Development and Validation of a Logistic-Bayes Method for In-Hospital Mortality Discrimination in Neonatal Heart Failure: A Real-World Study
5 affiliations
- 1Department of Cardiac Pacing and Electrophysiology, The First Affiliated Hospital of Xinjiang Medical University, Urumqi, Xinjiang, People’s Republic of China
- 2Xinjiang Key Laboratory of Cardiac Electrophysiology and Cardiac Remodeling, The First Affiliated Hospital of Xinjiang Medical University, Urumqi, Xinjiang, People’s Republic of China
- 3Department of Cardiovascular Medicine, The First People’s Hospital of Kashgar Prefecture, Kashgar, Xinjiang, People’s Republic of China
- 4Department of Cardiovascular Medicine, People’s Hospital of Gongliu County, Gongliu, Xinjiang, People’s Republic of China
- 5Department of Cardiovascular Medicine, The First People’s Hospital of Huocheng County, Huocheng, Xinjiang, People’s Republic of China
Abstract
Background: Neonatal heart failure is a severe condition with high mortality, posing a significant burden on families and healthcare systems. Accurately discriminating whether a neonate with heart failure will die in hospital is essential for improving outcomes.
Objective: To develop and validate a discriminant model for identifying neonates with heart failure who will experience in-hospital death.
Methods: This real-world study included 600 neonates with heart failure treated at the First Affiliated Hospital of Xinjiang Medical University between 2012 and 2025. Demographic, clinical, perinatal, laboratory, and first-24-hour medication data were extracted. Candidate variables were screened by LASSO regression and between-group comparisons; eight discriminators retained by multivariable logistic regression were used to construct the Logistic-Bayes model, and the cohort was randomly split in a 7:3 ratio into training and internal validation sets.
Results: The final discriminators were urea, fibrinogen, poor postnatal response, oliguria, cedilanid use, dopamine use, epinephrine use, and digoxin use. The model achieved area under the receiver operating characteristic curve (AUC) values of 0.87 (95% CI 0.82–0.91) in the training set and 0.86 (95% CI 0.78–0.92) in the internal validation set, with accuracy of 0.87 in both.
Conclusion: On internal validation, the model demonstrated high specificity and high negative predictive value, supporting early identification of low-risk neonates with heart failure; prospective multicenter validation is required before routine clinical use.
Related
- Development and validation of a novel metabolic-inflammatory health-related nomogram for predicting in-hospital mortality in older patients with type 2 diabetes mellitus combined with cardiovascular disease
- Development and validation of an interpretable machine learning model for predicting in-hospital mortality among critically ill patients with liver cirrhosis
- Prediction of in-hospital mortality at a 72-hour intensive care landmark in patients with sepsis and pre-existing hypertension: development and external validation
- Time-zero multimodal risk stratification for in-hospital mortality in acute aortic dissection: two-centre development, temporal and external validation, and occult shock phenotyping
- Data-driven reweighting of the Charlson and Elixhauser comorbidity indices using German claims data: development for in-hospital mortality only, external validation across clinical and health-economic outcomes.
BibTeX
@article{zhang2026development,
title = {{Development and Validation of a Logistic-Bayes Method for In-Hospital Mortality Discrimination in Neonatal Heart Failure: A Real-World Study}},
author = {Xiaoting Zhang and Meng Wei and Heyin Yang and Jixing Wu and Binghui Liu and Yanmei Lu and Baopeng Tang},
journal = {International Journal of General Medicine},
year = {2026},
month = oct,
doi = {10.2147/ijgm.s627564},
url = {https://doi.org/10.2147/ijgm.s627564},
note = {CC BY-NC 4.0},
}
RIS
TY - JOUR
TI - Development and Validation of a Logistic-Bayes Method for In-Hospital Mortality Discrimination in Neonatal Heart Failure: A Real-World Study
AU - Xiaoting Zhang
AU - Meng Wei
AU - Heyin Yang
AU - Jixing Wu
AU - Binghui Liu
AU - Yanmei Lu
AU - Baopeng Tang
PY - 2026
DA - 2026/10/30
JO - International Journal of General Medicine
DO - 10.2147/ijgm.s627564
UR - https://doi.org/10.2147/ijgm.s627564
AB - Background: Neonatal heart failure is a severe condition with high mortality, posing a significant burden on families and healthcare systems. Accurately discriminating whether a neonate with heart failure will die in hospital is essential for improving outcomes. Objective: To develop and validate a discriminant model for identifying neonates with heart failure who will experience in-hospital death. Methods: This real-world study included 600 neonates with heart failure treated at the First Affiliated Hospital of Xinjiang Medical University between 2012 and 2025. Demographic, clinical, perinatal, laboratory, and first-24-hour medication data were extracted. Candidate variables were screened by LASSO regression and between-group comparisons; eight discriminators retained by multivariable logistic regression were used to construct the Logistic-Bayes model, and the cohort was randomly split in a 7:3 ratio into training and internal validation sets. Results: The final discriminators were urea, fibrinogen, poor postnatal response, oliguria, cedilanid use, dopamine use, epinephrine use, and digoxin use. The model achieved area under the receiver operating characteristic curve (AUC) values of 0.87 (95% CI 0.82–0.91) in the training set and 0.86 (95% CI 0.78–0.92) in the internal validation set, with accuracy of 0.87 in both. Conclusion: On internal validation, the model demonstrated high specificity and high negative predictive value, supporting early identification of low-risk neonates with heart failure; prospective multicenter validation is required before routine clinical use.
SN - 1178-7074
ER -