Abstract
This study aimed to investigate the factors associated with hospitalization for emergency ankle fractures and to develop a predictive model based on these findings. A retrospective study design was employed to collect clinical data from patients with ankle fractures admitted to our hospital between January 2022 and December 2024. Single-factor and multifactor logistic regression analyses were conducted to identify independent factors influencing hospitalization decisions for emergency ankle fracture patients. Based on the selected factors, a nomogram predictive model was established for visual representation. The model's discriminatory ability was evaluated using receiver operating characteristic curves and their areas under the curve, while model fit was assessed via the Hosmer-Lemeshow goodness-of-fit test. Internal validation was performed using the Bootstrap method combined with 10-fold cross-validation, and the clinical application value of the predictive model was assessed through calibration curve and decision curve analysis. A total of 568 patients with ankle fractures were included, among whom 168 were hospitalized, resulting in a hospitalization rate of 29.58%. Through univariate and multivariate logistic regression analysis, 5 independent factors were identified that influence hospitalization in emergency ankle fracture patients: open fractures (OR = 1.436, 95% CI = 1.065-2.194), initial presentation at a nonaffiliated hospital (OR = 3.001, 95% CI = 2.123-4.781), high-energy injury (OR = 1.357, 95% CI = 1.025-2.175), multiple fractures (OR = 1.806, 95% CI = 1.520-3.200), and transport by emergency ambulance (OR = 2.818, 95% CI = 2.317-3.240). Through receiver operating characteristic curves and the Hosmer-Lemeshow goodness-of-fit test, the results showed that the model has good discriminative ability and fitting performance. Combined with Bootstrap internal validation and 10-fold cross-validation results, the model demonstrates good stability and has high clinical application value. The hospitalization need prediction model for emergency ankle fracture patients constructed in this study is based on 5 independent related factors and has good discriminatory ability and fitting degree. This model can be used as a clinical auxiliary tool to help emergency physicians scientifically assess the hospitalization needs of ankle fracture patients, optimize resource allocation, and improve treatment efficiency.