This computer-aided model can predict sepsis risk by analysing early symptoms

LONDON: Scientists in the UK have developed a computer-aided model that may help predict sepsis by routinely collecting data to identify early symptoms of the life-threatening condition.

Sepsis is a major cause of death in hospitals, and early detection is key to preventing deaths, said researchers from the University of Bradford in the UK.

The condition is caused by the body's response to an infection, triggering changes that can damage multiple organ systems.


Every hour of delay is linked to a seven per cent reduction in survival, but delays in detection are common, according to the research published in Canadian Medical Association Journal.

Several scores exist to help identify patients with sepsis, including the National Early Warning Score (NEWS) used in the UK's National Health Service hospitals.



Researchers developed the computer-aided National Early Warning Score (cNEWS) to determine if it could enhance the accuracy of predicting sepsis.

"The main advantage of these computer models is that they are designed to incorporate data that exist in the patient record, can be easily automated and place no extra burden on the hospital staff to collect additional information," said Professor Mohammed A Mohammed from the University of Bradford.

The cNEWS score can trigger screening for sepsis usually within 30 minutes of admission once routinely collected information has been electronically entered into the patient's medical record.

"These risk scores should support, rather than replace, clinical judgment. We hope they will heighten awareness of sepsis with additional information on this serious condition," Mohammed said in a statement.

Researchers said cNEWS may be introduced carefully into hospitals with appropriate information technology infrastructure and evaluated.