Risk prediction for 30-day heart failure-specific readmission or death after discharge: Data from the Korean Acute Heart Failure (KorAHF) registry

Background Identifying patients with acute heart failure (HF) at high risk for readmission or death after hospital discharge will enable the optimization of treatment and management. The objective of this study was to develop a risk score for 30-day HF-specific readmission or death in Korea. Methods We analyzed the data from the Korean Acute Heart Failure (KorAHF) registry to develop a risk score. The model was derived from a multiple logistic regression analysis using a stepwise variable selection method. We also proposed a point-based risk score to predict the risk of 30-day HF-specific readmission or death by simply summing the scores assigned to each risk variable. Model performance was assessed using an area under the receiver operating characteristic curve (AUC), the Hosmer–Lemeshow goodness-of-fit test, the net reclassification improvement (NRI), and the integrated discrimination improvement (IDI) index to evaluate discrimination, calibration, and reclassification, respectively. Results Data from 4566 patients aged ≥40 years were included in the analysis. Among them, 446 (9.8%) had 30-day HF-specific readmission or death. The final model included 12 independent variables (age, New York Heart Association functional class, clinical history of hypertension, HF admission, chronic obstructive pulmonary disease, etiology of cardiomyopathy, systolic blood pressure, left ventricular ejection fraction, serum sodium, brain natriuretic peptide, N-terminal prohormone of brain natriuretic peptide at discharge, and prescription of β-blockers and angiotensin-converting enzyme inhibitors or angiotensin II receptor antagonists at discharge). The point risk score showed moderate discrimination (AUC of 0.710; 95% confidence interval, 0.685–0.735) and good calibration ( χ 2 =8.540, p =0.3826). Conclusions The risk score for the prediction of the risk of 30-day HF-specific readmission or death after hospital discharge was developed using 12 predictors. It can be utilized to guide appropriate interventions or care strategies for patients with HF.