This study develops a machine learning model to predict osteoporosis in Chinese postmenopausal women. The model was trained using the largest nationwide cohort and externally validated. It showed AUCs of 0.744–0.798 and demonstrated prognostic value for 5-year fracture risk, supporting population-level screening where DXA resources are limited. Osteoporosis is highly prevalent among Chinese postmenopausal women, yet underdiagnosis remains due to limited DXA availability. We aimed to develop and validate a machine learning–based model for osteoporosis prediction using easily obtainable variables. The model was trained using data from the largest nationwide cohort (China Osteoporosis Prevalence Study, n = 6,574) and externally validated in two independent cohorts: the China Vertebral and Osteoporosis Study (n = 1,758) and the Peking Vertebral Fracture Study baseline (n = 1,439). Five predictors selected via LASSO regression were used to train the prediction model. Model performance was evaluated by area under the curve (AUC) and calibration plot. To evaluate its ability to predict future fractures, the model was evaluated in a 5-year follow-up cohort (n = 795). A web-based calculator was developed for public use. The model showed strong discrimination (AUC: 0.798 in the training cohort; 0.775 in the internal validation cohort; 0.750 and 0.744 in the external validation cohorts) and good calibration across all cohorts. A rule-out threshold (≥ 0.098) demonstrated high sensitivity (94.7
This study estimated FRAX®-based intervention thresholds for initiating osteoporosis treatment in Chinese postmenopausal women, using real-world data from the largest nationally representative osteoporosis survey in China and a validated Markov microsimulation model. Denosumab became cost-effective at a 10-year major osteoporotic fracture probability of 7
Purpose:To investigate the links between abdominal obesity and skeletal outcomes in a population-based cohort in China. Methods:8,251 participants from the COPS cohort were enrolled in this study and categorized by quartiles of abdominal obesity indices. Abdominal obesity was determined using sex-specific waist circumference (WC) cutoffs (≥ 90 cm for males, ≥ 85 cm for females), or a waist-to-height ratio (WHtR) ≥0.5. Physical performance and balance were assessed using the Five-Repetition Sit-to-Stand Test (5R-STS) and the Sharpened Romberg test. Vertebral fractures were identified by spine X-ray examination, while information on clinical fractures in recent 5y were collected by a self-report questionnaire. Multivariate regression models were employed with covariate adjustments. Predictive capacity of adiposity metrics for fractures was assessed through ROC analysis. Results:Abdominal obesity was linked to lower bone turnover rate and higher BMD. However, it was also significantly associated with impaired balance, evidenced by prolonged 5R-STS times and higher rates of positive Sharpened Romberg tests. The prevalence of vertebral fractures and clinical fractures in recent 5y increased across quartiles of WC, WHtR, and Chinese visceral adiposity index (CVAI) (all p< 0.001). An increment of one standard deviation (SD) in these indices was linked to a 19%-30% greater prevalence of fractures. These findings remained robust after limiting analysis to Genant grade ≥2 vertebral fractures. BMI-stratified analysis revealed that abdominal obesity independently increased the prevalence of vertebral fractures regardless of BMI. Notably, normal-weight individuals with abdominal obesity had a 1.75-fold higher prevalence of vertebral fractures (p = 0.022), while BMI alone showed no significant effect. Moreover, WHtR demonstrated superior predictive capacity for vertebral fractures and Genant grade ≥2 vertebral fractures vs BMI (AUC 0.62 vs 0.52), with optimal performance when combined with age, sex, fracture history, and BMD (AUC = 0.80). Conclusion:Abdominal obesity was independently associated with impaired balance and elevated prevalence of vertebral fractures and clinical fractures, even in normal-weight individuals. WHtR demonstrated superior discriminative power for vertebral fractures.
Although obesity is generally associated with increased bone mass, recent data have challenged its potential protective effect. Our study found that increased BMI showed beneficial effects on BMD in a non-linear way. However, individuals with obesity, especially women, were more likely to have vertebral fractures. There was a U-shaped relationship between BMI and the prevalent fractures. To estimate the association of obesity with the prevalence of fractures in the Mainland Chinese population. A total of 8251 individuals from the COPS cohort were enrolled in this cross-sectional study and grouped by BMI level. The Five-Repetition Sit-to-Stand Test (5R-STS) and the Sharpened Romberg test were used to evaluate the balance ability. Vertebral fractures (VFs) were confirmed by spine X-ray examination. Prevalent fractures were defined by a self-report questionnaire which happened in the recent 5 years. The restricted cubic spline (RCS) was used to explore the non-linear relationship. Multiple linear regression and multivariable logistic regression were conducted to adjust the covariates. Obesity was correlated with a reduced bone turnover rate and increased BMD. Nevertheless, there were significant non-linear correlations between BMI and BMD, with a rapid increase and plateau at extremely high BMI levels (p for non-linear < 0.001 for all). Individuals with obesity were associated with a longer time of the 5R-STS and more likely to have a positive Sharpened Romberg test, especially in women. Compared with the normal weight group, the likelihood of prevalence of VFs and the prevalent fractures were significantly increased in the obesity group, independent of the Sharpened Romberg test and lumbar spine BMD (VFs: OR = 1.88 [95
