Robert Kneschke -

EHR Tool Predicts Hospital Readmission Rates for Diabetes Patients

Researchers at Temple University developed an EHR tool to cut costs and predict hospital readmission rates for patients with diabetes.

Researchers at the Lewis Katz School of Medicine (LKSOM) at Temple University received a $2.5-million grant from the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) to advance EHR tools that predict hospital readmission among diabetes patients.

Over a million diabetes patients are readmitted to the hospital for a diabetes-related illness and most are readmitted within 30 days of their last hospitalization. These readmissions are costly for the patient, and multiple readmissions can significantly impact a patient’s health.

Researchers from Temple developed a model called the Diabetes Early Readmission Risk Indicator (DERRI). Now, the research group aims to optimize this base model using EHR data, to predict 30-day readmission risk among diabetes patients.

"The DERRI app originally was developed for point-of-care use, with patients manually inputting information on a small number of variables, such as insulin use, co-existing conditions and recent hospital discharge," said Daniel J. Rubin, MD, associate professor of Medicine at LKSOM and chair of the glycemic control taskforce at Temple University Hospital.

"Now, we want to upgrade DERRI to build models that bypass manual input and instead draw directly on EHR data, allowing for faster and more accurate prediction."

The Temple research group upgraded the DERRI model to eDERRI, which will integrate data from the PaTH Clinical Data Research Network (CDRN), a co-member of the National Patient-Centered Clinical Research Network. The study author said the researchers would optimize predictive modeling by applying state-of-the-art deep-learning methods.

Following optimization, the researchers will translate the models into an EHR readmission risk prediction tool that utilizes EHR data. The group predicts the tool will predict readmission risk and accurately identify risk factors, wrote the study author.

This tool intends to advance patient care to those with diabetes who are especially vulnerable to severe illness.

The research team plans to test eDERRI on diabetes patients at Temple University Hospital.

"Our proposed eDERRI tool could be widely used for predicting risks, reducing costs and improving care," Rubin concluded. "We are building and testing it at Temple, but the EHR component will be built in EPIC, which is used at numerous institutions across the country. Our tool could be rolled out to any hospital or treatment center that uses EPIC systems."

Seamless data exchange and interoperability are essential for making that level of care coordination and follow-up care occur, but it is not always the case for many outpatient clinicians. Outpatient clinicians do not always have real-time access to patient data from recent hospitalizations.

According to a recent study, clinicians with access to a shared inpatient-outpatient EHR were more likely to schedule a telehealth follow-up appointment or conduct laboratory monitoring, rather than an in-person visit.

Additionally, enhanced interoperability and patient data exchange can boost follow-up care efficiency.

The researchers also said the study shows the importance of patient data access across a number of providers between facilities, especially for patients with diabetes.

“Our findings from patients with diabetes also complement findings of previous studies in the same integrated delivery system, in patients with diabetes and in general patient populations, in which both providers and patients reported that EHR use facilitated care coordination both by providing informational continuity among providers and by supporting direct communication between clinicians and medical staff through electronic messaging tools,” explained the study authors.

While some studies showed in-person visits resulted in better patient outcomes, researchers said there was little evidence of worse outcomes in this study. Furthermore, researchers said patient data access may have decreased unnecessary or duplicate testing.

Telehealth is beneficial for the current need for social distancing to mitigate the spread of COVID-19, and it is also more convenient for patients.

Dig Deeper on Interoperability in healthcare

Cloud Computing
Mobile Computing