Introduction: The Scope and Nature of the Problem
A disaster is an unplanned event in which the needs of the affected community outweigh the available resources. A disaster occurs somewhere in the world almost daily, but these events vary considerably in scope, size, and context. Large-scale disasters with numerous casualties are relatively unusual events. Certain widely publicized disasters, including events such as the terrorist attacks on September 11, 2001, Hurricanes Katrina and Sandy, and the Boston Marathon bombing, have focused people’s attention on disaster planning and preparedness. Disasters are becoming more frequent, and the number of persons affected is also increasing. This greater morbidity is attributable not only to the greater number of events, but also to population dynamics, location, and susceptibilities.
While these incidents have led to an increase in general disaster awareness, the relative infrequency of major catastrophes affecting defined populations leads to a certain degree of complacency and underestimation of the impact of such an event. In the wake of a large-scale event, public attention focuses on disaster planning and preparedness and the resources dedicated to improving response and resiliency surge. This phenomenon has been described as "issue salience" and has then been noted to decline over time as communities return to their quotidian lives.  The result of complacency is relative reluctance to devote the necessary resources for adequate disaster preparedness. Understanding the effects of a serious event on the community may indicate that the best time to propose major changes for disaster preparedness, including funding, is immediately following a widely publicized disaster, even if the event has occurred in a remote location.
In the United States, large, multiple-casualty events have been rare. Only 10 disasters in US history have resulted in more than 1,000 casualties (see Table 1). The vast majority of events have resulted in fewer than 40 fatalities. According to data from the US Centers for Disease Control and Prevention (CDC), the September 11 attacks caused 2,819 deaths.  This number is a fraction of the 44,065 deaths from motor vehicle accidents in the United States in 2002. The dramatic nature of disasters, however, with a relatively high death toll and psychological impact for a short period, can overwhelm an unprepared health and emergency response system and create chaos in the affected community and surrounding regions.
Table 1. US Disasters With Greater Than 1,000 Casualties* (Open Table in a new window)
|1875||Forest fire, Wisconsin||1,182|
|1906||San Francisco earthquake||>3,000|
|1941||Pearl Harbor Attack||2,403|
|2001||September 11 Attack||2,819|
*Exact death tolls can be difficult to calculate, and some of these numbers are estimates.
**Even in modern times, death tolls can be difficult to establish. Debate still exists about the actual number of people who died during Hurricane Katrina and its aftermath. An article by Brunkard et al published in Disaster Medicine and Public Health Preparedness in August 2008, puts the Louisiana death toll at 971 plus another 15 deaths among evacuees.  An Associated Press article from 2006 claims the total number of bodies recovered from Louisiana and Mississippi was more than 1300.  There is also ongoing investigation into the possibility that the storm caused deaths during subsequent months and years due to myriad causes (eg, inadequate medical care, relocation stresses).
When a disaster strikes, the general population expects public service agencies, emergency response agencies, and other branches of the local, state, or federal government to rapidly mobilize to help the injured and the broader community in general. Preservation of life and health are of paramount importance to those individuals injured in disasters. For this, among other reasons, medical professionals must be included in all phases of disaster planning as well as immediate response to these events. Medical professionals have a unique expertise and knowledge of local and regional healthcare systems, which can assist in disaster mitigation and planning. Persons affected by a disaster also rely on their skills and treatment in the immediate response to an event. Adequate preparation and planning, as well as provider and agency education and training, have become more relevant following the visibly problematic responses to many events, including the response to Hurricane Katrina.
Types of disasters (natural vs technological)
Disasters are classified in a variety of ways. A common system divides incidents into natural and technological (human-made) disasters. For planning purposes, this distinction is of little conceptual help, as there are frequent crossovers. Natural and technological disasters are often intertwined, as in the case of a building collapse as the result of a hurricane or earthquake. One example is the emergency personnel working in the areas affected by Hurricane Katrina, who had to contend with structural fires while rescuing people from flooded areas. Human actions can increase the likelihood of certain types of disasters, such as deforestation leading to landslides, or their impact on the population.
In addition, most disasters require similar types of organization and emergency and rescue personnel. Thus, an all-hazards approach that includes both natural and technological disasters may be the most efficient system to manage incidents. An all-hazards approach entails developing general disaster plans that can be applicable to any type of event that occurs, as opposed to plans written for each specific type of possible incident.
Certain generalizations may be made about natural disasters. [5, 6] Tornadoes may be quite lethal but are generally short-lived. Hurricanes cut a wider swath than tornadoes, tend to last longer, and have more long-term recovery effects. They are, however, more predictable than other types of disasters. Wildfires may persist for months and cause significant long-term damage. Volcanoes lead to a high number of fatalities but have become more predictable in recent years. Many natural disaster events tend to occur in certain geographical locations, and advanced methodologies have lead to earlier prediction and warning.
One of the most devastating natural phenomena, with regard to the numbers of fatalities, may be earthquakes. Earthquakes tend to remain unpredictable, and populations have no time to evacuate or prepare for an impending event. In addition, local healthcare structures and hospitals probably will be affected by the earthquake. Specialized training and operations plans have been advocated to reduce morbidity and mortality in earthquake-prone regions. 
Advanced warning systems, structural and design improvements, and disaster planning may decrease the devastation caused by many natural disasters. As populations occupy and develop areas that are at greater risk of specific types of natural disasters, however, the human and economic impact of these incidents is still likely to rise.
Technological disasters tend to be more contained but can also deliver a significant impact on life and property. Structural fires have caused some of the largest numbers of casualties in this country. Toxic spills and nuclear mishaps have caused mortality and serious injuries. Major transportation accidents such as train derailments and airplane crashes may quickly overwhelm the existing, local emergency response system.
Other incidents with the potential for mass casualties include war and terrorism. Since the 9/11 attacks on the World Trade Center and Pentagon, terrorism has become a major focus of disaster response and preparedness. Even incidents that have a smaller number of casualties can cause considerable social disruption and significantly stress the healthcare network.
Bombings and blast injuries are increasing in frequency with larger numbers of injuries and fatalities. Blasts also have the potential to involve radiological dispersion devices, or so-called "dirty bombs." Chemical weapons have emerged as a serious potential threat, along with biological agents. Although the world has yet to experience a terrorist-related nuclear disaster, the potential for such an incident exists. No geographical location is immune from the devastating effects of terrorism. These activities have become more frequent and lethal in recent years and offer no forewarning, as evidenced by the 9/11 attacks and Boston Marathon bombings.
Disasters are often classified by the resultant anticipated necessary response. A level I disaster is one in which local emergency response personnel and organizations are able to contain and effectively deal with the disaster and its aftermath. A level II disaster requires regional efforts and mutual aid from surrounding communities. A level III disaster is of such a magnitude that local and regional assets are overwhelmed, requiring statewide or federal assistance.
This classification reflects a tiered response, which is a fundamental principle of the National Response Framework, a component of national disaster response planning in the United States. Disasters are managed locally, and, as additional assistance and resources are required, higher levels of government and non-government resources become involved.
Definitions and Terminology
Because of the varying contexts and environments in which disasters occur, as well as the number and type of resources available, disaster medicine can be difficult to conceptualize. Typically, disasters and disaster medicine are defined in broad terms. The World Health Organization defines disaster as a "sudden ecological phenomenon of sufficient magnitude to require external assistance." The American College of Emergency Physicians (ACEP) states that a disaster has occurred "when the destructive effects of natural or man-made forces overwhelm the ability of a given area or community to meet the demand for health care." Other definitions exist, but the common denominator calls for a disruption of such magnitude that the organization, infrastructure, and resources of a community are unable to return to normal operations following the event without outside assistance.
To further clarify the contrast between normal emergencies and disasters, the ACEP states "emergency medical services routinely direct maximal resources to a small number of individuals, while disaster medical services are designed to direct limited resources to the greatest number of individuals." This shift in priorities may represent a challenge for emergency services and physicians who are accustomed to dedicating all available resources to the most critical patient. During disasters, damage control procedures may be more valuable than definitive diagnostics and treatments. While a number of rapid triage systems exist, physicians and emergency healthcare workers may be unfamiliar with how to prioritize care during a rapid surge of potentially critical patients.
In contrast to disasters, multiple or mass casualty incidents have as their primary effects morbidity and mortality to individuals, while the community infrastructure remains relatively intact. A passenger train accident with 500 injured or dead occupants is considered a mass-casualty incident. However, if this morbidity and mortality were the result of the release of chlorine gas from a hazardous material accident, a much higher potential for additional casualties would exist. Normal operations and activities of daily living would be disrupted for a longer period, which would be considered a disaster by most experts.
Events involving multiple patients may be described using a number of different terms. A multiple-patient incident is an event involving several patients that can be managed by available, local resources, as opposed to a mass casualty incident that begins to overwhelm local resources. Depending on the context and the environment, such terminology varies and can be inconsistent. Various methods have been developed to assist planners in disaster preparation. [8, 9]
The potential injury-creating event (PICE) system was designed to standardize nomenclature in discussing the potential scope of disasters.  The PICE methodology is a modification of the Injury Severity Score. The modifiers include the potential for additional casualties (A), the degree to which local resources are disrupted (B), the geographical boundaries of involvement (C), and crisis staging, or the likelihood of needing outside assistance to augment local resources (see Table 2). The PICE system is designed to identify common aspects of a disaster and of response capabilities. Such systems may be valuable tools in planning for disasters. However, the PICE system has yet to be widely adopted by many professionals in the field of disaster medicine and management. 
Table 2. Potential Injury-Creating Event Algorithm (Open Table in a new window)
Phases of Disaster Response
A disaster cycle has 4 phases, and all events progress through each: (1) mitigation, (2) planning, (3) response, and (4) recovery. Pitfalls during transitions can occur throughout the phases. Some authorities may also include a fifth phase, prevention. Inordinate focus may be placed on certain aspects of the cycle, such as response, to the detriment of other phases, which may increase the harmful impact of events. Generalized awareness, proper planning, and contingencies may reduce the overall effect of any specific inadequacy of response.
In certain cases, some of the devastating effects of disasters can be reduced before the actual event. For example, evacuations may be orchestrated before hurricanes and floods. Early warning allows residents to seek shelter from tornadoes and approaching hurricanes. Sprinkler systems in business and homes can reduce overall risk of total fire destruction, and specific types of building construction may be implemented in hurricane- or earthquake-prone regions. While this phase of a disaster is often overlooked due to the fact that an impending incident has not raised an awareness of the need for mitigation, it may have a significant effect on the impact that a disaster will have on a region and must remain an important consideration.
Disaster planning is more thoroughly discussed in the section External and Internal Planning. It cannot be stressed enough, however, that a disaster plan is not synonymous with disaster planning. Many communities have detailed, "paper" plans, which, when tested, are found to be either based on faulty assumptions or to be totally unworkable in the context of an initial response. In addition to having a specific, written plan, disaster planning involves exercising, practicing, and revising the plan, including community and emergency resources, which will likely need to coordinate during an actual event.
The response phase of the disaster cycle tends to gather the most attention of all of the disaster phases. An effective, coordinated response, however, depends on the other aspects of the disaster cycle. A number of events occur during the initial response to a disaster. If there is forewarning, certain portions of the response may take place even before the event. Unfortunately, significant forewarning is rare.
Notification and initial response
During this phase, organizations involved in disaster response and the potentially affected populations are notified. In the event that the disaster is anticipated, this phase takes place even before the disaster. Many locations in hurricane areas, for example, require more than 24 hours for full evacuation. Certain types of incidents may be better managed by having the community shelter-in-place, as opposed to attempting evacuation. Again, planning plays a significant role in how the actual activation and notification unfold.
Organization of command and scene assessment
Once the activation phase has begun, the prearranged command staff and structure (for details, see Incident command system in External and Internal Planning) for responding to the disaster should be assembled and initial communication nets established. The incident command system (ICS) is an organizational and management tool used during disaster situations and emergency response operations. Establishing a command structure is one of the most crucial steps to take once the disaster occurs.
Historically, valuable time may be lost during a disaster response while the central system coordinating the response effort is being prepared. During this phase, initial reports leading to overall scene assessment begin to arrive. For static disasters, required response assets may need to be determined. Often, the only initially known fact is that the disaster is an ongoing process. Even this fact, however, is important in determining whether outside assistance is needed, leading to timely activation of those resources. There may be a delay in obtaining outside resources, and early assessment of the incident and arrangements for arriving aid prevent long periods of waiting for their arrival.
Search and rescue
Depending on the structure and function of the ICS, search and rescue may fall under the direction of fire, emergency medical services (EMS), police or security forces, or even federal assets such as the Federal Emergency Management Agency’s (FEMA) Urban Search and Rescue teams. In contained, geographically localized incidents, the search and rescue effort is likely to be fairly straightforward. In larger disasters, especially ones that are ongoing or may involve terrorist activities, a cooperative approach is necessary and the very act of search and rescue must be highly organized to ensure adequate and complete coverage of all areas.
Extrication, triage, stabilization, and transport
Extrication has evolved into a fire services function in most of the country. In addition to specialized technical and trench rescue teams, fire services have more experience with building collapse and secondary hazards such as floods and fires than do other organizations.
The concept of triage involves sorting victims in order to determine who needs what type of assistance in a particular timeframe. Typically, triage involves providing the greatest good for as many as possible, although this may lead to profound ethical questions. It also temporally prioritizes treatment and transport. Medical personnel are accustomed to providing extensive, definitive care to every patient, and often to the most critically ill or injured first. When confronted by numerous patients simultaneously in a disaster situation, it is easy to become overwhelmed, even for an experienced disaster worker. The manner in which patients are triaged, transported, and treated depends on the type of incident, the number of patients, the available resources, existing infrastructure, and the overall context of the disaster. Triage must occur at multiple levels, and patients must be reassessed during every step of the process.
There are several algorithms for triage in mass casualty incidents that have been shown to have acceptable sensitivity and specificity in detecting severely injured patients. [12, 13] Most common triage algorithms use respirations, perfusion, and mental status in order to categorize the severity of victims' conditions and then assign color-coded tags to designate priority in treatment. At least one commonly used triage system, however, may over-triage more than half the patients.  Over-triage may cause misallocation of valuable resources in disasters and may lead to worse outcomes among those affected.  Most important is that scene workers and healthcare providers be familiar and train often with the system they are using.
In some scenarios, such as extensive earthquakes, infrastructure may be damaged or destroyed to the extent that definitive care, even from outside resources, is not available for several days. In this austere setting, dynamic treatment and recurrent triage of patients, such as that taught by the Medical Disaster Response training curriculum, should occur until other sources of medical care become available. 
Transport must be both organized and orchestrated to equitably distribute victims to capable receiving facilities. Many of the less critically injured will self-extricate and arrive at the nearest medical facility by their own means.  Often, the more critical patients arrive after the first wave of so-called "walking wounded." Most patients are not transported by EMS, so it is even more important to distribute these patients to appropriate receiving facilities with the capacity to care for them. This process lessens the overwhelming impact a disaster may have on the closest facility and likely improves the care medical personnel are able to provide to the victims.
The nature of the event dictates what type of treatment and transport the victims require. In the wake of a terrorist incident, the exact nature of the attack may be unknown. Victims may require decontamination prior to transport in order to prevent the spread of a hazardous material or threat. Receiving facilities must be made aware of any exposures and should have preestablished plans in place for decontaminating victims. In fact, hospitals may be the first point of contact for victims, especially those who self-extricate. 
Procedures for decontamination of chemical, biological, and radiological exposures must be identified and reviewed by healthcare facilities. Associated concerns, such as surveillance, control of runoff, isolation, and risk communications, should be addressed.  The concern for assessing and treating patients potentially infected with Ebola virus highlighted some aspects of infection control and education that could be improved in US hospitals and demonstrated the need for recurrent training and reassessment. In addition, healthcare providers who may be involved in decontamination require training in appropriate decontamination and personal protective equipment. Standardization of essential levels of personal protective equipment for staff on a national level may be beneficial. 
Decontamination is shown in the image below.
Definitive scene management
While scene control and containment may be relatively simple in a local, static disaster, dynamic disasters and those that paralyze the response systems may take several days or longer to contain and stabilize. As the length of time of the disaster increases, additional resources must be made available, as rescue crews reach exhaustion, supplies are depleted, and additional hazards develop.
The recovery phase is frequently underemphasized in disaster plans, but it is crucial for the affected community. During this phase, some semblance of order is restored, public utilities are reestablished, and infrastructure begins to operate effectively. Scene withdrawal and a return to normal operations usually occur simultaneously. Planning for withdrawal of resources should begin from the onset of deployment. Rebuilding and restructuring may include mitigation measures in order to prevent as severe of an impact during the next event. Treatment of the responders is also vitally important during this phase for critical stress debriefing and other support services that have evolved for this purpose.
Debriefing may teach planners valuable lessons. It is of utmost importance to obtain as much information as possible from all parties involved in the disaster response effort. Without full disclosure, similar pitfalls may impede future efforts. Debriefing aids in future mitigation and planning efforts and, in essence, closes the loop of the disaster cycle.
External and Internal Planning
To the extent possible, disaster planning should incorporate formal disaster research findings. Disaster plans sometimes rely on faulty assumptions that do not prove true in actual disasters. For example, planners may logically assume that the sickest patients are transported first during a disaster, when in reality, this may not happen in many instances. Other realities that have been noted in prior events are that the initial search and rescue begins with victims and bystanders and not trained rescue teams, the majority of patients arrive to hospitals without use of the EMS system and have not been triaged or decontaminated, and there is often lack of communication between healthcare facilities and the scene.  These characteristics are likely to occur in future disasters and should be incorporated into realistic disaster plans.
A disaster plan encompassing both local and regional areas must focus on the following 3 possible scenarios:
The disaster occurs within the region and is confined and controlled with area resources.
The disaster occurs in a neighboring region and regional assets are requested through mutual aid agreements.
The disaster area is the region and requires state or federal assistance for an effective response.
In developing a disaster plan, leaders should remember that it is impossible to plan for all contingencies; therefore, plans must be relatively general and expandable. Most disasters that can be contained using local or regional resources have fewer than 100 fatalities and fewer than 500 casualties. Plans developed for larger-scale disasters should focus on the first 48 hours of the disaster until state and federal assistance teams can arrive and the need to address high initial fatality rates during the first 24 hours. Mutual aid agreements or contracts with other existing area resources also are important to establish before an actual incident, as well as planning for funding and reimbursement.
Incident command system
After a series of forest fires in California in the 1970s, the Fire Fighting Resources of Southern California Organized for Potential Emergencies (FIRESCOPE) developed the ICS concept to organize an effective response to major emergencies. Despite adequate resources, loss of life and property due to the fires was substantial. This fact was attributable to poor overall management of the incidents. Some of the myriad flaws included multiple organizations working together at incidents who had not previously trained together, disparate organizational structures that made effective leadership challenging, an unwieldy span of control, poor coordination and resource allocation, and breakdown of intra-agency communication. [19, 20] ICS was designed to minimize these issues and has gained widespread use in the emergency management community.
The National Incident Management System (NIMS), which is a national guideline for managing emergency incidents, includes ICS as a critical component. Compliance with NIMS is a requirement for federal disaster preparedness assistance under Homeland Security Presidential Directive 5 since 2004. The ICS structure includes 5 functional units: command, operations, logistics, planning, and finance (see the image below). Most emergency operations plans include similar organizational structures that are often modified depending on normal operations. Some of the attributes of ICS are its modular structure, scalability, use of common terminology, clear chain of command, and manageable span of control.
Patients and hospitals require personnel, supplies, facilities, and an incident management structure. Hospital emergency planners must take into account the scenarios previously described, including the possibility that the disaster may involve the hospital. For such rare events, aspects of hospital involvement such as mass decontamination, multiple triage and staging areas within the confines of the hospital, recall of critical personnel, and provisioning of adequate supplies and resupply must be anticipated. The ability for a healthcare facility to quickly expand its ability to treat a large influx of patients is referred to as surge capacity. In order to accommodate a surge in patients, hospitals require personnel, supplies, facilities, and an incident management structure.  A goal capacity that has been advocated is to expand a hospital by 20-25% within the first 24 hours.
The Joint Commission (formerly Joint Commission on Accreditation of Hospitals [JCAHO]) requires hospitals to exercise emergency operations plans periodically and to form emergency management committees. These committees should comprise key departments within the hospital, including administration, nursing, security, communications, laboratory, physician services (including but not limited to emergency medicine, general surgery, and radiology), medical records, maintenance, and engineering. Hospital emergency planners should plan using the Hospital Incident Command System (HICS), which is a modified version of the ICS and maintains many of its attributes.
The hospital emergency operations plan should include protocols and policies that meet the following needs:
Recognition and notification
Assessment of hospital capabilities
Personnel recall and emergency credentialing
Establishment of a facility control center
Maintenance of accurate records
New, more stringent requirements for health care organizations by the Joint Commission include the requirements to integrate hospital disaster planning into community plans, to ensure that disaster programs address all phases of the disaster cycle, and to have the capability to evacuate the entire hospital staff and patients and relocate and operate from an independent facility. [22, 23] Evacuations of hospitals after the Northridge, California earthquake in 1994 demonstrated that large numbers of patients could be evacuated without the use of specialized equipment and in an efficient and safe manner. 
Requirements concerning decontamination, policies, and training in response to terrorist activities involving chemical, biological, radiological (see image below), nuclear, and explosive agents may be strengthened in the future in response to concerns over these potential threats.
Disaster planning must be a cooperative effort. Every jurisdiction should plan for mass casualty incidents and disasters. All plans must be simple and based on normal daily operations of the various components involved in the emergency operations plan. Personnel potentially involved must be familiar with the emergency operations plan. It should be exercised frequently, even if only by tabletop or functional exercises. Contingency plans for mutual assistance and state or federal response also must be considered and reviewed. Planning is one important phase in the disaster cycle that ultimately influences the impact of a disaster on a community.