Oklahoma weather. Mass casualty incident (MCI). Tornado, right? That may seem like the best guess, but Oklahoma EMTs and paramedics faced an unusual MCI on Oct. 17, 2007, when high, straight-line winds caused a circus-sized tent to collapse at Tulsa’s Oktoberfest.
The National Weather Service forecasted two thunderstorm fronts, separated by about two hours, would impact the Tulsa area during the late afternoon to evening hours. Because Oklahoma EMS leaders know the accuracy and value of these forecasts, standard mid-afternoon activities were quickly replaced with planning weather-related contingencies, deploying additional ambulances and increasing trauma-related supplies on mobile restocking vehicles.
The Metropolitan Medical Response System (MMRS) director advised field crews to be wary if sunny, warm weather followed the first front, signaling that a stronger front would likely be forming. The first thunderstorm passed through at 1600 hrs, followed by warm, sunny skies, which was, ironically, bad news for the thousands of people heading to Oktoberfest Corporate Night.
Soon, 7,000 corporate patrons, media personnel and local dignitaries were celebrating the event’s start. The second storm front arrived after 1900 hrs, and those not already under tent cover soon sought refuge from heavy rain and hail. When winds upward of 70 mph began destabilizing the main tent around 1920 hrs, an estimated 2,500 people were under its canvas.
And Down It Came
Eyewitness reports from off-duty police, fire and EMS personnel in the crowd described winds pulling up the 400 lb. steel poles in the center of the tent and swinging their ends three to four feet above the ground. The corner anchor poles held, giving the tent a momentary parachute shape before its collapse. During this brief instant, some attendees were able to flee through the tent’s multiple exit points before being trapped under the heavy canvas.
Multiple cellular calls to 9-1-1 were placed in the next two to three minutes. Ambient crowd noise, continuing winds and caller requests for care of single patients without mention of a tent collapse made the initial calls hard to understand. Dispatchers eventually realized a major medical event was occurring and initiated an “all call” page to EMS leaders and supervisors.
The MMRS director, typically mobile during local severe weather events, arrived concurrent with the first EMS Authority (EMSA) mobile intensive care unit (MICU) at 1926 hrs. This was less than five minutes after the first calls for assistance. Ingress quickly proved difficult due to the self-extricating masses. Many arriving responders said it looked like a scene from “Night of the Living Dead,” with multiple people bleeding from head wounds and wearing dazed expressions as they made their way to the parking lots.
Medical command was quickly established and expanded with arriving Tulsa police and fire personnel to a NIMS-modeled unified incident command. The MMRS director had mobile Internet access and verified the second weather front was indeed the last anticipated within radar range of 50 or more miles. He decided to set up the treatment area in an uncollapsed tent after a quick structural stability assessment. This setup allowed crews to provide on-scene care while being protected from continuing rain fall and 1/2″ hail.
Although Internet access proved helpful in designating a safe treatment sector, it didn’t help much otherwise with event operations. The Oklahoma Oktoberfest home page featured an outdated 2006 tent map, and several tents had been repositioned in 2007 to accommodate additional activities and larger crowds. Event operations therefore had to be designated by their site’s relation to the treatment and transport sector locations.
The first of two red/immediate priority patients (head injury) was transported at 2003 hrs, 37 minutes after first EMS arrival. Several factors contributed to this interval:
> The initial MICU became embedded in the event and couldn’t be used for transport.
> Nighttime operations provided suboptimal lighting.
> Weather inclemency continued.
> Multiple clusters of injuries existed: There were at least three areas to which triage teams were being simultaneously directed by bystanders.
> Delays in additional responding personnel occurred: The event was between a two-lane road to the west and a riverbank to the east. There was no north or south access to the event grounds due to surrounding structures. Patrons leaving in private vehicles created a traffic volume that severely hampered fire engine and MICU access to the scene. Tulsa police officers proved essential in creating effective ingress and egress for these additional responders.
The remaining red/immediate priority patient (spinal cord injury with paralysis) was transported within the next two minutes, giving an “Arrival to All Immediates Transported” interval of 39 minutes. A total of 23 patients — the two red/immediate priority, six yellow/delayed priority and 15 green/minor priority — were transported in a well-dispersed manner among Tulsa’s six acute-care hospitals.
Nine MICUs and a city transit bus were used for patient transport. Many of the MICUs were able to return quickly for a second transport due to hospital proximity. Four patients were transported on the city bus (two with minor head injuries, two with leg injuries). Three EMS personnel accompanied these patients, bringing along a cache of portable medical equipment in case of declining patient status. Each of these patients proved stable throughout bus transport to the hospital furthest from the scene.
Hospitals were alerted of a potential MCI within five minutes of the first 9-1-1 calls and provided with updates by Tulsa_s Medical Emergency Response Center (MERC) via a proprietary Web-based system for MCIs and hospital status reports.
The last EMSA patient transport occurred at 2041 hrs, resulting in an “Arrival to All Patients Transported” interval of 75 minutes. A far-first transport pattern was utilized in anticipation that many self-extricating injured patients would utilize the two hospitals nearest the event. However, contrary to self-referral patterns after an MCI, most of the night’s attendees resided in South Tulsa and chose to go to hospitals in that area.
EMS was released from the scene at 2130 hrs, little more than two hours after arrival. An additional 35 self-referred patients presented to emergency departments that evening. Numerous patients with minor injuries either self-cared at home or sought care through their personal physicians, making it impossible to determine the total number of people injured at this incident.
Think Ahead
Pre-planning for this kind of major incident is critical in areas with large-scale events. EMS leaders should discuss the following questions with their jurisdictional government leaders and public-safety colleagues to produce answers specific to their area.
> To obtain an event license, your city likely requires event planners to have event mapping. Are they also required to have a medical evacuation plan? Will you have access to these maps and plans prior to the event, as well as on site, and how will they be made available to you?
> EMS doesn’t have authority to cancel privately organized events in the Tulsa area. Who in your city has the authority to cancel a privately organized public event due to public safety concerns? What input and criteria should be utilized to issue a cancellation?
> Can your field personnel get real-time weather maps and hospital status reports during initial MCI response?
> Would self-referral patients seek medical care at hospitals near or far from an MCI in your city?
In Closing
Tulsa’s EMSA, fire and police professionals successfully responded to this challenging event. Use their experiences to prepare your system to respond to similar incidents. Make sure you have contingency plans ready before your city’s next big event.
Jeffrey M. Goodloe, MD, FACEP, NREMT-P, is the associate EMS medical director for Tulsa and Oklahoma City, EMS program director for the Oklahoma Institute for Disaster and Emergency Medicine, and clinical associate professor of emergency medicine at the”žUniversity of Oklahoma College of Medicine-Tulsa. Contact him at jeffrey-goodloe@ouhsc.edu.
Charles E. Stewart, MD, PhD, FACEP, is a clinical associate professor of emergency medicine at the University of Oklahoma College of Medicine-Tulsa and the research director for the Oklahoma Institute for Disaster and Emergency Medicine.
Carolyn K. Synovitz, MD, MPH, FACEP, is a clinical associate professor of emergency medicine at the University of Oklahoma College of Medicine-Tulsa and the emergency medicine residency director for the Oklahoma Institute for Disaster and Emergency Medicine.
Kelly Deal, NREMT-P, is the Tulsa Metropolitan Medical Response System’s director and a faculty member of the Oklahoma Institute for Disaster and Emergency Medicine.