Development and validation of a nomogram for predicting unplanned readmission within 30 days after discharge in patients with chronic heart failure

Abstract

Objective: To develop a visual assessment tool for predicting unplanned readmission within 30 days after discharge in patients with chronic heart failure (CHF).

Methodology: This retrospective cohort study enrolled data from 535 patients with CHF who received treatment in Chifeng Municipal Hospital from January 2020 to September 2025. Patients’ records were divided into a training cohort (n=321) and a validation cohort (n=214) at a predefined ratio of 6:4 using simple random sampling. The Least Absolute Shrinkage and Selection Operator (LASSO) algorithm and logistic regression analysis were applied to perform feature analysis on the training cohort, which was then converted into a nomogram risk model for unplanned readmission. Comparisons were conducted using the calibration curves, receiver operating characteristic (ROC) curves, and decision curve analysis (DCA).

Results: The incidence of unplanned readmission within 30 days was 19.1% (102/535). Seven identified independent risk factors for predicting unplanned readmission included age, chronic kidney disease (CKD), anemia, atrial fibrillation (AF), levels of homocysteine (Hcy), length of hospital stay (LOS), and New York Heart Association (NYHA) classification. The constructed nomogram model exhibited sufficient predictive accuracy, with area under the curve (AUC) values of 0.884 (95% CI: 0.838-0.930) in the training cohort and 0.874 (95% CI: 0.814-0.933) in the validation cohort, respectively. The Hosmer-Lemeshow (H-L) test indicated good calibration of the model.

Conclusions: The predictive model for unplanned readmission within 30 days after discharge in patients with CHF developed in this study demonstrates good predictive value, serving as a reliable clinical tool for risk stratification and the early identification of high-risk populations.