Data-Driven Machine Learning Approaches to Cut Hospital Readmissions in USA

Tahmidur Rahman Chowdhury, Mizanur Rahman, Faysal Ahmed, Shamima Afrose and Md Adnan Sami Bhuiyan
Additional contact information
Tahmidur Rahman Chowdhury: Ambassador Crawford College of Business and Entrepreneurship, Kent State University, Kent, Ohio
Mizanur Rahman: Institute of Health Economics, University of Dhaka
Faysal Ahmed: Wright State University, Dayton, Ohio
Shamima Afrose: Govt. College of Applied Human Science, University of Dhaka
Md Adnan Sami Bhuiyan: DePauw University, Greencastle, IN

International Journal of Research and Innovation in Applied Science, 2025, vol. 10, issue 9, 896-908

Abstract: Readmissions in hospitals are a major challenge to healthcare systems globally and lead to increased cost, burden on clinical resources, and poor patient outcomes. Conventional methods of identifying readmission risk that commonly rely on rule-based models and clinician judgment have demonstrated poor predictive validity. New developments in machine learning (ML) offer potent alternatives, via the utilization of vast amounts of structured and unstructured healthcare information to detect intricate patterns, related to the risk of readmission. This paper will discuss the use of machine learning models, including logistic regression, random forests, gradient boosting and deep learning, to predict hospital readmissions in a variety of patients. We speak about the contribution of electronic health records (EHRs), demographic factors, comorbidities, medication adherence, and post-discharge follow-up variables to the enhanced model performance. Explainable AI methods are given a special focus to make the model prediction transparent and trusted by clinicians. Other important challenges that are identified in the review are data quality, class imbalance, bias, and generalizability in healthcare settings. Case studies reveal how predictive models can be used to initiate specific interventions, including improved discharge planning, telehealth monitoring, and tailored care coordination, which can in turn lead to a reduction in avoidable readmissions and eventually lead to better patient outcomes. A combination of machine learning with clinical workflows will enable healthcare organizations to transition to more proactive, data-driven, and cost-beneficial care. The results highlight the disruptive power of machine learning to solve the long-standing problem of hospital readmissions and define the areas of future research in designing ethical applications, interoperability, and policy.

Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
https://www.rsisinternational.org/journals/ijrias/ ... 6-908-202510_pdf.pdf (application/pdf)
https://www.rsisinternational.org/journals/ijrias/ ... readmissions-in-usa/ (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:bjf:journl:v:10:y:2025:i:9:p:896-908

Access Statistics for this article

International Journal of Research and Innovation in Applied Science is currently edited by Dr. Renu Malsaria

More articles in International Journal of Research and Innovation in Applied Science from International Journal of Research and Innovation in Applied Science (IJRIAS)
Bibliographic data for series maintained by Dr. Renu Malsaria ().