Annals of Internal Medicine Perspective Early Disaster Response in Haiti: The Israeli Field Experience Yitshak Kreiss, MD, MHA, MPA; Ofer Merin, MD; Kobi Peleg, PhD, MPH; Gad Levy, MD; Shlomo Vinker, MD; Ram Sagi, MD; Avi Abargel, MD, MHA; Carmi Bartal, MD, MPH; Guy Lin, MD; Ariel Bar, MD, MHA; Elhanan Bar-On, MD; Mitchell J. Schwaber, MD, MSc; and Nachman Ash, MD, MS

The earthquake that struck Haiti in January 2010 caused an esti- prepared and trained medical unit maintained on continuous alert. mated 230 000 deaths and injured approximately 250 000 people. The prompt deployment of advanced-capability field is The Israel Defense Forces Medical Corps was fully essential in disaster relief, especially in countries with minimal med- operational on site only 89 hours after the earthquake struck and ical infrastructure. The changing medical requirements of people in was capable of providing sophisticated medical care. During the 10 an earthquake zone dictate that field hospitals be designed to days the hospital was operational, its staff treated 1111 , operate with maximum flexibility and versatility regarding , hospitalized 737 patients, and performed 244 operations on 203 staff positioning, treatment priorities, and hospitalization policies. patients. The field hospital also served as a referral center for Early coordination with local administrative bodies is indispensable. medical teams from other countries that were deployed in the surrounding areas. Ann Intern Med. 2010;153:45-48. www.annals.org The key factor that enabled rapid response during the early For author affiliations, see end of text. phase of the disaster from a distance of 6000 miles was a well- This article was published at www.annals.org on 4 May 2010.

n earthquake measuring 7.0 on the Richter magnitude including computer and communication specialists, secu- Ascale struck close to Port-au-Prince, Haiti, on 12 Jan- rity staff, kitchen staff, carpenters, plumbers, mechanics, uary 2010. The official death toll was set at 230 000, and electricians, and a burial team. local authorities estimated that 250 000 people were in- A total of 1111 patients, 44% male, passed through jured (1). This catastrophic event galvanized a strong and our triage point (Appendix Table 2, available at www rapid response worldwide, and the Israeli government .annals.org), of whom 737 patients (63%) were hospital- quickly decided to launch a medical humanitarian mission ized. We performed a total of 244 surgical procedures on to provide medical care as advanced as possible under the 203 patients (Appendix Table 3, available at www.annals circumstances. .org). Trauma accounted for 66% of the admissions; most Whereas the fate of patients with life-threatening cases involved open or closed limb fractures. Many patients internal-organ injuries is determined within the first hours sustained local or systemic infections involving inade- of a disaster, early provision of treatment for the multi- quately treated deep wounds, including necrotizing soft- tudes of patients with open fractures can prevent life- tissue infections (such as gas gangrene) (Appendix Table 4, threatening sepsis and limb-threatening infections. In ad- available at www.annals.org). During the first 3 days, ap- dition, situations involving substantial casualties combined proximately 80% of the patients seen had had trauma with extensive damage to local medical facilities and infra- (Figure 1). Hence, all of the hospital resources were ini- structure highlight the need for a resourceful, experienced, tially dedicated to orthopedic procedures (debridement of and trained medical team backed by a logistics contingent. necrotic tissues in life-threatening infections; fracture fixa- The Israel Defense Forces Medical Corps (IDF-MC) Field tion for salvageable limbs; and amputation of nonsalvage- Hospital comprises such a unit (2–5). able limbs, with the aim of saving life). This protocol grad- The field hospital staff consisted of 121 servicemen ually changed given the dynamic nature of the needs of and servicewomen (Appendix Table 1, available at www patients in the earthquake region, resulting in other types .annals.org) and was organized into medical, surgical, or- of patients being hospitalized (Figure 1). thopedic, pediatric, gynecologic, and ambulatory care divi- The first goal of the IDF-MC is to arrive at a very sions, as well as auxiliary units (Appendix Figure, available early stage after a disaster. To do so after the Haitian earth- at www.annals.org), with a capacity of 60 inpatient beds quake, every effort was made to follow a protocol that had that could be expanded to 72. been established as a result of our experiences after the To ensure maximum logistic independence and to earthquakes in Turkey (Adapazari, 1999) and India (Gu- shorten the time to deployment, we brought all hospital supplies; a fully stocked pharmacy, including sufficient oral antibiotics to be distributed on discharge; imaging machin- See also: ery; a laboratory that could perform blood tests and urine chemistry, hematology, blood gases, and microbiology Web-Only analyses; and autoclaves for sterilization. Energy sources Appendix Tables (generators) and accommodations (tents and latrines) were Appendix Figure also brought from Israel. This crucial effort was carried out Conversion of graphics into slides by a highly trained, skilled logistics unit of 109 personnel,

© 2010 American College of 45

Downloaded From: http://annals.org/ on 11/01/2016 Perspective Early Disaster Response in Haiti

maintain constant readiness in dispatching a field hospital Figure 1. Distribution of trauma versus nontrauma cases. to anywhere in the world with minimal notice. Major time-saving factors were the high level of preparedness, 100 Trauma expressed by staff experience; the “on-alert” mentality of Nontrauma the IDF-MC reserve personnel; intensive training; and that % 80 established procedures were followed. Moreover, in the Haiti mission, quick decision making and personal in-

60 volvement of senior army and government leadership re- moved substantial obstacles, such as switching from smaller military planes usually deployed to the larger commercial 40 aircrafts required for such a massive and remote disaster. The quick departure of the forward-assessment team and 20 its actions on the ground were also crucial. Their success in

Total Daily Triage Admissions, Daily Triage Total enabling our landing in Port-au-Prince and preparing a suitable deployment area saved valuable time. 0 1 2 3 4 5 6 7 8 9 10 11 We needed to decide which would be the optimal type Day of Activity of field hospital to erect in response to the earthquake. The options included a “light” hospital that would provide pri- The graph demonstrates the gradual change in case mix over time. Care mary treatment to many patients, or alternatively a more of trauma patients made up most of the hospital activity in the early days, whereas the percentage of nontrauma patients gradually increased sophisticated hospital with advanced capabilities from a over time. wide variety of specialties that would limit the number of potentially treatable victims. Our choice of the latter op- jarat, 2001) and in humanitarian missions to Rwanda tion was based on experience with disasters in general and (1994) and Kosovo (Brazda, Macedonia, 1999) (2–5). We earthquakes in particular. Local hospital infrastructure is covered the 6000-mile distance to set up a fully operational often destroyed, as documented in earthquakes in Tang- field hospital within only 89 hours of the event. A special shan, China (1976) (6); Armenia (1988) (7); and Kobe, assessment team was en route to Haiti 11 hours after news Japan (1995) (8, 9). Many victims who need hospital-level of the earthquake reached Israel (Figure 2). At the same care require evacuation either to facilities outside the af- time, we began organizing the field hospital according to fected region or to mobile hospitals deployed locally that established protocol and landed in Haiti on the evening of can handle such injuries. In addition, given the nature of 15 January 2010. We chose a deployment area in a soccer earthquake-related injuries, specialized field hospital ser- field near the airport, and the hospital was fully operational vices that become available only 1 week or more after an 8 hours after all of the equipment was delivered to the site earthquake tend to be ineffective in reducing mortality (10, (Figure 3). 11). The United Nations Disaster Assessment and Coordi- Although several factors in the Israeli model enabled nation (UNDAC) reports and personal communications this rapid and effective response, the primary factor was the clearly indicated that there was no local medical infrastruc- Israeli government’s strategic decision that the IDF-MC ture to serve as a referral center and that no field hospital

Figure 2. Timeline from earthquake to operational deployment.

Special Assessment Team Personnel Dispatch Cargo Plane First Plane Takeoff Start of Landing Admission Start of Preparations Personnel Cargo Plane Hospital Plane Landing Earthquake Takeoff Assembly

89 Hours 12 Jan 2010 16 Jan 2010 4:53 p.m. 10:00 a.m.

Dates and times shown are according to Haitian time. The first patient was admitted 89 hours after the earthquake.

46 6 July 2010 Annals of Internal Medicine Volume 153 • Number 1 www.annals.org

Downloaded From: http://annals.org/ on 11/01/2016 Early Disaster Response in Haiti Perspective

Figure 3. An aerial photograph of the field hospital.

The tents on the periphery of the setup were used as accommodations for members of the mission. ICU ϭ ; Lab ϭ laboratory; OB and GYN ϭ obstetrics and gynecology; OR ϭ operating room.

already deployed in Haiti had advanced capabilities (such tremely frustrating situation, we accepted new patients as as intensive care unit, imaging, and laboratory facilities). soon as space became available, performed essential sur- Furthermore, we understood that delegations expected to gery, and discharged the patients sooner than we would arrive would be bringing light hospitals and clinical facili- have wanted to make room for new arrivals. Because there ties; therefore, we decided to transport a sophisticated field was no centralized triage mechanism that could direct pa- hospital capable of providing advanced care. We expected tients to one facility versus another, and being well aware that the light hospitals and clinics could care for the vast of the risks of not providing adequate postoperative care, majority of patients and that we could provide most of the we notified each light hospital and other health facilities functions of an acute-care hospital. That decision turned that for every patient referred to us for a higher level of out to be appropriate in Haiti, but in retrospect, we rec- care, we would expect the referring facility to be willing to ommend that for optimal efficiency, such decisions should accept one of our patients for immediate postoperative lie in the hands of UNDAC or another responsible agency management in exchange. This policy enabled us to max- with access to immediate input on all types of medical imize the throughput of our operating room by increasing facilities and personnel deployed to specific disaster areas. the number of operations and procedures that we were in a Another challenge we faced was how to deal with the unique position to perform, while ensuring that our pa- need for surge capacity in our field hospital. As soon as the tients were not abandoned. This kind of optimization of hospital was deployed, patients began to arrive, as news of resources, with centralized triage and coordination mecha- the existence of our facility began to spread—largely by nisms, should also ideally have been under the auspices word of mouth. In fewer than 2 days, the hospital was at of an all-encompassing administrative body, such as full capacity—more quickly than we had anticipated—in UNDAC. part because it began to serve also as a referral center for We coped with changes in the medical requirements the other primary care teams that were deployed in the of an earthquake zone over time. The first few days re- surroundings and because of the scarcity of advanced med- quired us to concentrate our efforts on treating injuries ical facilities. We could not continue to admit patients who caused directly by the earthquake, and so we transformed needed surgery or advanced procedures after our hospital- one orthopedic treatment station into a surgical unit with ization capability was fully utilized. To cope with this ex- full anesthetic and monitoring capabilities, thus doubling

www.annals.org 6 July 2010 Annals of Internal Medicine Volume 153 • Number 1 47

Downloaded From: http://annals.org/ on 11/01/2016 Perspective Early Disaster Response in Haiti

our surgical capacity. We also shifted medical staff mem- and Chaim Sheba Medical Center, Tel Hashomer; School of Public bers, especially nurses, from nonsurgical units to general Health and Sackler School of Medicine, Tel Aviv University, Tel Aviv and orthopedic surgery units. At the time of peak pressure, Sourasky Medical Center, Assaf Harofeh Medical Center, and National Center for Infection Control, Israel Ministry of Health, Tel Aviv; Soroka a Colombian military surgical team was incorporated into University Medical Center and Ben-Gurion University of the Negev, our staff; as a result, 3 to 4 operating tables were occupied Beer-Sheva; The Western Galilee Hospital, Nahariya; and Schneider around the clock. A few days later, when patients with less Children’s Medical Center, Petah Tikva, Israel. urgent medical needs began arriving, we again readjusted staff assignments, organization of the units, and the policy Acknowledgment: The authors thank Ms. Esther Eshkol for editorial of hospitalization. Such intramission adaptability, we have assistance, Ms. Dorit Tzur and Ms. Estela Drezena-Simhoni for data learned, is possible only when there is maximal versatility analysis, and Lieutenant Colonel Dr. Daniel Segura Sanchez of the Na- in staff selection and training, implementation of equip- tional Army of Colombia for clinical assistance and collegiality. ment, and organizational planning. Disaster medicine always involves ethical issues, and Potential Conflicts of Interest: None disclosed. Forms can be viewed at these were especially challenging in Haiti given the huge www.acponline.org/authors/icmje/ConflictOfInterestForms.do?msNumϭM10 discrepancy between available resources and human need. -0522. We confronted numerous clinical and ethical dilemmas that we rarely encounter in daily clinical practice. They Requests for Single Reprints: Yitshak Kreiss, MD, MHA, MPA, 182 began as early as triage—for example, in deciding which Kedem Street, Shoham, PO Box 1773, Israel; e-mail, ykreiss@gmail .com. patients we could accept with our limited resources—and continued throughout our stay (12). The World Medical Association recommendations on medical ethics in the Current author addresses are available at www.annals.org. event of disasters include the following statement: “The must act according to the needs of patients and References the resources available. He/she should attempt to set an 1. Me´decins Sans Frontie`res. Haiti: from one emergency to the next. London: order of priorities for treatment that will save the greatest Me´decins Sans Frontie`res; 12 February 2010. Accessed at www.msf.org.uk number of lives and restrict morbidity to a minimum” /haiti_one_month_on_20100212.news on 15 April 2010. (13). This guideline led us to prioritize the provision of 2. Amital H, Alkan ML, Adler J, Kriess I, Levi Y. Israeli Defense Forces Medical Corps humanitarian mission for Kosovo’s refugees. Prehosp Disaster Med. 2003; treatment to Haitian patients with life-threatening condi- 18:301-5. [PMID: 15310041] tions in a way that would allow us to extend our resources 3. Heyman SN, Eldad A, Wiener M. Airborne field hospital in disaster area: to the maximum number of people in need. Specifically, lessons from Armenia (1988) and Rwanda (1994). Prehosp Disaster Med. 1998; on the basis of this approach, patients receiving care were 13:21-8. [PMID: 10187022] 4. Bar-Dayan Y, Leiba A, Beard P, Mankuta D, Engelhart D, Beer Y, et al. A not necessarily the most severely injured but were those multidisciplinary field hospital as a substitute for medical hospital care in the deemed most likely to benefit from treatment. Medical aftermath of an earthquake: the experience of the Israeli Defense Forces Field leaders and personnel must be prepared to confront these Hospital in Duzce, Turkey, 1999. Prehosp Disaster Med. 2005;20:103-6. complex ethical decisions. [PMID: 15898489] 5. Bar-Dayan Y, Beard P, Mankuta D, Finestone A, Wolf Y, Gruzman C, et al. During our deployment in Haiti, humanitarian dele- An earthquake disaster in Turkey: an overview of the experience of the Israeli gations from many other countries, nongovernmental or- Defence Forces Field Hospital in Adapazari. Disasters. 2000;24:262-70. [PMID: ganizations, and the United Nations continued to arrive. 11026159] The increasing hospitalization capacity of The General 6. Sheng ZY. Medical support in the Tangshan earthquake: a review of the management of mass casualties and certain major injuries. J Trauma. 1987;27: Hospital operated by the Red Cross (the largest local hos- 1130-5. [PMID: 3312621] pital), the arrival of the USNS COMFORT, 7. Noji EK, Kelen GD, Armenian HK, Oganessian A, Jones NP, Sivertson KT. and the establishment of a medical facility by the Univer- The 1988 earthquake in Soviet Armenia: a case study. Ann Emerg Med. 1990; sity of Miami allowed provision of longer-term medical 19:891-7. [PMID: 2142590] 8. Tanaka K. The Kobe earthquake: the system response. A disaster report from care, thereby permitting us to coordinate an trans- Japan. Eur J Emerg Med. 1996;3:263-9. [PMID: 9056140] fer of patients and departure from Haiti with a sense of our 9. Aoki N, Nishimura A, Pretto EA, Sugimoto K, Beck JR, Fukui T. Survival mission having been accomplished. It is our hope that the and cost analysis of fatalities of the Kobe earthquake in Japan. Prehosp Emerg Care. 2004;8:217-22. [PMID: 15060860] lessons we learned in responding to the Haitian disaster 10. Adler J, Eldar R. [Recommendations for earthquake preparedness in Israel]. will be translated into internationally accepted recommen- Harefuah. 2001;140:877-82, 893. [PMID: 11579743] dations, and that these in turn will never in the future need 11. Schultz CH, Koenig KL, Noji EK. A medical disaster response to reduce to be implemented. immediate mortality after an earthquake. N Engl J Med. 1996;334:438-44. [PMID: 8552147] 12. Merin O, Ash N, Levy G, Schwaber MJ, Kreiss Y. The Israeli field hospital From the Israel Defense Forces Medical Corps Field Hospital, Home in Haiti—ethical dilemmas in early disaster response. N Engl J Med. 2010;362: Front Command, and Israel Defense Forces Medical Corps Surgeon e38. [PMID: 20200362] General; Shaare Zedek Medical Center, Hebrew University, Jerusalem; 13. World Medical Association. Statement on Medical Ethics in the Event Israel National Center for Trauma and Emergency Medicine Research, of Disasters. 14 October 2006. Accessed at www.wma.net/en/30publications The Gertner Institute for Epidemiology and Health Policy Reasearch, /10policies/d7/index.html on 15 April 2010.

48 6 July 2010 Annals of Internal Medicine Volume 153 • Number 1 www.annals.org

Downloaded From: http://annals.org/ on 11/01/2016 Annals of Internal Medicine

Current Author Addresses: Dr. Kreiss: 182 Kedem Street, Shoham, PO Box 1773, Israel. Appendix Table 2. Distribution of Patients, by Sex and Age Dr. Merin: POB 539 Mevaseret, Zion 90805, Israel. Dr. Peleg: 4 Tomer Street, Reut 71908, Israel. Variable Patients, n (%) Dr. Levy: 54 Halivne Street, Givaat-Ada, Israel. Sex* Dr. Vinker: POB 14238 Ashdod, 77041, Israel. Male 459 (44.0) Dr. Sagi: 26 Mishol Haahava Street, Kfar-Sava 44601, Israel. Female 582 (56.0) Dr. Abargel: 21 Hayasmin, Tel-Mond, Israel. Dr. Bartal: 10 Yamit Street, Beer-Sheba 84803, Israel. Age† Dr. Lin: Kibbutz Rosh-Hanikra, West Galilee 22825, Israel. 0–2 y 112 (11.4) Ͼ2–18 y 251 (25.5) Dr. Bar: 110 Emek Ayalon Street, Modi’in 71700, Israel. Ͼ18–50 y 574 (58.4) Dr. Bar-On: Pediatric Orthopedic Unit, Schneider Children’s Medical Ͼ50 y 46 (4.7) Center, 14 Kaplan Street, Petah Tikva 49202, Israel. Dr. Schwaber: National Center for Infection Control, 6 Weizmann * Data were available for 1041 of 1111 triage admissions. Street, Tel Aviv 64239, Israel. † Data were available for 983 of 1111 triage admissions. Dr. Ash: 1st Avital Street, Rosh Haain 48631, Israel.

Appendix Table 1. Composition of Hospital Personnel

Sector Personnel, n Physicians 44 Orthopedic surgery 7 General surgery 5 Pediatric surgery 1 Ear, nose, and throat 1 Ophthalmology 1 Anesthesiology 4 Obstetrics and gynecology 3 Internal medicine 7 Appendix Table 3. Key Operational Parameters Infectious diseases 2 Pediatrics 7 Family medicine 4 Parameter Total, n (%) Diagnostic radiology 1 Triage admissions 1111 (100.0) Psychiatry 1 Hospitalizations 737 (66.3*) Nurses 27 Surgical patients 203 (27.5†) and medics 21 Operations 244 (NA) Medical engineers 2 In-hospital deaths 17 (2.3†) Public health practitioners 2 Ambulatory and care 374 (33.7‡) Pharmacists 2 Pregnant patients 24 (2.16*) Radiology technicians 2 Deliveries 16 (66‡) Laboratory technicians 3 Premature infants 5 (NA) Social worker 1 Medical informatics personnel 2 NA ϭ not applicable. Command and logistics personnel 15 * Percentage of total triage admissions. Total 121 † Percentage of all 737 hospitalizations. ‡ Percentage of all 24 pregnant patients.

www.annals.org 6 July 2010 Annals of Internal Medicine Volume 153 • Number 1 W-25

Downloaded From: http://annals.org/ on 11/01/2016 Appendix Figure. Organizational chart of the field hospital.

Hospital Management and Command

Obstetrics Ambulatory Auxiliary Medicine Surgery Orthopedics Pediatrics and Clinic Services Gynecology

ED Triage ED ED Laboratory

Medical Inpatients ED Inpatients Inpatients Supplies and Engineering Neonatal Intensive Labor Intensive Care Room Care Imaging

Operating Mental Room Support Team

Postoperative Recovery Informatics

Logistics

The hospital was divided into 7 major divisions: medicine, surgery, orthopedics, pediatrics, obstetrics and gynecology, ambulatory outpatient clinic, and an auxiliary services unit. The surgical division was also responsible for staffing the triage point. The pediatric division included a neonatal intensive care unit, and the obstetrics and gynecology division operated a labor room that also functioned as an obstetric operating room. ED ϭ emergency department.

Appendix Table 4. Occurrence of the Top 10 Most Frequent Trauma Diagnosis Groups*

ICD-9-CM Diagnosis Patients, n (%)† Fractures 265 (38.74) Open wounds 188 (27.49) Superficial injuries 120 (17.54) Crush injuries 107 (15.64) Contusions with intact skin surface 89 (13.01) Burns 16 (2.34) Dislocations 12 (1.75) Internal injuries of the chest, abdomen, and pelvis 12 (1.75) Sprains and strains of joints and adjacent muscles 11 (1.61) Intracranial injury (including skull fracture) 8 (1.17)

ICD-9-CM ϭ International Classification of Diseases, Ninth Revision, Clinical Modification. * Diagnostic data were available for 1041 patients. † Many patients had Ͼ1 diagnosis.

W-26 6 July 2010 Annals of Internal Medicine Volume 153 • Number 1 www.annals.org

Downloaded From: http://annals.org/ on 11/01/2016