Why the study?
COPD is a leading cause of readmissions, but few existing tools use EHR data to forecast readmission risk during index hospitalizations.
Does a random forest machine learning model improve prediction of 90-day readmission risk compared to the HOSPITAL score in patients hospitalized for AE-COPD?
Population
3238 adult patients admitted for AE-COPD at the University of Chicago Medicine
Comparison
Random forest readmission risk model vs HOSPITAL score
Design
Retrospective cohort study
Follow-up
90 days
Authors
Loading...
ML models modestly outperformed HOSPITAL score for 90-day AE-COPD readmission prediction; hypothesis-generating for targeted interventions pending prospective validation.
Does a random forest machine learning model improve prediction of 90-day readmission risk compared to the HOSPITAL score in patients hospitalized for AE-COPD?
A random forest machine learning model using in-hospital EHR data outperformed the HOSPITAL score in predicting 90-day readmission risk among patients hospitalized for COPD exacerbation.
Bonomo et al. (2022) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: