Abstract
Background: Osteoporosis represents a critical public health challenge, particularly among postmenopausal women, characterized by reduced bone mineral density (BMD) and increased fracture risk. Traditional obesity metrics such as body mass index (BMI) and waist circumference (WC) have limitations in assessing bone health due to their inability to differentiate lean and fat mass. The weight-adjusted waist index (WWI), a novel anthropometric indicator, offers a more nuanced approach to evaluating body composition and metabolic risks.
Objective: The WWI is a novel obesity metric that demonstrates superior accuracy in evaluating both muscle mass and adiposity compared to conventional anthropometric measurements. This cross-sectional study examined the relationship between WWI and BMD at the femoral neck in a nationally representative sample of American postmenopausal women.
Methods: A cross-sectional analysis was conducted using data from the National Health and Nutrition Examination Survey (NHANES), including 3198 postmenopausal women aged 40 and older. WWI was calculated by dividing WC by the square root of body weight. BMD was assessed using dual-energy X-ray absorptiometry (DXA). Multivariate regression analyses were performed, adjusting for potential confounders including age, race, height, BMI, and metabolic markers.
Results: Multiple regression analyses revealed a significant negative correlation between WWI and femoral neck BMD. A critical threshold of 10.32 was identified, beyond which the relationship with BMD shifted. Below this threshold, higher WWI levels showed a protective effect on BMD (β = 0.03, p=0.0265), while above the threshold, WWI exhibited a significant negative influence (β = −0.01, p < 0.0001). Stratified analyses demonstrated variations in the WWI-BMD relationship across different subgroups, including age, race, and menopause status.
Conclusion: This cross-sectional analysis revealed a significant negative correlation between the WWI and femoral neck BMD in a nationally representative cohort of American postmenopausal women. Notably, the association demonstrated a nonlinear inverted U-shaped pattern with an identifiable threshold effect.
Keywords: bone mineral density, NHANES, osteoporosis, postmenopausal women, weight-adjusted waist index
1. Background
Osteoporosis, characterized by reduced bone mineral density (BMD) and increased fracture risk, represents a critical public health burden, particularly among postmenopausal women, with estrogen deficiency significantly accelerating bone loss [1, 2]. Traditional obesity metrics such as body mass index (BMI) and waist circumference (WC) have been extensively studied for their associations with BMD; however, these measures have limitations due to their failure to differentiate between lean mass and fat mass, which raises concerns about their reliability in assessing bone health [3, 4]. The weight-adjusted waist index (WWI), a novel anthropometric indicator that integrates weight and height, aims to better reflect central obesity and associated metabolic risks compared to conventional indices [5, 6]. Preliminary findings show that WWI has demonstrated superior performance in predicting cardiovascular and metabolic disorders, suggesting its potential utility in interdisciplinary health assessments [7, 8].
Emerging evidence suggests that fat distribution and metabolic dysfunction, rather than total adiposity alone, may exert differential effects on bone remodeling. Recent studies emphasize that localized fat distribution, especially visceral fat, negatively impacts bone health, independent of overall body weight. For instance, investigations highlight how metabolic dysfunction can lead to a state where increased fat mass does not equate to improved BMD, particularly in populations such as postmenopausal women who experience hormonal changes affecting bone density [9, 10]. Despite the significant implications of these findings, studies evaluating the WWI, which emphasizes the waist-to-weight ratio, in the context of osteoporosis are scarce. This oversight in research prompts a critical examination of WWI's prognostic value for skeletal health, particularly in light of contemporary understandings of how body composition affects osteoporosis risk [11].
The National Health and Nutrition Examination Survey (NHANES) provides a unique opportunity to address this gap in research literature. NHANES offers nationally representative data alongside standardized dual-energy X-ray absorptiometry (DXA) measurements for assessing BMD, as well as detailed anthropometric records that could facilitate robust analysis of WWI's role in bone health [12, 13].
This study aims to investigate the association between WWI and BMD in postmenopausal women using NHANES data, hypothesizing that higher WWI values correlate with lower BMD, independent of traditional obesity indices. By elucidating WWI's utility as a predictor of skeletal deterioration, this research seeks to advance understanding of body composition metrics in osteoporosis risk stratification, offering insights for targeted screening and preventive strategies in a high-risk population.
2. Methods
2.1. Survey Description and Study Participants
The National Center for Health Statistics (NCHS) is conducting a nationally representative cross-sectional study, the NHANES [14]. This survey employs a stratified multistage probability sampling method to collect comprehensive data and is updated every 2 years. NHANES provides valuable secondary analysis data on health and nutrition in the United States. NHANES data can be accessed on the NHANES website (https://www.cdc.gov/nchs/nhanes.htm; accessed March 2, 2025). The NCHS Research Ethics Review Board has approved the NHANES research protocol, and written informed consent was obtained from participants before their involvement in the study [15]. The inclusion criteria for the NHANES 2013–2014 and 2017–2020 cohorts required participants to be postmenopausal women aged 40 and older, with complete data on WWI and femoral neck BMD. Exclusion criteria included pregnant and breastfeeding women. After applying the inclusion and exclusion criteria, the final number of participants in the study was 3198 (Figure 1).
Figure 1.
Flowchart of the sample selection.
2.2. Study Variables
The WWI is calculated by dividing the waist measurement in centimeters by the square root of body weight in kilograms, serving as a method to evaluate body fat and muscle mass. Participants' weights and waist measurements were recorded by certified health professionals at the mobile examination unit. By removing their shoes and heavy clothing, the participants' weights were assessed. To estimate the WC, a horizontal line was drawn above the highest lateral edge of the right iliac bone, and a tape measure was placed where the lines intersected [16]. Femoral neck BMD was assessed using a DXA scan. Covariates included age, income-to-poverty ratio (PIR), standing height, BMI, total femur BMD, femoral neck BMD, alkaline phosphatase (ALP), blood urea nitrogen, globulin, glucose, triglycerides, uric acid, race, education level, and menopause status. The menopausal status is confirmed through self-reported reproductive health questionnaires in the NHANES survey. Women are classified as postmenopausal if they answer “no” to the question, “Have you had a menstrual period in the past 12 months? (do not include bleeding due to medical conditions, hormone therapy, or surgery)” and select “hysterectomy” or “menopause/life changes” as the reason for not having menstrual periods. Detailed information about the questionnaire can be found in the NHANES database and is accessible through the official website.
2.3. Statistical Analysis
The statistical analyses followed NCHS guidelines, using sample weights and considering the intricate multistage cluster survey design. Participant demographics by WWI tertile were assessed using the weighted chi-square and t-tests. The analysis of collinearity was conducted to assess and remove variables with collinearity, where a variance inflation factor (VIF) greater than 10 was deemed collinear. In addition, the relationship between WWI and BMD was resolved in this study by employing smooth and generalized additive models. When nonlinearity is identified, the research will utilize a recursive algorithm to find the inflection point. Next, a two-stage piecewise linear regression model was utilized to study the connection between WWI and BMD on both sides of the inflection point. Subgroup analysis was carried out to investigate the relationship between WWI and BMD in various subpopulations based on factors such as age, race, standing height, BMI, and menopause status. Continuous variables with missing values were appropriately categorized into tertiles or quartiles based on their distributions, with missing observations assigned to a dedicated “Missing” category. Categorical variables with missing data were handled as a separate category to avoid bias from arbitrary classification. Sensitivity analyses were conducted to evaluate the potential impact of missing data on the results. The analyses were performed using R (https://www.Rproject.org) and EmpowerStats (https://www.empowerstats.com), with a p value of less than 0.05 deemed statistically significant.
3. Results
3.1. Baseline Characteristics
This study enrolled 3198 participants, divided into three tertiles (low, middle, and high), with 1066 individuals in each tertile. Table 1 presents the baseline characteristics of the participants. Significant differences were observed across tertiles in terms of demographic characteristics, physical traits, and metabolic indicators. The high tertile had the oldest average age (65.33 ± 10.31 years), while the low tertile had the youngest (57.17 ± 10.53 years). Standing height decreased with tertile, with the high tertile being the shortest (156.63 ± 6.82 cm) compared to the low tertile (162.14 ± 6.47 cm). Conversely, BMI increased across tertiles, reaching 31.85 ± 6.66 kg/m2 in the high tertile, significantly higher than the low tertile's 26.41 ± 5.85 kg/m2. Femoral neck BMD was lowest in the high tertile (0.71 ± 0.14) compared to the low tertile (0.75 ± 0.14). Metabolic markers such as ALP, blood urea nitrogen, glucose, triglycerides, and uric acid were significantly higher in the high tertile. Race distribution and educational attainment also varied significantly across tertiles, with non-Hispanic Whites being the largest group in the high tertile (44.37%) and the highest proportion of participants with less than a high school education (28.08%). Menopause status differed notably, with the high tertile having the highest proportion of postmenopausal individuals (87.71%) compared to the low tertile (24.67%) (Table 1).
Table 1.
Basic characteristics of participants by weight-adjusted waist index tertile.
| Characteristics | Weight-adjusted waist index (cm/√kg) | p value | ||
|---|---|---|---|---|
| Low (9.02–11.12) | Middle (11.12–11.78) | High (11.78–14.14) | ||
| N = 1066 | N = 1066 | N = 1066 | ||
| Age (years) | 57.17 ± 10.53 | 61.49 ± 10.37 | 65.33 ± 10.31 | < 0.001 |
| PIR | 3.06 ± 1.66 | 2.68 ± 1.61 | 2.32 ± 1.50 | < 0.001 |
| Standing height (cm) | 162.14 ± 6.47 | 159.23 ± 6.56 | 156.63 ± 6.82 | < 0.001 |
| BMI (kg/m2) | 26.41 ± 5.85 | 29.63 ± 6.51 | 31.85 ± 6.66 | < 0.001 |
| Total femur BMD | 0.88 ± 0.14 | 0.88 ± 0.15 | 0.86 ± 0.15 | 0.035 |
| Femoral neck BMD | 0.75 ± 0.14 | 0.73 ± 0.14 | 0.71 ± 0.14 | < 0.001 |
| ALP (IU/L) | 70.41 ± 24.51 | 77.84 ± 25.84 | 83.53 ± 31.22 | < 0.001 |
| BUN (mmol/L) | 4.96 ± 1.89 | 5.26 ± 2.10 | 5.63 ± 2.31 | < 0.001 |
| Globulin (g/L) | 29.22 ± 4.60 | 29.74 ± 4.68 | 30.45 ± 4.99 | < 0.001 |
| Serum glucose (mmol/L) | 5.40 ± 1.79 | 5.93 ± 2.29 | 6.30 ± 2.68 | < 0.001 |
| Triglycerides (mmol/L) | 1.30 ± 0.76 | 1.69 ± 2.29 | 1.83 ± 1.04 | < 0.001 |
| Uric acid | 274.07 ± 71.59 | 301.71 ± 79.78 | 319.02 ± 80.71 | < 0.001 |
| Race (%) | < 0.001 | |||
| Mexican American | 61 (5.72%) | 136 (12.76%) | 144 (13.51%) | |
| Other Hispanic | 98 (9.19%) | 122 (11.44%) | 126 (11.82%) | |
| Non-Hispanic White | 451 (42.31%) | 408 (38.27%) | 473 (44.37%) | |
| Non-Hispanic Black | 270 (25.33%) | 246 (23.08%) | 183 (17.17%) | |
| Other race | 186 (17.45%) | 154 (14.45%) | 140 (13.13%) | |
| Education level (%) | < 0.001 | |||
| Less than high school | 121 (11.35%) | 195 (18.31%) | 299 (28.08%) | |
| High school | 215 (20.17%) | 269 (25.26%) | 281 (26.38%) | |
| More than high school | 730 (68.48%) | 601 (56.43%) | 485 (45.54%) | |
| Menopause status (%) | < 0.001 | |||
| No | 263 (24.67%) | 135 (12.66%) | 74 (6.94%) | |
| Yes | 749 (70.26%) | 886 (83.11%) | 935 (87.71%) | |
| Other or refused | 54 (5.07%) | 45 (4.22%) | 57 (5.35%) | |
Note: Mean ± SD for continuous variables: the p value was calculated by the weighted linear regression model. Percentage (%) for categorical variables: the p value was calculated by the weighted chi-square test. Among the 3198 patients, the amount of missing value for the covariates were 347 (10.9%) for PIR, 3 (0.09%) for standing height, 3 (0.09%) for BMI, 156 (4.9%) for ALP, 156 (4.9%) for BUN, 158 (4.9%) for globulin, 156 (4.9%) for serum glucose, 158 (4.9%) for triglycerides, 157 (4.9%) for uric acid, and 156 (4.9%) for menopause status. ALP, alkaline phosphatase; PIR, income-to-poverty ratio.
Abbreviations: BMD, bone mineral density; BMI, body mass index; BUN, blood urea nitrogen.
3.2. Association Between WWI and Femoral Neck BMD
Table 2 presents the connection between WWI and femoral neck BMD. The multiple regression analysis confirmed the significant impact of WWI on femoral neck BMD after adjusting for confounders. In the nonadjusted model, WWI showed a negative trend (β = −0.02, p=0.0003). After adjusting for age, race, standing height, and BMI (Adjust I), the negative association strengthened (β = −0.03, p < 0.0001). The Adjust II model, which included additional variables such as total femur BMD and metabolic markers, further confirmed the negative effect (β = −0.02, p < 0.0001) (Table 2 and Figure 2).
Table 2.
Association between weight-adjusted waist index (cm/√kg) and femoral neck BMD in postmenopausal women.
| Exposure | Model 1 | Model 2 | Model 3 |
|---|---|---|---|
| WWI (tertile) | β (95% CI) p value | β (95% CI) p value | β (95% CI) p value |
| Low (9.02–11.12) | Reference | Reference | Reference |
| Middle (11.13–11.78) | −0.01 (−0.02, 0.01) 0.4419 | −0.01 (−0.02, −0.00) 0.0136 | −0.01 (−0.02, −0.00) 0.0029 |
| High (11.79–14.14) | −0.03 (−0.04, −0.01) 0.0001 | −0.03 (−0.04, −0.02) < 0.0001 | −0.02 (−0.03, −0.01) < 0.0001 |
Note: Model 1: no covariates were adjusted. Model 2: age, race, standing height, and BMI were adjusted. Model 3: age, race, standing height, BMI, total femur BMD, ALP, BUN, globulin, serum glucose, triglycerides, and uric acid were adjusted.
Figure 2.
Association between weight-adjusted waist index (cm/√kg) and femoral neck BMD in postmenopausal women. Age, race, standing height, BMI, total femur BMD, ALP, BUN, globulin, serum glucose, triglycerides, and uric acid were adjusted (the solid red line indicates the smooth curve fit between variables, and the blue bands denote the 95% confidence interval from the fit).
3.3. Threshold Effect Analysis
The threshold effect analysis (Table 3) identified a critical threshold of WWI at 10.32, beyond which the relationship with femoral neck BMD shifted. Below this threshold, higher WWI levels were associated with a protective effect on BMD (β = 0.03, p=0.0265). Above the threshold, WWI exhibited a significant negative influence (β = −0.01, p < 0.0001). The difference in effects across the threshold was statistically significant (effect difference = −0.04, p=0.0023), highlighting the nonlinear nature of WWI's impact on BMD.
Table 3.
Threshold effect analysis of WWI (cm/√kg) on femoral neck BMD (g/cm2) in postmenopausal women.
| WWI (cm/√kg) | Femoral neck BMD |
|---|---|
| Fitting by the standard linear model | −0.01 (−0.01, −0.01) < 0.0001 |
| Fitting by the two-stage piecewise linear model | |
| Inflection point | 10.32 |
| < 10.32 | 0.03 (0.00, 0.06) 0.0265 |
| > 10.32 | −0.01 (−0.02, −0.01) < 0.0001 |
| Log-likelihood ratio | 0.002 |
Note: Age, race, standing height, BMI, total femur BMD, ALP, BUN, globulin, serum glucose, triglycerides, and uric acid were adjusted. The log-likelihood ratio (0.002) indicates significantly improved model fit (p < 0.05) when using the two-stage piecewise linear model over the standard linear model, supporting the threshold effect at WWI = 10.32. ALP, alkaline phosphatase.
Abbreviations: BMI, body mass index; BUN, blood urea nitrogen.
3.4. Stratified Analysis
The stratified analysis revealed variations in the association between WWI and femoral neck BMD across subgroups. Age-stratified results showed a positive association in the 40–55 year-old group (β = 0.01, p=0.0773) and a negative association in the 66–80 year-old group (β = −0.01, p=0.0056). Racial differences were observed, with non-Hispanic Blacks having higher BMD compared to Mexican Americans (β = −0.00, p=0.8111), while other racial groups showed a mild negative association (β = −0.02, p=0.0055). Standing height and BMI categories indicated taller individuals, and those with higher BMI generally had better BMD, though BMI's positive effect diminished at higher levels. Menopause status also played a role, with premenopausal women showing higher BMD (β = 0.03, p=0.0016) and the “Other or refused” category presenting a notable negative effect (β = −0.06, p < 0.0001) (Table 4).
Table 4.
Stratified analyses of the association between weight-adjusted waist index (cm/√kg) and femoral neck BMD.
| β (95% CI) | P for interaction | |
|---|---|---|
| Stratified by age | 0.4654 | |
| 40–55 | 0.01 (−0.00, 0.02) 0.0773 | |
| 56–65 | −0.00 (−0.01, 0.01) 0.7527 | |
| 66–80 | −0.01 (−0.02, −0.00) 0.0056 | |
| Stratified by race | 0.3257 | |
| Mexican American | −0.03 (−0.05, −0.01) 0.0005 | |
| Other Hispanic | −0.03 (−0.05, −0.01) 0.0014 | |
| Non-Hispanic White | −0.02 (−0.03, −0.02) < 0.0001 | |
| Non-Hispanic Black | −0.00 (−0.02, 0.01) 0.8111 | |
| Other race | −0.02 (−0.04, −0.01) 0.0055 | |
| Stratified by standing height (cm) | 0.2034 | |
| 135.3–156.1 | −0.01 (−0.02, −0.00) 0.0294 | |
| 156.2–162.2 | −0.01 (−0.02, 0.00) 0.1349 | |
| 162.3–187.8 | −0.00 (−0.01, 0.01) 0.5265 | |
| Stratified by BMI (kg/m2) | < 0.0001 | |
| 14.2–25.7 | −0.05 (−0.06, −0.04) < 0.0001 | |
| 25.8–31.2 | −0.05 (−0.06, −0.04) < 0.0001 | |
| 31.3–63 | −0.06 (−0.07, −0.05) < 0.0001 | |
| Stratified by menopause status | 0.0451 | |
| No | 0.03 (0.01, 0.04) 0.0016 | |
| Yes | −0.01 (−0.02, −0.01) < 0.0001 | |
| Other or refused | −0.06 (−0.08, −0.03) < 0.0001 | |
| Stratified by total femur BMD | 0.964 | |
| 0.362–0.806 | −0.02 (−0.02, −0.01) < 0.0001 | |
| 0.807–0.93 | −0.01 (−0.02, −0.01) < 0.0001 | |
| 0.931–1.384 | −0.02 (−0.02, −0.01) 0.0003 | |
| Stratified by ALP (IU/L) | 0.0613 | |
| 14–63 | −0.03 (−0.04, −0.01) < 0.0001 | |
| 64–82 | −0.02 (−0.03, −0.01) 0.0002 | |
| 83–501 | −0.02 (−0.03, −0.01) 0.0027 | |
| Stratified by BUN (mmol/L) | 0.2045 | |
| 0.36–3.93 | −0.02 (−0.03, −0.00) 0.0064 | |
| 4.28–5.36 | −0.01 (−0.02, −0.00) 0.0174 | |
| 5.71–22.85 | −0.03 (−0.04, −0.02) < 0.0001 | |
| Stratified by globulin (g/L) | 0.0849 | |
| 16–27 | −0.03 (−0.04, −0.02) < 0.0001 | |
| 28–30 | −0.03 (−0.04, −0.02) < 0.0001 | |
| 31–65 | −0.02 (−0.03, −0.01) < 0.0001 | |
| Stratified by serum glucose (mmol/L) | 0.1271 | |
| 2.61–5 | −0.01 (−0.03, −0.00) 0.0185 | |
| 5.05–5.55 | −0.03 (−0.05, −0.02) < 0.0001 | |
| 5.61–32.03 | −0.03 (−0.04, −0.02) < 0.0001 | |
| Stratified by triglycerides (mmol/L) | 0.0032 | |
| 0.305–1.061 | −0.02 (−0.03, −0.01) 0.0021 | |
| 1.073–1.694 | −0.03 (−0.04, −0.01) < 0.0001 | |
| 1.705–68.384 | −0.03 (−0.04, −0.02) < 0.0001 | |
| Stratified by uric acid (μmol/L) | 0.0013 | |
| 41.6–255.8 | −0.02 (−0.03, −0.01) < 0.0001 | |
| 261.7–315.2 | −0.03 (−0.04, −0.02) < 0.0001 | |
| 321.2–701.9 | −0.03 (−0.05, −0.02) < 0.0001 |
Note: In subgroup analyses stratified by age, race, standing height, BMI, total femur BMD, ALP, BUN, globulin, serum glucose, triglycerides, and uric acid, the model adjusted for covariates such as age, race, standing height, BMI, total femur BMD, ALP, BUN, globulin, serum glucose, triglycerides and uric acid, but the model did not adjust for the stratification variables themselves. ALP, alkaline phosphatase.
Abbreviations: BMI, body mass index; BUN, blood urea nitrogen.
Supporting information containing additional threshold effect analyses (Supporting Table S1) and smoothed curve fitting results (Supporting Figure S1) are available online. These analyses further validate the robustness of the observed nonlinear relationship between WWI and femoral neck BMD.
4. Discussion
This study aimed to investigate the relationship between the WWI and BMD in postmenopausal women, utilizing data from the NHANES. A cross-sectional analysis was conducted, including 3198 eligible postmenopausal participants. We employed DXA to assess BMD and performed multiple regression analyses to control for confounding factors such as age, race, height, and BMI. Our findings revealed a significant negative correlation between WWI and femoral neck BMD, particularly pronounced when WWI exceeded a threshold of 10.32. This highlights WWI as a critical indicator for assessing skeletal health risks, suggesting that the complexities of body composition should be considered in screening and preventive strategies for high-risk populations.
The association between the WWI and BMD or osteoporosis has been explored in several studies, highlighting its significance across different age groups. One study demonstrated a negative correlation between WWI and BMD among adults, suggesting that higher WWI levels are associated with lower BMD, which could potentially increase the risk of osteoporosis [17]. This inverse relationship was also observed in adolescents, where a significant negative association between WWI and total BMD was identified, emphasizing the importance of monitoring WWI in younger populations to prevent future bone health issues [18]. In older adults, the relationship between WWI and BMD continues to be significant. A study involving older adults revealed that higher WWI scores were associated with increased odds of frailty, which is often linked to decreased bone density and higher osteoporosis risk [19]. This underscores the importance of considering WWI as part of a comprehensive assessment of bone health in the elderly.
In addition, the impact of WWI on bone health has been compared with other obesity indicators, such as BMI and WC [20, 21]. Traditional obesity indices such as BMI and WC often fall short in capturing the nuances of fat distribution and metabolic risk. In contrast, WWI has emerged as a significant anthropometric measure in assessing body composition, particularly in relation to sarcopenia and visceral fat. Several studies have explored the utility of WWI in different populations, highlighting its potential as an integrated index for assessing fat, muscle, and bone health. A study conducted using data from the Korean NHANES demonstrated that higher WWI values are associated with unfavorable body composition outcomes, such as increased fat mass and decreased muscle and bone mass. This study found that WWI was positively correlated with total body fat percentage and inversely correlated with appendicular lean mass and bone mineral density, suggesting that WWI can serve as a comprehensive indicator of body composition [21]. Another study corroborated these findings by showing that WWI positively correlates with total abdominal fat area and visceral fat area, while negatively correlating with appendicular skeletal muscle mass, further supporting its role in reflecting fat and muscle mass in opposite directions in older adults [22].
The menopause-related endocrine shift significantly amplifies the impact of weight and body fat distribution on skeletal health. The withdrawal of estrogen during menopause is well-documented to accelerate bone resorption, which refers to the loss of bone density that occurs due to the increased activity of osteoclasts, the cells responsible for bone breakdown [23, 24]. Concurrently, the hormonal changes associated with menopause lead to fat redistribution toward visceral depots [25, 26]. This phenomenon creates a negative feedback loop: increased visceral adiposity not only exacerbates bone loss by enhancing systemic inflammation and altering hormonal balance but also mitigates the mechanical loading benefits that are usually associated with greater body weight [24, 27]. For instance, the dual pathway hypothesis proposed in some studies suggests that the adipose tissue surrounding the abdomen secretes proinflammatory cytokines and adipokines, which adversely affect bone density [28]. This distinction between visceral and subcutaneous fat is essential when assessing bone health in postmenopausal women, since traditional obesity metrics, such as BMI, often do not account for the specific risks conferred by visceral fat accumulation [29, 30]. Research has shown that visceral fat correlates negatively with BMD, while subcutaneous fat tends to have a neutral or even positive correlation with bone health due to its role as an extraovarian source of estrogen, which can have protective effects against bone resorption [27, 30, 31].
Clinically, our findings suggest that WWI could enhance risk stratification for osteoporosis in postmenopausal women, particularly for normal-weight individuals with central adiposity, who might be overlooked by conventional screening paradigms. Future research should explore whether WWI-guided interventions can mitigate bone loss progression in this vulnerable population. In addition, mechanistic studies investigating WWI's relationship with bone turnover markers and trabecular microarchitecture could further elucidate its pathophysiological role.
This study has several strengths. First, the large, nationally representative NHANES sample enhances generalizability, while rigorous adjustment for confounding variables reduces the likelihood of spurious associations. Second, methodologically, the use of DXA-derived BMD measurements strengthens the clinical relevance of our findings compared to studies relying solely on quantitative ultrasound. In addition, we identified a nonlinear relationship and threshold effect between WWI and BMD using generalized additive models.
However, several limitations warrant consideration. The cross-sectional design precludes causal inference, and residual confounding by unmeasured factors cannot be excluded. Furthermore, WWI's predictive value for fracture outcomes remains to be validated in longitudinal studies. Another limitation of our research is the absence of data for some variables. However, we used contemporary methods to handle the missing data and reduce bias.
5. Conclusion
A cross-sectional study of U.S. postmenopausal women revealed a significant inverse correlation between WWI and femoral neck BMD, showing a nonlinear inverted U-shaped relationship with a threshold effect. Further longitudinal studies with larger cohorts are needed to confirm these associations.
Acknowledgments
We thank all the participants in NHANES for their selfless contributions to this study.
Nomenclature
- WWI
Weight-adjusted waist index
- BMD
Bone mineral density
- PIR
Income-to-poverty ratio
- BMI
Body mass index
- LDL-C
Low-density lipoprotein cholesterol
- HDL-C
Direct high-density lipoprotein cholesterol
- AST
Aspartate transaminase
- ALT
Alanine transaminase
- BUN
Blood urea nitrogen
Funding Statement
The authors confirm that there was no financial support for the research, writing, or publication of this article.
Data Availability Statement
Publicly available datasets were used for analysis in this study. The data are located at https://www.cdc.gov/nchs/nhanes/.
Ethics Statement
The National Center for Health Statistics Research Ethics Review Board approved the studies involving human subjects. These studies adhered to local laws and institutional guidelines. Participants gave their consent.
Consent
Please see the Ethics Statement.
Conflicts of Interest
The authors declare no conflicts of interest.
Author Contributions
Q.J.: writing–original draft. Y.T.: writing–review and editing.
Supporting Information
Additional supporting information can be found online in the Supporting Information section.
Supporting Table S1. Threshold effect analysis of WWI (cm/√kg) on femoral neck BMD (g/cm2) in postmenopausal women.
Supporting Figure S1. Association between WWI (cm/√kg) and femoral neck BMD in postmenopausal women.
References
- 1.Peng X., Wu X., Zhang J., Zhang G., Li G., Pan X. The Role of CKIP-1 in Osteoporosis Development and Treatment. Bone & Joint Research . 2018;7(2):173–178. doi: 10.1302/2046-3758.72.bjr-2017-0172.r1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Prakash K. O., Choudhary R., Singh G. Lean Body Mass, Body Fat Percentage, and Handgrip Strength as Predictors of Bone Mineral Density in Postmenopausal Women. Journal of Mid-Life Health . 2021;12(4):299–303. doi: 10.4103/jmh.jmh_21_21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Salamat M. R., Salamat A. H., Janghorbani M. Association Between Obesity and Bone Mineral Density by Gender and Menopausal Status. Endocrinol Metab (Seoul) . 2016;31(4):p. 547. doi: 10.3803/enm.2016.31.4.547. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Peppa M., Stefanaki C., Papaefstathiou A., Boschiero D., Dimitriadis G., Chrousos G. P. Bioimpedance Analysis Vs. DEXA as a Screening Tool for Osteosarcopenia in Lean, Overweight and Obese Caucasian Postmenopausal Females. Hormones . 2017;16(2):181–193. doi: 10.14310/horm.2002.1732. [DOI] [PubMed] [Google Scholar]
- 5.Zhang Y., Jia X., Liu X., An W., Li J., Zhang W. Relationship Between Different Body Composition and Bone Mineral Density in Qinhuangdao City. Revista da Associação Médica Brasileira . 2022;68(4):445–449. doi: 10.1590/1806-9282.20210669. [DOI] [PubMed] [Google Scholar]
- 6.Zierfuss B., Hobaus C., Herz C. T., Pesau G., Koppensteiner R., Schernthaner G. Predictive Power of Novel and Established Obesity Indices for Outcome in PAD During a Five-Year Follow-Up. Nutrition, Metabolism, and Cardiovascular Diseases . 2020;30(7):1179–1187. doi: 10.1016/j.numecd.2020.03.019. [DOI] [PubMed] [Google Scholar]
- 7.Ono Y., Kasukawa Y., Sasaki K., Miyakoshi N. Association of the Bioimpedance Phase Angle and Quality of Life in Postmenopausal Osteoporosis. Medical Principles and Practice . 2022;32(1):71–76. doi: 10.1159/000528004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Karasik D., Hsu Y. H., Zhou Y., Cupples L. A., Kiel D. P., Demissie S. Genome-Wide Pleiotropy of Osteoporosis-Related Phenotypes: The Framingham Study. Journal of Bone and Mineral Research . 2010;25(7):1555–1563. doi: 10.1002/jbmr.38. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Park S. E., Park C. Y., Choi J. M., et al. Depot-Specific Changes in Fat Metabolism With Aging in a Type 2 Diabetic Animal Model. PLoS One . 2016;11(2):p. e0148141. doi: 10.1371/journal.pone.0148141. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Vondracek S. F., Voelkel N. F., McDermott M. T., Valdez C. The Relationship Between Adipokines, Body Composition, and Bone Density in Men With Chronic Obstructive Pulmonary Disease. International Journal of Chronic Obstructive Pulmonary Disease . 2009;4:267–277. doi: 10.2147/copd.s2745. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Loonstra F. C., Falize K., de Ruiter L. R. J., et al. Adipokines in Multiple Sclerosis Patients are Related to Clinical and Radiological Measures. Journal of Neurology . 2022;270(4):2018–2030. doi: 10.1007/s00415-022-11519-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Motyl K. J., Bishop K. A., DeMambro V., et al. Altered Thermogenesis and Impaired Bone Remodeling in Misty Mice. Journal of Bone and Mineral Research . 2013;28(9):1885–1897. doi: 10.1002/jbmr.1943. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Lü X., Zhao H., Wang E. A High-Fat Diet Induces Obesity and Impairs Bone Acquisition in Young Male Mice. Molecular Medicine Reports . 2013;7(4):1203–1208. doi: 10.3892/mmr.2013.1297. [DOI] [PubMed] [Google Scholar]
- 14.Luo C., Cheng H., He X., Tan X., Huang X. Association Between Serum Uric Acid and Female Infertility: a cross-sectional Study of National Health and Nutrition Examination Survey (NHANES) 2013–2018. BMC Women’s Health . 2023;23(1):p. 224. doi: 10.1186/s12905-023-02376-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Cheng L., Wang S. Correlation Between Bone Mineral Density and Sarcopenia in US Adults: A Population-Based Study. Journal of Orthopaedic Surgery and Research . 2023;18(1):p. 588. doi: 10.1186/s13018-023-04034-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Ouyang Y., Quan Y., Guo C., et al. Saturation Effect of Body Mass Index on Bone Mineral Density in Adolescents of Different Ages: A Population-based Study. Frontiers in Endocrinology . 2022;13:p. 922903. doi: 10.3389/fendo.2022.922903. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Zhang Y., Wu H., Li C., et al. Associations Between Weight-Adjusted Waist Index and Bone Mineral Density: Results of a Nationwide Survey. BMC Endocrine Disorders . 2023;23(1):p. 162. doi: 10.1186/s12902-023-01418-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Wang X., Yang S., He G., Xie L. The Association Between Weight-Adjusted-Waist Index and Total Bone Mineral Density in Adolescents: NHANES 2011–2018. Frontiers in Endocrinology . 2023;14:p. 1191501. doi: 10.3389/fendo.2023.1191501. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Jia S., Huo X., Sun L., Yao Y., Chen X. The Association Between the Weight-Adjusted-Waist Index and Frailty in US Older Adults: A Cross-Sectional Study of NHANES 2007-2018. Frontiers in Endocrinology . 2024;15:p. 1362194. doi: 10.3389/fendo.2024.1362194. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Tao Z., Zuo P., Ma G. The Association Between Weight-Adjusted Waist Circumference Index and Cardiovascular Disease and Mortality in Patients With Diabetes. Scientific Reports . 2024;14(1):p. 18973. doi: 10.1038/s41598-024-69712-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Kim K. J., Son S., Kim K. J., Kim S. G., Kim N. H. Weight‐Adjusted Waist as an Integrated Index for Fat, Muscle and Bone Health in Adults. Journal of Cachexia Sarcopenia Muscle . 2023;14(5):2196–2203. doi: 10.1002/jcsm.13302. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Kim N. H., Park Y., Kim N. H., Kim S. G. Weight-Adjusted Waist Index Reflects Fat and Muscle Mass in the Opposite Direction in Older Adults. Age and Ageing . 2021;50(3):780–786. doi: 10.1093/ageing/afaa208. [DOI] [PubMed] [Google Scholar]
- 23.Jang S. A., Hwang Y. H., Kim T. S., Lee A., Ha H. Anti-Osteoporotic and Anti-Adipogenic Effects of the Water Extract of Drynaria Roosii Nakaike in Ovariectomized Mice Fed a High-Fat Diet. Molecules . 2019;24(17):p. 3051. doi: 10.3390/molecules24173051. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Silva T. R., Franz R. F., Maturana M. A., Spritzer P. M. Associations Between Body Composition and Lifestyle Factors With Bone Mineral Density According to Time Since Menopause in Women from Southern Brazil: A Cross-Sectional Study. BMC Endocrine Disorders . 2015;15(1):p. 71. doi: 10.1186/s12902-015-0072-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Papadakis G., Hans D., Rodriguez E. G., et al. Menopausal Hormone Therapy is Associated With Reduced Total and Visceral Adiposity: The Osteolaus Cohort. Journal of Clinical Endocrinology and Metabolism . 2018;103(5):1948–1957. doi: 10.1210/jc.2017-02449. [DOI] [PubMed] [Google Scholar]
- 26.Sengar M., Srivastava K., Saxena R. Correlation of Menopausal Status With Body Composition and Abdominal Fat Distribution. National Journal of Community Medicine . 2022;13(4):203–206. doi: 10.55489/njcm.1342022431. [DOI] [Google Scholar]
- 27.Bredella M. A., Torriani M., Ghomi R. H., et al. Vertebral Bone Marrow Fat is Positively Associated With Visceral Fat and Inversely Associated With IGF‐1 in Obese Women. Obesity . 2011;19(1):49–53. doi: 10.1038/oby.2010.106. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Wang L., Wang W., Xu L., et al. Relation of Visceral and Subcutaneous Adipose Tissue to Bone Mineral Density in Chinese Women. International Journal of Endocrinology . 2013;2013:1–5. doi: 10.1155/2013/378632. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Begic Z., Rizvanovic M. Bone Density and Body Fat Distribution in Postmenopausal Women. Materia Socio Medica . 2021;33(1):p. 41. doi: 10.5455/msm.2021.33.41-44. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Fu X., Ma X., Lu H., He W., Wang Z., Zhu S. Associations of Fat Mass and Fat Distribution With Bone Mineral Density in Pre- and Postmenopausal Chinese Women. Osteoporosis International . 2010;22(1):113–119. doi: 10.1007/s00198-010-1210-9. [DOI] [PubMed] [Google Scholar]
- 31.Griffith J. F., Yeung D. K. W., Ma H. T., Leung J., Kwok T., Leung P. C. Bone Marrow Fat Content in the Elderly: A Reversal of Sex Difference Seen in Younger Subjects. Journal of Magnetic Resonance Imaging . 2012;36(1):225–230. doi: 10.1002/jmri.23619. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Additional supporting information can be found online in the Supporting Information section.
Data Availability Statement
Publicly available datasets were used for analysis in this study. The data are located at https://www.cdc.gov/nchs/nhanes/.