Preparedness to respond effectively to natural or manmade

The Journal of TRAUMA Injury, Infection, and Critical Care AQ: 1 Do Trauma Centers Have the Capacity to Respond to Disasters? Frederick P. Rivara, MD, MPH, Avery B. Nathens, MD, PhD, MPH, Gregory J. Jurkovich, MD, and Ronald V. Maier, MD Background: Concern has been raised about the capacity of trauma centers to absorb large numbers of additional patients from mass casualty events. Our objective was to examine the capacity of current centers to handle an increased load from a mass casualty disaster. Methods: This was a cross-sectional study of Level I and II trauma centers. They were contacted by mail and asked to respond to questions about their surge capacity as of July 4, 2005. Results: Data were obtained from 133 centers. On July 4, 2005 there were a median of 77 beds available in Level I and 84 in Level II trauma centers. Fifteen percent of the Level I and 12.2% of the Level II centers had a census at 95% capacity or greater. In the first 6 hours, each Level I center would be able to operate on 38 patients, while each Level II center would be able to operate on 22 patients. Based on available data, there are 10 trauma centers available to an average American within 60 minutes. Given the available bed capacity, a total of 812 beds would be available within a 60-minute transport distance in a mass casualty event. Conclusions: There is capacity to care for the number of serious non-fatally injured patients resulting from the types of mass casualties recently experienced. If there is a further continued shift of uninsured patients to and fiscally driven closure of trauma centers, the surge capacity could be severely compromised. Key Words: Trauma, Capacity, Terrorism, Disaster. J Trauma. 2006;61:. Submitted for publication January 11, 2006. Accepted for publication March 7, 2006. Copyright 2006 by Lippincott Williams & Wilkins, Inc. From the Harborview Injury Prevention and Research Center (F.P.R., A.B.N., G.J.J., R.V.M.), and the Departments of Pediatrics (F.P.R.), Epidemiology (F.P.R.), and Surgery (A.B.N., G.J.J., R.V.M.), University of Washington, Seattle, Washington. Address for reprints: Frederick P. Rivara, Harborview Injury Prevention and Research Center, Box 359960, 325 Ninth Ave, Seattle, WA 98104; email: fpr@u.washington.edu. DOI: 10.1097/01.ta.0000219936.72483.6a Preparedness to respond effectively to natural or manmade disasters (such as a terrorist bombing) is of grave concern to the public, medical, and public health communities alike. The September 11, 2001 bombings in New York City and the Pentagon, the devastation of Hurricane Katrina, and the train bombings in Madrid and London emphasize the problems of dealing with mass casualties. In these disasters, the immediate concern is first triage, and then provision of definitive care for injured patients. Ideally, the provision of definitive care is best delivered in designated trauma centers where there is clear evidence of lower mortality compared with outcomes after care in non-designated centers. 1,2 Concern has been raised, however, about the capacity of the nation s current regional trauma centers to absorb large numbers of additional patients. 3 There were over 1.8 million people hospitalized for injuries in the United States in 2001, 4 of whom about 45% were treated in Level I or II trauma centers. 5 There are currently approximately 450 Level I or Level II trauma centers in the United States. 6 The current health care economic environment with its large proportion of under- or uninsured patients and poor reimbursement rates associated with either state or federal health care coverage has resulted in these centers assuming a large proportion of the care for indigent patients, both related and unrelated to trauma. As a result, many of these centers are overcrowded and struggling financially 7 and are operating at capacity. Thus, the ability of urban trauma centers to absorb large numbers of patients injured in a disaster might be compromised. To explore the extent to which existing trauma centers might contribute to care in the face of either manmade or natural disasters, we conducted a survey of Level I and II trauma centers in the United States to identify potential capacity to care for a large number of additional trauma patients on the Fourth of July, traditionally the busiest day of the year for trauma centers. METHODS Level I and II trauma centers were defined as those that were designated as such by a state or local regulatory body, or those verified by the American College of Surgeons Committee on Trauma. The centers were identified by contacting these designating agencies. Cover letters and a brief questionnaire were sent to the trauma director or trauma nurse coordinator at each institution. The study was approved by the Human Subjects Committee of the University of Washington. The respondent at each institution was asked to report on the number of licensed beds, the number of beds for which the hospital would be able to staff, the number of treatment and number of beds dedicated to observation in the emergency department (ED) and in the postanesthesia recovery unit, and the number of operating rooms. We asked for the inpatient census on the morning of July 4, 2005, the number Volume 61 Number 4 1 <zjs;original Article> <zjss;original Article> <zdoi;10.1097/01.ta.0000219936.72483.6a>

The Journal of TRAUMA Injury, Infection, and Critical Care F1 of ventilators that would be available that day, and the number of ED visits for the 24-hour period of July 4 to July 5. In Washington State, the mean time in the ED for trauma patients subsequently requiring admission was 5.0 hours. We assumed the diagnostic evaluation of the injured patient would be abbreviated with resource constraints and the pressure of time. Thus we examined the ED capacity under the scenarios that the ED triage and evaluation time could be shortened to at maximum two hours and, at minimum, one hour. Because there are no national standards for ED evaluation time, this range was used as a form of sensitivity analysis. Military field hospitals project approximately 2 hours per operation for damage-control procedures. 8 At the beginning of the Iraq war, the 212th Mobile Army Surgical Hospital averaged 100 minutes for an exploratory laparotomy, 120 minutes for thoracic procedures, and a mean of 85 minutes for placement of an external fixator. 9 These damage control procedures have been adopted by civilian hospitals as well 10,11 and in civilian mass casualty events. 12 These estimates of procedural duration were used in the calculation of operating room capacity. We assumed turnaround time between patients would be 30 minutes. We also assumed that no operating rooms would be available in the first hour because these would already be in use and would take time to clear for arrival of mass casualties. RESULTS Surveys were sent to 460 institutions; surveys were returned by 133 centers. The response rate was 31.2% for Level I centers and 27.6% for Level II centers. The responding centers were more likely to be ACS-verified than were nonresponding centers; responding and nonresponding Level I centers were of similar size, while responding Level II centers were somewhat smaller than nonresponding centers. On July 4, 2005, there were a median of 77 beds available in the 59 Level I trauma centers and 84 in the 74 Level II centers (Fig. 1). Fifteen percent of the Level I and 12.2% of the Level II centers had an inpatient census at 95% capacity or greater. Table 1 Median Emergency Department (ED) Capacity Level I Level II ED beds 39 25 ED patients 175 119 24-hour capacity 5 hours/patient 187 120 2 hours/patient 468 450 1 hour/patient 936 600 6-hour capacity 2 hours/patient 117 75 1 hour/patient 234 150 The median number of ED beds was 39 in Level I centers and 25 in Level II centers (Table 1). If the diagnostic evaluation of the injured patient in the ED could be shortened to 2 hours per patient, then the ED throughput would be 117 patients in the first 6 hours after a mass casualty event in Level I trauma centers (TC) and 75 patients over the same time interval in Level II centers (Table 2). If ED evaluations could be abbreviated to less than 1 hour per patient, then Level I centers could treat 234 patients and Level II centers 150 patients in the first 6 hours. The median number of operating rooms was 19 in Level I centers and 11 in Level II centers (Table 2). If we were to consider that the mean duration of an operative procedure is approximately 2 hours with a 30-minute room turnaround time, then operating room (OR) throughput in Level I centers would be 38 and 22 patients in Level I and II centers, respectively in the first 6 hours. These numbers would increase to 175 patients in Level I centers and 101 patients in Level II centers for operations within the first 24 hours. However, the number of ventilators immediately available at these centers is much less than the number that would be needed for postoperative care of this number of seriously injured trauma patients (Table 2). Many centers reported that additional ventilators were available from rental companies on short notice; these numbers were not included in our calculations. There are currently 189 Level I and 261 Level II TC in the US. 13 Branas et al. 14 report that there are 10 Level I and II centers accessible to an average American by helicopter or ground transport within 60 minutes. Assuming the same distribution as for an average American, these data suggest there would be 4 Level I and 6 Level II centers available to treat trauma patients within the golden hour after serious trauma in a metropolitan area (Table 3). Given the available bed T1 AQ: 2 T2 T3 Table 2 Median Operating Room (OR) and Ventilator Capacity Level I Level II Fig. 1. Bed capacity in Level I and Level II trauma centers on July 4, 2005. Operating rooms 19 11 24-hour OR capacity at 2 hours/operation 175 101 6-hour OR capacity at 2 hours/operation 38 22 Available ventilators 20 12 2 October 2006

Disaster Capacity of Trauma Centers Table 3 Capacity of Trauma Centers in an Average Metropolitan Area Number of Centers Within 6 ED Capacity Within 24 Within 6 OR Capacity Within 24 Inpatient Bed Capacity Level I centers 4 468 936 152 700 308 Level II centers 6 450 900 132 606 504 Total 10 918 1,836 284 1,304 812 capacity at these centers, a total of 812 beds would be available for inpatients within a 60-minute transport distance. DISCUSSION Historically, terrorist attacks have nearly all consisted of events that resulted in blunt and penetrating trauma, rather than biological or chemical injury. Prior studies indicate that an average of 34% of injured patients in mass casualties require hospital admission. 15 Recent terrorist attacks indicate that, while there are many individuals who seek ED care 12 and many with immediate mortality especially in cases of bombings with building collapse, the number of seriously injured patients requiring operative care and inpatient hospitalization is much smaller. 15 For example, the number of individuals with serious injuries was 76 in the London transit bombings, 16 89 in the Madrid train bombings, 17 83 in the Oklahoma City Federal Building bombing 18 and 135 after the New York City September 11, 2001 attack. 19 Only 40 patients were admitted to the trauma center designated as the main receiving hospital for casualties from the Twin Tower attacks. 20 In the period immediately after Hurricane Katrina, 552 people were admitted to hospitals with injuries and/or illness; individuals with injuries accounted for about onequarter of health-related events recorded and many were stable enough for ground or air transport to additional institutions outside the immediate disaster area. 21 The only Level I trauma center in the region, Charity Hospital, was flooded and six months after the storm, remained closed with an uncertain future. The closest Level I trauma center remaining open was 350 miles away, resulting in lack of adequate trauma care for many people. 22,23 Data providing estimates of the ability of trauma centers to respond to mass casualties in the United States have been limited; a recent simulation study was based on capability of only one trauma center. 24 The results of this single center study suggested that the surge capacity was slightly lower than that estimated through this survey of American trauma centers. Our study indicates that the theoretical capacity of trauma centers to provide operative care and inpatient beds for an average American would not be exceeded by the number of serious non-fatally injured patients resulting from the types of terrorist attacks experienced in the United States and Europe in the last five years. The two largest U.S. attacks (Oklahoma City bombing and the New York City 9/11 attack) produced large numbers of fatalities but relatively few seriously injured patients because of building collapse, which tends to have the highest immediate mortality rate of any kind of terrorist bombing. 15 The data on access to trauma centers is based on work from Branas et al. 14 This study calculated the access to trauma centers for an average American, based on a weighted population average. This estimate might be very different from the access in an average American metropolitan area. In some areas, such as cities in the Northeast, there may actually be more than 10 centers available within 60 minutes. For other areas of the country, such as the Pacific Northwest, there are likely to be far fewer centers accessible within 60 minutes. Access to this number of trauma centers also critically hinges on the ability to move seriously injured patients out of a city to centers in other cities. The immediate capability to do this on a large scale within the time frame needed may not be feasible. For example, in the model used by Hirshberg and colleagues 24 the mean time between patients arriving at the trauma center was only 7.2 minutes, indicating that the trauma patients from a mass casualty incident would arrive at trauma centers during a short period of time, stressing the capabilities to transfer patients elsewhere. This analysis indicates that there is the theoretical capacity of trauma centers as described previously to absorb the number of injured patients likely resulting from a mass casualty event. Utilization of these limited resources will greatly depend upon adequate coordination of the prehospital delivery system to appropriately triage patients in the field and get patients to the available beds in the appropriate level facility, and to mobilize the human resources needed for the staffing of these beds. Our study assumes that personnel will not be a limiting resource and that trained personnel will be prepared to deal with a mass casualty event. However, this may be optimistic in light of a recent study of trauma surgeons indicating that many may not be trained to treat the nontrauma aspects of weapons for mass destruction. 25 Thus, current focus of disaster response should be on the training of prehospital and trauma personnel and development and testing of coordination and communication systems. This study also assumes that the majority of injuries will be blunt trauma. If a mass casualty event resulted in a large number of burns, the capability to manage a large number of burn victims would quickly be overwhelmed, especially at Level II centers. In this study, we assume that initial operative interventions will consist of damage-control surgery. Based on their experience at a civilian trauma center, Eiseman and col- Volume 61 Number 4 3

The Journal of TRAUMA Injury, Infection, and Critical Care leagues found that nearly half of patients receiving damage control surgery will require definitive surgical repair within 12 hours and three-quarters within 24 hours. 10 For such an approach to be feasible, some patients will need to be transported to other facilities in different cities because the index trauma centers will still be engaged in the initial surgery on the victims. There might be several bottlenecks in the evaluation and management of the critically injured trauma patients. For example, we did not evaluate CT scan throughput and/or the availability of other critical components of the diagnostic evaluation. However, we did identify that the ventilator availability is another potential bottleneck. The need for additional ventilators beyond that usually present in most centers is apparent because on average hospitals will have only 15 ventilators available. Many hospitals reported that their disaster plans did include the availability of additional ventilators, but the extent to which their availability would be increased was not readily available. We chose to examine only total bed capacity rather than specific critical care unit capacity because in a true disaster, noncritical care beds can be used for critically ill patients with the availability of ventilators and portable monitoring devices. There is a significant proportion of both Level I and II centers that usually operate at near capacity during the summer trauma season, and thus would have limited capacity to absorb additional trauma patients. A coordinated response may involve transferring many of the medical and recovering surgical patients to other facilities to allow open capacity in the centers best able to care for trauma. The response rate of trauma centers to the study was low. Because this was conducted during the busiest time of the year for trauma centers, there may be a response bias in those centers working at capacity may have had less time to respond, thus potentially biasing the results to indicate higher surge capacity than might exist. We focused on the availability of physical resources in the context of conventional disasters. There was no assessment of human resources to respond, nor the availability of resources that might be required in the case of unconventional disasters where decontamination or isolation of victims is necessary. This study indicates that there does not appear to be a deficiency in the capacity of trauma centers in most cities to absorb the number of patients that require admission and surgical care after contemporary mass casualty disasters. However, failure to address relatively simple bottlenecks in the system might severely compromise a system that might otherwise be very effective. There might be further compromise if there is continued use of trauma centers as regional safety nets for the un- and underinsured. Increasing the number of uninsured at trauma centers will result in the closure of some centers, discourage other centers from seeking certification, and reduce the surge capacity of current centers. If the trauma systems that have evolved in this country become fragmented because of fiscal pressures, then the ability to cope with mass casualty events will be dramatically curtailed. In addition, the trauma surge capacity in any city clearly depends on the ability to transfer patients from the scene to facilities within a transport time of one hour. This requires a functioning communication and coordination system. As the events after Hurricane Katrina demonstrated, the current system with its heavy reliance on federal support is inadequate. REFERENCES 1. MacKenzie EJ, Rivara FP, Jurkovich G, et al. A national evaluation of the effect of trauma center care on mortality. New Engl J Med. 2006;354:366 378. 2. Demetriades D, Martin M, Salim A, et al. The effect of trauma center designation and trauma volume on outcome in specific severe injuries. Ann Surg. 2005;242:512 517. 3. Romano M. At capacity and beyond. Modern Healthcare. 2005; 6 7, 16. 4. Heinen H, Hall MJ, Boudreault MA, Fingerhut LA. National Trends in Injury Hospitalization, 1979 2001. Hyattsville, MD: National Center for Health Statistics, 2005. 5. Nathens AB, Jurkovich GJ, MacKenzie EJ, Rivara FP. A resourcebased assessment of trauma care in the United States. J Trauma. 2004;56:173 178. 6. MacKenzie EJ, Hoyt DB, Sacra JC, et al. National inventory of hospital trauma centers. JAMA. 2003;289:1515 1522. 7. Taheri PA, Butz DA, Lottenberg L, Clawson A, Flint LM. The cost of trauma center readiness. Am J Surg. 2004;187:7 13. 8. US Army. Army Field Manual 4-02.25. Employment of Forward Surgical Teams Tactics, Techniques, and Procedures, 2003. 9. Cho JM, Jatoi I, Alarcon AS, et al. Opeartion Iraqi Freedom: Surgical experience of the 212th Mobile Army Surgical Hospital. Milit Med. 2005;170:268 272. 10. Eiseman B, Moore EE, Meldrum DR, Raeburn C. Feasibility of damage control surgery in the management of military combat casualties. Arch Surg. 2000;135:1323 1327. 11. Holcomb JB, Helling TS, Hirshberg A. Military, civilian, and rural application of the damage control philosophy. Mil Med. 2001; 166:490 493. 12. Almogy G, Belzberg H, Mintz Y, et al. Suicide bombing attacks: update and modifications to the protocol. Ann Surg. 2004;239: 295 303. 13. American Trauma Society. Vol. 2005, 2005. 14. Branas CC, MacKenzie EJ, Williams JC, et al. Access to trauma centers in the United States. JAMA 2005;293:2626 2633. 15. Arnold JL, Tsai MC, Halpern P, et al. Mass-casualty, terrorist bombings: epidemiological outcomes, resource utilization, and time course of emergency needs (Part I). Prehospital Disaster Med. 2003; 18:220 234. 16. Ryan J, Montgomery H. The London attacks preparedness: Terrorism and the medical response. N Engl J Med. 2005;353: 543 545. 17. Gutierrez de Ceballos JP, Turegano Fuentes F, et al. Casualties treated at the closest hospital in the Madrid, March 11, terrorist bombings. Crit Care Med. 2005;33:S107 S112. 18. Teague DC. Mass casualties in the Oklahoma City bombing. Clin Orthop Relat Res. 2004;77 81. 19. Cushman JG, Pachter HL, Beaton HL. Two New York City hospitals surgical response to the September 11, 2001, terrorist attack in New York City. J Trauma. 2003;54:147 154. AQ: 3 4 October 2006

Disaster Capacity of Trauma Centers 20. Feeney JM, Goldberg R, Blumenthal JA, Wallack MK. September 11, 2001, revisited. A review of the data. Arch Surg. 2005;140:1068 1073. 21. Centers for Disease Control and Prevention. Surveillance for illness and injury after hurricane Katrina New Orleans, Louisiana, September 8 25, 2005. MMWR Morb Mortal Wkly Rep. 2005; 54:1018 1021. 22. Kaiser Commission on Medicaid and the Uninsured. Addressing the health impact of Hurricane Katrina. Vol. 2006: Kaiser Family Foundation, 2005. 23. Rosenbaum S. US health policy in the aftermath of Hurricane Katrina. JAMA. 2006;295:437 440. 24. Hirshberg A, Scott BG, Granchi T, et al. How does casualty load affect trauma care in urban bombing incidents? A quantitative analysis. J Trauma. 2005;58:686 693. 25. Ciraulo DL, Frykberg ER, Feliciano DV, et al. A survey assessment of the level of preparedness for domestic terrorism and mass casualty incidents among Eastern Association for the Surgery of Trauma members. J Trauma. 2004;56:1033 1039. Volume 61 Number 4 5

JOBNAME: AUTHOR QUERIES PAGE: 1 SESS: 1 OUTPUT: Fri Aug 18 13:28:52 2006 /balt5/zta ta/zta ta/zta01006/zta9129 06z AUTHOR QUERIES AUTHOR PLEASE ANSWER ALL QUERIES 1 AQ1 No running head was provided. Please confirm or change as needed. AQ2 Please confirm expanision of TC. AQ3 Please provide complete reference.