Related Experiment Video
Updated: May 16, 2026
Establishing a Competing Risk Regression Nomogram Model for Survival Data
Published on: October 23, 2020
Development and validation of a nomogram predicting osteoporosis risk in rheumatoid arthritis
Lujing Wang1, Yueting Gu2, Xiaochun Zhang3
1Department of Rheumatology, Taizhou Second People's Hospital Affiliated to Yangzhou University, Taizhou, Jiangsu, China.
Background:
Rheumatoid arthritis (RA) increases the risk of osteoporosis, but tools that integrate RA-specific clinical and metabolic factors to predict osteoporosis risk are limited. We aimed to develop and validate a practical risk prediction nomogram for osteoporosis in RA patients.
Methods:
In this single-center retrospective study, 349 RA patients with available DXA data were analyzed; 132 (37.8%) had osteoporosis. A training cohort (n = 250; osteoporosis = 92) and a temporal validation cohort (n = 99; osteoporosis = 40, enrolled later in the study period) were used. Candidate predictors included clinical, functional, and laboratory variables. Stepwise backward logistic regression identified independent predictors that were incorporated into a nomogram. Model performance was assessed by discrimination (AUROC), calibration (calibration curve and Hosmer-Lemeshow test), decision curve analysis (DCA), and risk stratification.
Results:
Female sex, higher health assessment questionnaire-disability index (HAQ-DI), elevated alkaline phosphatase (ALP), increased ApoA1/ApoB ratio, higher free fatty acids (FFA), and lower body mass index (BMI) were independent predictors of osteoporosis and were included in the nomogram. The model yielded AUROCs of 0.812 (training) and 0.788 (validation), showed good calibration (Hosmer-Lemeshow p > 0.05), and provided positive net benefit across a range of threshold probabilities in DCA. Nomogram-based risk strata (low/medium/high) discriminated osteoporosis risk with statistically significant odds ratios for medium and high groups.
Conclusion:
The proposed nomogram, built from readily available clinical and laboratory measures, demonstrates good discrimination, calibration, and clinical utility for identifying RA patients at elevated risk of osteoporosis, and may facilitate targeted screening and early intervention. However, the model's performance in diverse populations remains unknown, and prospective multicenter external validation is essential before any clinical application.
