28 Nov Electronic Medical Records: Tool to Identify Readmission Risk
MedicalResearch.com Interview with:
Craig A Umscheid, MD, MSCE, FACP
Assistant Professor of Medicine and Epidemiology
Director, Center for Evidence-based Practice
Medical Director, Clinical Decision Support
Chair, Department of Medicine Quality Committee
Senior Associate Director, ECRI-Penn AHRQ Evidence-based Practice Center
University of Pennsylvania Philadelphia, PA 19104
MedicalResearch.com: What are the main findings of the study?
Dr. Umscheid: We developed and successfully deployed into the electronic health record of the University of Pennsylvania Health System an automated prediction tool which identifies newly admitted patients who are at risk for readmission within 30 days of discharge. Using local data, we found that having been admitted to the hospital two or more times in the 12 months prior to admission was the best way to predict which patients are at risk for being readmitted in the 30 days after discharge. Using this finding, our automated tool identifies patients who are “high risk” for readmission and creates a “flag” in their electronic health record (EHR). The flag appears next to the patient’s name in a column titled “readmission risk.” The flag can be double-clicked to display detailed information relevant to discharge planning. In a one year prospective validation of the tool, we found that patients who triggered the readmission alert were subsequently readmitted 31 percent of the time. When an alert was not triggered, patients were readmitted only 11 percent of the time. There was no evidence for an effect of the intervention on 30-day all-cause readmission rates in the 12-month period after implementation.
This is the first study in a general population of hospitalized patients to describe the development, validation and impact of an automated readmission risk assessment tool on readmission rates. The simplicity of the prediction rule and the integration of it into a commonly used commercial electronic health record make these findings generalizable to other EHRs and healthcare populations.
MedicalResearch.com: Were any of the findings unexpected?
Dr. Umscheid: A distinctive characteristic of our prediction model is its simplicity. We were cognizant of the realities of running a prediction model in a high-volume production environment and the diminishing returns of adding more variables.
We were not particularly surprised that readmission rates did not change significantly during the study period. This likely reflects the reality that providing readmission risk assessment alone is not sufficient to influence readmission rates. Interventions and organizational changes targeting those at high risk for readmission need to be implemented and routinely performed to reduce readmissions. Thus, as the readmission risk flag becomes more routinely used, and those interventions necessary to impact readmission rates of those defined as high risk are implemented and performed, we believe readmission rates will decrease.
MedicalResearch.com: What should clinicians and patients take away from your report?
Dr. Umscheid: An automated prediction model can be effectively integrated into an existing EHR and identify patients on admission who are at risk for readmission within 30 days of discharge.
MedicalResearch.com: What recommendations do you have for future research as a result of this study?
Dr. Umscheid: Future work will aim to further examine the impact of the flag on readmission rates, further refine the prediction model, and gather data on how providers and care teams use the information provided by the flag.
A link to a press release is below:
Citation:
J Hosp Med. 2013 Nov 13. doi: 10.1002/jhm.2106. [Epub ahead of print]
Last Updated on July 17, 2015 by Marie Benz MD FAAD