Three years after the death of its major proponent, the five-level Emergency Severity Index (ESI) has passed major tests of its validity, and is available as a handbook from the Emergency Nurses Association. Already, said those who helped develop the formula and keep it alive, its use is spreading across the nation.
When Richard Wuerz, MD, of Harvard Medical School died in 2000, several of his collaborators and colleagues picked up the torch of the five-level triage system he had espoused, and brought it to clinical fruition. A report in the October issue of Academic Emergency Medicine notes that a test of the index in seven emergency departments has proved it safe and reliable (2003;10:1070).
“We know about the national stockpile, but what do you do until the stockpile gets there?”
Dr. Earl Siegel
“It's designed to be simple. An experienced emergency room nurse can read it in an hour.”
Dr. Thomas Stair
“There's a meeting of the minds across emergency care that five-level is the way to go.”
Dr. Debbie Travers
A combined task force from the American College of Emergency Physicians and the Emergency Nurses Association are evaluating the literature surrounding five-level triage systems already in use, and are expected to recommend that U.S. institutions adopt such triage systems. Five-level triage based on a variety of scales is currently used in Canada, Australia, and New Zealand as well as the United Kingdom.
The five-level system, unlike the traditional three-level system, uses a specific triage algorithm to categorize patients from the most urgent (level 1) to those requiring few resources (level 5). The algorithm takes into account not only the severity of the patient's injuries on admittance to the emergency department, but also the number of hospital sources that will be needed to treat the patient.
The index is constantly being improved and updated, said Thomas Stair, MD, the director of the emergency department at the Brigham and Women's Hospital and an associate professor at Harvard Medical School. “The basic algorithm has been copyrighted, but other than that, I think it's just like any other clinical protocol.
“It's going to continue to be evolved and elevated,” said Dr. Stair. The second version of the Emergency Severity Index (the version involved in the recent report in Academic Emergency Medicine) includes pediatric vital signs and some modification of vital signs that would upgrade some patients from level 3 to level 2. A third version is already being fine-tuned.
“People are working on various sorts of five-level triage programs,” he said. “Its time has come. We met with some people from HCA [Hospital Corporation of America, Inc.], and they will go to five-level triage in all their hospitals, although they haven't settled on which one yet. I still think the great thing about ESI is that it is validated and standard. It is easy to train people to use it. It helps you determine where the patients need to go.”
Easy to Use
That issue is not as important for the patients classified at level 1 or 2; they are going to be treated quickly, he said, but it is vital for patients at levels 3, 4, and 5. Most people can follow the ESI algorithm in their heads, he said. “It's designed to be simple,” Dr. Stair said. “There are 20 standard training set cases to go through. An experienced emergency room nurse can read it in an hour.”
Debbie Travers, PhD, RN, of the University of North Carolina at Chapel Hill, is a member of the joint task force on triage. “Both organizations passed resolutions recently backing five-level triage,” she said, adding that a white paper evaluating the need for five-level triage is in the works.
“There's a meeting of the minds across emergency care that five-level is the way to go, and that's from the research on ESI and the Canadian system,” said Dr. Travers, who helped develop the ESI. “A couple of papers put out recently show they have good reliability. The Australians continue to generate information on their system. Five-level is superior to three-level, but which five-level has yet to be sorted out.”
“The Emergency Severity Index is basically rooted in an industrial engineering way of thinking,” said David Eitel, MD, MBA, a member of the core faculty of the York Hospital emergency medicine residency program in York, PA. “You get the right patients to the resources at the right place at the right time. Triage is fixing the front end of the system.”
Not only does the index allow the efficient operation of the emergency department, it gives hospital managers reliable information about case mix. Eventually, it will allow modeling of what kinds of patients hospitals can expect when.
“The research team has no financial or entrepreneurial interest in ESI. ESI is copyrighted, but it's acquired with acquisition of the book from ENA,” said Dr. Eitel.
“We are terribly interested in the reliable and reproducible implementation of the tool.”
Dr. David Eitel
“We are terribly interested in the reliable and reproducible implementation of the tool,” he said. It may become more important as the Joint Commission on the Accreditation of Healthcare Organizations becomes more interested in simulation modeling as a way to deal with overcrowded hospitals and emergency departments, he said.
Predicting Patient Flow
Using the Emergency Severity Index would allow hospital managers to predict patient flow through the hospital, he said, meaning they could determine the resources and personnel needed during the year. “We need to know the ranges and surges,” Dr. Eitel said. “If they are predictable, we can get ready for them. That's the light we need to turn on in health care. Once we do that, we can prepare to manage.” The Agency for Healthcare Research and Quality is interested in turning the research into application, he said, noting that some nurses on the ESI team will produce a video to be distributed with the handbook.
Dr. Stair said the data generated by ESI could provide important information for other kinds of research. “It's on paper now, and often does not make it into the electronic record,” he said. “We need to add a computer screen at triage to capture those data.”
Already, nurses use computerized equipment to take blood pressure, pulse, and temperature, he said. Those data could be collected automatically. “There's a move toward 150 standard chief complaints, which will be valuable to clinical research,” Dr. Stair said. “Public health can look at those data and use them to recognize outbreaks. It would be a key part of biosurveillance.
“The whole triage will be legible. Now it just gets lost. In fact, the triage information and the EMS run sheets are on paper and not electronic,” he said. Finding ways to put that information into the electronic medical record will be important to patient flow, patient safety, and other parts of clinical management, he said.