Public Health Policy and Disaster Performance Discussion

Emerg Med Clin N Am 24 (2006) 749–768 Unstable Ethical Plateaus and Disaster Triage Matthew D. Sztajnkrycer, MD, PhDa,*, Bo E. Madsen, MDa, Amado Alejandro Báez, MD, MScb a Department of Emergency Medicine, Mayo Clinic, 200 First Street SW, Rochester, MN 55905, USA b Department of Emergency Medicine and Division of Trauma, Burns and Surgical Critical Care, Brigham and Women’s Hospital, Neville House, Room 226, 75 Francis Street, Boston, MA 02115, USA Disasters are defined medically as mass casualty incidents in which the number of patients presenting during a given time period exceeds the capacity of the responders to render effective care in a timely manner. During such circumstances, triage is instituted to allocate scarce medical resources. Current disaster triage attempts to do the most for the most, with the least amount of resources. This article reviews the nature of disasters from the standpoint of immediate medical need, and places into an ethics framework currently proposed utilitarian triage schema for prioritizing medical care of surviving disaster victims. Specific questions include whether resources truly are limited, whether specific numbers should dictate disaster response, and whether triage decisions should be based on age or social worth. The primary question the authors pose is whether disaster triage, as currently advocated and practiced in the western world, is actually ethical. The key concepts of this article are as follows: ! Disasters are defined medically in terms of relative scarcity of medical resources, as opposed to absolute patient numbers. ! Subsequent disaster triage decisions are inherently utilitarian in nature, attempting to do the most for the most, with the least resources. The content of this publication solely represents the views and opinions of the authors and does not necessarily reflect the official views, policies, or position of the National Disaster Medical Service, US Department of Homeland Security, or the US Government. * Corresponding author. Department of Emergency Medicine, Mayo Clinic, 200 First Street SW, Rochester, MN 55905. E-mail address: sztajnkrycer.matthew@may.edu (M.D. Sztajnkrycer). 0733-8627/06/$ – see front matter ! 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.emc.2006.05.016 emed.theclinics.com 750 SZTAJNKRYCER et al ! Most modern triage schemes use a tiered response, in which one group is deemed expectant and, therefore, not deserving of resuscitation because of consumption of scarce resources. ! Data from recent mass casualty events seem to contradict the concept of scarce resources, and suggest that these expectant patients can be managed aggressively. ! Because current triage schemes essentially constitute a societally mandated Do Not Resuscitate order, broad-level discussions involving all elements of the community should be undertaken to determine the appropriateness of these decisions. The numbers are staggering. In the past 30 years, millions of lives have been lost to disasters, and billions of lives affected [1,2]. Approximately 62,000 people per annum die as a result of large-scale global disasters [3]. As defined by the Merriam-Webster dictionary [4], a disaster is ‘‘a sudden calamitous event bringing great damage, loss, or destruction.’’ From a global perspective, the World Health Organization defines a disaster as a ‘‘sudden ecological phenomenon of sufficient magnitude to require external assistance’’ [5]. However, this definition does little to provide insight and guidance into the specific medical needs of a disaster. Public Health Policy and Disaster Performance Discussion

The American College of Emergency Physicians’ definition of disasters as ‘‘situations in which destructive effects of an event provoked by nature or human beings exceed the available resources required by a community or region in need of medical care’’ once again provides very little guidance to the medical community [6]. At what point are resources exceeded, for example? Moreover, a disaster may result in mass fatalities but few patients. At the institutional level, a working definition of a disaster might be a situation in which ‘‘the number of patients presenting within a given time period are such that the emergency department (or field responding units) cannot provide care for them without external assistance’’ [7]. More precisely, care cannot be rendered in a timely manner. For the remainder of this article, the authors will use this working definition in discussing medical care during disasters. The principle underlying these definitions is the concept of relative scarcity of available resources. For example, a small community hospital may have fewer resources available to manage a multi-vehicle accident involving multiple victims than a tertiary care referral center. What constitutes a disaster for the former may in fact be routinely managed by the latter. In this way, a mass casualty incident can be distinguished from a multiple casualty incident by virtue of the former, either by the number of patients or by the nature of their injuries, exceeding the capability of the facility or responding services to adequately render care to the victims [8]. In contrast to these more subjective definitions, prehospital emergency medical services frequently define a mass casualty incident as an event involving three or more patients, or two or more responding ambulances [9]. Alternatively, a tiered system based on number of reported victims 751 UNSTABLE ETHICAL PLATEAUS AND DISASTER TRIAGE has been used by larger municipalities [10]. The advantage of this predefined approach is that responders can concentrate immediately on patient care, rather than on determining available resources. Most disasters are brief, self-limited events that preserve the community infrastructure. In other words, the ability to care for patients remains intact, albeit stressed by the rate of casualty presentation (Fig. 1) [11,12].

One could further define a catastrophe or a catastrophic disaster as a circumstance where the destruction is so overwhelming that the infrastructure itself collapses, or as an ‘‘utter failure’’ [4]. Such a term was applied during the cold war to the results of a potential nuclear conflagration [13–15]. In more recent history, initial reports in the lay press after the 2004 South Asia tsunami indicated that all 569 medical facilities in Sri Lanka were destroyed [16]. Under such circumstances, even external assistance may be nonexistent, and care for the injured may not be an initial priority. Moreover, care for such victims may not occur in the traditional health care setting [11,14,15]. Regardless of the definition, many questions arise concerning appropriate medical care during disasters. How should we allocate resources and how do we justify these decisions? Will we treat patients with the most serious illnesses and injuries first? Will we treat on a first-come, first-served basis? Will we treat on the basis of individual ‘‘worth’’ to society? Will we treat the most people with the available resources, allowing that otherwise salvageable individuals will be allowed to die? The purpose of this article is to review the nature of disasters from an immediate medical need standpoint, and to place currently proposed utilitarian triage schema for prioritizing the medical care of surviving disaster victims into an ethics framework. The goal is not necessarily to provide definitive answers, but rather to raise 35 Number of Patients Daily 30 MCI 25 20 15 10 5 0 1 3 5 7 9 11 13 15 17 19 21 23 Time of Day Fig. 1. Patient encounter compression during a mass casualty incident. The solid line (Daily) represents the average number of patients per hour registering to be seen in an emergency department operating under principles of daily triage. The dashed line (MCI) represents a hypothetical scenario in which an additional 30 patients present to the emergency department after a self-limited mass casualty incident. The result is a sudden spike in patient encounters over a brief period of time. 752 SZTAJNKRYCER et al questions and encourage debate. The primary question the authors pose is whether disaster triage, as currently advocated and practiced in the western world, is actually ethical. Triage in the disaster setting The term triage, derived from the French verb trier (to sort), refers to the rapid sorting and prioritizing of patients. It is the first principle in mass casualty and disaster care [14]. The origin of modern triage is rooted in military medicine, and the work of Baron Dominique Jean Larrey, surgeon general to Napoleon’s Army of the Rhine [17–19]. Larrey first implemented a system in which the wounded were prioritized for care based on need rather than rank. Public Health Policy and Disaster Performance Discussion

Moreover, initial care of the wounded occurred while on the battlefield, before transportation to a site of definitive care. In 1846, British Naval Surgeon John Wilson developed the principles of modern mass casualty triage [18]. He postulated that in order for medical care to be truly life saving, it needed to be provided to those most in need. As a consequence, care was withheld from those for whom it was either futile or could be delayed until a later time, thereby forming the first tiered triage system. This concept in casualty care, and the term triage, was introduced subsequently to United States physicians serving in Europe in World War 1 [19,20]. In modern medical usage, the term triage more commonly refers to the concept of daily triage. From an emergency medicine perspective, this represents the prioritization of patient care during periods of the day when emergency department resources (typically beds) are scarce. It must be noted that no form of triage is needed if there is no resource limitation. Depending on the triage acuity system, patients’ injuries or illnesses may range from nonurgent to life threatening [21–24]. The purpose of daily triage is to identify the latter, so as to rapidly expedite care for those most in acute need. The highest level of care is provided to these patients, even if they have a low probability of survival. A classic example would be the patient presenting in blunt traumatic arrest. In contrast to daily triage, mass casualty or disaster triage refers to a system that occurs when available resources are insufficient to provide for the needs of all patients. Accurate casualty triage is viewed as the most important initial medical function during a mass casualty event [14,25]. It may be divided further into primary, secondary, and tertiary triage.

Primary and secondary triage constitute field or prehospital triage. Primary triage occurs during the initial assessment of victims at the disaster site. Although many systems have been proposed for primary triage, the most commonly encountered in the United States is the Simple Triage and Rapid Treatment (START) system (Fig. 2) [17,18,26–32]. Regardless of the system used, patients are stratified typically into one of four categories, based on physiologic parameters (Table 1) (see Refs. 753 UNSTABLE ETHICAL PLATEAUS AND DISASTER TRIAGE Begin Triage Yes Ambulatory? GREEN No Breathing? No Open Airway Yes RR > 30 Yes No RED No Breathing? BLACK Radial Pulse Yes RED Absent Present RED Follows Simple Commands? No RED Yes YELLOW Fig. 2. START mass casualty triage algorithm. The START algorithm categorizes individuals into four categories, based on physiologic parameters and ambulatory status. RR, respiratory rate. [12,18,26,28,30,33]). Using START terminology, ambulatory individuals are classified as GREEN. Individuals who have sustained injuries deemed not compatible with life under current resource constraints are triaged as expectant (BLACK). Immediate (RED) casualties have abnormal respiratory, perfusion, or mentation status, but are deemed as salvageable by the triage officer (see Fig. 2). Delayed individuals (YELLOW) are unable to ambulate but have normal physiologic parameters. Public Health Policy and Disaster Performance Discussion

Although they have the potential for decompensation, they require no immediate life-sustaining interventions, and treatment can be delayed typically for 4 to 6 hours [33]. Table 1 Prioritization of mass casualty victim medical need according to triage methodology Triage methodology Patient category Time to treatment START Triage SIEVE Homebush triage John Wilson (1846) Immediate need Urgent need Delayed/ minor need Dying/ expectant Within 1 h I (RED) I I (RED) I 4–6 h PRN II (YELLOW) III (GREEN) II III II (GOLD) III (GREEN) II II PRN dependent on resources N/A IV (BLACK) N/A IV (WHITE) III IV (BLACK) IV V (BLACK) III Deceased Abbreviations: N/A, not considered in the context of the triage scheme; PRN, pro re nata (as needed). 754 SZTAJNKRYCER et al Secondary triage occurs either in the field, at the site of the disaster, or at the casualty collection point, and, as such, is a field triage schema. Depending on the speed with which victims are transported to sites of definitive care, secondary triage may not be needed. Primary triage has been well described in the literature, whereas much less attention has been placed on secondary triage. Typically, such patients are either simply rescreened using a primary triage tool to assess for changes in condition, or triaged according to published trauma center referral guidelines [8].

Two dedicated secondary triage schema have been developed, Secondary Assessment of Victim Endpoint (SAVE) and Triage Sort [31,34], but have not been studied or used widely. The SAVE system was developed to identify those patients at a casualty collection point most likely to benefit from care under austere field conditions [31]. The key assumption in using SAVE is that transportation from the casualty collection point will be delayed and, therefore, definitive care is not immediately available. SAVE might therefore best be applied to catastrophic disaster circumstances. Triage Sort, as practiced in the United Kingdom, uses the revised trauma score to prioritize patient care, and assumes that care evacuation will be immediate [27,34,35]. Tertiary triage refers to patient sorting at the receiving hospital. Classically, this triage determines need for immediate resuscitation, immediate operative intervention, or intensive care unit admission, and then for admission versus discharge. The remainder of this article focuses primarily on the implications and consequences of field triage in the disaster setting. Western medical ethical principles and principles of justice Ethics, from the Greek ethike (character), may be viewed as the discipline of virtue and righteous action, an attempt to define moral principles and thereby resolve moral dilemmas [6,36,37]. As a consequence, ethics reflects the underlying moral values of society, and therefore is not absolute [38]. For example, in the context of medical ethics, if society views no moral right or value for health care, then subsequent ethical issues may be moot [39]. For the purposes of this article, the authors discuss ethics in the context of modern western society, understanding therefore that this work is not universally applicable [40]. Within western society, the concept of an underlying right to life has been upheld for several hundred years [41,42]. Although a principle of right to life may exist, the right to life in itself does not imply right to a specified length of life, nor does it imply a right or claim on resources, especially when such resources are scarce [41]. The fundamental principles of western medical ethics are respect for patient autonomy, beneficence, and nonmalfeasance [33,43–46]. A fourth principle of medical ethics is distributive justice, henceforth referred to as justice [45], which refers to the fair distribution of scarce or limited resources. The basic principle, described by Aristotle as the principle of formal justice, is that equals should be treated equally, and unequals, unequally [44]. UNSTABLE ETHICAL PLATEAUS AND DISASTER TRIAGE 755 Commonly, four ethical theories are applied to problems of justice: utilitarian, egalitarian, libertarian, and communitarian [44,47]. Utilitarian theory derives from the works of Bentham and subsequent philosophers and economists, especially Mill [48]. The underlying principle, as described by Mill in 1861, is the greatest happiness principle [48,49]. The concept is that, ‘‘Actions are right in proportion as they tend to promote happiness.’’ The goal is actually the greatest total happiness, not necessarily the greatest happiness for the greatest number, as is often stated [44]. The most just decisions increase net usefulness to society, by both maximizing societal benefit and minimizing harm.

The value of individual freedom is sacrificed for the common good. This ethical concept may be reflected in such medical terms as quality-adjusted life years and disability-adjusted life years. Central to egalitarian theory is the concept of an equal distribution of scarce goods [48]. The theory may be refined further into strong and maximin egalitarianism. Strong egalitarianism refers to all individuals receiving an identical share. In contrast, the maximin principle, advanced by Rawls [50], accepts inequalities, provided that they benefit those worst off or that it is no longer possible to improve the lot of those worst off. The egalitarian theory articulates the concepts of fairness, equality, and opportunity in defining the ideals of a just community. Libertarian theory may best be viewed as analogous to an economic free market concept [44,47]. The basic theory, with respect to medical ethics, postulates that the individual is the best judge of his/her own health needs. Priorities are self-determined, and reflect a capacity to pay. Public Health Policy and Disaster Performance Discussion

Lastly, communitarian theory holds forth the concept of community standards that attempt to define virtue and good within the context of cultural traditions and society [47]. The general focus is toward creating a ‘‘good’’ society, and as a consequence, of selecting the right individuals to further this goal. A central conflict of this theory is who defines the concepts of ‘‘good’’ and ‘‘virtue.’’ The relativist view is that each community defines its own norms within the societal context [51]. In contrast, the universalist view believes in a global concept of a good society, regardless of local cultural beliefs. A nonmedical example of the universalist view would be the importance of female education and literacy, regardless of societal context and belief. Ethical basis of current disaster triage schema Triage may be viewed as the means by which scarce medical resources are allocated. Unfortunately, most ethical debates concerning the triage of scarce medical resources refer to nonemergent circumstances, such as the allocation of scarce intensive care unit beds or solid organs for transplantation, or public access to hemodialysis [41,46,52–56]. Although there is an underlying concept of need in each of these cases, triage decisions affecting survival measure life span in days to months [41]. In contrast, in the disaster setting, death may be imminent, and measured in minutes to hours. 756 SZTAJNKRYCER et al As a consequence, the traditional ethical decisions related to triage may not be applicable to the disaster setting [56].

The current concept of mass casualty triage, as developed from the work of Wilson, is clearly and unequivocally based on the concept of utilitarianism [18,56]. ‘‘Do as little as possible, for as many as possible, as quickly as possible’’ [57]. The intent is to maximize casualty survival, and therefore benefit society as a whole, at the expense of individual needs. The question is whether this is fair and acceptable in modern society. In essence, first responders who are typically sworn to protect and uphold the s … Public Health Policy and Disaster Performance Discussion