7

Disaster Resilience and Animal Research Programs

The purpose of this chapter is to inform the reader about some of the special considerations that arise when vertebrate animals used in research are present during a disaster, to examine the current status of requirements for disaster plans for animal research programs at academic research institutions, and to propose suggestions for strengthening resilience and improving comprehensive planning.

THE PRESENCE OF ANIMALS IN A DISASTER

Disasters are extreme events which elicit strong emotional reactions in both survivors and first responders, such as stress, grief, compassion, fatigue, and burnout (Ligenza et al., 2010). When animals are present during a disaster, humans who have strong attachment bonds with animals are reported to experience more intense emotional reactions and higher levels of stress (Hall et al., 2004).

It has been documented that the presence of animals during a disaster alters response and recovery operations following a disaster—so much so, in fact, that the Robert T. Stafford Disaster Relief and Emergency Assistance Act¹ was amended in 2006 to add the Pets Evacuation and Transportation Standards (PETS) Act² (Hall et al., 2004; FHWA, 2017). The PETS Act was enacted after analyses of the response to Hurricane Katrina found that human and animal fatalities were caused by the unwillingness of residents to abandon their pets and evacuate. Hurricane Katrina brought the importance of the human–animal bond during disasters into the national spotlight, and the PETS Act legitimized animal disaster response in the United States. Provision for the evacuation and sheltering of pets during a disaster is now a required function of local emergency management and government. Caring for both people and animals is an essential part of disaster response and recovery (FEMA, 2015a; FHWA, 2017; Hall et al., 2004). Today, the public expects animals to be rescued during disasters, and the academic biomedical research community cannot ignore this societal trend.

It is well recognized that some people who have close affiliations with animals are willing to place themselves in imminent danger to save the animals—by failing to evacuate, by making unauthorized reentry attempts into hazardous areas to retrieve animals, or by conducting unsafe rescue attempts (Hall et al., 2004; Heath and Linnabary, 2015). Interestingly, research has shown there does not have to be a specific attachment bond between the person and the animal that he or she is trying to save. Many people will attempt to rescue animals they do not even know and will even

___________________

¹ Robert T. Stafford Disaster Relief and Emergency Assistance Act of 1988, P.L. 93-288.

² Pets Evacuation and Transportation Act of 2006, P.L. 109-308.

attempt to rescue wild animals (FEMA, 2015a; Hall et al., 2004; Heath and Linnabary, 2015).

Similarly, the attachment bonds between research animals and the humans who care about them (e.g., researchers, veterinarians, and animal care staff) have been described as being identical to the bonds that develop between people and their pets. Indeed, people who are drawn to careers in the field of animal research are known to have great regard and compassion for animals (Bayne, 2002; Chang and Hart, 2002). Researchers, veterinarians, and animal care staff may experience strong emotional reactions during disasters and may place themselves in danger trying to rescue research animals. The outcome may be that unsafe actions are taken by untrained rescuers. This could interfere with the overall response operations and create a situation where the animal rescuers may have to be rescued.

Because animal ownership is so common in the United States, it is likely that many first responders, administrators, researchers, veterinarians, and animal care staff will be psychologically affected if there are abandoned, injured, or dead research animals in the disaster area (AVMA, 2012). Disaster survivors who have strong attachment bonds to animals warrant additional consideration because they have been subjected to the added psychological stresses that arise as an outcome of the human–animal bond in addition to the stresses associated with the disruptions of their daily lives (Heath, 2000; Ikeda, 2012). Sometimes the loss of animals is unavoidable; the disaster may have struck without warning and quickly killed large numbers of animals, an outbreak of an infectious disease may necessitate quick en masse depopulation, or euthanasia might have to be done for humane reasons. Severe emotional reactions are common in humans who witness massive animal casualties; these reactions have been reported to include profound guilt, shame, helplessness, failure, anger, and hostility (Hall et al., 2004). Severe states of anxiety and distress can also progress to more serious forms of mental illness. For example, increased rates of depression were reported in individuals who witnessed the depopulation of food animals during the outbreak of foot and mouth disease in Great Britain and the Netherlands (Mort et al., 2005; Van Haaften et al., 2004). Similar stress reactions have been anecdotally reported in animal research care staff (Ikeda, 2012). To help researchers, veterinarians, and animal care staff work through their grief and anxiety over the loss of their research animals, memorial services have been conducted to honor the memory of the animals and acknowledge their loss (Goodwin and Donaho, 2010; Ikeda, 2012).

REGULATORY REQUIREMENTS AND GUIDELINES

The Animal Welfare Act (AWA) is a federal law that prescribes minimum standards for the housing, care, feeding, transportation, and use

of regulated vertebrate species in biomedical research.³ Institutions that receive any funding from the Public Health Service (PHS) (which includes funding from National Institutes of Health [NIH]) must also comply with the PHS Policy on Humane Care and Use of Laboratory Animals (PHS Policy), which uses the Guide for the Care and Use of Laboratory Animals (the Guide) as its standards document (NIH, 2015). The PHS Policy and the associated standards that are outlined in the Guide apply to all vertebrate animals that are used in research. The government agency that oversees the implementation of the PHS Policy at research institutions is the NIH Office of Laboratory Animal Welfare. Both the AWA and the PHS Policy require that each institution have an institutional animal care and use committee (IACUC) that oversees all animal use activities that take place at the institution. The institutional official (IO), the IACUC, and the attending veterinarian are charged with the responsibilities of monitoring the health and the welfare of the vertebrate animals that are used in research at the institution (NRC, 2011). In addition to this regulatory oversight, many institutions also participate in a voluntary accreditation program under the Association for Assessment and Accreditation of Laboratory Animal Care (AAALAC) International. AAALAC International evaluates the care and use of vertebrate animals and invertebrate animals when they are the subjects of the research that is being conducted at the institution (AAALAC International, 2017b). In the United States, AAALAC International uses the Guide as a basic standards document, but it also uses the Guide for the Care and Use of Agricultural Animals in Research and Teaching (AAALAC International, 2016).

The importance of disaster planning is reflected in the guidelines and regulations that govern the conduct of animal research in the United States. Every academic research institution that conducts animal research and receives federal funding from PHS must have a disaster plan as a required component of its animal research program (NRC, 2011). The components of the disaster plan, which are outlined in the Guide, give each animal research program the flexibility to decide what specific actions and procedures would be appropriate for its animal facilities. The components of animal research program disaster plan, as described in the Guide, are the following (NRC, 2011, p. 35):

• The disaster plan should describe the actions that will be taken to prevent pain, distress, and deaths due to loss of life-sustaining systems such as heating, ventilation, and air-conditioning (HVAC) and potable water.

___________________

³ Animal Welfare Act of 1966, P.L. 89-544.

• The disaster plan should, if possible, describe how the facility will preserve animals that are necessary for critical research activities or are irreplaceable.
• The disaster plan should be developed in conjunction with the responsible investigators, taking into consideration both the priorities for triaging animal populations and the institutional needs and resources. Animals that cannot be relocated and would be subjected to adverse consequences because of the disaster should be humanely euthanized.
• Responding personnel from the animal facility should be identified and designated as essential, should be trained in advance of a disaster, and should be given access to the site during or immediately after a disaster.
• The disaster plan for the animal facility should be part of an overall institutional disaster plan that is approved by senior administration.
• Local law enforcement and emergency personnel should be provided with a copy of the plan for comment and integration into broader, area-wide planning.

The components of the disaster plan discussed in the Guide are mainly centered on actions that should be taken during response, with some consideration of mitigation, and the areas of prevention, protection, and recovery, as discussed in Chapters 4–6, are not fully considered. Any disaster planning for animal research programs that refers only to the guidance stated in the Guide is not sufficient because it is focused mainly on response. This planning weakness was apparent, for example, in the disaster plan at the University of Texas Health Science Center at Houston (UTHSC-H) during Tropical Storm Allison. Despite there being a disaster plan in place, thousands of animals drowned in basement facilities (Goodwin and Donaho, 2010):

Furthermore, animal research programs at AAALAC International–accredited institutions must also have a disaster plan that follows the Guide (AAALAC International, 2017b). AAALAC International also requires accredited programs to describe the availability of standby power in each vivarium and to address any animal welfare concerns that might

occur during power outages. The disaster plan contents and the capacity of the animal research program to provide for animal well-being during power outages are evaluated as part of the accreditation process (AAALAC International, 2016, pp. 19, 37).

A second set of regulatory requirements for animal disaster plans was proposed by the Department of Agriculture’s Animal and Plant Health Inspection Service (USDA–APHIS) in 2012 as an amendment to the AWA, Handling of Animals; Contingency Plans (USDA–APHIS, 2012). The contingency plan was proposed because USDA–APHIS inspectors found that the level of emergency preparedness in the regulated community was low and that at least some of the animal welfare issues that arose during disasters could be mitigated with better preparation (USDA–APHIS, 2012). Each AWA-regulated animal facility was to prepare a contingency plan that contained four basic elements:

• It addressed both common emergencies (such as structural fire, HVAC failure, power outages, and animal escapes) and the relevant natural disasters (tornado, flood, earthquake, etc.) that were likely to occur;
• It included both an evacuation plan and a shelter-in-place plan to manage animals during emergencies and disasters, and it included a plan for the recapture of escaped animals;
• It defined the chain of command for implementing the contingency plan; and
• It included the materials and resources needed for response and recovery on hand or else had a plan to obtain these.

The target date for full implementation of the contingency plan rule was September 2013 (USDA–APHIS, 2013). However, objections from the regulated community resulted in the issuance of an indefinite stay on July 31, 2013.

Currently, the Guide remains the primary guidance document addressing disaster planning for the animal research community. The Guide includes the disaster plan for the animal research program as a part of the institution’s overall animal care and use program (NRC, 2011). The responsibility for determining the adequacy of the animal disaster plan thus falls to the IACUC, the attending veterinarian, and the IO, who are tasked with reviewing it during their required semiannual evaluations of the animal care and use program. In the committee’s experiences it is common for IACUC members, attending veterinarians, and IOs to have no formal education or training in emergency management or disaster planning. Relying solely on the information in the Guide to develop a disaster plan will not achieve resilience in animal research programs at academic research institutions

because the Guide does not direct the planners for the animal research program to use a comprehensive planning approach that is congruent with the National Preparedness System (see Chapters 4–6). The suggestions for developing a disaster plan in the Guide focus primarily on response, and according to animal disaster planning expert Sebastian Heath:

FACTORS THAT AFFECT THE SUCCESS OF PLANS

The Complexity of the Research Animal Housing Environment

Several factors make animal research particularly vulnerable to loss during disasters. Many animal models are unique and irreplaceable (e.g., laboratory-generated genetically modified rodents and fishes). Experiments that use animals are often conducted over a long period of time, and disruption can set a research project back by months or even years (see Chapter 2). The catastrophic loss of research animals could terminate all future research in a specific area of scientific inquiry. Research animals impose both ethical and practical considerations during disasters because caged animals cannot voluntarily escape and save themselves, and they remain totally dependent upon humans for their survival. Environmental conditions, such as ventilation and ambient room temperature, must be able to be maintained within reasonable limits at all times (NRC, 2011; Animal Welfare Act⁴).

Experiment-specific hazards also are common in academic animal research programs. Some experiments involve radioactivity, biohazards, or chemical carcinogens and therefore require special procedures for isolating and handling animals and for disposing of carcasses (Swearengen et al., 2010). Animal research may also be conducted under conditions where increased security is necessary (e.g., research that uses select agents or a cesium irradiation source).

With the emergence of genetically modified rodents and fishes as powerful research tools, the number of these species being used in research has increased exponentially. To accommodate this increase, animal research programs have had to rapidly develop the capacity to house them (Ormandy, 2012). For rodents, this has led to the widespread use of individually ventilated caging systems (IVCs) as a practical way to increase housing capacity and provide comfortable and healthy living conditions as

___________________

⁴ The Animal Welfare Act of 1966, P.L. 89-544.

well as to maintain safe environmental working conditions for staff and to control the labor costs associated with the care (Lipman, 2009). These cages supply air to each individual cage in either a positive- or negative-pressure condition, depending upon the experimental need. The air supply to the cage may be dependent upon a rack-mounted ventilating unit powered by electricity or the building’s HVAC system. With some IVCs, the mechanical ventilation system that supplies filtered air to each cage will not operate without electric power and the air quality can rapidly deteriorate to fatal conditions (Höglund and Renström, 2001). For example, mice experienced severe hypoxia within 6 hours and rats died within 1 hour of loss of the mechanical ventilation systems that supplied air to their cages (Huerkamp et al., 2003; Nagamine et al., 2012). When IVCs are in use and do not have a fail-safe design,⁵ the rodents are placed at risk if a power interruption or outage occurs and the emergency power supply stops working. However, not all IVCs are unsafe during a power interruption or outage, and several manufacturers advertise that some of their cage designs are fail-safe (Lab Products, 2017; Phoenix Controls Corporation, 2002).

Similar power-dependent housing systems are also used for fish. The ability to oxygenate water and supply it to the tanks is dependent upon the system having a constant supply of power (Helfrich and Libey, 2013; Lawrence and Mason, 2012). Large volumes of water are preconditioned to achieve the desired levels of oxygenation and salinity, and the water is filtered at multiple levels before being pumped and circulated into individual fish tanks. The racks that hold the tanks containing the fish are commonly bolted to the floor of the room, potentially making them immoveable.

For both the IVCs and fish housing systems, the dependency upon a power source and the lack of a fail-safe design can lead to problems when a disaster occurs.

Consolidation of Research Animals into Centralized Facilities

In the past, research animals were often spread out and housed among smaller research facilities that were in close proximity to their principal investigators’ laboratories. Today it has become more common for the majority of an academic research institution’s research animals to be housed

___________________

⁵ The general use of the term “fail-safe design” usually suggests that building or other engineered systems incorporate components or processes that will mitigate losses caused by system or component failures. The design assumption is that if failure occurs (e.g., in the air-handling system, the environmental cooling system for animal care, small-animal housing equipment, electricity, or water supply systems to the vivarium), then the device, system, or process will fail in a safe manner which will result in continued life support for the research animals. This fail-safe design also suggests a level of system redundancy; during the study process it was defined as N+1 or N+2.

within a single vivarium or a few large vivaria that are managed with centralized care and oversight (NRC, 2000, 2011). This consolidation has many advantages for the institution’s animal research program:

• Facilitates better oversight of the research animals’ health and welfare by the veterinary care staff.
• Improves the post-procedural care and the ability of the veterinary staff to provide timely care.
• Takes advantage of economies of scale, which improves efficiency and lowers the operating costs for the animal research program.

However, the downside of this approach is that larger numbers of research animals are congregated in fewer locations. For an academic research institution that has only a single, large vivarium, the institution is essentially “placing all of its eggs in one basket.” Should that single large vivarium be destroyed, all of the research animals are at risk of being lost. When an academic research institution chooses to place its research animals within centralized facilities that will house large numbers of animals, the planning actions must be impeccable and the facility design must be fail-safe (NRC, 2011).

The Prevailing Attitude That a Disaster Will Not Occur

The inherent belief that a disaster will not happen is a normal human feeling, and it can be a difficult hurdle to overcome when undertaking disaster resilience efforts (Kapucu, 2008). When institutional leadership and the individuals responsible for the animal research program think that a disaster will never befall them, they are more likely to view disaster resilience planning as a waste of time and resources (Mortell and Nicholls, 2013). This issue is compounded by the economic constraints that most academic research institutions operate under on a daily basis. When financial resources are already limited, the academic research institution often focuses its efforts on fixing the problems at hand. This mindset blocks rational discussions about the predictable impacts of a disaster and the dynamic planning that needs to take place proactively to ensure the continuity of operations (Mortell and Nicholls, 2013).

This belief can influence the quality of the disaster plan because the individuals responsible for executing the animal research program’s disaster plans are less likely to engage in adequate training and the testing of the disaster plans may be limited to a tabletop exercise—which will not reveal flaws in the response procedures when live animals and equipment must actually be mobilized (Mortell and Nicholls, 2013). The committee emphasizes that conducting periodic functional drills and participating in full-scale

exercises that mimic real disaster scenarios (when available) are essential to keeping the plan fresh in the minds of the individuals of the academic research institution who will be responsible for implementing it during a disaster (see Chapter 4). The focus of AAALAC International on training in and practicing on the disaster plans at accredited institutions during site visits is a positive step forward (AAALAC International, 2017a).

PLANNING FOR ANIMAL RESEARCH PROGRAMS

Prevention, Protection, and Mitigation Planning

Academic research institutions should plan to prevent the worst-case scenario: a catastrophic loss of research animals. As part of the disaster resilience planning undertaken by the research enterprise, strategies can be developed to enable the replacement of the animal models and the restarting of the research. To achieve this objective, institutions could consider identifying and prioritizing their animal models and then work with researchers to devise plans that would enable them to salvage their animal models should a catastrophic loss befall them (Donahue et al., 2012). This can be accomplished through several mechanisms. Critical breeding pairs can be kept in geographically separate facilities that are owned by the institution, animal models can be distributed to other researchers, animal models can be sent to federally funded research cores for redistribution and cryopreservation (e.g., the Mutant Mouse and Rat Research Centers), or germplasm can be preserved using onsite cryopreservation (MMRRC, 2017).

A critical, yet underappreciated mitigation strategy is to improve the design and construction of vivaria in research buildings (Vogelweid et al., 2009). Saving research animals is more likely if the vivarium is located in a safe area of the building, if relevant hazard-resistant features are incorporated into the building construction, and if there is a provision to maintain the essential utilities servicing the vivarium (NRC, 2000). To enhance the disaster resilience of the built environment of animal research facilities, animal research professionals and design professionals should adopt disaster resilience as a mindset at the outset of the planning process for a new research building. Guidance about resilience planning for the built environment is outlined in the National Institute of Standards and Technology (NIST) Community Resilience Planning Guide for Buildings and Infrastructure Systems (NIST Planning Guide) (NIST, 2015). Refer to Chapter 8 in this report for a more detailed discussion of design criteria for new vivaria and how the NIST Planning Guide can be used by academic research institutions.

There may be knowledge gaps among design professionals and animal research professionals with regard to the level of protection that is needed

in a building versus the level of protection that the building will provide. Animal research professionals might assume that a new building that meets the current building code standards will be safe and will retain some functionality after a disaster. Design professionals know that compliance with current building codes means that the building will not collapse until sufficient time has elapsed for the human occupants to evacuate during a predesignated hazard event of a specified magnitude (Vogelweid et al., 2009). On the other hand, design professionals do not always remember that placing vivaria in a research building means that there will be living, permanent occupants in that building that cannot voluntarily evacuate in an emergency. Many building code standards are designed for humans and do not consider animals’ safety. Both design professionals and animal research professionals may not fully appreciate that post-disaster access to the research animals in the building will be denied if the inspection of the building uncovers significant safety hazards for humans. The outcome then becomes forced abandonment of the research animals, with their resultant suffering and, potentially, deaths.

The unique requirements that arise when animals are placed inside structures that are shared with human occupants have been addressed in a standards document from the National Fire Protection Association, NFPA 150: Standard on Fire and Life Safety in Animal Housing Facilities (NFPA Standard 150) (NFPA, 2016). The NFPA stated in the explanatory material on NFPA Standard 150 that the following principles underpinned the development of this standard:

• Animals are sentient beings with a value greater than property.
• Animals, both domesticated and feral, lack the ability of self-preservation when housed in buildings and other structures.
• Current building, fire, and life safety codes do not address the life safety of animal occupants.

NFPA Standard 150 outlines recommendations for facility design and construction that will permit the rapid egress of animals should an evacuation become necessary, and it provides recommendations to construct fire-rated separations around animal housing areas to slow the spread of fire to permit time for evacuation, relocation, or defending in place (NFPA, 2016). NFPA Standard 150 also addresses disaster planning and emphasizes the importance of conducting effective exercises to ensure that the animals can be successfully moved during a real emergency.

Vivaria are typically dependent upon emergency power systems to maintain adequate living conditions for animals (Lipman, 2009; Stitch, 2013). Ensuring that emergency power systems have redundancy is important. The need to provide a continuous, uninterrupted power supply

to certain types of IVCs and fish housing systems can be thought of as being analogous to the need to provide a continuous, uninterrupted power supply to life-sustaining equipment that supports patients in a hospital. In hospitals, emergency power systems must be reliable and meet specific, standard design criteria (NFPA Standard 99) (NFPA, 2015). Similar to the case with patients on life-sustaining equipment, an uninterrupted power supply to IVCs and fish housing systems is often critical to the survival of the animals residing on those systems. For the emergency power system to remain operational after a disaster, the generators, electrical switchgear, and fuel must be protected from incapacitation during whatever disaster has occurred, such as having been inundated with floodwaters if the components are located in the basement of the building. The reliable performance of an emergency power system is also dependent upon periodic testing and careful maintenance. Hospitals are required to test their standby power systems every month (NFPA, 2015). The idea to build to a higher standard than life safety code was recommended by Vogelweid et al. (2009), who suggested that health care code requirements would afford research animals better protection during disasters. Animal research professionals should have a good understanding of a building’s mechanical systems and their redundancy.

There are also several common design errors that could affect the safety of animals in vivaria during disasters (Lehner and Hessler, 2009). For example, doorframes that are too short or too narrow to accommodate the cages and racks will impede an evacuation of the animals. Installing solenoid valves for animal room terminal reheats that fail in the open position may cause lethal overheating of the animal room when the HVAC malfunctions occur during a disaster. An inadequate emergency power supply will compromise a facility’s ability to maintain suitable environmental conditions in the animal rooms, and power system failure may cause rapid mortality of animals housed in some types of IVCs (Lehner and Hessler, 2009). To prevent these types of problems from occurring, the design team for a new animal research building must include knowledgeable animal research professionals and design professionals with previous experience in successful vivarium design. Conducting a system failure analysis as part of the design process can be an important step in strengthening disaster resilience.

In the committee’s judgment, when a new research building is proposed, the first cost and design stage of the planning process for the new building is the best place to address hazard mitigation in the vivarium. Actions for consideration may include

• The selection of the building site could be derived from a risk assessment, and a site with the least risk for natural disasters should be chosen.

• Programs could use strategies to stratify risk across the animal research program so that the likelihood of catastrophic loss is minimized.
• The vivarium could be located on a safe floor level within the building, and routine assignment to the basement should be avoided;
• The vivarium could have fail-safe design for animal housing systems and power supply.
• The design could permit the emergency operations plan to accomplish both evacuation and shelter in place.
• The emergency generator, switchgear, and fuel supply could not be in the basement.

Response and Recovery Planning

The details for developing an emergency operations plan (EOP) (see Chapter 6 for a broad discussion on response) for animal research facilities have been described in several publications (Goodwin and Donaho, 2010; Heath, 2000). It is important that every animal research program address both evacuation and shelter-in-place procedures in the EOP because the nature of the disaster will dictate which strategy holds the most promise for being able to protect and save the research animals (Yard, 2007). Animal research facilities that house potentially dangerous species or aggressive animals should consider keeping species-appropriate capture equipment on hand, such as nets or chemical immobilization devices (e.g., darts or guns). As part of the planning process, the facility’s administrative staff can work with their communications and public affairs department to prepare statements about their various species of animals that could be given to the media in the event that escapes occur. By doing so, the academic research institution can minimize the perpetuation of false information that might cause undue concern or panic among members of the general public.

Evacuation is an ideal response strategy when a disaster can be forecasted (e.g., a hurricane or riverine flood) and there is adequate logistical support and time available to move the research animals before the disaster (FEMA, 2013). However, evacuation is not an ideal response strategy for a no-notice disaster such as an earthquake or tornado or when a disruptive event occurs and the movement of people is prohibited. Plans that propose to evacuate research animals after the disaster strikes are going to be more difficult to execute because the manpower, supplies, and logistical support required to move the research animals are likely not going to be available. After a disaster, transportation is going to be disrupted, access to the animal research facilities will be restricted, many employees will not be available to work, and there will be intense competition for the limited resources that are present in the area. In the chaos that follows a disaster, it is highly

unlikely that large numbers of research animals will be able to be moved safely (Goodwin and Donaho, 2010).

A shelter-in-place response strategy is based on the concept that it will be safer for the research animals to remain at their current location during the disaster and that it will be best to care for them there after the disaster (FEMA, 2015a).This is the default response strategy when a disaster occurs without forewarning. It is also the preferred response strategy when the movement of people is prohibited by local authorities, such as following terrorist attacks, hazardous materials spills, and radiation accidents. Shelter in place is a practical way to provide care for research animals after a disaster, but this type of strategy is subject to catastrophic failure if the building sustains damage that leads to it being deemed unsafe for humans to reenter (Durkee, 2007). The success or failure of shelter in place is dependent, first and foremost, on ability of the building’s exterior and interior components to resist the impacts of the natural disasters that are common in its geographic area. For sheltering-in-place to be successful, the animal research facility’s building exterior must remain structurally intact, and the building’s interior must not collapse or sustain so much damage as to cause a cascading event such as a fire or internal flood (Vogelweid et al., 2009). Life-sustaining building systems that supply vivaria must either remain available or be readily restored in the immediate post-disaster period (i.e., there must be an ability to establish at least limited control of temperature and ventilation and to restore power). General recommendations for the selection of an emergency power generator for vivaria have been published (Stitch, 2013). Enough potable water, food, and husbandry supplies to provide care for the animals for 7 to 14 days should be onsite, or else there should be a plan to expediently acquire these items (FEMA, 2015a). Furthermore, the animal research facility must have euthanasia equipment and supplies on-hand (Durkee, 2007).

For both evacuation and shelter-in-place procedures, training in and practicing of the procedures at regular intervals is essential to a successful implementation during a disaster. The amount of manpower needed to accomplish evacuation and shelter-in-place procedures must be known pre-disaster so that essential functions can be identified and personnel trained and credentialed by their academic research institution or local government to ensure that they will be given post-disaster access into the animal research facility. The importance of training and credentialing personnel from the animal research program and forming partnerships between the academic research institution and the community response organizations has been previously noted (Wingfield et al., 2010). The attending veterinarian, IACUC, and IO should play key roles during the planning process to ensure that the training and materials required for successful execution of the EOP for the animal facility are prioritized and to ensure that there is

appropriate alignment and integration of the animal facility plans into the overall response plans for the entire institution.

Important actions to take in the immediate post-disaster recovery period include prompt notification of appropriate regulatory agencies (the NIH–OLAW and USDA–APHIS) and accrediting organizations (e.g., AAALAC International) about the status of the institution and the welfare of the animals (Goodwin and Donaho, 2010). Contact should also be made with the authority having jurisdiction (AHJ) for animals, typically the Office of the State Veterinarian. The AHJ may be able to expedite the delivery of life-saving resources for the animals at the institution. The IACUC needs to be functioning in the immediate post-disaster recovery period in order to promptly and appropriately address any animal welfare issues that arise as a consequence of the disaster. Veterinary care must be promptly re-established. In cases where shelter-in-place strategies are chosen to protect the research animals, the planners should also consider having a post-disaster evacuation plan in place to relocate animals in the event conditions deteriorate or it becomes too difficult to obtain critical supplies. The rescue of surviving animals can be undertaken by trained personnel as soon as it is safe to do so. It is important for the damages to be documented before any cleanup begins. The condition of the vivarium should be assessed to determine its suitability for continued use as a sheltering site for the research animals. In cases where a vivarium is severely damaged and is unsuitable for use, the research animals can be relocated to safer temporary housing areas. Temporary housing locations, standard operating procedures for moving animals, and memorandums of agreement with participating suppliers and partner organizations can be developed pre-disaster, approved by the IACUC, and included in the institution’s resilience planning activities. This level of advanced preparation would speed the delivery of life-sustaining supplies and quickly restore more suitable living conditions for the animals. Refer to Chapter 6 for a more detailed explanation of recovery planning activities.

In contrast to the widespread planning that supports the transportation, care, and sheltering of pets following a disaster, the availability of outside assistance to provide similar services for research animals following a disaster is uncertain. At the present time, the primary federal resources for treating ill or injured animals following a disaster are the national veterinary response teams (NVRTs), which are a part of the National Disaster Medical System (HHS, 2015). The NVRTs’ missions have so far focused on supplementing veterinary services for pets and agricultural animals, but there are laboratory animal veterinarians who are members of some of these teams. The American Veterinary Medical Association supports teams of volunteer veterinary disaster responders, the veterinary medical assistance teams (VMATs) (AVMA, 2017). Many states have local disaster animal

response teams that can provide assistance for companion and agricultural animals and that are organized at either the county or state levels. For example, the Missouri Volunteer Veterinary Corps (http://www.movma.org/?78) is a statewide response group of veterinarians and veterinary technicians who are trained and credentialed for animal disaster response in Missouri. For a research facility to obtain assistance from a NVRT, a VMAT, or a local disaster response team, a request would have to be made through the AHJ for animals, typically the Office of the State Veterinarian.⁶

The need to provide improved care for research animals during disasters is clear. USDA–APHIS is planning to convene a multiagency working group specifically to address the acknowledged gap in research animal welfare during disasters (McCann, 2017). The proposed Laboratory Animal Best Practices Initiative would bring together animal research professionals, federal government animal and disaster response agency representatives, and private-sector experts to identify successful strategies for improving the planning and management of research animals during disasters. This initiative would be modeled after a similar initiative for zoos and aquariums, known as the Zoo and Aquarium All Hazards Preparedness, Response, and Recovery (ZAHP) Fusion Center (ZAHP, 2017). In 2014, the ZAHP Fusion Center was created to provide the greater managed wildlife community with information on prevention, protection, mitigation, and response and to connect the greater managed wildlife community with the emergency management sector.

In the committee’s judgment, the following criteria could be considered when developing emergency operations plans (EOP) for animal research programs:

• Use the incident command system (ICS), following the principles as outlined in the National Incident Management System (NIMS). The attending veterinarian and IACUC chair can be members of the command staff. An example of a NIMS-compliant ICS structure for an animal research facility has been published in Vogelweid (2013).
• Include both evacuation and shelter-in-place procedures as well as procedures in the event research animals escape.
• Address regulatory requirements to provide humane care and timely euthanasia.
• Understand the importance of local resource availability. The plan could draw on local resources obtained from near-by colleagues. Mutual aid agreements (MAAs) and memorandums of understand-

___________________

⁶ Personal communication, Sebastian Heath, Federal Emergency Management Agency, December 15, 2016.

• ing (MOUs) could be drafted and approved at the institutional level to ensure that partner institutions will be able to provide reciprocal assistance in the event of a disaster.

• Consider working to credential first responders from the academic research institution as first responders in the community and integrated into the local disaster response network. These first responders could be given access to their animal facility by both the institution and the local jurisdictional authorities. Because the credentials are recognized by jurisdictional authorities, these first responders could also provide reciprocal support to colleagues at nearby institutions, should that need arise.
• Address the mental health needs of survivors and first responders. Training the first responders from the animal research community in psychological first aid would help them to recognize the common signs of stress and distress in themselves and other survivors and direct them to seek help from mental health professionals.

CONCLUDING OBSERVATIONS

Gaining an appreciation of the influence of animals in a disaster area has practical implications for disaster resilience. The current guidelines available to the animal research community to guide disaster planning activities are incomplete and do not align with effective planning principles as outlined in the National Preparedness Sysyem (FEMA, 2015b). The disaster plans developed for research animals may be ineffective because they were developed by animal research professionals who may have inadequate knowledge of emergency management principles compounded by a lack of integration and input from the professional emergency managers at the institution; the deficiencies in the plans remain hidden because the mandated reviewers of the plan may also lack sufficient knowledge about planning principles and optimal response procedures. The committee identifies this as a significant gap in resilience planning.

Significant gains in strengthening resilience can be made by considering prevention, protection, mitigation and recovery, in addition to response. The committee suggests that resilience can be achieved in the animal research community by (1) use of the National Preparedness System to develop comprehensive plans that address prevention, protection, mitigation, response, and recovery for animal research programs; (2) strengthening the built environment that supports animal research programs; and (3) ensuring that disaster resilience planning for animal research programs is valued and appropriately integrated into the institution’s disaster resilience planning. Consultation should be sought from the institution’s professional emergency management planners, when available.

The disaster planning criteria described in the proposed USDA–APHIS contingency plan (currently on indefinite stay) are more in alignment with the National Preparedness System than with the existing recommendations for disaster planning stated in the Guide. If the contingency rule was in effect, the disaster plans from institutions that house regulated species would also be subject to review by USDA–APHIS inspectors during their routine visits to the facility. With adequate training of USDA–APHIS inspectors and the regulated community, the public could be assured that institutions are doing their best to protect and ensure the welfare of their research animals. By removing the stay on the contingency plan rule and requiring regulated entities to comply with its requirements, USDA–APHIS could move the animal research community into alignment with national planning practices. The ideal time to begin enforcement of the contingency plan rule would be after accurate training and guidance materials have been available to the animal research community for a reasonable period of time.

The Guide requires prompt reporting of adverse events that compromise animal welfare and animal mortalities to the IACUC, which will then determine if these events must be reported to oversight agencies, such as NIH–OLAW (NRC, 2011). Similarly, adverse events and animal mortalities must also be promptly reported to USDA–APHIS, the enforcing agency for the AWA (USDA–APHIS, 2016). Institutions that have accredited programs with AAALAC International also report adverse events to this voluntary accreditation organization (AAALAC-International, 2017b). Thus, there are several entities that collect firsthand reports of the adverse events and animal mortalities that occur during disasters, but this information is not currently shared with the animal research community at large. The absence of credible, published information about what contributes to the success or failure of disaster plans for research animals is unfortunate. The communication of best practices could be used to improve many plans, and the acknowledgment of lessons learned from unanticipated failures could minimize the propagation of errors.

A progressive step toward a safer future for research animals can be made if the animal research community will adopt disaster resilience as a mindset and apply it at each opportunity to purchase a housing system, build a new vivarium, or renovate an existing one. Animal research professionals at institutions must play a key role in defining and communicating the level of protection that is necessary for a vivarium to the architects and engineers on the design team.

Building site selection should be based on results of a comprehensive risk analysis, with the safest site available being selected. Hazard-resistant construction features can then be incorporated into the proposed construction to mitigate the remaining relevant hazards. The selection of the floor level for the vivarium is very important, and a vivarium should never be located in the basement of a building that is constructed at a site that is susceptible to flooding (Vogelweid et al., 2009).

It must be recognized that research animals are permanent occupants of the structure and that they cannot voluntarily evacuate to save themselves. Essential facilities must be designed to maintain their operations during and following a disaster, and facilities that handle or store hazardous materials must be required to be designed to maintain containment (refer to Chapter 8) (VA, 2016). Therefore, the academic research institution should consider its vivaria as essential facilities and work to incorporate fail-safe design criteria. If housing systems are dependent upon a continuous supply of electric power to sustain the animals’ lives, then these housing systems must be on emergency power systems, and the established protocol for testing emergency power systems for essential facilities should also be applied to the vivarium (NFPA, 2015).

• Conducting comprehensive planning for the animal research program by a multidisciplinary planning group that is integrated with the institutional planning group.
• Identifying a method of engagement with external community partners, such as the local emergency planning committee, emergency management, law enforcement, fire, public works, weather service, department of transportation, and others, to communicate the unique public health and safety issues of the animal research program.
• Developing evacuation and shelter-in-place procedures, as well as procedures in the event research animals escape, in emergency operations plans for animal facilities. Facilities maintenance staff should be involved in the planning process so that they are aware of the power and other utilities requirements for the vivarium post-disaster for successful sheltering in place. Plans should include contact information for the people who can facilitate the acquisition of outside assistance and help meet regulatory reporting requirements. The Office of the State Veterinarian (or the authority having jurisdiction for animals) is the point of contact for obtaining any outside assistance for animals that might be available at the local, state, or federal levels. Institutions that receive Public Health Service funds are required to contact NIH–OLAW; those with species regulated under the Animal Welfare Act are required to contact USDA–APHIS; and accredited organizations are required to contact AAALAC International.
• Incorporating fail-safe criteria in vivarium design, as appropriate for each animal research program. Examples include (a) designing and testing emergency power systems on a schedule that is similar to that required for a hospital; and (b) ensuring that the valves controlling reheat coils on heating, ventilation, and air-conditioning systems fail in the closed position.
• Basing the vivarium location on a threat and hazard identification and risk assessment. A safe location within the building should be selected. A vivarium should never be placed in flood-prone areas within a building.

REFERENCES

———. 2017a. Position statements. https://www.aaalac.org/accreditation/positionstatements.cfm (accessed March 1, 2017).

———. 2017b. Rules of accreditation. https://www.aaalac.org/accreditation/rules.cfm (accessed January 27, 2017).

AVMA (American Veterinary Medical Association). 2012. U.S. pet ownership statistics. https://www.avma.org/KB/Resources/Statistics/Pages/Market-research-statistics-US-pet-ownership.aspx (accessed March 2, 2017).

———. 2017. AVMA’s veterinary medical assistance teams. https://www.avma.org/ProfessionalDevelopment/TrainingAndService/VMAT/Pages/default.aspx (accessed March 1, 2017).

Bayne, K. 2002. Development of the human–research animal bond and its impact on animal well-being. ILAR Journal 43(1):4–9.

Chang, F. T., and L. A. Hart. 2002. Human–animal bonds in the laboratory: How animal behavior affects the perspective of caregivers. ILAR Journal 43(1):10–18.

Donahue, L. R., M. Hrabe de Angelis, M. Hagn, C. Franklin, K.C.K., Lloyd, T. Magnuson, C. McKerlie, N. Nakagata, Y. Obata, S. Reed, W. Wurst, A. Hörlein, and M. T. Davisson. 2012. Centralized mouse repositories. Mammalian Genome 23(0):559–571.

Durkee, S. 2007. Disaster planning. Animal Lab News 6(3):21–24.

FEMA (Federal Emergency Management Agency). 2013. Evacuating v. shelter-in-place. https://www.fema.gov/faq-details/Evacuating-v-Shelter-in-place-1370032121004 (accessed March 1, 2017).

———. 2015a. IS-10.A: Animals in disasters: Awareness and preparedness. https://training.fema.gov/is/courseoverview.aspx?code=is-10.a (accessed December 8, 2016).

———. 2015b. National Preparedness Goal, 2nd ed. http://www.fema.gov/media-library-data/1443799615171-2aae90be55041740f97e8532fc680d40/National_Preparedness_Goal_2nd_Edition.pdf (accessed September 6, 2016).

FHWA (Federal Highway Administration). 2017. Evacuating people with special needs: Routes to effective planning primer series. Annex 2: Legislation, regulations, and policy. https://ops.fhwa.dot.gov/publications/fhwahop09022/sna2_legislation.htm (accessed February 28, 2017).

Goodwin, B. S., and J. C. Donaho. 2010. Tropical storm and hurricane recovery and preparedness strategies. ILAR Journal 51(2):104–119.

Hall, M. J., A. Ng, R. J. Ursano, H. Holloway, C. Fullerton, and J. Casper. 2004. Psychological impact of the animal–human bond in disaster preparedness and response. Journal of Psychiatric Practice 10(6):368–374.

Heath, S. E. 2000. Disaster planning for research and laboratory animal facilities. Animal Welfare Information Center Bulletin 11(1–2):1–8.

Heath, S. E., and R. D. Linnabary. 2015. Challenges of managing animals in disasters in the U.S. Animals 5(2):173–192.

Helfrich, L. A., and G. Libey. 2013. Fish farming in recirculating aquaculture systems (RAS). Virginia Tech. http://fisheries.tamu.edu/files/2013/09/Fish-Farming-in-Recirculating-Aquaculture-Systems-RAS.pdf (accessed March 1, 2017).

HHS (Department of Health and Human Services). 2015. National veterinary response team (NVRT). https://www.phe.gov/preparedness/responders/ndms/teams/pages/nvrt.aspx (accessed January 18, 2017).

Höglund, A. U., and A. Renström. 2001. Evaluation of individually ventilated cage systems for laboratory rodents: Cage environment and animal health aspects. Laboratory Animals 35(1):51–57.

Huerkamp, M. J., W. D. Thompson, and M. D. Lehner. 2003. Failed air supply to individually ventilated caging system causes acute hypoxia and mortality in rats. Contemporary Topics in Laboratory Animal Science 42(3):44–45.

Ikeda, T. 2012. Crisis management and recovery from the damage to the laboratory animal production facility due to the Great East Japan earthquake. Japanese Association for Laboratory Animal Facilities 61(1):1–11.

Kapucu, N. 2008. Collaborative emergency management: Better community organizing, better public preparedness and response. Disasters 32(2):239–262.

Lab Products, Inc. 2017. Balanced airflow. http://labproductsinc.com/product/balanced-airflo/ (accessed March 1, 2017).

Lawrence, C., and T. Mason. 2012. Zebrafish housing system: A review of basic operating principles and considerations for design and functionality. ILAR Journal 53(2):179–191.

Lehner, N. D. M., and J. R. Hessler. 2009. Common facility design errors and problems. In Planning and design research animal facilities, edited by J. R. Hessler and N. P. Lehner. Amsterdam: Elsevier, Inc.

Ligenza, L., M. Robinson, and H. Oglesby. 2010. Resiliency in the face of disaster: Strategies for survivors and responders. PowerPoint presented at FEMA IA ESF-6, San Diego, CA. https://www.fema.gov/pdf/conferences/iaconference/2010/tuesday_1045am_resiliency_in_face_of_disaster.pdf (accessed December 8, 2016).

Lipman, N. S. 2009. Rodent facilities and caging systems. In Planning and designing research animal facilities, edited by J. R. Hessler and N. P. Lehner. Amsterdam: Elsevier, Inc. Pp. 265–287.

McCann, A. 2017. USDA Animal and Plant Health Inspection Service response to committee questions. Available by request through the National Academies’ Public Access Records Office.

MMRC (Mutant Mouse Resource and Research Centers). 2017. Mutant Mouse Resource and Research Centers supported by NIH. https://www.mmrrc.org/about/organization.php. (accessed February 7, 2017).

Mort, M., I. Convery, J. Baxter, and C. Bailey. 2005. Psychosocial effects of the 2001 UK foot and mouth disease epidemic in a rural population: Qualitative diary based study. BMJ 331(7527):1234.

Mortell, N., and S. Nicholls. 2013. Practical considerations for disaster preparedness and continuity management in research facilities. Lab Animal 42(10):F18–F24.

Nagamine, C. L., C. T. Long, G. P. McKeon, and S. A. Felt. 2012. Carbon dioxide and oxygen levels in disposable individually ventilated cages removed from mechanical ventilation. Journal of the American Association of Laboratory Animal Science 51(2):155–161.

NFPA (National Fire Protection Association). 2015. NFPA 99: Health care facilities code. http://www.nfpa.org/codes-and-standards/all-codes-and-standards/list-of-codes-andstandards?mode=code&code=99 (accessed January 18, 2017).

———. 2016. NFPA 150: Standard on fire and life safety in animal housing facilities. http://www.nfpa.org/codes-and-standards/all-codes-and-standards/list-of-codes-andstandards?mode=code&code=150 (accessed January 18, 2017).

NIH (National Institutes of Health). 2015. Public Health Service policy on humane care and use of laboratory animals. https://grants.nih.gov/grants/olaw/references/phspolicylabanimals.pdf (accessed January 27, 2017).

NIST (National Institute of Standards and Technology). 2015. Community resilience planning guide for buildings and infrastructure systems: Volume 1. http://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.1190v1.pdf (accessed September 7, 2016).

NRC (National Research Council). 2000. Strategies that influence cost containment in animal research facilities. Washington, DC: National Academy Press.

———. 2011. Guide for the care and use of laboratory animals. 8th ed. Washington, DC: The National Academies Press.

Ormandy, E. H. 2012. The use of animals in science: Trends and public attitudes. Ph.D. dissertation, The University of British Columbia, Vancouver.

Phoenix Controls Corporation. 2002. Cage rack ventilation options for laboratory animal facilities. https://www.phoenixcontrols.com/en-US/solutions/bi/pharmaceutical/Documents/Cage%20Rack%20White%20Paper%20(MKT-0101).pdf (accessed March 1, 2017).

Stitch, E. B. 2013. Considerations for the selection of a generator as a back-up power source for an animal facility. Lab Animal 42(10):F25–F27.

Swearengen, J., K. Varga, M. Tate, and N. S. Linde. 2010. Disaster preparedness in bio-containment animal research facilities: Developing and implementing an incident response plan (IRP). ILAR Journal 51(2):120–126.

USDA–APHIS (U.S. Department of Agriculture Animal and Plant Health Inspection Service). 2012. Handling of animals; Contingency plans. Federal Register 7(250):76815–76824. https://www.federalregister.gov/d/2012-31422 (accessed December 8, 2016).

———. 2013. Handling of animals: Contingency plans; stay of regulations. https://www.regulations.gov/document?D=APHIS-2006-0159-0214 (accessed January 27, 2017).

———. 2016. Code of Federal Regulations: Title 9—Animals and animal products. https://www.gpo.gov/fdsys/pkg/CFR-2009-title9-vol1/xml/CFR-2009-title9-vol1-chapI-subchapA.xml (accessed January 30, 2017).

VA (Department of Veterans Affairs). 2016. Seismic design requirements. H-18-8. https://www.cfm.va.gov/til/etc/seismic.pdf (accessed March 4, 2017).

Van Haaften, E. H., M. Olff and P. H. Kersten. 2004. The psychological impact of the foot and mouth disease crisis on Dutch dairy farmers. NJAS–Wageningen Journal of Life Sciences 51(4):339–349.

Vogelweid, C. M. 2013. Using principles from emergency management to improve emergency response plans for research animals. Lab Animal 42(10):F1–F7.

Vogelweid, C., J. B. Hill, and R. A. Shea. 2009. Hazard-resistant building construction. In Planning and designing research animal facilities, edited by J. Hessler and N. Lehner. London: Academic Press. Pp. 135–149.

Wingfield, W. E., B. E. Rollin, and A. B. Bowen. 2010. You have a disaster plan but are you really prepared? ILAR Journal 51(2):164–170.

Yard, C. R. 2007. Evacuation vs. shelter in place: A situational comparison. http://www.drj.com/journal/summer-2000-volume-13-issue-3/evacuation-vs-shelter-in-place-a-situational-comparison.html (accessed March 1, 2017).

ZAHP (Zoo and Aquarium All Hazards Preparedness). 2017. ZAHP Fusion Center: Preparedness, response, and recovery. http://zahp.aza.org/ (accessed March 1, 2017).

This page intentionally left blank.