Predictive Analysis of Thirty-Day Hospital Readmission Among Heart Failure Patients Using Gradient Boosting on Nationwide Electronic Health Records
Keywords:
Heart Failure, Hospital Readmission, Gradient Boosting, Machine Learning, Electronic Health Records, Predictive ModelingAbstract
Hospital readmissions among heart failure (HF) patients remain a critical burden on healthcare systems, leading to adverse patient outcomes and inflated costs. Predictive modeling using machine learning on electronic health records (EHRs) offers a data-driven pathway to address this challenge. This study applies Gradient Boosting Machines (GBM) to a nationwide EHR dataset to predict 30-day readmission risk among heart failure patients. The model demonstrated superior performance over traditional logistic regression and random forest baselines, achieving an AUC of 0.83 and sensitivity of 76%. The results underscore the potential of GBM in proactive risk stratification for heart failure readmissions.
References
Chen, Tianqi, and Carlos Guestrin. "XGBoost: A Scalable Tree Boosting System." Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 2016, pp. 785–794.
Desai, Rohan J., et al. "Predicting 30-Day All-Cause Readmissions Using Claims and Electronic Health Record Data." Medical Care, vol. 56, no. 2, 2018, pp. 150–157.
Krumholz, Harlan M., et al. "An Administrative Claims Measure Suitable for Profiling Hospital Performance on the Basis of 30-Day All-Cause Readmission Rates Among Patients with Heart Failure." Circulation: Cardiovascular Quality and Outcomes, vol. 2, no. 5, 2009, pp. 429–437.
Mortazavi, Bita J., et al. "Analysis of Machine Learning Techniques for Heart Failure Readmissions." Circulation: Cardiovascular Quality and Outcomes, vol. 9, no. 6, 2016, pp. 629–640.
Rajkomar, Alvin, et al. "Scalable and Accurate Deep Learning with Electronic Health Records." NPJ Digital Medicine, vol. 1, 2018, pp. 1–10.
Ahmad, Tariq, et al. "Clinical Implications of Chronic Heart Failure Phenotypes Defined by Cluster Analysis." Journal of the American College of Cardiology, vol. 64, no. 17, 2014, pp. 1765–1774.
Amarasingham, Ruben, et al. "Using Electronic Health Record Data to Identify Patients at High Risk for Readmission." BMJ Quality & Safety, vol. 20, no. 9, 2011, pp. 747–752.
Obermeyer, Ziad, and Ezekiel J. Emanuel. "Predicting the Future — Big Data, Machine Learning, and Clinical Medicine." The New England Journal of Medicine, vol. 375, no. 13, 2016, pp. 1216–1219.
Frizzell, John D., et al. "Prediction of 30-Day All-Cause Readmissions in Patients Hospitalized for Heart Failure: Comparison of Machine Learning and Other Statistical Approaches." JAMA Cardiology, vol. 2, no. 2, 2017, pp. 204–209.
Golas, Sara B., et al. "Machine Learning Models for Predicting Post-Hospital Discharge Complications Using Structured and Unstructured Data." PLOS ONE, vol. 13, no. 12, 2018, e0208211.
Dey, Neelesh, et al. "Predicting Readmission Risk Following Coronary Artery Bypass Surgery Using Gradient Boosting." Journal of Biomedical Informatics, vol. 84, 2018, pp. 57–65.
Steyerberg, Ewout W., et al. "Assessing the Performance of Prediction Models: A Framework for Traditional and Novel Measures." Epidemiology, vol. 21, no. 1, 2010, pp. 128–138.
Downloads
Published
Issue
Section
License
Copyright (c) 2023 Ijtihadi Phyu Thwe (Author)
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
