Electronic health record–derived data and novel analytics, such as machine learning, offer promising approaches to identify high-risk patients and inform nursing practice.
The aim was to identify patients at risk for readmissions by applying a machine-learning technique, Classification and Regression Tree, to electronic health record data from our 300-bed hospital.
We conducted a retrospective analysis of 2165 clinical encounters from August to October 2017 using data from our health system's data store. Classification and Regression Tree was employed to determine patient profiles predicting 30-day readmission.
The 30-day readmission rate was 11.2% (n = 242). Classification and Regression Tree analysis revealed highest risk for readmission among patients who visited the emergency department, had 9 or more comorbidities, were insured through Medicaid, and were 65 years of age and older.
Leveraging information through the electronic health record and Classification and Regression Tree offers a useful way to identify high-risk patients. Findings from our algorithm may be used to improve the quality of nursing care delivery for patients at highest readmission risk.
Center for Health Outcomes and Policy Research, and University of Pennsylvania Leonard Davis Institute of Health Economics, University of Pennsylvania School of Nursing, Philadelphia (Drs Brom and Brooks Carthon); and Biostatistics Evaluation Collaboration Consultation Analysis (BECCA) Lab, University of Pennsylvania School of Nursing, Philadelphia (Messrs Ikeaba and Chittams).
Correspondence: Heather Brom, PhD, APRN, Center for Health Outcomes and Policy Research, University of Pennsylvania School of Nursing, 418 Curie Blvd, Rm 388R, Philadelphia, PA 19104 (firstname.lastname@example.org).
The authors acknowledge the University of Pennsylvania LDI Work group on Socially at-Risk Patients and Ms Chidinma Wilson for her work as a research assistant.
This work was supported by the University of Pennsylvania Leonard Davis Institute (LDI) of Health Economics (Brooks Carthon, PI) and the National Institute of Nursing Research (T32-NR0714, L. Aiken, PI).
The authors declare no conflicts of interest.
Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal's Web site (www.jncqjournal.com).
Accepted for publication: March 25, 2019
Published ahead of print: May 24, 2019