Background
Frailty is common in people with arthritis and may result in a range of adverse consequences. This study aimed to investigate risk factors for frailty in people with arthritis and to develop and validate a nomogram prediction model.
Methods
The study used data from the 2015 China Health and Retirement Longitudinal Study (CHARLS). This study analyzed 36 indicators including socio-demographic, behavioral, and health status factors. Participants were randomly included in training and validation sets in a ratio of 7:3. The least absolute shrinkage and selection operator (LASSO) regression was used on the training set to screen the best predictor variables of the model through 10-fold cross-validation. Binary logistic regression was used to explore the related factors of frailty in people with arthritis. Construct nomograms to develop prediction models. Use receiver operating characteristic (ROC) curves to evaluate the discrimination ability of the model, Calibration curves to evaluate calibration, and decision curve analysis (DCA) to evaluate clinical validity.
Results
A total of 6209 people with arthritis were included in this study, of whom 952 (15.3%) suffered from frailty. The nomogram model includes 9 predictive factors: age, gender, activities of daily living (ADL), waistline, cognitive function, depressive symptoms, hearing status, self-perceived health status, and inpatient needs. The model shows good consistency and accuracy. The AUC values for the model in the training set and validation set are 0.866 (95% CI = 0.852-0.880) and 0.854 (95% CI = 0.832-0.876) respectively. The calibration curves showed good accuracy between the nomogram model and actual observations. ROC and DCA showed that the nomogram had good predictive performance.
Conclusions
The frailty risk prediction model constructed in this study has good discrimination, calibration, and clinical validity in people with arthritis. It is a promising and convenient tool that can be used as an objective guide for the clinical screening of high-risk populations.