Electronic health record data could hold the key to predicting the onset of sepsis, an immune system response to infection that kills more than 258,000 Americans annually.

Delayed diagnosis and treatment of sepsis have made it one of the most deadly and costly medical conditions confronting U.S. hospitals, with more than 1 million cases each year. However, researchers at North Carolina State University are working to overcome these challenges by integrating EHR data and clinical expertise in major hospital systems to provide an evidence-based framework to diagnose and accurately risk-stratify patients.

Funded by the National Science Foundation, the research project is a collaborative effort with the Mayo Clinic in Rochester, Minn., and Christiana Care Health System in Newark, Del. Researchers are leveraging engineering and computer science methodologies to analyze patient-level EHR data from the two large-scale healthcare facilities to inform clinical decision making for sepsis.

“We decided to focus on that patient population to understand how sepsis manifests in patients, the evolution of sepsis over time within a patient, and how to appropriately respond in a timely manner so that adverse outcomes can be prevented,” says Julie Ivy, associate professor at NCSU’s Edward P. Fitts Department of Industrial and Systems Engineering and a faculty fellow in health systems engineering.

Time is of the essence in diagnosing sepsis. According to Ivy, every one hour delay in treatment of severe sepsis/shock with antibiotics decreases a patient’s survival probability by 10 percent. “It’s a very rapid progression,” she adds.

Among the objectives of the research project are to develop:

  • Data-driven models to classify patients according to their clinical progression to diagnose sepsis and predict risk of deterioration.
  • Personalized intervention policies for patients within the sepsis spectrum.
  • Decision support systems for personalized interventions focusing on resource implications and usability within real hospital settings.

“It’s something that could potentially be applied to other health systems,” says Ivy. “We want to understand what are the features that you can observe about patients through their vital signs and labs, as well as things that happen during clinical care that can be used to help more quickly identify a patient at risk for sepsis or who has sepsis before it progresses to organ failure.”

Also See: Alert system uses predictive analytics to combat sepsis

She sees EHRs as clinical decision making tools and as such believes they are fundamental to the effort. Researchers will use data gathered from EHR systems at Mayo Clinic and Christiana Care, including body temperature, heart rate, respiratory rate, white blood cell count, medication lists, blood culture reports, diagnostic imaging, as well as clinician notes, to identify when/how to intervene in real time.

“We want to use the data from what they observe during clinical care,” concludes Ivy. “Our overarching goal is to combine this rich data with clinical expertise—as part of this interdisciplinary collaboration between industrial and systems engineering and computer science—to improve sepsis diagnosis and treatment.”

Currently, the project is completing the first year of research with two more years of funding to follow.

Register or login for access to this item and much more

All Health Data Management content is archived after seven days.

Community members receive:
  • All recent and archived articles
  • Conference offers and updates
  • A full menu of enewsletter options
  • Web seminars, white papers, ebooks

Don't have an account? Register for Free Unlimited Access