Triage Study Challenges Notions of Emergency Medical Response to Disaster -- Overtriage Alone May Not Be Culprit

Dr. Nathaniel Hupert

NewYork-Presbyterian Hospital/Weill Cornell Medical Center Study Predicts Trauma System Performance for Mass Casualty Incidents



NEW YORK (Oct. 31, 2007) — In the face of terrorism and catastrophic natural disasters, modern regional trauma systems that improve survival for critically injured patients are more vital than ever.

Yet many fundamental assumptions underlying these systems—such as the notion that it is imperative to send the sickest patients to the hospital first—have rarely been subjected to rigorous scientific scrutiny. Now, for the first time, researchers at NewYork-Presbyterian Hospital/Weill Cornell Medical Center have created a computer simulation model of trauma system response to mass casualty incidents involving dozens or hundreds of injured victims. The study shows that the best response depends more on the capability of regional hospitals to treat critically injured victims than on the ability to accurately identify those victims in the field.

"There's been the notion gleaned from prior studies that 'overtriage'—letting some people into emergency care who might not actually need it—usually ends up costing lives, with deaths rising as overtriage rates increase. But our new model demonstrates that overtriage alone is unlikely to be the culprit," says lead researcher Dr. Nathaniel Hupert, assistant professor of public health and medicine at Weill Cornell Medical College and assistant attending physician at NewYork-Presbyterian Hospital/Weill Cornell Medical Center.

Instead, levels of overtriage can be beneficial, harmless or detrimental, depending on complex factors included in the researchers' model, he says. Those factors include the capacity of medical facilities to deal with the wounded and the time it takes to process and care for patients.

"No triage system is 100-percent accurate, so the key issue to define from an outcomes perspective is, 'How good is good-enough?'" Dr. Hupert says. "Our study suggests that pre-disaster planning can begin to address this question systematically, using modeling that takes into account local resources and response times, as well as specific types of mass casualty events."

The study is published in a special October supplement on mass casualty incidents of the journal Disaster Medicine and Public Health Preparedness, which focuses on the Virginia Tech shooting.

According to Dr. Hupert, experts in the field of traumatology and disaster preparedness have tended to rely on historical or anecdotal evidence to describe the downside of overtriage. "A number of studies released over the past decade have bolstered the notion that overtriage stretches limited medical resources during mass casualty events and ends up costing lives," Dr. Hupert says. "This was thought to happen in a linear fashion: More overtriage, more unnecessary deaths."

Overtriage can be valuable, however, because it helps ensure that critically injured people who do require speedy, lifesaving care aren't missed. In fact, guidelines from the American College of Surgeons support a limited amount of overtriage in emergency care.

To determine how much overtriage matters to patient outcomes, Dr. Hupert, along with engineers Eric Hollingsworth and Dr. Wei Xiong, Instructor in the Department of Public Health, created a discrete event simulation model representing the size and type of mass casualty event, the accuracy of field triage, and the treatment capability of the regional hospital trauma system. The model included a number of key variables that had never been brought together in one unified framework.

"We included the ability of responders to triage patients, either in the field or at the site of care; the capacity of local centers to care for incoming wounded and then recycle resources to care for new patients; the time needed to process and treat patients; and the window of time in which it was assumed critically injured patients might die," Hollingsworth explains.

The team found that, contrary to prior reports, the relationship between overtriage and critical patient mortality is not linear, and is highly dependent on whether there is a surplus or shortage of trauma bays in regional emergency departments.

"In some cases—for example, when the risk of death over the short term is high but you have a really large capacity to care for the injured—we can now show how overtriage may actually be a good thing, because you get more people into emergency care than you would otherwise," Dr. Hupert says. "On the other hand, if you have a more limited capacity, overtriage can be much less valuable, and perhaps harmful."

The model also addresses another relative unknown in mass casualty response: the impact of large numbers of "walk-in" patients on outcomes of those most critically injured. As with overtriage, the model suggests that the "walking wounded" can have a variable impact on the relationship between resources and mortality. By giving planners the ability to quantify these effects, however, the model advances the current state of disaster response logistics.

The bottom line, according to the researchers, is that the "best" triage strategy during a mass casualty event is probably one that takes into account a variety of local and regional factors, which means that use of the model should spur regional collaboration and information sharing. "We hope that this type of modeling study will be used to help disaster response teams plan effective strategies to deal with mass casualty events occurring in their own specific regions," Dr. Hupert says. "We can never predict when or where disaster will strike, but with models like these we can hope to be more prepared."

Now that they have a better understanding of the relationship between triage and outcomes, the researchers are continuing to build the model. "In our current work, we are incorporating details such as the spectrum of likely injuries from a particular event, the diagnostic tests needed to sort out critical from noncritical patients, and transportation times from the site of injury to the care center," Dr. Xiong says.

"This work and other projects that Dr. Hupert and his team are conducting clearly demonstrate the value of combining methods from public health and operations research. Insights are provided that can directly lead to ways to improve the effectiveness and efficiency of our health-care system," says Dr. Alvin Mushlin, professor and chairman of the Department of Public Health and Professor of Medicine at Weill Cornell Medical College, and Public Health Physician-in-Chief at NewYork-Presbyterian Hospital/Weill Cornell Medical Center.

This work was funded by a contract with the U.S. Department of Health and Human Services' Agency for Healthcare Research and Quality (AHRQ).

NewYork-Presbyterian Hospital/Weill Cornell Medical Center


NewYork-Presbyterian Hospital/Weill Cornell Medical Center, located in New York City, is one of the leading academic medical centers in the world, comprising the teaching hospital NewYork-Presbyterian and its academic partner, Weill Cornell Medical College. NewYork-Presbyterian/Weill Cornell provides state-of-the-art inpatient, ambulatory and preventive care in all areas of medicine, and is committed to excellence in patient care, research, education and community service. NewYork-Presbyterian, which is ranked sixth on the U.S.News & World Report's list of top hospitals, also comprises NewYork-Presbyterian Hospital/Columbia University Medical Center. For more information, visit www.nyp.org.
Office of Public Affairs
pr@nyp.org

Weill Cornell Medicine
Office of External Affairs
Phone: (646) 962-9476