Skip to page content


Reducing Sepsis Mortality by 18% Using Machine Learning

North Oaks Health System predicts patients’ risk of sepsis with Epic to intervene sooner

A minute can be the difference between life and death for a septic patient, so North Oaks Health System uses machine learning in Epic to find and treat at-risk patients 30 minutes sooner and reduce sepsis mortality by 18%.

“By embedding machine learning into the existing workflow, minutes can be saved,” said Seth Hain, Epic’s director of analytics and machine learning. “The results are put directly in front of clinicians, giving them the cue to intervene earlier and help patients.”

North Oaks uses machine learning predictions to give clinicians timely and detailed guidance. Machine learning searches the patient’s record for signs of sepsis as soon as the patient arrives. When symptoms indicate high risk of sepsis, clinicians can quickly order interventions and see reminders about next steps.

“Patients are now receiving antibiotics about 25 percent faster—or 30 minutes sooner—than they were before,” said North Oaks CIO Dr. Herbert Robinson. “Combined with the clinical expertise of our doctors and nurses, our approach is saving lives.”

Read more from Health IT Analytics. Epic community members can learn more from North Oaks’ UGM slides and audio.