May 6, 2003

PHONE: 410-955-1534

Johns Hopkins Researchers Devise Methods To Evaluate Disaster Drills

Researchers at Johns Hopkins and Loma Linda universities have published what is believed to be the first peer-reviewed set of standards for planning and evaluating disaster drills anywhere in the world.

"Disaster simulation drills are widely used throughout the world and are considered a fundamental tool to evaluate and improve local disaster response," says lead author Gary B. Green, M.D., M.P.H., an associate professor of emergency medicine at Johns Hopkins. "Despite this, however, no generally accepted methodology had existed to evaluate the medical response to a disaster. We created a set of tools that are flexible enough to be practical for disaster planners worldwide."

The methods, published in the May issue of the Annals of Emergency Medicine, break evaluation of drills into four parts: command and control (including the time from initial notification of a disaster until the paramedics arrive on scene); out-of-hospital care; in-hospital care; and system integration, including public safety response and interagency communications. They are designed to be especially helpful for governmental organizations and medics in developing countries, which are disproportionally affected by disasters.

"They really have a chronic, ongoing disaster every day, and are often so caught up in trying to survive and care for patients that they don't have time to focus on drills," Green says. "As a result, many physicians have no specific training in disasters. Or, they may have a plan on paper but don't know how to use it."

In the past decade, an estimated 2 billion people worldwide have been affected by disasters, with natural disasters alone causing 666,000 deaths, Green says. The frequency of disasters is increasing, he says, with approximately one disaster a week somewhere in the world requiring major relief activities. Direct losses attributed to natural disasters have risen to more than $100 billion per year.

Green and colleagues worked with health officials in Guatemala to test the methods during a disaster drill in the capital, Guatemala City. In the scenario they chose, a propane tank exploded at an open-air food vendor located adjacent to a busy government building. Staged casualties included victims with direct blast injuries, trauma from flying debris and severe burns. Adding further complications, the explosion took place in a walled courtyard with only two entrances, located on a narrow, one-lane street several blocks from the nearest main street. Local officials and mock patients rehearsed and prepared before the drill was implemented.

Within five minutes of starting the drill, the first units of the Guatemalan Red Cross responded, quickly followed by municipal and volunteer firefighter and police units. In total, there were 40 health care providers in 12 vehicles, caring for 49 victims. Evaluators measured the times until patients were triaged, and observed the actions in caring for those patients.

Evaluators noticed some areas to improve, including a better need for security and crowd control at the hospital, and better education for some health care providers who were unsure of their role. But Green says overall they were relatively efficient: "Paramedics responded quickly and the providers were highly motivated to do their jobs."

The research team is continuing to test their methods. Another drill in Panama occurred in January.

The study was supported by The International Foundation and Emergency International. Co-authors were Surbhi Modi, M.P.H.; Kevin Lunney, M.H.S.; and Tamara Thomas, M.D.


- -JHMI- -

Green, Gary B. et al, "Generic Evaluation Methods for Disaster Drills in Developing Countries," Annals of Emergency Medicine, May 2003, Vol. 41, pages 689-699.


To see a table of evaluation methods, click here

Johns Hopkins' Department of Emergency Medicine

Annals of Emergency Medicine

Emergency International