In the Journals

Computer-assisted model accurately predicts sepsis

A model that included routinely collected patient information accurately predicted sepsis among hospitalized patients aged 16 years and older in England, according to research published in CMAJ.

Mohammed A. Mohammed, PhD, professor of health care quality and effectiveness and deputy director of the Bradford Institute of Health Research at the University of Bradford, and colleagues wrote that in England, a standard assessment called the National Early Warning Score (NEWS) measures patients’ respiration rate, oxygen saturations, any supplemental oxygen, temperature, systolic blood pressure, heart rate and level of consciousness. A score is assigned to patients based on clinical observations, with higher scores indicating more critical illnesses. These assessments are usually collected within 30 minutes for most patients.

Mohammed and colleagues developed and compared three computer-assisted NEWS models: NEWS alone (M0), NEWS plus age and sex (M1) and M1 plus subcomponents of NEWS and diastolic blood pressure. Patients aged 16 years and older who presented to EDs and were admitted to York Hospital (YH; n = 36,751) and Goole National Health Service Foundation Trust (NH; n = 37,100) were included in the analysis.

The researchers defined a sepsis case as at least one organ failure or septic shock based on 84 ICD-10 codes.

Hospitalized child 
Source: Shutterstock

The M2 model most accurately identified patients with sepsis who had a NEWS threshold score of 4 or greater (sensitivity = 63.22%; specificity = 75.55%) and a NEWS threshold score of 5 or greater (sensitivity = 52.69%; specificity = 82.77%). The researchers wrote that the M2 system may now be “carefully introduced and evaluated in hospitals with sufficient informatics infrastructure.”

“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,” Mohammed said in a press release.

“These risk scores should support, rather than replace, clinical judgment,” Mohammed added. “We hope they will heighten awareness of sepsis with additional information on this serious condition.” – by Katherine Bortz

Disclosures: The authors report no relevant financial disclosures.

A model that included routinely collected patient information accurately predicted sepsis among hospitalized patients aged 16 years and older in England, according to research published in CMAJ.

Mohammed A. Mohammed, PhD, professor of health care quality and effectiveness and deputy director of the Bradford Institute of Health Research at the University of Bradford, and colleagues wrote that in England, a standard assessment called the National Early Warning Score (NEWS) measures patients’ respiration rate, oxygen saturations, any supplemental oxygen, temperature, systolic blood pressure, heart rate and level of consciousness. A score is assigned to patients based on clinical observations, with higher scores indicating more critical illnesses. These assessments are usually collected within 30 minutes for most patients.

Mohammed and colleagues developed and compared three computer-assisted NEWS models: NEWS alone (M0), NEWS plus age and sex (M1) and M1 plus subcomponents of NEWS and diastolic blood pressure. Patients aged 16 years and older who presented to EDs and were admitted to York Hospital (YH; n = 36,751) and Goole National Health Service Foundation Trust (NH; n = 37,100) were included in the analysis.

The researchers defined a sepsis case as at least one organ failure or septic shock based on 84 ICD-10 codes.

Hospitalized child 
Source: Shutterstock

The M2 model most accurately identified patients with sepsis who had a NEWS threshold score of 4 or greater (sensitivity = 63.22%; specificity = 75.55%) and a NEWS threshold score of 5 or greater (sensitivity = 52.69%; specificity = 82.77%). The researchers wrote that the M2 system may now be “carefully introduced and evaluated in hospitals with sufficient informatics infrastructure.”

“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,” Mohammed said in a press release.

“These risk scores should support, rather than replace, clinical judgment,” Mohammed added. “We hope they will heighten awareness of sepsis with additional information on this serious condition.” – by Katherine Bortz

Disclosures: The authors report no relevant financial disclosures.