsp;1.5 SD (depending on the SD of the DXA reference database). Such a range far exceeds the level of error acceptable for clinical diagnosis or longitudinal monitoring. Because these discrepancies exceed the least significant change (LSC) of DXA, BIA-derived whole-body BMD is not suitable for patient follow-up and should not be used for any diagnostic or therapeutic decision-making. Errors of this magnitude are sufficient to cause clinically meaningful misclassification, including misidentifying normal individuals as osteoporotic, classifying osteoporotic patients as normal, or incorrectly categorizing individuals within the osteopenic range. Even if the mean bias appears small, the large individual-level variability considerably limits the reliability of BIA-based BMD for diagnostic classification.\u003c/p\u003e \u003cp\u003eThe primary comparison metric in this study was the absolute difference in BMD between DXA and BIA. Absolute differences were used for the main analyses, including Bland\u0026ndash;Altman plots, ICC, and MAPE. Percentage differences, which exhibit different distributional properties due to their dependence on the denominator, were not used as a basis for the primary inferences. Therefore, percentage differences were presented only in the supplementary materials for exploratory purposes.\u003c/p\u003e \u003cp\u003eAccording to the WHO and major clinical guidelines (e.g., ISCD, NOF), osteoporosis diagnosis should primarily rely on Z-scores derived from the lumbar spine (L1\u0026ndash;L4), femoral neck, or total hip [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e, \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e]. Among these, femoral neck BMD is most predictive of hip fracture risk [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e], while lumbar spine BMD is particularly sensitive to vertebral fracture risk [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]. Lumbar BMD can detect early bone loss at the onset of aging but may be falsely elevated in older individuals due to degenerative changes such as disc calcification, osteoarthritis, or scoliosis, potentially leading to underestimation of osteoporosis risk [\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e, \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e]. Whole-body BMD, by averaging across skeletal regions, can mitigate localized biases and is particularly useful in epidemiological and body composition studies (e.g., sarcopenia, obesity, and bone health) [\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e]. However, whole-body Z-scores are not currently recommended as a diagnostic criterion by WHO or clinical practice [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e, \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eBecause averaging may dilute localized bone loss, whole-body BMD has lower predictive power for fracture risk compared with site-specific measures [\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e, \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e]. According to WHO and ISCD guidelines, whole-body BMD is not an accepted diagnostic criterion for osteoporosis. Whole-body BMD values are generally higher than those of the lumbar spine and hip and are less sensitive to localized bone loss.\u003c/p\u003e \u003cp\u003eAlthough BIA shows a high correlation with whole-body DXA BMD, this does not imply suitability for clinical diagnosis. Lumbar spine and femoral neck BMD remain the required anatomical sites for diagnostic classification. Given its portability and radiation-free nature, BIA may serve as a preliminary screening tool to identify high-risk individuals who can then be referred for confirmatory DXA assessment. The lumbar spine and femoral neck are metabolically active regions with high fracture relevance, whereas whole-body BMD represents an average across multiple skeletal sites, thereby reducing sensitivity to localized bone loss. Consequently, whole-body BMD cannot be used for clinical interpretation or diagnostic decision-making。\u003c/p\u003e \u003cp\u003e According to WHO and ISCD guidelines, the diagnosis of osteoporosis must be based on BMD measured at the lumbar spine, femoral neck, or total hip, rather than whole-body BMD. Whole-body BMD values are typically higher and less sensitive to localized bone loss. Consequently, applying whole-body Z-score thresholds to BIA-derived measurements will inherently yield higher rates of abnormal classification, largely because whole-body averages dilute focal reductions in bone density. In addition, foot-to-foot BIA estimation of whole-body BMD may differ substantially from clinically relevant skeletal sites such as the hip and lumbar spine. At present, no validated clinical diagnostic thresholds exist for whole-body BMD\u0026mdash;whether measured by BIA or DXA. Therefore, even if whole-body BMD can be obtained, the absence of standardized cut-offs precludes its use for clinical diagnosis. Because this study assessed only whole-body BMD, rather than lumbar spine or hip BMD, the findings should not be applied to osteoporosis diagnosis nor to clinical fracture-risk stratification.\u003c/p\u003e \u003cp\u003eAge and sex are major determinants of BMD; thus, the wide age range and mixed-sex composition of the sample (25\u0026ndash;86 years) naturally produced substantial inter-individual variability. Under such conditions, even a method that merely approximates the age-related decline in BMD may appear to differentiate individuals with lower BMD. Therefore, the identification of 17 low-BMD cases in this study may largely reflect the structure of the sample itself rather than the precision of the estimation method.\u003c/p\u003e \u003cp\u003eSeveral factors may explain why the device used in this study could appear more accurate: the prediction model was directly calibrated against whole-body BMD measured by DXA, and the use of multiple impedance frequencies may capture richer information on tissue properties. However, several factors may also reduce accuracy. The current pathway does not encompass clinically relevant skeletal sites such as the hip and lumbar spine. Age-related changes in hydration, muscle mass, and limb geometry may violate key model assumptions. In addition, the algorithm has not been validated in older adults or osteoporotic populations, which may introduce bias.\u003c/p\u003e \u003cp\u003eAlthough this study employed stringent pre-measurement standardization procedures\u0026mdash;such as fixed measurement time, bladder emptying, avoidance of exercise and food intake, hydration control, and standardized posture and electrode contact\u0026mdash;these conditions are difficult to replicate consistently in real-world clinical settings. Therefore, in practical applications, the error and repeatability limitations of BIA-derived whole-body BMD may be more pronounced than those observed in this controlled research environment.\u003c/p\u003e \u003cp\u003eThe SA201 utilizes a four-electrode, foot-to-foot configuration with dual-frequency measurements and a dedicated algorithm specifically developed for estimating whole-body BMD using DXA-calibrated data. Unlike conventional BIA systems designed for general body composition assessment, the SA201 algorithm incorporates the relationship between lower-limb impedance and trunk composition. However, because the electrical current primarily travels through the lower limbs, it cannot directly capture characteristics of axial skeletal sites such as the spine and hip.\u003c/p\u003e \u003cp\u003eIn summary, BIA technology is gaining recognition not only for estimating BMD but also for providing additional insights into cellular health, quality of life, and functional outcomes. The present findings reinforce the potential of BIA as a practical tool for osteoporosis screening in both clinical and community settings, complementing but not replacing DXA.\u003c/p\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eLimitations and Future Directions\u003c/h2\u003e \u003cp\u003eThis study has several limitations. First, the sample size was relatively modest, which may have limited the statistical power and generalizability of the findings. Second, the study population was drawn from a single geographic region, resulting in limited heterogeneity, which may restrict extrapolation to other populations. Third, the analysis focused solely on whole-body BMD, without comparing site-specific measurements (e.g., lumbar spine and femoral neck), which are of greater clinical importance in the diagnosis of osteoporosis. Finally, as BIA is a novel device, further validation across multiple centers and diverse populations is necessary to confirm its accuracy and reproducibility.\u003c/p\u003e \u003cp\u003eFuture studies should address these limitations by: (1) recruiting larger and more diverse cohorts encompassing different age groups, sexes, and clinical conditions to improve the representativeness of findings; (2) evaluating the validity and agreement of BIA with DXA at clinically critical skeletal sites (e.g., femoral neck, lumbar spine); and (3) investigating the utility of BIA in community screening, longitudinal monitoring, and epidemiological research. Such studies will be essential to establish the broader clinical applicability of BIA in osteoporosis prevention and management. The absence of fracture outcomes represents a major limitation of this study. Clinically meaningful validation should assess whether BIA-derived measures can improve the accuracy of FRAX or other fracture-risk models. Future research should include prospective follow-up and comparisons with site-specific DXA measurements to determine the predictive value and clinical utility of this approach.\u003c/p\u003e \u003c/div\u003e"},{"header":"Conclusion","content":"\u003cp\u003eBIA demonstrated a strong correlation and good agreement with DXA in whole-body BMD measurement, as well as excellent discriminatory ability in the diagnosis of osteoporosis (AUC\u0026thinsp;\u0026gt;\u0026thinsp;0.9). However, the results also revealed considerable individual-level error (approximately 6\u0026ndash;8%), very wide limits of agreement, and limited repeatability. In addition, accurate estimation requires strict pre-measurement standardization, and the Passing\u0026ndash;Bablok regression indicated substantial estimation uncertainty. BIA is therefore not interchangeable with DXA, with the overly wide LOA being the primary limiting factor. Consequently, BIA is suitable as an initial clinical screening tool and for epidemiological research, whereas DXA remains the essential standard for the diagnostic confirmation of osteoporosis.\u003c/p\u003e "},{"header":"Declarations","content":"\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eData availability statement\u003c/h2\u003e \u003cp\u003eThe raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eEthics statement\u003c/h2\u003e \u003cp\u003eThe studies involving humans were approved by Institutional Review Board (IRB) of the Ministry of Health and Welfare Nantou Hospital (IRB-111047 and IRB-113002). The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.\u003c/p\u003e \u003c/div\u003e\u003cp\u003e \u003ch2\u003eConflict of interest\u003c/h2\u003e \u003cp\u003eK-CH was employed by a commercial company, StarBIA Meditek Co. Ltd., during this study.\u003c/p\u003e \u003c/p\u003e\u003ch2\u003eFunding\u003c/h2\u003e \u003cp\u003eThe author(s) declare that financial support was received for the research, authorship, and/or publication of this article. Funding for this work has been provided by Ministry of Health and Welfare Hospital Research and Development plan.\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eH.-K.L. conceived the study, designed the methodology. H.-K.L. and T.-J.H. conducted the investigations. C.-L.L. and W.-C.T. performed data curation, while C.-L.L. was responsible for funding acquisition. C.-W.L. and A.-C.H. performed the formal analysis. K.-C.H. provided the software. H.-K.L. and C.-L.L. carried out the validation of the results. A.-C.H. prepared the visualizations. H.-K.L. and C.-L.L. supervised the project. H.-K.L. wrote the original draft of the manuscript, and H.-K.L. and K.-C.H. reviewed and edited the final text. All authors reviewed the manuscript.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eThe raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eNIH Consensus Development Panel on Osteoporosis Prevention. Diagnosis, and Therapy. Osteoporosis prevention, diagnosis, and therapy. \u003cem\u003eJAMA\u003c/em\u003e \u003cb\u003e285\u003c/b\u003e, 785\u0026ndash;795 (2001).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRachner, T. D., Khosla, S. \u0026amp; Hofbauer, L. C. Osteoporosis: now and the future. \u003cem\u003eLancet\u003c/em\u003e \u003cb\u003e377\u003c/b\u003e, 1276\u0026ndash;1287 (2011).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWorld Health Organization. Assessment of fracture risk and its application to screening for postmenopausal osteoporosis. WHO Technical Report Series, No. 843. WHO, (1994).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCummings, S. R. \u0026amp; Melton, L. J. Epidemiology and outcomes of osteoporotic fractures. \u003cem\u003eLancet\u003c/em\u003e \u003cb\u003e359\u003c/b\u003e, 1761\u0026ndash;1767 (2002).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKanis, J. A. et al. European guidance for the diagnosis and management of osteoporosis in postmenopausal women. \u003cem\u003eOsteoporos. Int.\u003c/em\u003e \u003cb\u003e30\u003c/b\u003e, 3\u0026ndash;44 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKanis, J. A. et al. The diagnosis of osteoporosis. \u003cem\u003eJ. Bone Min. Res.\u003c/em\u003e \u003cb\u003e9\u003c/b\u003e, 1137\u0026ndash;1141 (1994).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMarshall, D., Johnell, O. \u0026amp; Wedel, H. Meta-analysis of how well measures of bone mineral density predict occurrence of osteoporotic fractures. \u003cem\u003eBMJ\u003c/em\u003e \u003cb\u003e312\u003c/b\u003e, 1254\u0026ndash;1259 (1996).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJohnell, O. \u0026amp; Kanis, J. A. An estimate of the worldwide prevalence and disability associated with osteoporotic fractures. \u003cem\u003eOsteoporos. Int.\u003c/em\u003e \u003cb\u003e17\u003c/b\u003e, 1726\u0026ndash;1733 (2006).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSeeman, E. \u0026amp; Delmas, P. D. Bone quality\u0026mdash;the material and structural basis of bone strength and fragility. \u003cem\u003eN Engl. J. Med.\u003c/em\u003e \u003cb\u003e354\u003c/b\u003e, 2250\u0026ndash;2261 (2006).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRaisz, L. G. Pathogenesis of osteoporosis: concepts, conflicts, and prospects. \u003cem\u003eJ. Clin. Invest.\u003c/em\u003e \u003cb\u003e115\u003c/b\u003e, 3318\u0026ndash;3325 (2005).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRiggs, B. L., Khosla, S. \u0026amp; Melton, L. J. 3rd. Sex steroids and the construction and conservation of the adult skeleton. \u003cem\u003eEndocr. Rev.\u003c/em\u003e \u003cb\u003e23\u003c/b\u003e, 279\u0026ndash;302 (2002).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCompston, J. E. Sex steroids and bone. \u003cem\u003ePhysiol. Rev.\u003c/em\u003e \u003cb\u003e81\u003c/b\u003e, 419\u0026ndash;447 (2001).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHolick, M. F. Vitamin D deficiency. \u003cem\u003eN Engl. J. Med.\u003c/em\u003e \u003cb\u003e357\u003c/b\u003e, 266\u0026ndash;281 (2007).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEastell, R. et al. Modern Spectrum Bone turnover markers: are they clinically useful? \u003cem\u003eEur. J. Endocrinol.\u003c/em\u003e \u003cb\u003e178\u003c/b\u003e, R19\u0026ndash;R31 (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMiller, P. D. et al. Clinical utility of bone mass measurements in adults: Consensus of an international panel. \u003cem\u003eSemin Arthritis Rheum.\u003c/em\u003e \u003cb\u003e25\u003c/b\u003e, 361\u0026ndash;372 (1996).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKanis, J. A. et al. The diagnosis of osteoporosis. \u003cem\u003eJ. Bone Min. Res.\u003c/em\u003e \u003cb\u003e9\u003c/b\u003e, 1137\u0026ndash;1141 (1994).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBlake, G. M. \u0026amp; Fogelman, I. The role of DXA bone density scans in the diagnosis and treatment of osteoporosis. \u003cem\u003ePostgrad. Med. J.\u003c/em\u003e \u003cb\u003e83\u003c/b\u003e, 509\u0026ndash;517 (2007).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEngelke, K. et al. Clinical use of quantitative computed tomography (QCT) of the hip in the management of osteoporosis in Adults: the 2015 ISCD official Positions \u0026ndash; Part I. \u003cem\u003eJ. Clin. Densitom\u003c/em\u003e. \u003cb\u003e18\u003c/b\u003e, 338\u0026ndash;358 (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLink, T. M. Osteoporosis imaging: state of the art and advanced imaging. \u003cem\u003eRadiology\u003c/em\u003e \u003cb\u003e263\u003c/b\u003e, 3\u0026ndash;17 (2012).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBousson, V. et al. Volumetric quantitative computed tomography of the proximal femur: relationships linking geometric and densitometric variables to bone strength. Role for compact bone. \u003cem\u003eOsteoporos. Int.\u003c/em\u003e \u003cb\u003e17\u003c/b\u003e, 855\u0026ndash;864 (2006).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMacNeil, J. A. \u0026amp; Boyd, S. K. Bone strength at the distal radius can be estimated from high-resolution peripheral quantitative computed tomography and finite element analysis. \u003cem\u003eBone\u003c/em\u003e \u003cb\u003e42\u003c/b\u003e, 1203\u0026ndash;1213 (2008).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKrieg, M. A. et al. Quantitative ultrasound in the management of osteoporosis: the 2007 ISCD Official Positions. \u003cem\u003eJ. Clin. Densitom\u003c/em\u003e. \u003cb\u003e11\u003c/b\u003e, 163\u0026ndash;187 (2008).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMoayyeri, A. et al. Quantitative ultrasound of the heel and fracture risk assessment: an updated meta-analysis. \u003cem\u003eOsteoporos. Int.\u003c/em\u003e \u003cb\u003e23\u003c/b\u003e, 143\u0026ndash;153 (2012).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNayak, S. et al. Meta-analysis: accuracy of quantitative ultrasound for identifying patients with osteoporosis. \u003cem\u003eOsteoporos. Int.\u003c/em\u003e \u003cb\u003e17\u003c/b\u003e, 801\u0026ndash;810 (2006).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKyle, U. G. et al. Bioelectrical impedance analysis\u0026mdash;part I: review of principles and methods. \u003cem\u003eClin. Nutr.\u003c/em\u003e \u003cb\u003e23\u003c/b\u003e, 1226\u0026ndash;1243 (2004).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKushner, R. F. \u0026amp; Schoeller, D. A. Estimation of total body water by bioelectrical impedance analysis. \u003cem\u003eAm. J. Clin. Nutr.\u003c/em\u003e \u003cb\u003e44\u003c/b\u003e, 417\u0026ndash;424 (1986).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChuang, C. L. et al. Comparison of whole body bone mineral density measurement between dual energy X ray absorptiometry and novel bioelectrical impedance analysis. \u003cem\u003eSci. Rep.\u003c/em\u003e 14, (待補頁碼或文章編號) (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBland, J. M. \u0026amp; Altman, D. G. Statistical methods for assessing agreement between two methods of clinical measurement. \u003cem\u003eLancet\u003c/em\u003e \u003cb\u003e1\u003c/b\u003e, 307\u0026ndash;310 (1986).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePassing, H. \u0026amp; Bablok, W. A new biometrical procedure for testing the equality of measurements from two different analytical methods. \u003cem\u003eJ. Clin. Chem. Clin. Biochem.\u003c/em\u003e \u003cb\u003e21\u003c/b\u003e, 709\u0026ndash;720 (1983).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBilić-Zulle, L. Comparison of methods: Passing\u0026ndash;Bablok regression. \u003cem\u003eBiochem. Med.\u003c/em\u003e \u003cb\u003e21\u003c/b\u003e, 49\u0026ndash;52 (2011).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKoo, T. K. \u0026amp; Li, M. Y. A guideline of selecting and reporting intraclass correlation coefficients for reliability research. \u003cem\u003eJ. Chiropr. Med.\u003c/em\u003e \u003cb\u003e15\u003c/b\u003e, 155\u0026ndash;163 (2016).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMartins, P. C., Moraes, M. S. \u0026amp; Silva, D. A. How is the phase angle associated with total and regional bone mineral density in university athletes? \u003cem\u003ePhysiol. Meas.\u003c/em\u003e \u003cb\u003e42\u003c/b\u003e, 085001 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNgai, H. H. Y., Cheung, C. L., Yao, T. J. \u0026amp; Kung, A. W. C. Bioimpedance: can its addition to simple clinical criteria enhance the diagnosis of osteoporosis? \u003cem\u003eJ. Bone Min. Metab.\u003c/em\u003e \u003cb\u003e27\u003c/b\u003e, 372\u0026ndash;378 (2009).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOno, Y., Kasukawa, Y., Sasaki, K. \u0026amp; Miyakoshi, N. Association of the Bioimpedance Phase Angle and Quality of Life in Postmenopausal Osteoporosis. \u003cem\u003eMed. Princ Pract.\u003c/em\u003e \u003cb\u003e32\u003c/b\u003e, 71\u0026ndash;76 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e\u0026Ouml;zt\u0026uuml;rk, N., Ozturk-Isik, E. \u0026amp; \u0026Uuml;lgen, Y. Screening post-menopausal women for bone mineral level by bioelectrical impedance spectroscopy of dominant arm. \u003cem\u003eJ. Electr. Bioimp\u003c/em\u003e. \u003cb\u003e9\u003c/b\u003e, 39\u0026ndash;47 (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLewiecki, E. M. et al. Official positions of the International Society for Clinical Densitometry. \u003cem\u003eJ. Clin. Endocrinol. Metab.\u003c/em\u003e \u003cb\u003e89\u003c/b\u003e, 3651\u0026ndash;3655 (2004).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCosman, F. et al. Clinician\u0026rsquo;s Guide to Prevention and Treatment of Osteoporosis. \u003cem\u003eOsteoporos. Int.\u003c/em\u003e \u003cb\u003e25\u003c/b\u003e, 2359\u0026ndash;2381 (2014).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRiggs, B. L. \u0026amp; Melton, L. J. The prevention and treatment of osteoporosis. \u003cem\u003eN Engl. J. Med.\u003c/em\u003e \u003cb\u003e327\u003c/b\u003e, 620\u0026ndash;627 (1992).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYu, W. et al. Spinal bone mineral assessment in postmenopausal women: comparison of dual-energy X-ray absorptiometry and quantitative computed tomography. \u003cem\u003eOsteoporos. Int.\u003c/em\u003e \u003cb\u003e5\u003c/b\u003e, 433\u0026ndash;439 (1995).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShepherd, J. A., Ng, B. K., Sommer, M. J. \u0026amp; Heymsfield, S. B. Body composition by DXA. \u003cem\u003eBone\u003c/em\u003e \u003cb\u003e104\u003c/b\u003e, 101\u0026ndash;105 (2017).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDe Laet, C. et al. Hip fracture prediction in elderly men and women: validation in the Rotterdam Study. \u003cem\u003eJ. Bone Min. Res.\u003c/em\u003e \u003cb\u003e13\u003c/b\u003e, 1587\u0026ndash;1593 (1998).\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Z-score, Dual-energy X-ray absorptiometry (DXA), Whole-body bone mineral density (BMD), Bioelectrical impedance analysis (BIA), Agreement","lastPublishedDoi":"10.21203/rs.3.rs-9314239/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9314239/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003eDual-energy X-ray absorptiometry (DXA) is the gold standard for diagnosing osteoporosis but is limited by high cost, bulky size, and lack of portability. The SA201 is a novel bioelectrical impedance analysis (BIA) device designed to estimate whole-body bone mineral density (BMD). This study aimed to evaluate the accuracy of SA201 measurements and their agreement with DXA.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003eA total of 178 participants (56 men and 122 women; mean age 52.6\u0026thinsp;\u0026plusmn;\u0026thinsp;14.8 years) underwent whole-body BMD assessments using both SA201 and DXA. Statistical analyses included descriptive statistics, Pearson correlation, intraclass correlation coefficient (ICC), Passing\u0026ndash;Bablok regression, Bland\u0026ndash;Altman analysis, paired t-tests, Wilcoxon signed-rank tests, coefficient of variation (CV%), mean absolute percentage error (MAPE), root mean square percentage error (RMSPE), and receiver operating characteristic (ROC) curve analysis.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eBIA-derived BMD was strongly correlated with DXA (r\u0026thinsp;=\u0026thinsp;0.819, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001; ICC (2,1)\u0026thinsp;=\u0026thinsp;0.896, indicating good reliability). Paired t-tests (p\u0026thinsp;=\u0026thinsp;0.603) and Wilcoxon tests (p\u0026thinsp;=\u0026thinsp;0.690) showed no significant mean differences. Bland\u0026ndash;Altman analysis demonstrated a mean difference of \u0026minus;\u0026thinsp;0.003 g/cm\u0026sup2; with 95% limits of agreement between \u0026minus;\u0026thinsp;0.166 and 0.160 g/cm\u0026sup2;. Passing\u0026ndash;Bablok regression produced the equation: SA201\u0026thinsp;=\u0026thinsp;0.361\u0026thinsp;+\u0026thinsp;0.661 \u0026times; DXA, with wide confidence intervals. Error metrics indicated CV% \u0026asymp; 7.7%, MAPE\u0026thinsp;=\u0026thinsp;6.5%, and RMSPE\u0026thinsp;=\u0026thinsp;8.3%. For osteoporosis diagnosis, ROC analysis yielded an AUC of 0.906, with an optimal cutoff of 0.969 g/cm\u0026sup2; (sensitivity 88.2%, specificity 87.0%).\u003c/p\u003e\u003ch2\u003eConclusions\u003c/h2\u003e \u003cp\u003eSA201 demonstrated strong correlation and good agreement with DXA for whole-body BMD measurement, as well as good discriminatory ability in osteoporosis screening (AUC\u0026thinsp;\u0026gt;\u0026thinsp;0.9). However, individual-level errors of 6\u0026ndash;8% and wide regression confidence intervals suggest limited interchangeability. BIA may serve as a practical tool for community screening and epidemiological research, while DXA remains indispensable for definitive clinical diagnosis of osteoporosis.\u003c/p\u003e","manuscriptTitle":"Application of a Standing Quad-Plate Bioelectrical Impedance Analyzer for Whole-Body Bone Mineral Density Measurement in Osteoporosis Screening","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-05-13 17:57:17","doi":"10.21203/rs.3.rs-9314239/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"reviewerAgreed","content":"114201658114512350232444410800956958729","date":"2026-05-07T07:33:20+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-07T02:16:59+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"264181171377200280669658649123456846580","date":"2026-05-06T15:52:27+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"272946400676242223417221583274267515316","date":"2026-05-05T08:25:12+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"121464614510344299937168894790573333518","date":"2026-05-05T07:56:41+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-05-05T07:05:01+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2026-04-13T10:10:57+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-04-07T11:16:20+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-04-07T11:15:45+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2026-04-03T14:57:54+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"92136c91-0070-4163-8226-e0811a3ab5d0","owner":[],"postedDate":"May 13th, 2026","published":true,"recentEditorialEvents":[{"type":"reviewerAgreed","content":"114201658114512350232444410800956958729","date":"2026-05-07T07:33:20+00:00","index":81,"fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-07T02:16:59+00:00","index":80,"fulltext":""},{"type":"reviewerAgreed","content":"264181171377200280669658649123456846580","date":"2026-05-06T15:52:27+00:00","index":79,"fulltext":""},{"type":"reviewerAgreed","content":"272946400676242223417221583274267515316","date":"2026-05-05T08:25:12+00:00","index":75,"fulltext":""},{"type":"reviewerAgreed","content":"121464614510344299937168894790573333518","date":"2026-05-05T07:56:41+00:00","index":74,"fulltext":""},{"type":"reviewersInvited","content":"13","date":"2026-05-05T07:05:01+00:00","index":"","fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":67730719,"name":"Health sciences/Biomarkers"},{"id":67730720,"name":"Health sciences/Diseases"},{"id":67730721,"name":"Health sciences/Endocrinology"},{"id":67730722,"name":"Health sciences/Health care"},{"id":67730723,"name":"Health sciences/Medical research"}],"tags":[],"updatedAt":"2026-05-13T17:57:17+00:00","versionOfRecord":[],"versionCreatedAt":"2026-05-13 17:57:17","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9314239","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9314239","identity":"rs-9314239","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}