A risk stratification model for frailty in rheumatoid arthritis patients based on clinical risk factors: a cross-sectional study

Rheumatoid arthritis patients (RA) are associated with frailty. Frailty significantly detrimentally affects patients’ physical and emotional wellness and lowers their standards of life. Establishing a risk stratification model for frailty patients with RA was the goal of this investigation. The study included 283 RA patients from Anhui Province. The risk factors are determined through Minimum Absolute Contraction and Selection Operator (lasso) analysis, random forest, and multivariate logistic regression. The risk stratification model has been developed via multivariate logistic regression. We employed correction curves to assess the nomogram model's accuracy. Area under curve (AUC) and decision curve analysis (DCA) were used to assess predictive performance. 188 patients (66.4%) with RA had frailty. Physical activity, depression, pain, and disease activity, Pittsburgh sleep quality index, were found to be risk factors of RA patients’ frailty. The internal validation set had an AUC of 0.825 (95% CI 0.730–0.920) and the risk stratification model had an AUC of 0.877 (95% CI 0.842–0.932). The findings of the H–L test indicated that the P ≥ 0.05. The nomogram model and the actual observation findings are in good agreement, as seen by the calibration curve. The nomogram performs well in terms of risk stratification, according to the ROC and DCA values. In order to screen high-risk groups for clinical care, the nomogram is a useful and promising method for determining the frailty risk of RA patients in this study.