The insidious onset of osteoporosis and the high cost of DXA examination make it urgent to develop suitable prediction or screening tools. The NHANES cohort contains standardized DXA-BMD results and comprehensive nutrition-related information. Therefore, this study aimed to develop and validate a nomogram clinical prediction model dedicated to predicting the exact probability of osteoporosis occurrence in the elderly population. Data of elderly participants were extracted from the NHANES database and categorized into the training (n = 3181) and validation (n = 1622) groups. Clinical characteristics and BMD results were obtained and analyzed. Univariate and multivariate logistic regression analyses were performed. General and dynamic nomogram clinical prediction models were constructed. The models were validated using ROC curves, calibration curves, DCA curves, and clinical impact curves. Based on 11 variables, including age, gender, race, poverty income ratio (PIR), waist circumference, DBP, physical exercise, protein intake, carbohydrate intake, caffeine intake, and fracture history, a nomogram clinical prediction model was constructed. This model exhibited moderate predictive value (AUC = 0.795), alongside good calibration, clinical benefit, and clinical impact. The constructed online dynamic nomogram (https://jialinwang.shinyapps.io/OP-Prediction-Model/) is interactive, accessible, and user-friendly. This nomogram prediction model and the web-based dynamic nomogram exhibit good practical application value within the U.S. elderly population. However, external validation in non-U.S. cohorts is necessary before widespread global promotion. Ultimately, this tool could facilitate the early prediction, diagnosis, and treatment of osteoporosis, thus contributing to the bone health of the elderly population and promoting the development of public health. This study constructed and validated a nomogram clinical prediction model to predict the probability of developing osteoporosis based on the NHANES cohort. By elucidating the patterns of osteoporosis incidence and its independent risk factors, this interactive tool assists primary care physicians in implementing rapid early screening, ensuring that appropriate interventions can be administered in a timely manner.
Wang et al. (Fri,) studied this question.