Data-Driven Model Can Help Predict Hospital Discharges

Dr. Sean Barnes Ph.D. Department of Decision, Operations & Information Technologies Robert H. Smith School of Business University of Maryland, College Park, MD Interview with:
Dr. Sean Barnes Ph.D.
Department of Decision, Operations & Information Technologies
Robert H. Smith School of Business
University of Maryland, College Park, MD


Medical Research: What is the background for this study? What are the main findings?

Dr. Barnes: Hospitals are continually being challenged to provide timely and efficient care in the face of increasingly constrained resources. One recent approach to help improve patient flow in hospitals is Real-Time Demand and Capacity Management, by which clinicians huddle each morning to predict the number of patients they expect to discharge on a given day (and hence the number of beds that will become available to potentially utilize for newly admitted patients). We proposed a data-driven method for predicting discharges–either on an individual or aggregate basis–and demonstrated that we could match or exceed the predictive accuracy of clinicians. In addition, we showed (with moderate success) that we could use this model to rank patients in order of their expected discharge times, which could be used to prioritize the remaining care tasks for specific subsets of patients.

Medical Research: What should clinicians and patients take away from your report?

Dr. Barnes: We believe that clinicians should take away the potential for employing data-driven predictive modeling into operational and clinical decision-making. These methods have been introduced into the healthcare field relatively recently (as opposed to other industrial fields), and although many studies have been published (in operational and healthcare literature), it remains to be seen whether these methods will be implemented broadly in practice. Patients should take away that researchers with expertise outside the field of medicine are leveraging their skills to bring improvements to operational and clinical decision-making that directly affect the efficiency and quality of the care they receive.

Medical Research: What recommendations do you have for future research as a result of this study?

Dr. Barnes: This study was conducted in an individual medical unit in a mid-Atlantic academic hospital. The model was designed to use readily available data from hospital information systems; however, it remains to be seen whether this modeling approach would be effective across an entire hospital or multiple hospitals. The next step for this research is to apply this methodology on a larger scale, potentially incorporating additional data to improve the predictive accuracy of patient discharges.


J Am Med Inform Assoc. 2015 Aug 7. pii: ocv106. doi: 10.1093/jamia/ocv106. [Epub ahead of print]

Real-time prediction of inpatient length of stay for discharge prioritization.

Barnes S1, Hamrock E2, Toerper M3, Siddiqui S4, Levin S5.

[wysija_form id=”5″] is not a forum for the exchange of personal medical information, advice or the promotion of self-destructive behavior (e.g., eating disorders, suicide). While you may freely discuss your troubles, you should not look to the Website for information or advice on such topics. Instead, we recommend that you talk in person with a trusted medical professional.

The information on is provided for educational purposes only, and is in no way intended to diagnose, cure, or treat any medical or other condition. Always seek the advice of your physician or other qualified health and ask your doctor any questions you may have regarding a medical condition. In addition to all other limitations and disclaimers in this agreement, service provider and its third party providers disclaim any liability or loss in connection with the content provided on this website. Interview with: Dr. Sean Barnes Ph.D. (2015). Data-Driven Model Can Help Predict Hospital Discharges 

Last Updated on September 1, 2015 by Marie Benz MD FAAD