Computer Algorithm Predicts Sepsis

Sepsis kills more Americans every year than AIDS and breast and prostate cancer combined. The search for early diagnosis of sepsis has moved forward as researchers at Johns Hopkins University have developed a more effective way to spot hospital patients at risk of septic shock.

The new computer-based method correctly predicts septic shock in 85 percent of cases, without increasing the false positive rate from screening methods that are now common. “It is critical to detect these patients early enough so that clinicians have time to intervene,” said Suchi Saria, Assistant Professor of Computer Science and Health Policy at JHU’s Whiting School of Engineering http://engineering.jhu.edu and just recently contributed the cover article in the journal “Science Translational Medicine” http://stm.sciencemag.org.

As explained in the article, more than two-thirds of the time, the computer-based method used was able to predict septic shock before any organ dysfunction which is a 60 percent improvement over existing screening protocols.

The study drew from EHRs of 16,234 patients admitted to intensive care units including medical, surgical, and cardiac units at Boston’s Beth Israel Deaconess Medical Center from 2001-2007. Researchers created an algorithm that combines 27 factors into a Targeted Real-time Early Warning Score (TREWScore) capable of measuring the risk of septic shock.

This method differs in several respects from previous attempts to predict septic shock as it is based on a larger data pool that takes into account more health indicators and factors in several elements that could influence the results.

Now the question is how can the TREWScore be used in a hospital or nursing home? David Hager, a Co-Author of the Study and Director of the Medical Progressive Care Unit at Johns Hopkins Hospital www.hopkinsmedicine.org, said “The algorithm could be programmed into an electronic health records system to alert doctors and nurses about a patient at risk of septic shock.”

The research team reports, “The tricky issue is how to provide the clinical team the information,” “A hospital’s EHR system could be set up to convey alerts to clinicians via pagers or cell phones at regular intervals. However, the information needs to be well integrated into the existing clinical workflow so that alarm fatigue does not occur.”