A team of researchers from Baltimore-based Johns Hopkins Medicine developed an electronic triage tool to risk-stratify patients in its emergency departments.
The team — led by Scott Levin, PhD, associate professor of emergency medicine at the Johns Hopkins University School of Medicine — built the clinical decision support tool using machine learning algorithms to predict the likelihood of adverse patient outcomes.
The researchers tested the e-triage tool, which leverages data related to vital signs, chief complaints and medical histories, on 172,726 visits to urban and community EDs. They compared the e-triage tool's results with those from the emergency severity index, the triage guide nurses and physicians typically use.
The e-triage tool demonstrated equivalent or improved ability to identify patient outcomes compared with the ESI, according the researchers' results, published in Annals of Emergency Medicine. It was particularly useful for differentiating patients who were triaged to a mid-level priority, identifying more than 10 percent of mid-level patients who necessitated more critical care.
"For patients at risk of having a critical care need, this technology is designed to detect them better and make sure they are seen quicker," Dr. Levin said. "The ultimate objective is patients should be waiting less in the emergency department."
The researchers will continue to evaluate the e-triage tool, which is used at The Johns Hopkins Hospital in Baltimore and Howard County General Hospital in Columbia, Md.