Objective To construct and evaluate a prediction model for in-hospital mortality in el-derly patients with heart failure(HF).Methods After the establishment of inclusion criteria,clinical data of 767 elderly HF patients were extracted from the MIMIC-Ⅳ database.They were randomly divided into a training set(n=628)and a validation set(n=139)in an 8∶2 ratio.Least Absolute Shrinkage and Selection Operator(LASSO)regression analysis was employed to screen variables and identify independent risk factors for in-hospital mortality.Subsequently,multivariate logistic regression analysis was utilized to develop a risk prediction model and create a nomogram.ROC curve analysis was conducted to evaluate the AUC value of the model,calibration curve was used to assess the calibration.Additionally,decision curve analysis(DCA)was conducted to evalu-ate the clinical applicability of the model.Results ROC curve analysis showed that the prediction model achieved an AUC value of 0.749(95％CI:0.701-0.800)in the training set and of 0.725(95％CI:0.622-0.829)in the validation set.DCA indicated that the nomogram prediction model demonstrated good clinical applicability.Conclusion Our nomogram effectively predicts in-hospital mortality in elderly HF patients.

Objective To construct and evaluate a prediction model for in-hospital mortality in el-derly patients with heart failure(HF).Methods After the establishment of inclusion criteria,clinical data of 767 elderly HF patients were extracted from the MIMIC-Ⅳ database.They were randomly divided into a training set(n=628)and a validation set(n=139)in an 8∶2 ratio.Least Absolute Shrinkage and Selection Operator(LASSO)regression analysis was employed to screen variables and identify independent risk factors for in-hospital mortality.Subsequently,multivariate logistic regression analysis was utilized to develop a risk prediction model and create a nomogram.ROC curve analysis was conducted to evaluate the AUC value of the model,calibration curve was used to assess the calibration.Additionally,decision curve analysis(DCA)was conducted to evalu-ate the clinical applicability of the model.Results ROC curve analysis showed that the prediction model achieved an AUC value of 0.749(95％CI:0.701-0.800)in the training set and of 0.725(95％CI:0.622-0.829)in the validation set.DCA indicated that the nomogram prediction model demonstrated good clinical applicability.Conclusion Our nomogram effectively predicts in-hospital mortality in elderly HF patients.