It is not very different from someone losing their entire home. For scientists, their research is their lifeline.
—Dafni Bar-Sagi (quoted in Sifferlin, 2013)
Academic research institutions are unique and specialized in terms of their physical research facilities. They tend to serve both research and teaching functions, provide employment and other economic value within their community and even region, and may also provide critical services like medical care and laboratory functions. Universities are essentially self-contained communities with housing facilities, food services, small businesses (from retail stores to printing presses), and hospitals (Comerio, 2000).
Tasked with describing the extent of the impact of prior disasters on the academic biomedical research community, the committee relates here selected events that have befallen academic research institutions in the past several decades, predominantly in the United States, but also drawing from overseas disasters where they are informative. This discussion demonstrates how the special features of the academic biomedical research community create unique vulnerabilities which may require specific mitigation efforts. The committee recognizes that effective mitigation efforts reduce the impact of a disaster and ease the response and recovery stages and that the lessons learned during a disaster may lead to further mitigation and preparedness efforts.
Many of the events described here are natural disasters such as hurricanes and earthquakes. Many of these have been covered in newspapers, journals, and after-action reports, and therefore much information is avail-
able about them. They have also been the drivers of an increased acknowledgment of the need for change and therefore merit specific attention. There are many other kinds of disasters, however, and so we also discuss animal rights terrorism, chemical, biological, radiological, and nuclear incidents, information technology outages, and cyber attacks. Each of these illustrations informs the committee’s findings.
The hazards that can affect the academic biomedical research community are the same as those that impact the wider community: natural threats from hurricanes, floods, tornadoes, high winds, earthquakes, and hazardous material releases; and intentional threats from arson, cyber intrusion, or chemical, biological, radiological, and nuclear terrorism. An examination of past disasters uncovers the specific insults and subsequent impacts large and small that have befallen researchers, institutions, and research sponsors. A consideration of low-probability but very high-consequence events, such as high-magnitude earthquakes of the sort that have not occurred in modern times, also reveals the potential for damage far beyond what has been planned for in many cases.
First and foremost, events that damage biomedical laboratories and the institutions that house them can have enormous impacts on the safety and well-being of humans and research animals. Furthermore, disasters can affect career trajectories, scientific progress, and financial stability at the individual and institutional levels. Disasters can very directly influence investigators’ ability to meet grant goals and requirements, in turn influencing metrics for and program outcomes of their sponsors.
Primary power outages compounded by generator failures; major flooding surpassing maximum predicted levels and overwhelming flood walls, pumps, and other physical barriers; an emphasis on emergency plans for hospitals but not for research centers—the surprises of disasters are numerous and far-reaching. A review of prior disasters reveals that the planning failures that lead to the greatest damage are often rooted in flaws that are systemic to an institution or to general practices across institutions: generators and other utilities stored on low floors; animal vivaria housed in basements; emergency plans that do not account for employees’ inability to reach the site and implement them, and so on. Such a review also reveals that failures of resilience in a single element can have a ripple effect on seemingly unrelated areas.
Stories of the human lives affected by disasters are compelling. For this reason, media reports in the aftermath of disasters often focus on the individual: the seasoned researcher who lost decades of work the instant
a unique mouse lineage was lost; the graduate student who never graduated; and the young investigator whose critical first years of publishing in a competitive field were undercut by a power outage that destroyed freezer samples. Some of these researchers saw their laboratories destroyed, while others were unable even to reach their place of work to assess the damage and begin to rebuild their research
Studies of business continuity show that natural disasters result in significant business interruptions. A study of the California Northridge Earthquake of 1994 reported that 57.5 percent of area businesses surveyed reported employees unable to get to work for some time post-earthquake (Tierney, 1997). Researchers and their institutions are subject to the same challenges. About 4 weeks after Hurricane Katrina’s rains subsided, Louisiana State University Health Science Center (LSUHSC) faculty were instructed to be available for work and to possibly be prepared to commute to Baton Rouge or other locations (Savoie et al., 2007). The institution warned that faculty who did not return as instructed could be discharged; even those who did were not assured their positions would ultimately be retained.
These kinds of stories were common in the early aftermaths of disasters. When Hurricane Ike overcame the University of Texas Medical Branch (UTMB) in 2008, the island nature of Galveston created a bottleneck (Goodwin and Donaho, 2010). On a normal day access was limited to three ingresses—a causeway, a toll bridge, and a ferry—and post-storm damage left the causeway as the only operational route onto the island. This prevented most staff from accessing the buildings until day 3 post-event, although a “rideout team” was present for that entire time (Goodwin and Donaho, 2010). Staff who were present or otherwise able to reach the facility focused on caring for the thousands of animals now living in buildings without power. Separate from the relief staff who provided necessary services from that point forward, UTMB put in place a work-from-home strategy that included online networking and streamlined purchasing procedures, allowing for resupply and payroll to move forward, if not bench research.
In October 2012, Hurricane Sandy rolled into New York Harbor with the rising tide. Floodwaters led to explosions and the shutdown of an electrical substation near New York University Langone Medical Center (NYU Langone). NYU Langone quickly fell to the mercy of dual insults: major flooding combined with power outages. With mass transit down, many staff were unable to reach the site, and many who did arrive had to come on foot. Just a mile or so up the island, Kyu Rhee’s laboratory at Weill Cornell Medical College in New York City was largely unaffected
by Hurricane Sandy, but his work was nevertheless significantly impeded when he was forced to evacuate his New Jersey home after a tree fell on his roof (Cossins, 2012). He and his family moved from hotel to hotel until they could locate more permanent accommodations. For a while he worked by telecommuting via coffee shops, which ultimately impacted the submission of time-sensitive manuscripts. Universities in New York City and other areas that have faced transit strikes, major winter weather, or other challenges to commuter transit at this point have some considerable experience dealing with staff inability to reach their worksite. A number of their emergency management plans now take this into consideration.
Many of the incidents the committee studied caused researchers to lose their work environment—their buildings, their laboratories, or their workstations. The downstream impacts of this were many, but one of the most immediately pressing was the need to find alternative workspaces in the near term to prevent research projects from meeting a premature end. This was not always a simple proposition: the competitive nature of science meant that pre-existing agreements were not typically in place for such situations. In addition, a major event that affected one institution often affected others nearby.
As a direct result of Tropical Storm Allison in 2001, 3,200 faculty, staff, and students at the University of Texas Health Science Center at Houston (UTHSC-H) were displaced for more than a month (Goodwin and Donaho, 2010). More than 1 million gross square feet of space were put out of commission, including research, animal care, and support functions. Ten million gallons of water had inundated the medical school basement alone. Similarly, Hurricane Ike’s effects in 2008 barred UTMB faculty from their buildings for nearly 2 months. In 2005, as a result of Hurricane Katrina, Tulane’s medical school held classes for two semesters in Houston, requiring some faculty to teach at a location where they lacked laboratory space (Simon, 2006). At NYU Langone, the Smilow Research Center experienced extensive damage, particularly to its vivarium, and the attached Medical Science Building also sustained substantial damage such that everything in its basement was destroyed (Bisceglio, 2012).
The loss or diminishment of the work environment created what David Gresham of NYU Langone described as an “astronomical” loss of time and effort after Hurricane Sandy (Mandavili, 2013). Nevertheless, the outreach from scientist to scientist and institution to institution in the aftermath of the disasters was in many ways remarkable. Many Langone researchers established makeshift, temporary laboratories in neighboring research institutions such as Weill Cornell Medicine and Rockefeller University or on other NYU
campuses not affected by the storm. At least 24 Langone scientists found temporary workspace at Rockefeller, in some cases even signing leases (Church, 2013). Rockefeller’s possession of biosafety level 3 facilities was an added bonus for those researchers who needed it. Not all Langone researchers were forced to leave their campus; some remained, using other buildings that were still operational. This is, perhaps, an example of “unplanned resilience,” in that there were positive outcomes for some scientists despite a lack of formalized and pre-existing understandings with partner institutions.
The loss of employment seen in the worst disasters adds further injury to that already caused by the loss of work environments. It is indeed one of the most significant impacts that can befall a researcher as a result of a disaster. Following a disaster, one of the first questions of employees is whether their paychecks may be disrupted. Graduate students, trainees, and early-career investigators may have limited savings or significant personal commitments that may be particularly stressful. Temporary disruptions may become permanent either based on an institutional decision to let people go, or of researchers taking the initiative to leave and restart their research programs elsewhere as soon as possible
After Hurricane Katrina, LSU continued to pay salaries and benefits to all faculty, even when the university was only partially reopened, into November 2005 (Savoie et al., 2007). On November 22, however, the LSU Board of Supervisors approved a “force-majeure exigency plan” for the Health Sciences Center. According to Savoie et al. (2007, p. 67) the board cited the disruption of “revenue streams which no longer exist because they were generated by hospitals and clinical practices in New Orleans which have been destroyed, closed, or are non-operational,” and the exigency plan declared the administration’s right to abrogate tenure and normal notice-of-termination protections. Some 51 full-time School of Medicine faculty and 34 part-time faculty were furloughed and removed from payroll as of December 1, and additonal furloughs followed.
The situation was similar at Tulane University. The board of administrators decided during that December to declare financial exigency, and at least 160 faculty received notifications of release; 90 additional faculty resigned or retired in the days that followed (Savoie et al., 2007). This included some faculty who had been very active in research and who were recruited elsewhere. The School of Medicine was hardest hit, with clinical faculty accounting for about 120 of those let go, among whom were 34 tenured faculty members (30 clinical and 4 basic science). In Tulane’s Department of Psychiatry and Neurology, approximately half of the 50 faculty were let go, including all but one of the tenured psychologists (Savoie et al.,
2007). This in turn decimated a psychology internship program, showing that the loss of senior faculty jobs can affect not only the livelihoods of those faculty but also the careers of those they mentor.
Hurricane Ike forced the closure of the UTMB university hospital and caused a major disruption to the clinics; because research was integrated with the clinical enterprise, the closure of the hospital and clinics significantly disrupted a source of UTMB’s income (Goodwin and Donaho, 2010). This necessitated a reduction in force as of November 2008 that cut 20 percent of UTMB’s staff, mostly clinical but also research. Some researchers left of their own will because they needed to get their research up and running somewhere else.
To prevent these kinds of employment losses, NYU Langone made a concerted effort to ensure that staff salaries were not interrupted after Hurricane Sandy. As a well-resourced private institution, Langone found this to be more feasible than other institutions did. Even at that, the long-term effects of reduced grant income did take its toll on staffing (Heins, 2014).
Hurricane Katrina provides one of the more striking examples of how both careers and personal lives can be affected simultaneously. When Hurricane Katrina struck New Orleans, Tulane University biochemist Dr. Arthur Lustig spent 4 days sequestered in his laboratory before being evacuated by helicopter to spend a night in a shelter (Kaiser, 2005). Dr. Lustig’s house was nearly lost to flooding, requiring the gutting and rebuilding of its lower level; in his lab, 80 percent of the yeast strains that had been developed over 20 years and used to study telomeres were lost to the power outage (Kaiser, 2005) At the LSUHSC, many researchers lost their homes and were forced to direct their attention toward finding or building new homes; those whose homes were intact but damaged often waited months for repairs because of intense competition for contractors (Kuehn, 2013).
The difficulties like these encountered by faculty, staff, and students after disasters may lead to mental states like depression or post-traumatic stress disorder; disasters may also make these states worse when they are pre-existing conditions. A prospective study of Stanford University students found that after the 1989 Loma Prieta Earthquake, students reported a variety of stressors, including inconvenience to daily life (the most common), damage to the area they were in when the earthquake occurred, damage or injury to family and friends, and damage to their dormitory residence (Nolen-Hoeksema and Morrow, 1991). Manifestations of stress
may be labeled easily, but they are less easily quantified or understood by those who lack the shared experience.
The far deeper harm defies quantification or physical description. For faculty and staff who lacked not only telephone and Internet access but also places to live after their homes had been destroyed, the measure of loss seems incalculable. For scientists who eventually returned to their flooded laboratories only to find that years—even decades—of research had been destroyed, the impact of the storm is well beyond even the most sympathetic conjecture. (Savoie et al., 2007, p. 62)
Many of the disasters described herein have destroyed animal research facilities and, with them, tens of thousands of animals. Eleven years after Hurricane Katrina, LSUHSC was still using some temporary animal research facilities and is just now beginning the construction of a new facility designed to replace that which flooded.1 The numerical losses of these animals are detailed in subsequent sections, but it is worth noting here that the losses affected the animals’ handlers in many ways. Bradford Goodwin described University of Texas Health Science Center at Houston’s (UTHSC-H) response to this element of the Tropical Storm Allison tragedy as requiring special attention to the mental and physical well-being of the animal-care-and-use staff:
The recognition of both physical and emotional loss was a key element of the personal recovery of dedicated staff members from both the CLAMC [Center for Laboratory Animal Medicine and Care] and research laboratories. Psychological counseling was made available and, about 5 months after the loss, a memorial service for the animals that died was attended by over 350 UTHSC-H faculty and staff. The memorial was a tribute to the animals that gave their lives to biomedical research without completing the goals and objectives of many long-term studies involving memory, learning, and autism. Each year on the anniversary of the storm a floral arrangement and tribute to the nonhuman primates that died during this disaster is presented by a nonhuman primate veterinary technician and displayed in the main lobby of the Medical School building. (Goodwin and Donaho, 2010, p. 108)
Upon learning of the loss of his research animals in the Hurricane Sandy floods that devastated NYU Langone, Gordon Fishell, the associate director of the NYU Neuroscience Institute, described an “awful sense of despair” (Fishell, 2013, p. 421). He grieved for the suffering and loss of the animals themselves, for the years of work lost, and for the impact that all of this would have on his laboratory staff, who had “put their hearts and souls
1 Personal communication, Prescott Deininger, Tulane University, February 6, 2017.
into their research. I mourned for 12 hours, then realized that I needed to work out how to move forward” (Fishell, 2013, p. 421). Even today, some researchers at NYU Langone become overwhelmed when speaking of the events that they experienced.
In some cases, psychological distress is a result of manmade events. In the early 2000s a trend of criminal targeting of researchers appeared in which researchers associated with animal research were subjected to violence and scare tactics whose aim was to intimidate them and their families (Bailey et al., 2010). Some experts cite a decrease in events at institutions, but also a continued vulnerability (Bailey et al., 2010). Experts also cite a shift in the targets of the extremists, from a predominance of universities in the 1990s to individuals in the 2000–2012 period (FASEB, 2014).
Even when external events such as terrorism and disasters have not forced researchers out of their fields, their place in the research hierarchy has not necessarily been secure. Altered perceptions on the part of sponsors and other research partners of the fitness of an institution within a disaster area may have a real effect on its future. Grant reviewers sometimes assume the same, even years after their recovery, opting against funding long-term projects in what they may still see as a disaster area.2 A successful recovery process addresses the full range of psychological needs of the people of the research enterprise as it recovers from the disaster through the provision of support, counseling, screening, and treatment as needed (FEMA, 2016).
The physical, financial, and psychological damage wrought by disasters can affect the trajectory of researchers’ careers. Researchers may even be confronted with the prospect of risking their own lives to save valuable experiments and the equipment on which their careers depend (McNutt and Leshner, 2013). The severity of the impact may depend on the stage of the researcher’s career. Graduate students, trainees, and early-career researchers may be especially vulnerable. Graduate students typically depend on financial support in the form of employment as graduate research assistants. If the research project or funding is disrupted, they may not have financial support for degree completion. Early-career investigators are also vulnerable because they need to establish competitive research programs. Disruption or loss of momentum can damage competitiveness. Even in cases when research projects were not completely destroyed, researchers subject to disasters describe a loss of momentum as they spend precious time acquiring new research animals and getting their research fully back on line, while simultaneously dealing with damage to their own homes (Ortolon, 2009).
Some early-career researchers have been forced to start over completely. Dafna Bar-Sagi, now senior vice president and vice dean for science at NYU Langone, said, “For someone who started 3 to 4 years ago and just got to a point to launch their research program, it’s time to rewind and start from fresh” (Sifferlin, 2013). These words could easily describe her colleague, Jeffrey Berger, who in 2012 was an early-career scientist at NYU Langone and had a burgeoning research program studying cardiology drugs.3 His career was interrupted by the floods that wiped out power to the freezer that contained his research samples. Research had to be restarted; publications were delayed. About $250,000 in lost resources had to be recouped, which took 6 months. (Not every project can survive a 6-month delay; in some cases, the competitive advantage disappears [Fishell, 2013].) One of the most vexing aspects of the situation—and one echoed in the stories of many other researchers—was what to do about the freezer samples. To what extent had they thawed, and for how long, and would this create a variable that influenced experimental results? Berger contacted other researchers in his field, and no one really knew whether the potential thaw would affect his results. He made the difficult decision not to use the samples for further study or for his publications. While this may have cost time and money, Berger said that there was benefit in being able to redo the science even better the second time around. Although it is essentially impossible to determine whether such redirection positively or negatively affects research outcomes, disasters do create opportunities for scientists to redesign their research or restructure their laboratories.
Hurricane Katrina hit graduate student research very hard. Vincent Shaw, a student at Tulane, was displaced from his lab, and, while he was able to relocate temporarily to a laboratory thousands of miles away at Brown University in Providence, Rhode Island, he was forced to leave behind elements that were critical to doing his job well. These included freezer samples, now thawed; experimental animals, now dead; and the analyses needed to finish a paper in press (Kaiser, 2005). LSUHSC noted issues with manuscript generation and publication delays as well as grant submissions and grant renewals, all of which had a lasting impact on the institution’s mission and the many contributions of the research programs in Louisiana sponsored by the National Institutes of Health (NIH) and the National Science Foundation (NSF) (Lanier, 2005).
The personal damages combined with the losses and disruptions of research programs often had a severe impact on career development for both students and seasoned faculty. After Hurricane Katrina, about 35 faculty members from the Tulane School of Medicine self-funded travel to Houston, Texas, to develop a program for medical students (Savoie et
3 Personal communications, Jeffrey Berger, NYU Langone, July 14, 2016.
al., 2007). One professor of 15 years acted as the interim clinical clerkship director and also handled other assignments in anticipation of previously scheduled accreditation-related visits; his position was later terminated (Savoie et al., 2007). Faculty found themselves taking on such volunteer efforts in attempts to get the school’s programs back on line. This undoubtedly meant sacrificing time that they could have spent restarting their own research.
About 400 faculty members, postdoctoral fellows, and graduate students were seriously affected by Tropical Storm Allison, with research projects and graduate training programs significantly delayed (Goodwin and Donaho, 2010). In June 2001, pathologist David Roth was preparing for a sabbatical in England, when the arrival of Tropical Storm Allison brought rainwater flooding into his laboratory at the Baylor College of Medicine (Matthews, 2012). After assessing the animals, the reagents, and the research lost, Roth decided his best option was to start anew elsewhere. “If you lose everything, you’re at least a couple of years behind,” he said. He moved to NYU Langone, where he ultimately became the director of the medical scientist training program. By the time Hurricane Sandy hit, Roth had moved on to the University of Pennsylvania, but his mice were still being used at NYU Langone, and history repeated itself when those mice were lost to the storm. The difference this time was that Dr. Roth had previously provided mouse strains from all ongoing experiments to a laboratory in Maine, making recovery far more manageable (Matthews, 2012).
The immediate pain of devastating events is often felt at the level of the researcher and also as a shared experience among the researchers within an affected department or building, but the pain eventually works its way up to the level of the entire academic research institution. From the buildings to the research animals, equipment, and critical infrastructure they contain, entire departments and institutions may be affected by disasters to the point of shutdown.
While institutions may be concerned with protecting the data or specimens contained within their buildings, these buildings may not have been designed for ensuring this purpose. Buildings are regulated via building codes and standards for the risk of structural failure, among other characteristics. Risk categories are used to designate the consequences of failure—such as loss of life or impact on economy—from lowest risk (Category I) to highest (Category IV). Institutes of higher education are generally regulated
as a Category II or III (NIST, 2016). Structural damage—even if it is not considered life-threatening—can create obstacles to functionality in the following weeks and months, as services can be temporality shut down or relocated during repair (Mitrani-Reiser et al., 2013). Furthermore, the lack of communication to building occupants and the community of what is considered life-threatening structural damage can lead to unnecessary facility closures.
In 1971, the magnitude-6.6 Sylmar–San Fernando earthquake struck California’s San Fernando Valley (USGS, 2016). The earthquake lasted about 60 seconds, took 64 lives, and caused an estimated $505 million in damage. There was widespread damage throughout the San Fernando Valley, including the destruction and collapse of major structures, such as the Olive View and Department of Veterans Affairs (VA) hospitals, and the collapse of multiple freeway overpasses.
The quake ultimately triggered the development of hundreds of new safety standards and programs across the state for public and private buildings, hospitals, dams, and freeways as well as land use. California developed a special code for the design and construction of hospitals and essential facilities. The Uniform Building Code adopted for use in the western United States was revised to require improved performance of buildings in the event of seismic events, with special consideration given to hospitals and other essential facilities (Bartholomew, 2016). Seismic evaluation and retrofit programs initiated by various public agencies and private owners also grew from the earthquake experience.
The programs instituted by the VA and the University of California provide examples. The collapse of two unreinforced masonry buildings that killed 49 people at the VA Campus in San Fernando as a result of the 1971 event triggered national legislation requiring the Secretary of the VA to assure that each medical facility was constructed to be resistant to fire, earthquake, and other natural disasters. This led to the creation of the Secretaries Advisory Committee on Structural Safety, which created and in 1975 formally approved H-08-8, the first special seismic design provisions for all VA facilities.
In 1987 the Whittier Narrows California earthquake caused severe damage to California State University, Los Angeles (Cal State LA) (Pine, 1988). The campus sustained more damage than any other large facility in the area, and the Department of Chemistry and Biochemistry, occupying four floors of the physical sciences building completed in 1972, was the most damaged department. Structural damage was minimal because the building had been designed to the California earthquake standards of the late 1960s, but losses of equipment and supplies were estimated at over $1 million. A fire in the organic chemistry laboratory damaged or destroyed much of the laboratory equipment, supplies, and utilities in the ceiling
utility area. It became clear that tethering equipment to benches and walls would be necessary to mitigate future earthquake impacts. Despite this damage, the Whittier Narrows earthquake was, in fact, a relatively small and local event.
Then, on January 17, 1994, Los Angeles found itself at the center of a devastating earthquake now known as the Northridge Earthquake of 1994 (Chavez and Binder, 1996). The Sepulveda VA Medical Center was located near the center. Electrical power and emergency generators there failed immediately; in an instant, hospital personnel became both victims and responders. The 6.7 magnitude quake forced the evacuation of that medical center, which included an animal research facility, as well as of several other hospitals (Schultz et al., 2003). Two major earthquake training exercises had been conducted in the area the year before. The first, in April, had been based on a scenario of a bus accident; the second, in November—just 2 months before the real event—posited a major fault rupture just north of the Sepulveda VA hospital. The exercise design included a complete evacuation of the hospital, which meant the evacuation of actual patients in the hospital at the time. Research clinicians used the exercise to identify hazards in their laboratories (Chavez and Binder, 1996). The Northridge earthquake led to significant revisions in the building codes and, as was the case in 1971, triggered the development of public and private seismic retrofit programs.
California is not the only state susceptible to earthquakes. A magnitude-5.7 quake that struck near Prague, Oklahoma, on November 2011 collapsed an old tower at St. Gregory’s University in Shawnee (Wertz, 2015). Damage from the quake was still evident on campus 4 years later. Overseas, earthquakes of extraordinary scale have devastated communities and the research institutions within them. The Christchurch, New Zealand, magnitude-6.3 quake of February 2011 led to a 13 percent decline in students at Canterbury University, and the school’s president stated in a letter to staff that April that it was likely to be several years before annual revenue covered the school’s operating costs (Law, 2011). The Great East Japan Earthquake of 2011 did major damage to a laboratory animal production facility; the impacts of this are discussed in the next section (Ikeda, 2012).
The Cascadia region of western Washington, Oregon, and coastal northern California is susceptible to great earthquakes with potentially devastating impact. Studies have shown that a magnitude 9 earthquake and tsunami were generated by the Cascadia subduction zone in 1700 A.D. Repeat of a similar earthquake today would damage critical infrastructure and buildings across the region, cause more than 10,000 deaths, and more than 30,000 injuries (CREW, 2013). Planning scenarios for a Cascadia region mega-quake do not specifically address impacts on academic re-
search institutions in the region, but the expected damage would be large and long-lasting.
That same year, the Federal Emergency Management Agency’s (FEMA’s) National Level Exercise simulated a catastrophic earthquake in the New Madrid seismic zone, which stretches from Missouri into the southern United States. Although Indiana has not experienced an earthquake recently, its proximity to the New Madrid and Wabash seismic zones necessitates earthquake preparedness. Indiana University published an Earthquake Preparedness for Laboratories guidance document to help investigators and staff prepare for and respond to earthquakes that may affect their laboratories (Kohler and Gray, 2014).
In many of the disaster events the committee reviewed, water has been one of the main causes of significant building damage, the loss of fire protection, and the loss of potable water. Tropical Storm Allison dumped 10 million gallons of water into the UTHSC-H Medical School basement, putting more than 1 million gross square feet of space out of service for months (Goodwin and Donaho, 2010). This space was used for a variety of functions ranging from teaching and research to animal care and support; the impacts of the damage were thus widely felt. Facility damage alone was estimated at $52 million.
Hurricane Ike’s biggest impact to UTMB was on its “critical core,” which included buildings whose function was deemed critical to remain operational (Maybauer, 2011). Eighty-nine percent (36 buildings) of the critical buildings in the research and health care complex core were damaged as a result of flooding.
Hurricane Katrina’s impacts were similar. Research animals, some generators, and some key laboratory infrastructure were housed on the lower floors of the LSUHSC (Kuehn, 2013). By the time small teams were able to enter the research facility safely, some refrigerators and freezers had reached temperatures of 100°F (Kuehn, 2013). LSUHSC was functioning by March 2006 (a 5-month downtime), although power interruptions and other challenges remained. It was a full year before all of the Health Science Center’s schools had reopened (Savoie et al., 2007).
Mitrani-Reiser et al. (2013) notes that is important to quantify how physical damage leads to loss of services typically provided by facilities and to identify factors that delay the recovery process. Chapter 8 discusses a resilient built environment for the research enterprise in more detail.
The impact of major disasters has often been borne not only by human researchers, but also by the thousands of animals that support their research efforts. In many of the natural disasters discussed here, the thou-
sands of animals housed at research facilities were unable to be evacuated with human evacuees, resulting in extensive loss of animal life. The toll of such losses can be measured in many ways: in the number of lives lost; in the stunting of biomedical research; in the cost to sponsors and insurers; and in the emotional impact on researchers and animal care staff. From storms to earthquakes to arson, animal facilities and the animals that live within them have been the subjects of devastation. In some cases, the damage has been extreme:
In inky blackness, the group stood at the brink of the animal section of the Smilow Research Center, where rodents for experiments had been kept, but they did not go inside. On Nov. 3, a memo sent to NYU Langone researchers said the animal section, or vivarium, was “completely unrecoverable.” (Hartocollis, 2012)
Two of NYU Langone’s vivarium facilities were severely damaged as a result of Hurricane Sandy. Although many animals were relocated in time to satellite facilities, with many rodents saved, a significant number of animals still perished: up to 35,000 mice in 7,000 pens (Ahlborn, 2017). NYU Langone estimated their value—including the cost to replace the animals and re-derive the lines—at $25-30 million.
During Tropical Storm Allison, basement vivaria at Texas Medical Center (TMC) component institutions, such as the Baylor College of Medicine, the Texas Heart Institute, and the UTHSC-H, suffered catastrophic flooding (Goodwin and Donaho, 2010). The flash floods drowned thousands of research animals because more than half were housed in basements. In all, about 35,000 animals died at TMC. Flooding hit the Center for Laboratory Animal Medicine and Care (CLAMC) of the UTHSC-H. Recovery operations ultimately pumped more than 10 million gallons of water from the basement. The 4,700 research animals that drowned included thousands of rodents, 300 rabbits, 78 nonhuman primates, and 35 dogs (Goodwin and Donaho, 2010).
There were many stories of human heroism related to the rescue of the privately owned cats, dogs, and horses that fell victim to the floods of Hurricane Katrina. Less well known are the similar stories of courage and the emotional distress of failed rescue efforts that took place behind the closed doors of research facilities. Goodwin said, for example, that many of TMC’s animal care providers attempted during Tropical Storm Allison to reach vivaria through waist-deep water and closed roadways (Goodwin and Donaho, 2010).
The staggering loss of animal life ultimately meant that all animal-based research emerging from the affected institutions was delayed (Goodwin and Donaho, 2010). Experiments on autism, gene therapy, cardiovascular dis-
eases, asthma, immune deficiency diseases, and neurological disorders that relied on research animals were postponed or destroyed because of the loss of the animals or even because their loss meant that entire genetic lines were now gone. The president of UTHSC-H, James Willerson, said, “The development of improved diagnostic tools; more effective treatments and drugs; and preventive measures for heart disease, stroke, respiratory diseases, and other killers has been seriously delayed” (Goodwin and Donaho, 2010, p.111).
When Hurricane Ike arrived 7 years later, it tested the improvements UTHSC-H had put in place after Tropical Storm Allison. The buildings withstood the high winds, and flooding and power outages were avoided (Goodwin and Donaho, 2010). But a different institution in the greater Houston area, UTMB, took a beating on the island of Galveston. John C. Donaho described it in this way:
This was only a Category 2 storm, yet the surge was already here 20 hours ahead of the storm. There was no wind or rain yet, just a slow relentless rise in water levels. I still had no idea on that Friday morning, 12 September, that our lives would be changed forever. I did comment in one of my emails that night as the waters swirled around the hospital and the winds started to pick up, that this was the one we have always worried about. And then the power and Internet went out. (Goodwin and Donaho, 2010, p. 106)
Virtually every building on campus sustained damage, save the newly constructed Galveston National Laboratory (Goodwin and Donaho, 2010). The hospital lost blood bank, pharmacy, radiology, and food services, and while the emergency room was spared, the failure of power and water meant its operations were unsupportable. Faculty were unable to enter buildings for nearly 60 days; schedules slipped, animals aged. Fortunately, animals in vulnerable low-floor vivaria had been moved to higher ground (Goodwin and Donaho, 2010).
History repeated itself as NYU Langone’s animal research centers—including one of the largest animal research facilities for heart, cancer, and brain research—were flooded (Sifferlin, 2013). The lack of a fresh supply of water meant that volunteers had to deliver fresh water manually to research aquaria by passing multigallon jugs from person to person from trucks outside to the aquarium inside. In addition, 751 mouse lines were lost, most of which needed to be re-derived (Ahlborn, 2017). This work was typically funded either through FEMA Community Assistance Program Grant proceeds, an NIH R24 (resource-related research projects) award funded by the Disaster Relief Appropriations Act (DRAA) (although expenses charged prior to the award start date could not be included), or through DRAA supplements to active NIH awards (Ahlborn, 2017).
Laboratories and the animals housed within them are susceptible to various types of disasters other than floods, of course. The Great East Japan Earthquake of 2011 shook the Tsukuba Breeding Center of Charles River Laboratories Japan in Ishioka, Ibaraki, Japan, sending animal cages falling from racks (Ikeda, 2012). The racks themselves were fixed by crossbars and bolted together, and although this prevented the racks from falling, connected racks moved up to 15 cm in some cases, and 2,500 mouse cages and 1,700 rat cages fell from the shelves. This freed 14,000 mice and rats, all of which were ultimately euthanized.
Natural disasters are not the only challenges that research facilities face with respect to protecting the animals and animal facilities in their care. A trend toward arson, break-ins, theft, and deliberate property damage at animal research facilities in the United States began in the 1980s (Bailey et al., 2010). These crimes have typically been committed by animal rights activists; their offenses have not generally harmed the animals themselves but rather the facilities in which they are housed. The criminals have also at times targeted personnel, using violent scare tactics to intimidate researchers and their families. Some of the higher profile attacks include
- April 1999: University of Minnesota, Minneapolis. The Lyons Research Building was broken into during World Laboratory Animal Liberation Week, resulting in extensive damage to equipment, computers, videotape, and research data and the theft of research animals involved in Alzheimer’s and Parkinson’s disease studies. Estimated damage: over $2 million (Bailey et al., 2010).
- September 2003: Louisiana State University, Baton Rouge. The Animal Liberation Front (ALF) claimed responsibility for a break-in at LSU’s Inhalation Toxicology Laboratory. Computers and other research equipment used in study of cardiovascular and respiratory disease were destroyed. Estimated damage: $250,000 (Bailey et al., 2010).
- November 2004: University of Iowa, Iowa City. ALF burglarized two laboratories, stealing 400 research animals and destroying computers, other equipment, and research data. Perpetrators dumped hazardous chemicals in offices and halls, resulting in building closures and class relocations for the duration of the semester. Estimated damage: $400,000 (Bailey et al., 2010).
In addition to affecting the personal safety of researchers and their staff and families, these kinds of extremist activities can result in the diversion of time and money from research and animal care; long-term damage to reputations; political interference with the scientific process; a loss of
community support and financial resources; and forgone research benefits or delays in medical progress (Bailey et al., 2010). Federal prosecutions and heightened facility security appear to have contributed to a drop in the frequency of these kinds of events (Bailey et al., 2010). The Animal Enterprise Terrorism Act, enacted in 2006, provided new authorities to the Department of Justice to apprehend, prosecute, and convict perpetrators of animal rights terrorism.4
The damage and destruction to buildings carry their impacts well beyond the structures and into the many instruments of science contained within them. These instruments include biological samples and reagents such as antibodies, antigens, cell lines, cell cultures, tissue samples, blood, proteins, viruses, and many other important elements of research studies. They also include expensive and finely calibrated research equipment.
The floods that took the lives of so many animals during and after Tropical Storm Allison in Houston destroyed the cages they lived in as well as an array of other equipment, some of it new and very expensive, including a 6-month-old magnetic resonance imaging instrument as well as computers and diagnostic instruments, which were housed in the basement (Goodwin and Donaho, 2010). It also destroyed cell and tissue cultures and volumes of research data. As a result of Hurricane Katrina, Arthur Haas lost a suite of samples reflecting 20 years of study (Kaiser, 2005). In New York many researchers came up against the same challenges. NYU’s financial stability put it in the enviable position of being able to release funds almost immediately to researchers to start ordering new equipment and supplies. Nevertheless, the storm surge from Hurricane Sandy destroyed four magnetic resonance scanners, a linear accelerator, and gamma knife surgery equipment, which were kept in the basement (Hartocollis, 2012). FEMA paid NYU Langone just over $92 million for lost biospecimens (Ahlborn, 2017). According to preliminary estimates by NYU Langone, $20-25 million in scientific equipment was lost.5
The nature of the research that occurs in biomedical laboratories necessitates the storage and use of a variety of reagents and other materials, including potentially hazardous substances ranging from the biological to the chemical and the radiological. These represent unique risks. Such agents
4 Animal Enterprise Terrorism Act of 2006, P.L. 109-374, 18 U.S.C. § 43.
5 Personal communication, Neil Rambo, NYU Langone, August 10, 2016.
may be released as a result of a physical insult, such as flood or fire, or via other accidents, and thereby endanger people and animals in the vicinity. The 1987 Whittier Narrows Earthquake, for instance, caused considerable chemical spills within the chemistry and biochemistry departments at Cal State LA (Pine, 1988). These agents may also be used as weapons against a variety of targets, including laboratories themselves.
Institutions with biomedical research programs may be subject to incidents involving the biological agents used or stored in their research facilities. Following the anthrax attacks of 2001, oversight of select agents was significantly strengthened, and regulations were put into place for the possession, use, and transfer of select agents (FSAP, 2016). The theft, loss, or release of a biological select agent or toxin (BSAT) requires immediate reporting to the Centers for Disease Control and Prevention (CDC) or the Department of Agriculture. In 2015, 199 potential occupational exposures to BSATs occurred in registered entities (FSAP, 2016). Registered entities include more than just academic laboratories; published reporting does not specify which laboratories reported potential exposures. These mishaps generally qualify as laboratory safety accidents. But the select agent regulations are also designed to mitigate security risks. The Federal Bureau of Investigation performs security risk assessments on individuals or entities registering for the program: in 2015, 16 individuals were flagged as security risks out of 4,426 applicants (FSAP, 2016). These included individuals convicted of felonies and misdemeanors, fugitives from justice, and an individual under felony indictment. These regulations are designed to prevent insider threats that could result in the theft and misuse of select agent material in bioterror attacks.
While no biological disasters have occurred, various accidents have nevertheless affected an institution’s research enterprise. In July 2007, CDC ordered the cessation of work on select agents and toxins at Texas A&M University while it investigated a mishap with Brucella and Coxiella burnetti. This cost Texas A&M $1 million from its research compliance funds in the form of a fine to the Department of Health and Human Services (HHS); the university proposed this dollar value itself in February 2008 so it could reach an agreement with HHS quickly and resume its biodefense research (Schnirring, 2008). In 2004, three Boston University researchers became ill with tularemia. They had been working with what they believed to be a live but attenuated strain of Francisella tularensis in the infectious disease laboratory on the sixth floor of the Evans Biomedical Research building (Barry, 2005). Seven researchers there were directly involved with the tularemia study, but 77 people worked on the floor in some capacity. Ultimately, no other cases of active disease were found. The incident highlighted how exposed academia can be to laboratory errors—it turned out that the researchers had been accidentally sent a virulent strain by one of
their suppliers. As a result of the outbreak, the Boston health department adopted regulations requiring anyone operating or planning to operate a biosafety level 3 (BSL3) or 4 (BSL4) research laboratory to apply for and receive a permit (Barry, 2005).
When the Whittier Narrows earthquake struck southern California, all physical sciences building utilities were lost, including emergency power (Pine, 1988). This meant that the initial cleanup and even damage assessment were hampered from the very beginning. Faculty volunteers fanned out to unplug essentially all electrical equipment in order to prevent electrical shorts and fires. Much of the building had electrical power restored later that same day, while broken pipes delayed water restoration until the next day. This was critical because fire officials were very concerned about fire safety in the absence of running water. The major utilities stored on the building roof were also severely damaged, including the air-handling system, hood fans, the distilled-water system, and a cooling tower.
Still, the extent of damage to utilities and other critical infrastructure can, in many scenarios, far exceed the damage experienced by Whittier Narrows. A comparative study on the seismic preparedness of health care facilities following disasters found that nonstructural damage was more common and widespread, and that failures of critical utilities (i.e., communications, power systems, and water systems) had the greatest impact on functionality (Mitrani-Reiser et al., 2013). Non–life-threatening damage can be disruptive in the response and recovery stages.
The National Institute of Standards and Technology published a planning guide that outlines the vulnerabilities of communities and offers guidance to making them more resilient (NIST, 2016). It describes dependencies throughout communities in energy infrastructure, in transportation, and in communication—all of the elements that an institution and its research enterprise will need to recover from a disaster.
The Southern San Andreas ShakeOut Scenario of 2008 examined the implications of a 7.8-magnitude earthquake in southern California (Jones et al., 2008). Its goal was to identify the physical, social, and economic consequences of a major quake in the region:
The major losses for this earthquake fall into four categories: building damages, non-structural damages, damage to lifelines and infrastructure, and fire losses. Within each category, the analysis found types of losses that are well understood—that have been seen in previous earthquakes and the vulnerabilities recognized but not removed—and types of losses that had been less obvious—where the type of failure is only recently understood or the extent of the problem not yet fully recognized. The study also found
numerous areas where mitigation conducted over the last few decades by state agencies, utilities, and private owners has greatly reduced the vulnerability. Because of these mitigation measures, the total financial impact of this earthquake is estimated to be “only” about $200 billion with approximately 1,800 fatalities. (Jones et al., 2008, p. 6)
The ShakeOut exercise predicted notional damage that impeded functionality at hospitals, where medical gas piping ruptured due to nonstructural wall failures and unrestrained medical equipment toppled, including medical gas and water filtration cylinders. Among the “expected damage” was significant disruption of utilities and other critical infrastructure. Water-line ruptures created grounds in buildings’ electrical distribution systems, tripping circuit breakers and causing near-total loss of functionality in some facilities; external supplies of potable water were lost for several days; and landlines, cellular systems, and private business exchanges were knocked out, leaving radio telephones as the only option in many locations. Most of the presumptive fires occurred in residential structures, but they were also reported in laboratories, chemical plants, and oil refineries. The ShakeOut scenario broadly tested southern California’s resilience to a major earthquake; while the after-action report does not reference specific impacts to academic research, the high concentration of research universities in the region implies that their potential role in fostering earthquake resilience for their own communities could be substantial.
The first ShakeOut exercise in 2008 was focused on southern California. Participation has grown to include many U.S. states and countries in seismically active areas of the world. In 2016, over 21 million people participated in the United States (ShakeOut, 2017). In the United States, ShakeOut drills actively engage educational institutions, including K–12 schools and universities. Many research universities with biomedical research labs and clinical care facilities, such as the University of California, Irvine (UCI), participate by encouraging faculty, students, and staff to practice drop-cover-hold to protect themselves from injury during shaking. Additionally, UCI tests its mass notification system, typically followed by a full-scale tabletop exercise with staff from the emergency operations center.6
A study of the 2012 ShakeOut drill participants in California involved collecting ShakeOut evaluation survey data from participating organizations. The surveyed organizations included 45 higher education institutions and 137 health care facilities (Halkia and Wood, 2015). The majority of health care organizations surveyed reported that they had disaster preparedness plans (Halkia and Wood, 2015). Broad ShakeOut participation by educational and health care sectors suggests that ShakeOut earthquake
6 Personal communication, Robert Simmons, UC Irvine, February 14, 2017.
preparedness drills are helping to foster earthquake resilience in California and could be adapted to academic research communities in other areas.
Utilities are perhaps the most notable elements of an institution’s critical infrastructure that suffer immediate insults from natural disasters, and they may also be subject to targeted damage from deliberate cyber intrusions. While any modern office is dependent on such systems as electricity, water, and heating, ventilation, and air-conditioning (HVAC), the loss of these kinds of assets becomes acutely felt in research laboratories. Animal rooms lose fresh water and air supplies; freezers storing irreplaceable samples lose power; humans who return to salvage these lives and assets place their own lives in danger. At the VA Medical Center in Sepulveda in 1971, water lines ruptured, resulting in immediate damage, including the loss of emergency power (Jones et al., 2008).
When Hurricane Ike raged through Houston, many critical city utilities failed, including water, sanitary sewer, power, and phones; the UTMB campus experienced failures in steam, chilled water, and some backup power due to switchgear flooding, and the campus network also failed (Goodwin and Donaho, 2010). Even though trailer-mounted auxiliary chillers and generators mounted in trailers had been pre-staged, they, like the rest of the area, became flooded. The campus’s most reliable generator was powered by natural gas, but the natural gas line from the mainland that supplied it failed. This cut the emergency power that would have been provided to a large research facility, resulting in the loss of decades’ worth of frozen specimens. The failure of generators was widespread, owing to such issues as clogged fuel filters, failed pneumatic controls, and the rerouting of a fuel delivery truck by unknown officials. Critically, funding challenges had delayed plans to raise low-lying generators to higher ground (Goodwin and Donaho, 2010).
The loss of electrical power runs like a theme throughout many of the disasters faced by academic research and medical campuses. Tropical Storm Allison’s floods quickly severed underground-routed electricity (RMS, 2001). Memorial Hermann Hospital, which housed a Level 1 trauma center, lost both primary and backup power. More than 1,000 patients were evacuated from the TMC during the night (RMS, 2001). NYU Langone was similarly forced to evacuate patients (more than 300) in darkness after emergency power was lost (Hartocollis, 2012). In the case of NYU Langone, most generators were located on high floors (although one was located in the basement of the Medical Sciences Building); fuel tanks were located in the basements of the various buildings on the campus (Ahlborn, 2017).
Clean water supplies also quickly became a major problem. NYU Langone’s drinking water comes from the city, which uses water tanks and pumps that feed into the East River. The storm surge situation sent water flooding back through the sewer line. According to one report, 15 million
gallons of contaminated water were deposited as a result of the storm damage, destroying utility lines and equipment such as MRI machines (Barbanel, 2016).
Even small-scale events can have impacts significant enough to warrant investigation, after-action reporting, and identification of preparedness needs. NYU Langone described an August 2015 fire in a transformer that affected six floors’ worth of laboratories (Ahlborn, 2017). The evacuation was not smooth, emergency contact information was not immediately available for each affected lab, and freezers were not labeled. A critical pressure point in this situation was also an issue at NYU Langone during Hurricane Sandy: communications. Communications infrastructure systems can fail in many ways. These systems are susceptible to direct physical insult as well as to cascade effects from the failure of other infrastructure, such as power, water, or transportation (NIST, 2016). Communications at Langone were challenging from perspectives both managerial (the handling of information) and technical (reaching people, given the power outages and the lack of sufficient contact information) (Ahlborn, 2017). Communications infrastructure is in some ways tenuous: cell phones are dependent on signal towers that cannot handle the massive volume of calls expected in mass emergencies, and flooding or other physical insults may ultimately put towers out of commission. Cell phones also require charging and thus a power source. Traditional analog phones still operate under such circumstances, but the trend toward replacing their traditional copper wires with fiber optics means that one of their major advantages—the ability to operate independent of power outages—is disappearing. During the 1994 Northridge earthquake, communication rapidly became a major problem for all affected hospitals (Chavez et al., 1996). In many cases, radios, telephones, and internal communications system were unusable. The public cellular system of the time was quickly overloaded. For many of the VA hospitals, sporadic connections via landlines were the only outside link in the first few hours of the disaster.
In 2010, a computer outage at Mount Sinai Medical Center resulted from an erroneous update sent from a software company to its users (Genes et al., 2013). The faulty software caused the computers to infinitely loop through shutdown and reboot, effectively causing an information technology (IT) shutdown and affecting all clinical computers until it was resolved the next day. Although this event affected a hospital and not a research facility, its lessons apply. After-action analyses determined that it was valuable to have telephones, cell phones, and two-way radios available and ready as a means to supplement the lost computer-based communication.
As this last example makes clear, a critical infrastructure element that warrants special attention because of its growing role in both academia and health care is IT.
Repeated flooding of Tulane’s server cable during Hurricane Katrina disrupted Internet access at the university, at least intermittently, for some time (Savoie et al., 2007). The storm surges from Hurricane Sandy completely flooded the basement floor on which the NYU Langone IT data center was located. Damage included 420 pieces of equipment, including high-performance computer clusters, fiber and copper cable connections for NYU Langone’s network, electrical panels, switches, batteries, computer air-conditioning units, and many other devices. The damage has been estimated at $33 million (Ahlborn, 2017). Most of the center had actually been relocated by the time of Hurricane Sandy; although what was there was destroyed and all functionality was immediately lost, no data or applications were lost permanently because most of it had been duplicated elsewhere.7
The trend toward cyber hacking creates a concerning new development as far as fiscal cost: that of ransom. Hospitals are becoming increasingly targeted by ransomware attacks that lock their data and demand a ransom payment to unlock the data (Wagstaff, 2016). Cyber attacks are also a tactic for animal rights actors. The most common cyber attacks related to animal rights have been the deletion of user accounts on targeted networks, e-mail demands flooding a company’s e-mail server with messages to the point of network shutdown (called denial of service), and otherwise forcing businesses to exhaust resources (Bailey et al., 2010).
Like any other entity in an affected community, the research enterprise can feel the effects of supply chain interruptions and the disruption of other critical services. The shipping of important materials to and from research centers is but one operation that may be interrupted. The University of California, Los Angeles, animal emergency plan notes that “road closures and inclement weather can prevent transportation of staff and supplies” and recommends that facilities that use research animals maintain sufficient food and water to support animals for 5 to 7 days (UCLA, 2013). The University of California, San Diego’s Continuity Plan for Labs goes further: the plan asks labs to state whether they have 1-month supply stock at all times for projects that may be affected if the supply chain is interrupted due to disaster (UCSD, 2016).
7 Personal communication, Neil Rambo, NYU Langone, August 10, 2016.
Academic research institutions planning for disaster resilience may be wise to implement recommendations like those cited above. In addition to interruptions in supply chains, the aftermath of disasters has shown that other problems with supply chains can occur as well. In some cases, a supply chain may remain in place after a disaster, but the demand for supplies may be too great to fill. In other cases, as documented in the aftermath of Hurricane Katrina, the demand for resources may be too great to process by understaffed organizations; a special report of the Senate Committee on Homeland Security and Governmental Affairs found that FEMA did not have enough staff to process requests for critical supplies even for humanitarian aid (Committee on Homeland Security and Governmental Affairs, 2006, p. 379), even though the number of requests received was later found to be fewer than initially thought (Holguín-Veras and Jaller, 2012). Iqbal et al. echo this in their Comparison of Disaster Logistics Planning and Execution for the 2005 Hurricane Season, noting that FEMA had “a wonderful operation plan on paper . . . [but] did not have the proper number of staff to implement the plan in a successful way” (Iqbal et al., 2007, p. 23).
Clinical laboratory services and biobanks are also important elements of modern medical care and rely on the same utilities as every other institution. Essential services may also be affected by disaster, even when the institution is not a hospital or other kind of direct patient care center. Hurricanes Katrina and Rita caused a 2-year interruption of in-state newborn screening services and a 6-month shutdown of clinical biochemical genetics laboratory services in Louisiana (Andersson et al., 2011). In 2012, the Harvard Brain Tissue Resource Center experienced a major failure in freezer functioning that destroyed approximately 139 postmortem human brains (Baird et al., 2013). Redundant alarm systems within the freezers had also failed.
Institutions often first feel the damage from major disasters in terms of the influence on human capital. Even before the monetary costs start adding up, an immediate personnel effect is recognized, and the numbers can be enormous: damage from Hurricane Katrina immediately displaced 84,000 students and 15,000 faculty members of higher education (Savoie et al., 2007). Many displaced faculty members ultimately lost their employment due to Hurricane Katrina or chose to move on to other institutions. At NYU Langone, facility closures resulted in the temporary displacement of almost all personnel; approximately 1,000 research faculty and staff were displaced for longer periods of time (Ahlborn, 2017). The uncertainties caused by disasters may be felt disproportionately by graduate students, trainees, and early-career investigators. These individuals are likely to have
fewer personal resources available to weather any disruption of employment and income. In addition, they may be facing significant disruptions that might impact their ability to meet necessary goals within a reasonable time frame for graduation or for promotion. Because of these factors, unless their institution provides rapid reassurance that their income will continue, or that there will be appropriate adjustments to their time line for achieving graduation or tenure, they may be forced to make difficult decisions to move on at a very disadvantageous time.
The effects of the institution’s handling of these delicate situations can cause further disruption to the human capital equation. An American Association of University Professors (AAUP) report on Hurricane Katrina stated:
The imperative that affected faculties be consulted and assume a meaningful role in making critical judgments reflects more than the values of collegiality; given the centrality of university faculties in the mission of their institutions, their meaningful involvement in reviewing and approving measures that vitally affect the welfare of the institution (as well as their own) becomes truly essential at such times. The Special Committee has been impressed with how deeply devoted the vast majority of faculty appeared to be to their institutions at a time of stress and, often, of significant personal economic loss. Administrators were able effectively to draw from that wellspring in dealing with the immediate aftermath of the disaster, in pulling their institutions together. (Savoie et al., 2007, p. 119)
Communication with faculty at Tulane was often very difficult, and many research faculty who were forced to temporarily move elsewhere felt very isolated. There was a strong feeling that once people started getting fired, things were being done in a very corporate fashion.8 The AAUP report elaborates:
Equally disturbing to the Special Committee was the general sense of betrayal that some faculty members said they initially felt, and continued to feel, because of the termination of their appointments. The chief harvest of the events of fall 2005, not only in the School of Medicine but also on the uptown campus, seems to be a pervasive mistrust. President Cowen and Board Chair Pierson, in their correspondence with the Association [AAUP], repeatedly and correctly pointed to the unprecedented disaster Hurricane Katrina represented for the entire city of New Orleans. (Savoie et al., 2007, p. 117)
Disasters can also influence an institution’s reputation and therefore its ability to recruit graduate students and researchers, with researchers choosing to go to other institutions where they might feel their futures are more
8 Personal communication, Prescott Deininger, Tulane University, October 4, 2016.
secure. For years following Hurricane Katrina, Tulane had significant difficulty recruiting senior faculty.9 In some cases other institutions recruited Tulane faculty away from Tulane, suspecting that the university was no longer functional.
The monetary costs of disasters for academic research institutions can be quite high and, indeed, can threaten the very survival of an institution. These costs may be the result of direct outlays for recovery elements such as emergency cleanup, repair, and rebuilding. The total cost of Tropical Storm Allison for the UTHSC-H Medical School alone was $205.4 million (Goodwin and Donaho, 2010). Only $50 million of this was covered by insurance. There were a variety of types of costs, including $25 million for emergency cleanup, business interruption, and temporary facilities; an estimated $68 million for remediation of critical areas including the entire animal care center (which had been demolished), the cyclotron, and the gross anatomy laboratory; and an estimated $7.4 million loss of research animals. With these figures in mind, it is not difficult to see how a disaster like Hurricane Katrina caused an estimated $500-600 million in direct damage costs to Louisiana’s institutions (Savoie et al., 2007). Institutional funds, insurance, state and federal money, and philanthropic donations may all play a role in mitigating these costs.
Incoming funding streams may also be adversely affected. Ongoing grants and contracts may be held up in the near- to mid-term aftermath of a disaster and even into the future as a result of the destruction of buildings or the research programs needed to administer the work. Tropical Storm Allison compromised an estimated $105 million in sponsored research awards at UTHSC-H (Goodwin and Donaho, 2010). At LSUHSC, the combination of property damage and lost revenues from Hurricane Katrina damage exceeded $200 million (Kuehn, 2013). As a result, LSUHSC was forced to downsize and lost one-fifth of its faculty and about $20 million in research funding. Revenue-generating elements of a research system, like university hospitals, may see sharp declines in incoming funds due to closed doors. Hurricane Katrina forced LSUHSC to suspend its operations, in part because the exodus of the city’s population dispersed students, staff, and faculty and led to a sharp decline in the potential patient pool, leading to an acute and immediate loss of revenue (Savoie et al., 2007). Louisiana’s public higher education institutions lost more than $150 million in revenue and tuition and absorbed $75 million in immediate budget cuts. Another assessment reported that Louisiana’s public institutions took a direct revenue loss
of $229 million. Comparable estimates emerged for private institutions (Savoie et al., 2007).
Comerio (2000) conducted a study to assess the potential financial loss, and the University of California, Berkeley’s capacity to recover after a damaging earthquake. It was found that 72 percent of all research income went to 25 research units, and 75 percent of research funds are expended in a total of 17 buildings which was one-third of the campus space. Based on the seismic ratings of 11 of these buildings, they would be closed for repairs for an extensive period in the event of an earthquake, and the potential loss suggests the viability of the ongoing research efforts would be threatened.
The total economic costs to institutions from the greatest disasters can be staggering. Tulane’s uptown campus reported property damage and operating losses for fiscal year 2005–2006 in excess of $450 million (Savoie et al., 2007). NYU Langone estimates its losses at more than $1.46 billion (Ahlborn, 2017). The dispersion of Langone’s patients also dispersed a major source of revenue and created the risk that they might not return. Ultimately, the patients did return, along with hospital revenue (Gooch, 2015). The loss of hospital patients at academic research institutions raises the additional problem of potential disruption of clinical trial research.
Litigation and other legal-related claims are another potential effect of disaster, but there is limited evidence that they are a major influence on academic research in most cases. Potential legal sequelae should, however, be considered and planned for at the institutional level as part of resilience initiatives. In the wake of disasters, institutions may face such legal actions as negligence assertions, unpaid insurance claims, and salary disputes. Loss of life is perhaps the most significant legal issue to arise from disasters, but the committee did not find examples of this in the context of events that have affected the academic biomedical research community.
Because of the enormous impact that Hurricane Katrina had on affected universities’ ability to remain operational, this disaster represents the most extreme example of the legal ramifications of a disaster in the context of the academic biomedical research community. Indeed, research buildings were damaged, and some were closed due to safety—or possibly liability—concerns. At Tulane, for instance, the closure of laboratories concerned researchers because it made it that much harder to recover their laboratories and research programs.10 In some cases, the complicated post-damage decisions to furlough personnel, including research faculty, were
not related to the research infrastructure itself or to loss of grant funding, but rather to the loss of hospital income:
Asserting the lack of an alternative to the prompt placement of scores of faculty on furlough, General Counsel Lamonica wrote that at the [LSU] Health Sciences Center there simply was not enough money to continue the same levels of employment as before the storms; there was not employment which produced the funds. Much of the faculty of [the Health Sciences Center] is devoted primarily to clinical duties, or to research which involves the treatment of patients in clinical settings. The patients were largely gone from New Orleans for months after the storms, and they still have not, and may never, return in the same numbers as before. LSU did not need, and could not afford to maintain, a faculty large enough to service a city of almost half a million people after the population dropped to something much less than that. Without work for them to do or money with which to pay them, LSU had no choice but to issue the furloughs it did. (Savoie et al., 2007, p. 72)
The LSUHSC furloughs and those of other universities were the subject of an after-action assessment by AAUP (Savoie et al., 2007). The report states that the appropriate use of “force-majeure,” a legal doctrine that releases a party from a contractual obligation due to unforeseen circumstances, which LSUHSC invoked as the basis for faculty furloughs, is ultimately a matter for judicial adjudication. The report notes, however, that even in a case where a formal ruling upholds the legality of such an action, “it is unlikely to undo the damage to the status and the careers of many of those faculty members most directly affected” and that its implementation is equally a consideration of academic policy and sound practice as it is a legal consideration (Savoie et al., 2007).
Lawsuits related to insurance claims were also seen after Hurricane Katrina. In one instance, a property insurer went to court to claim that its liability to Tulane for the school’s Hurricane Katrina-related losses was limited to $100 million, arguing that its policy excluded flood losses (Mcleod, 2006). These suits can tie up universities in court for years. Again, this is an institutional burden and not one by which the research enterprise is likely to be directly impacted.
Animal rights activity has also created legal headaches for research centers. Infiltration by activists of a Silver Spring, Maryland, laboratory in the early 1980s led to a 13-year saga with extensive repercussions, including the first state conviction of a researcher for animal cruelty (later reversed), the abandonment of some research by the laboratory, and lawsuits at the state and federal level seeking custody of the animals in question (Bailey et al., 2010).
Although researchers and the institutions they work for bear the brunt of disasters’ impacts, the organizations that sponsor their research feel it, too. Sponsors may be private philanthropies, or federal agencies such as the NIH or NSF. All have a vested interest in seeing their investments succeed.
In direct response to Hurricane Sandy, Congress appropriated $148.8 million to NIH. Nearly all of this—$147 million—was for grantee research programs (Jarman, 2015). The appropriation of emergency funding can provide significant relief, and yet its disbursement can also create an unaccounted-for administrative burden. Disasters may make it necessary for research sponsors to increase staff hours directed at managing affected research. Staff time is money, and it is also time not spent on other projects. At NIH, these staff hours may be spent on redirecting funding within a grant to pay salaries and benefits, assisting with animal welfare issues, providing extensions, or publishing new funding opportunities targeted to institutions in affected areas (Bundesen, 2016). Sponsors may also have to pay again for work and equipment already paid for once before. As a result of Hurricane Sandy, NIH reinvested millions to reinitiate research programs: $75.9 million for research restoration; $49.2 million for construction and renovation; and $1.7 million for safety training (Bundesen, 2016). NIH does not appear to track the amount of funding that it has spent as a result of disasters or what proportion comes from existing resources versus additional appropriations. The committee was also unable to determine the extent of NSF’s investment in disaster response and recovery efforts for its grantees.
The major disasters discussed in this chapter at times required researchers or their funders to make difficult choices. Sometimes this meant that a researcher or funder felt compelled to drop a research project altogether. At other times it meant that research was revamped or redirected. This is one impact that might actually be viewed as a benefit to the research.
For instance, Gordon Fishell of NYU Langone wrote that NIH allowed him to rewrite some of the aims in his open grants, allowing him to redirect the money toward new and different projects (Fishell, 2013). The redirection of research funding may occur not only because a research laboratory has actually been decimated and no longer able to perform its function, but simply because of a perception on the part of the sponsor that this is so. After Hurricane Katrina, some grant reviewers began to question
whether Tulane was sufficiently recovered and whether it was appropriate to do some sorts of research in that environment post-disaster; this put an increased burden on scientists to counter misconceptions.11 This affects not only the researchers and their academic research institutions, but also the research sponsors in that it narrows their own opportunities for funding good work.
While it may be possible to quantify the number of freezers that went offline, the number of animals lost, or the number of researchers affected by a disaster, it is essentially impossible to quantify the ripple effects of these losses.
When the power went out at NYU Langone, Mary Helen Barcellos-Hoff’s researchers trekked up and down eight flights of stairs to save specimens that had been stored in now-warming freezers (Sifferlin, 2013):
We work on those for years, and those are the most precious things in our laboratory. We can always replace equipment, we can always replace reagents, we can even replace mice, but replacing an experiment that took you 2 years to complete is really tedious. (Sifferlin, 2013)
Ultimately, Sifferlin’s laboratory was able to get some experiments quickly back on track, but she had to terminate others. Identifying the lost benefits from that which never came to be is perhaps impossible. Impacts on the science are largely intangible. Barcellos-Hoff said that her team could never account for the effects that the dramatic change in available light to the mice for more than a week may have had on their results.
Impacts on the science may be real or potential. Michael R. Blackburn, a lung disease researcher at UTHSC-H, lost 200 cages of mice and faced a 6-month delay before becoming functional and productive again (Kuehn, 2013). Despite the hardship of such instances, some researchers describe a sort of forced impetus to do things better the second time around. Blackburn said that because rebuilding mice colonies would take time, his laboratory had to diversify its scientific approach and began studies using cell cultures and human tissue. It is impossible to know the true effects of this shift, but they may not have all been negative—perhaps new discoveries were made in this way, or new techniques developed. Berger of NYU Langone described how the devastation of his laboratory essentially gave him a chance to redo his study and to do it better the second time around.12
12 Personal communications, Jeffrey Berger, NYU Langone, July 14, 2016.
This review of disasters that have affected the academic biomedical research community reveals the kinds of major hazards that tend to affect them and the most vulnerable elements of their operations. The number of multi-billion-dollar disasters is on the rise, and the recent disasters and their consequences—from the loss of intellectual property in many forms that may be extremely difficult, perhaps impossible, to replicate to the general economic loss given the density of economic value tied up in the academic biomedical research community—is a signal for every research institution to begin discussions about how to prepare for the worst, as the worst clearly can happen (McNutt and Leshner, 2013).
Conclusion: For the events reviewed, there are common impacts, some of which are unique to the academic biomedical research community: destruction to the physical laboratory workspace, loss of supportive utilities and specialized equipment, and deaths of research animals have been among the most salient immediate impacts. Further effects on the lives, livelihoods, and contributions to science by the researchers responsible for implementing grant awards, and to the financial viability of the academic research institutions that employ them, are extensive.
The most destructive hazards have been weather related: rains and storm surges from hurricanes and other major storms have caused substantial damage. Earthquakes have also taken a considerable toll, although the worst expected damage has not been experienced yet in modern times. Animal rights terrorism has been the most noteworthy among the intentional acts that have disrupted research operations and has often resulted in property damage from fire and other forms of vandalism. Academic research institutions have not been the target of chemical, biological, radiological, or nuclear terrorism, but programs in place to prevent infiltration of these centers are indicative of security concerns that must be taken seriously. Even if they were not targeted for destruction but only as a resource for the theft of stored materials, their operations could be suspended while investigations ensue.
Some disasters come with virtually direct notice: an impending hurricane almost always provides at least a few days’ warning, although its path can shift unexpectedly. Others provide no notice: an earthquake, a nuclear plant meltdown, an act of arson. In between exist the kind of disasters that occur without direct notice but which may be predictable based on risk assessment: an academic research institution located in a high-risk earthquake zone or floodplain, an information technology system lacking optimized firewalls to cyber intrusions. Old risk assessments may need to be updated: the August 2016 floods in Louisiana, which flooded houses and businesses that had never seen floodwaters and were outside FEMA’s 100-
year floodplains, illustrate how areas that perhaps considered themselves in safe zones can still be subject to unanticipated disasters (Terrell, 2016). Following a disaster, institutions can develop an after-action report to look at how the disaster affected all aspects of the institution and document corrective actions aimed at improving resilience (NCCPS, 2016).
Academic research institutions are hubs of education, employment, economic productivity, and biomedical progress. In 2013, the 62 research institution members of the Association of American Universities had combined operating budgets of $152 billion, issued 3,460 patents and executed 3,068 licensing agreements, and initiated 479 start-up companies (AAU, 2017) The disruptions to the work of the academic biomedical research community described herein therefore reach beyond even the researchers, students, their institutions, and their sponsoring agencies, to affect their local economies. The emphasis of the academic biomedical research community is, by definition, on research, and losses in research-related assets translate into the loss of patents, new start-up companies never developed, or stunted economic growth in the private sector. An interruption of the research process has a multiplier effect in the economic sector (Comerio, 2000). Its further scientific impacts on society are less calculable, but just as important. Insurance cannot be purchased for the intangibles, but the intangibles may be where the academic biomedical research community and the nation by extension would suffer the greatest loss.
Conclusion: It is not simple, or perhaps possible, to quantify how the loss of research samples and data translates into scientific losses for the public. Given universities’ central role in basic research and also in applied research, delayed or arrested discoveries are a likely consequence.
Nongovernmental organizations, including academic research institutions, also play important roles in the public health system, providing health services, education, and research capacity, which may be interrupted during and after disasters (IOM, 2015). Public health academia, for instance, engages its community via community service to state and local health departments and by the provision of policy guidance to inform public debates (IOM, 2003).
Finally, while impact may be somewhat narrowly defined as the kinds of physical, fiscal, or scientific destructions or disruptions that have occurred in the events reviewed here, disasters can also lead to the promulgation of laws, regulations, or policies designed to prevent or mitigate future occurrences. As described, the Sylmar–San Fernando earthquake of 1971 triggered the development of new safety standards and programs, special codes for the design and construction of hospitals and essential facilities, the revision of
the Uniform Building Code, and the promulgation of a statute directed at the VA preparedness for fires, earthquakes, and other natural disasters. The Animal Enterprise Terrorism Act followed decades of animal rights extremism incidents. Special appropriations, such as the DRAA, while temporary, may also be a reaction to disasters and may affect the academic biomedical research community’s response, recovery, and mitigation capacities.
Conclusion: Governing bodies, including local, state, and federal governments, and professional associations that promulgate standards, can affect the disaster resilience of the academic biomedical research community through the development of laws, regulations, policies, standards, requirements, and other instruments.
AAU (Association of American Universities). 2017. Economic impacts of AAU universities. https://www.aau.edu/research/article.aspx?ID=9266 (accessed March 6, 2017).
Ahlborn, L. 2017. New York University response to committee questions. Available by request through the National Academies’ Public Access Records Office.
Andersson, H. C., W. Perry, B. Bowdish, and P. Floyd-Browning. 2011. Emergency preparedness for genetics centers, laboratories, and patients: The Southeast Region Genetics Collaborative strategic plan. Genetics in Medicine 13(10):903–907.
Bailey, M. R., B. A. Rich, and B. T. Bennett. 2010. Crisis planning to manage risks posed by animal rights extremists. ILAR Journal 51(2):138–148.
Baird, P. M., F. M. Benes, C. H. Chan, C. B. Eng, K. H. Groover, Z. Prodanovic, M. RawleyPayne, R. Kizza, C. Hia, and A. Abouhamze. 2013. How is your biobank handling disaster recovery efforts? Biopreservation and Biobanking 11(4):194–201.
Barbanel, J. 2016. NYU’s hospital super-stormproofs itself: Four years after Sandy, new measures at NYU Langone Medical Center guard against the effects of storms and hurricanes. Wall Street Journal, October 28.
Barry, A. 2005. Report of pneumonic tularemia in three Boston University researchers. http://cbc.arizona.edu/sites/default/files/Boston_Univerity_Tularemia_report_2005.pdf (accessed September 7, 2016).
Bartholomew, D. 2016. Sylmar-San Fernando earthquake: 45 years ago Tuesday, 64 killed. Los Angeles Daily News, February 8.
Bisceglio, P. 2012. Of mice and floods: Researchers at NYU pull together to save lab animals. Our Town. http://www.ourtownny.com/of-mice-and-floods-researchers-at-nyu-pull-to-gether-to-save-lab-animals/ (accessed March 2, 2017).
Bundesen, L. Q. 2016. Strengthening the disaster resilience of academic biomedical research communities. Presentation to the Committee on Strengthening the Disaster Resilience of Academic Research Communities. Washington, DC, March 2. http://www.nationalacademies.org/hmd/~/media/Files/Activity%20Files/PublicHealth/Academic%20Resilience/Liza%20Bundesen%20NIH%20Sponsor%20Presentation.pdf (accessed October 17, 2016).
CREW (Cascadia Region Earthquake Workgroup). 2013. Cascadia subduction zone earthquakes: A magnitude 9.0 earthquake scenario. Washington Division of Geology and Earth Resources Information Circular 116.
Chavez, C. W., and B. Binder. 1996. A hospital as victim and responder: The Sepulveda VA Medical Center and the Northridge Earthquake. Journal of Emergency Medicine 14(4):445–454.
Church, L. 2013. Scientists displaced by Sandy take refuge at Rockefeller. http://benchmarks.rockefeller.edu/2013/04/19/scientists-displaced-by-sandy-take-refuge-at-rockefeller/ (accessed September 7, 2016).
Comerio, M. C. 2000. Economic benefits of a disaster resistant university: Earthquake loss estimation for UC Berkeley. Berkeley, CA: Institute of Urban and Regional Development.
Committee on Homeland Security and Governmental Affairs. 2006. Hurricane Katrina: A nation still unprepared. Special Report 109-322. 109th Congress, 2nd Session. https://www.congress.gov/109/crpt/srpt322/CRPT-109srpt322.pdf (accessed December 9, 2016).
Cossins, D. 2012. NYC science stunned by Sandy. The Scientist, November 2. http://www.the-scientist.com/?articles.view/articleNo/33109/title/NYC-Science-Stunned-by-Sandy/ (accessed September 6, 2016).
FASEB (Federation of American Societies for Experimental Biology). 2014. The threat of extremism to medical research: Best practices to mitigate risk through preparation and communication. https://www.faseb.org/Portals/2/PDFs/opa/2014/Animal%20Extremism%20Report%20Final.pdf (accessed September 6, 2016).
FEMA (Federal Emergency Management Agency). 2016. National Disaster Recovery Framework, 2nd ed. https://www.fema.gov/media-library-data/1466014998123-4bec8550930f774269e0c5968b120ba2/National_Disaster_Recovery_Framework2nd.pdf (accessed September 6, 2016).
Fishell, G. 2013. Hurricane Sandy: After the deluge. Nature 496:421–422.
FSAP (Federal Select Agent Program). 2016. 2015 Annual report of the Federal Select Agent Program. http://www.selectagents.gov/resources/FSAP_Annual_Report_2015.pdf (accessed September 7, 2016).
Genes, N., M. Chary, and K. Chason. 2013. An academic medical center’s response to widespread computer failure. American Journal of Disaster Medicine 8(2):145–150.
Gooch, K. 2015. Returning an organization to profitability: Q&A with NYU Langone Medical Center CFO Michael Burke. Becker’s Hospital Review. http://www.beckershospitalre-view.com/finance/returning-an-organization-to-profitability-q-a-with-nyu-langone-medi-cal-center-cfo-michael-burke.html (accessed March 2, 2017).
Goodwin, B. S., and J. C. Donaho. 2010. Tropical storm and hurricane recovery and preparedness strategies. ILAR Journal 51(2):104–119.
Halkia, G., and M. M. Wood. 2015. The great California ShakeOut: Findings from the 2012 California earthquake drill. PowerPoint presented at Southern California Earthquake Center Annual Meeting, Los Angeles, CA, September 12–15. https://www.scec.org/meetings/2015am (accessed March 2, 2017).
Hartocollis, A. 2012. A flooded mess that was a medical gem. New York Times, November 10.
Heins, S. 2014. NYU medical professors suffer pay cut thanks to Hurricane Sandy damage. Gothamist. http://gothamist.com/2014/07/19/nyu_profs_suffer_pay_cut_after_alle.php (accessed September 7, 2016).
Holguín-Veras, J., and M. Jaller. 2012. Immediate resource requirements after Hurricane Katrina. Natural Hazards Review 13:117–131.
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.
IOM (Institute of Medicine). 2003. The future of the public’s health in the 21st century. Washington, DC: The National Academies Press.
———. 2015. Healthy, resilient, and sustainable communities after disasters: Strategies, opportunities, and planning for recovery. Washington, DC: The National Academies Press.
Iqbal, Q., K. Mehler, and M. B. Yildirim. 2007. Comparison of disaster logistics planning and execution for 2005 hurricane season. Midwest Transportation Consortium. http://www.intrans.iastate.edu/reports/disaster-management.pdf (accessed March 6, 2017).
Jarman, G. L. 2015. New York University School of Medicine budgeted costs that were appropriate and claimed allowable Hurricane Sandy Disaster Relief Act funds. Office of the Inspector General. https://oig.hhs.gov/oas/reports/region2/21402011.pdf (accessed October 19, 2016).
Jones, L. M., R. Bernknopf, D. Cox, J. Goltz, K. Hudnut, D. Mileti, S. Perry, D. Ponti, K. Porter, M. Reichle, H. Seligson, K. Shoaf, J. Treiman, and A. Wein. 2008. The ShakeOut scenario: U.S. Geological Survey open-file report 2008-1150 and California Geological Survey preliminary report 25, Version 1.0. https://pubs.usgs.gov/of/2008/1150/of2008-1150small.pdf (accessed March 6, 2017).
Kaiser, J. 2005. Displaced researchers scramble to keep their science going. Science 309: 1980–1981.
Kohler, C. E., and W. E. Gray. 2014. Earthquake preparedness for laboratories. Indiana University. http://ehs.iu.edu/docs/Earthquake-Preparedness-for-Laboratories.pdf (accessed April 7, 2017).
Kuehn, B. M. 2013. Rebuilding after disaster: Recovering research takes time, planning. JAMA 309(2):123–124.
Lanier, S. M. 2005. Hurricane Katrina—Summary of impact on LSUHSC and focus on LSUHSC department of pharmacology. Louisiana State University Health New Orleans. http://www.medschool.lsuhsc.edu/pharmacology/katrina_report.pdf (accessed March 2, 2017).
Law, T. 2011. Canterbury Uni invites staff to resign. Stuff. http://www.stuff.co.nz/the-press/news/5726614/Canterbury-Uni-invites-staff-to-resign (accessed October 27, 2016)
Mandavili, A. 2013. One year after Hurricane Sandy, kindness buoys New York labs. Spectrum. https://spectrumnews.org/news/one-year-after-hurricane-sandy-kindness-buoys-new-york-labs/ (accessed September 7, 2016).
Matthews, S. 2012. Need for mouse backup, but costs present challenges. Nature Medicine 18(12):1724–1725.
Maybauer, M. O., M. Megna, S. Asmussen, and D. M. Maybauer. 2011. Hurricane evacuations of the University of Texas Medical Branch at Galveston. In Recent hurricane research: Climate, dynamics, and societal impacts, edited by A. Lupo. Rijeka, Croatia: InTech.
Mcleod, D. 2006. Allianz fights school’s claim over damage from Katrina. Business Insurance, April 9. http://www.businessinsurance.com/article/20060409/ISSUE01/100018635/allianz-fights-schools-claim-over-damage-from-katrina (accessed September 6, 2016).
McNutt, M., and A. Leshner. 2013. Preparing for disasters. Science 341(6146):592.
Mitrani-Reiser, J., C. Jacqes, T. Kirch, with M. Comerio, S. Mahin, W. T. Holmes, S. Giovinazzi, J. McIntosh, and T. Wilson. 2013. A comparative study on the seismic preparedness of hospitals following recent damaging earthquakes, Report to the California Seismic Safety Commission, Pacific Earthquake Engineering Research Center. http://www.seismic.ca.gov/meeting_info/July11_2013/05-Item%20V%20Hospital%20Evac%20Survey.pdf (accessed January 21, 2017).
NCCPS (National Center for Campus Public Safety). 2016. National Higher Education Emergency Management Program needs assessment. http://www.nccpsafety.org/news/articles/national-higher-education-emergency-management-needs-assessment (accessed January 21, 2017).
NIST (National Institute of Standards and Technology). 2016. Community resilience planning guide for buildings and infrastructure systems, volume I. NIST Special Publication 1190. https://www.nist.gov/sites/default/files/community-resilience-planning-guide-volume-1_0.pdf (accessed March 6, 2017).
Nolen-Hoeksema, S., and J. Morrow. 1991. A prospective study of depression and posttraumatic stress symptoms after a natural disaster: The 1989 Loma Pieta earthquake. Journal of Personality and Social Psychology 61(1):115–121.
Ortolon, K. 2009. Rising from the storm: UTMB struggles to recover after Hurricane Ike. Texas Medicine 105(12):12–20.
Pine, S. H. 1988. Safety in chemical laboratory. Journal of Chemical Education 65(4):A98–A99.
RMS (Risk Management Solutions). 2001. Tropical Storm Allison, June 2001: RMS event report. http://forms2.rms.com/rs/729-DJX-565/images/tc_2001_tropical_storm_allison.pdf (accessed October 20, 2016).
Savoie, E. J., R. O’Neil, and D. Rabban. 2007. Report of an AAUP special committee: Hurricane Katrina and New Orleans universities. Academe 93(3):59–126.
Schnirring, L. 2008. Texas A&M fined $1 million for lab safety lapses. Center for Infectious Disease Research and Policy. February 21, 2008. http://www.cidrap.umn.edu/news-perspective/2008/02/texas-am-fined-1-million-lab-safety-lapses (accessed April 7, 2017).
Schultz, C. H., K. L. Koenig, and R. J. Lewis. 2003. Implications of hospital evacuation after the Northridge, California earthquake. New England Journal of Medicine 348(14):1349–1355.
ShakeOut. 2017. Great ShakeOut earthquake drills. http://www.shakeout.org (accessed March 2, 2017).
Sifferlin, A. 2013. Three months after Sandy: Inside the rebuilding of New York University’s research labs. Time: Healthland. http://healthland.time.com/2013/02/14/three-months-after-sandy-inside-the-rebuilding-of-new-york-universitys-research-labs/ (accessed November 10, 2016).
Simon, F. 2006. Doctors without quarters. Tulanian. https://www2.tulane.edu/news/tulanian/doctors_without_quarters.cfm (accessed November 10, 2016).
Terrell, D. 2016. The economic impact of the August 2016 floods on the state of Louisiana. Lewis Terrell and Associates, LLC. http://gov.louisiana.gov/assets/docs/RestoreLA/SupportingDocs/Meeting-9-28-16/2016-August-Flood-Economic-Impact-Report_09-01-16.pdf (accessed February 12, 2017).
Tierney, K. J. 1997. Business impacts of the Northridge earthquake. Journal of Contingencies & Crisis Management 5(2):87–97.
UCLA (University of California, Los Angeles). 2013. UCLA animal emergency plan. http://surgery.ucla.edu/workfiles/research/Animal_Disaster_Plan_Template.pdf (accessed December 9, 2016).
UCSD (University of California, San Diego). 2016. Continuity plan for labs. https://blink.ucsd.edu/_files/finance-tab/accountability/Lab_Question_Set.pdf (accessed December 9, 2016).
USGS (U.S. Geological Survey). 2016. Historic earthquakes, San Fernando, California 1971. https://earthquake.usgs.gov/earthquakes/states/events/1971_02_09.php (accessed on December 11, 2016).
Wagstaff, K. 2016. Big paydays force hospitals to prepare for ransomware attacks. NBC News. http://www.nbcnews.com/tech/security/big-paydays-force-hospitals-prepare-ransomware-attacks-n557176 (accessed September 6, 2016).
Wertz, J. 2015. Years after earthquake, Oklahoma College still shaken by cracked budgets and broken buildings. StateImpact Oklahoma. https://stateimpact.npr.org/oklahoma/2015/07/16/years-after-earthquake-oklahoma-college-still-shaken-by-cracked-budgets-and-broken-buildings/ (accessed October 27, 2016).