As health systems continue to navigate the complexities of patient care, many are turning to custom-built applications within their Epic EHR systems to address specific needs.
Three health systems — Yale New Haven Health, Emory Healthcare and UChicago Medicine — have recently announced the development of their own Epic applications.
Yale New Haven Health: Predicting risk
Researchers at Yale New Haven (Conn.) Health have developed an Epic application that predicts the risk of gastrointestinal bleeding in patients.
This machine learning model analyzes EHR data, including symptoms, lab results and medical history, to generate a clinical risk score for GI bleeding.
The researchers noted that building the tool within Epic, the nation's most widely used EHR system, allows other health systems to tailor the data fields to fit their specific patient populations, enhancing its adaptability and impact across different care settings.
Emory Healthcare: Optimizing sepsis
Atlanta-based Emory Healthcare developed a custom Epic application that leverages an artificial intelligence algorithm to alert clinicians when patients with sepsis may benefit from an alternative treatment.
The AI-driven tool specifically identifies patients who might be better suited for balanced crystalloids instead of traditional saline IV fluids at the time of the fluid order. According to Emory, more than 70% of its clinicians have accepted these recommendations within Epic, resulting in treatment adjustments that could improve patient outcomes.
UChicago Medicine: Streamlining documentation
UChicago Medicine recently developed an application within Epic, now deployed across its EHR, with the potential to be integrated into Epic systems nationwide.
The feature, known as AutoDx, automatically populates patients' conditions and risk factors into the EHR's note template, ensuring clinicians capture critical details and minimizing the need for coding queries. According to Cheng-Kai Kao, MD, chief medical information officer at UChicago Medicine, this tool helps reduce the risk of important information being overlooked, streamlining clinical documentation.