Evaluation of a Site-Specific Risk Assessment for the Department of Homeland Security's Planned National Bio- and Agro-Defense Facility in Manhattan, Kansas (2010)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

2 Consideration of Site-Specific Risk and Mitigation Factors A seminal risk assessment report Risk Assessment in the Federal Government: Managing the Process by the National Research Council (NRC, 1983) and a recent publication on Quantitative Microbial Risk Assessment (Haas et al., 1999) define risk assessment as “the qualitative or quantitative characterization and estimation of potential adverse health effects associated with exposure of individuals or populations to hazards (materials, situations, or physical, chemical, or microbial agents).” Risk assessment can be used as a tool to estimate the probability of an adverse event in a specified period. Risk management—which takes into account risk assessment and engineering, economic, legal, and political issues—is process of controlling risks, weighing alternatives, and selecting appropriate actions (NRC, 1983; Haas et al., 1999). The Department of Homeland Security’s (DHS’s) site-specific risk assessment (SSRA) identified the following objectives for its report: • “To inform the requisite design and engineering controls for the National Bio- and Agro-Defense Facility (NBAF). • “To inform emergency response plans with city, regional, and state officials in the event of an accidental release of a pathogen. • “To assist in the development of the operational protocols needed to safely and securely operate the facility. • “To include an integrated set of analyses using plume modeling and epidemiological impact modeling.” The NRC report Review of the Department of Homeland Security’s Approach to Risk Analysis recently evaluated the quality of the current DHS approach to estimating risk and applying those estimates in its many management, planning, and resource-allocation activities (NRC, 2010a). This report’s evaluation of the SSRA is consistent with the findings, conclusions, and recommendations of that report. 19 

EVALUATION OF THE NBAF SITE-SPECIFIC RISK ASSESSMENT 20 SITE-SPECIFIC CHARACTERISTICS THAT AFFECT RISK Factors Identified by the Department of Homeland Security High-Risk Tornado Area On average, 1,000 tornado events (ranging from 700-1,400) occur annually in the contiguous United States (Lott et al., 2000). Cold dry fronts coming over the Rocky Mountains in the spring travel south or east and then collide with moist air coming north from the Gulf of Mexico and cause severe thunderstorms (NOAA, 2010). The thunderstorms can develop into tornadic storms. A vast majority of U.S. tornadoes occur in the region of the Midwest that extends from northwest Texas to the Great Lakes in an area called Tornado Alley, which includes Manhattan, Kansas. Vicinity as a Hub for Transportation of Cattle and Other Livestock Key site-specific characteristics with respect to conducting research on foot-and-mouth disease (FMD) are the large number of susceptible animals in the immediate vicinity and the percentage of the U.S. herd in the region. Those issues are addressed by examining the animal population within a radius of 200 miles around Manhattan, Kansas. The area in question is primarily a beef cattle region that includes some dairy cattle and swine operations (USDA- NASS, 2009). A 200-mile radius includes most of Kansas, large parts of Nebraska and Missouri, western Iowa, and northern Oklahoma. Roughly 9.5% of the U.S. cattle inventory is in the area, and it includes substantial regions of swine production and several meat packing plants. In 2009, the total U.S. inventory for cattle and swine were 94.5 million head and 66.7 million head, respectively (USDA-NASS, 2009). The SSRA recognizes that the area studied is “a hub of animal movement for the entire United States” and that “in reality, as infected animals are moved throughout the country, pockets of infection would be expected to occur great distances from the initial focus of infection” (page 176). If FMDv escapes from the NBAF, it is likely to cause a widespread and economically devastating outbreak of FMD because of the facility’s proximity to a hub of animal production and transportation. It could cause major damage to the credibility of all federal agencies involved,1 especially if rigorous and robust mitigation strategies have not been put in place. Such an event would probably have major consequences for future operation of the facility and the ability to work with FMDv on the mainland. Given that FMD is a highly contagious disease and that the chance of FMDv escape is not zero, it is essential that rigorous and robust regional and national mitigation strategies that address an extensive outbreak of FMD be put into place before the NBAF opens. Factors Not Identified by the Department of Homeland Security The Plum Island, New York, and Manhattan, Kansas, locations differ in proximity to human and animal populations. The Plum Island Animal Disease Center (PIADC) is on an uninhabited island, whereas the NBAF will be in a suburban-rural area that has a large human 1 Credibility could also be jeopardized in the event of escape of a zoonotic pathogen if the infected workers lacked access to top-quality infectious disease expertise and facilities. 

CONSIDERATION OF SITE-SPECIFIC RISK AND MITIGATION FACTORS 21 population and that is very close to susceptible animals. Those characteristics introduce site- specific risks that are important to consider for the NBAF but were not included in the SSRA. Proximity of Metropolitan Area The NBAF site in Manhattan, Kansas, is in a corner of the Kansas State University (KSU) campus directly adjacent to student housing, campus recreational facilities, and the university football stadium, which has a capacity of 52,200. Figure 2-1 shows the proximity of the planned facility to the local population at KSU. The large population that gathers for football games and other events is potentially susceptible to infections with a zoonotic agent; additionally, the presence of large numbers of vehicles during public events increases the odds that some will transport a released pathogen outside of the area, increasing the potential for spread and complicating mitigation. Although many high-containment laboratories are located in highly-populated areas, the SSRA failed to adequately account for such populations and the large animal aspects of the NBAF’s work in its risk analysis. Figure 2-1 Map of the planned NBAF and the surrounding Manhattan, Kansas, vicinity. Kansas State University occupies most of the area shown in the map. The planned site of the NBAF is outlined in red. SOURCE: Riley County, October 2010. Map used with permission. Proximity of Other Animal Facilities The NBAF will be adjacent to the KSU Biosecurity Research Institute (BRI) and the KSU College of Veterinary Medicine. The purpose of the proximity is to facilitate interaction of 

EVALUATION OF THE NBAF SITE-SPECIFIC RISK ASSESSMENT 22 scientists in the NBAF with scientists in those two facilities, other KSU scientists, and those in the recently established DHS Center of Excellence for Emerging and Zoonotic Animal Diseases. Locating the NBAF on the KSU campus would facilitate opportunities for training in foreign animal and zoonotic diseases. It is expected that the NBAF will continue the PIADC responsibility of providing foreign animal disease training to qualify veterinarians as foreign animal disease diagnosticians. The NBAF will also provide opportunities for training KSU graduate students and postdoctoral research associates. However, access to high-containment laboratories and animal facilities requires extensive training and certification, and this will limit the opportunities for those in training positions. Proper training, education, and monitoring of skills would reduce the risk of human error causing a release from the NBAF. Yet increasing the number of scientists, students, and postdoctoral scholars who have access to high-containment facilities would in all likelihood increase the risk of accidental spread of infectious agents via fomites, the primary identified risk in the SSRA. One overlooked characteristic that would contribute to risk is the NBAF’s proximity to KSU’s College of Veterinary Medicine. Many susceptible animals at the college would be adjacent to the NBAF. Some of the animals may be housed on the campus, and others may travel to the campus for diagnosis and treatment and return home with their owners. If an agent were released from the NBAF, those animals could readily become infected and serve as a conduit for amplification and transport. The National Centres for Animal Disease in Winnipeg, Canada, and the Australian Animal Health Laboratory in Geelong, Australia, have comparable biosafety level 3 agriculture (BSL-3Ag) and BSL-4 animal housing capabilities but on a smaller scale compared to the proposed NBAF. With regard to the risks associated with FMD research, there are major differences between these facilities and the proposed NBAF. The laboratory in Winnipeg conducts research on FMD, but the laboratory in Geelong is not allowed to conduct research on FMD. The Winnipeg laboratory has 14 BSL-3Ag rooms for housing FMD-infected animals. Thirteen of the rooms can each hold two head of cattle, and one room can hold four head of cattle. However, the Winnipeg laboratory typically does not have more than 8 animals infected with FMD at any one time. The laboratory also has one BSL-4 animal room which can hold two cattle. The nearest FMD-susceptible animals are in a zoo approximately 6-8 km away from the Winnipeg facility, and the nearest susceptible agricultural species are estimated to be 8-10 km from the Winnipeg facility2 (Soren Alexandersen, National Centres for Animal Disease, personal communication, September 22, 2010). SITE-SPECIFIC FACTORS THAT AFFECT MITIGATION PLANS Clinical Capability for Handling BSL-4 Infections The NBAF will handle zoonotic pathogens that cause disease in humans, including BSL- 4 agents that do not have available treatment and can cause severe illness and death in humans. Clinical facilities and physician expertise will be important for responding to occupational 2 The 8-10 km distance to the nearest susceptible agricultural species is significant because a radius of at least 10 km is the minimum distance that would be established for a control area around an FMD-infected premises in Canada (CFIA, 2008). A control zone with a radius of at least 10 km (6.2 miles) would also be established around each infected premises in the United States (Jon Zack, USDA-APHIS-VS, personal communication, January 26, 2010). 

CONSIDERATION OF SITE-SPECIFIC RISK AND MITIGATION FACTORS 23 exposures (such as those leading to laboratory-acquired infections) and to releases outside the facility (such as those leading to community-acquired infections). That will require the presence of a sophisticated occupational medicine program to assess and manage potential exposures as varied as needlesticks and fomite and aerosol exposures. This may include hospitalization, isolation, and critical care of infected people and quarantine of contacts. The SSRA provides a detailed analysis of the capabilities of the sole medical center that is close to the NBAF. Mercy Regional Health Center (MRHC), the major medical provider in the area, has a 150-bed hospital (normal census, 75-100 inpatients) that has nine isolation rooms (with negative pressure) in various locations around the facility, and it operates outpatient and related services in the municipality. The hospital medical staff includes a single infectious disease physician. There is also an Occupational Health Services group that provides services to KSU and BRI, and offers services such as respirator-fit testing and baseline serology assays. MRHC does not have the appropriate level of clinical isolation facilities, diagnostic laboratory capability, or world-class infectious disease clinicians experienced in diagnosing and treating for exposure to BSL-4 pathogens. It does not meet the standards of performance found at BSL-4 research facilities at the Centers for Disease Control and Prevention, USAMRIID, and the University of Texas Medical Branch (see NRC, 2010b for similar issues regarding laboratory- acquired infections in USAMRIID and the community’s impact in developing response measures). Regional Surveillance and Response Capability Once the risk is estimated, prevention and mitigation strategies and preparedness and response plans will need to be developed as part of an overall risk management program. Rapid detection and response are important for preventing or minimizing spread in the event of a release of an infectious agent. The local and regional emergency response community will bear major responsibility in the event of an accidental or intentional pathogen release. Thus, the emergency management and response, medical, diagnostic, and surveillance capabilities of the NBAF and the surrounding community will serve as critical factors in response plans. Human Health On the basis of conversations with DHS and a meeting with local first responders and emergency management officials, the SSRA concludes that the area’s emergency response capabilities are strong in a number of ways, such as a long history of cooperation, collaboration, and communication and a well-built emergency management organizational structure throughout the state of Kansas. The risk assessment team was in contact with representatives of MRHC, and the report contains a list of known and expected gaps in MRHC’s readiness to deal with medical and other emergency responses for events at the NBAF. Furthermore, the risk assessment identified a gap in communication with health and medical providers, especially physicians. A possible way to improve communications is to use and expand the existing statewide communications network (the Health Alert Network) to send messages to key healthcare providers. The SSRA states that the Manhattan, Kansas, area does not have adequate resources or capabilities to undertake all the prevention, mitigation, preparedness, response, and recovery activities necessary to develop and implement the emergency and contingency plans needed for 

EVALUATION OF THE NBAF SITE-SPECIFIC RISK ASSESSMENT 24 the NBAF. Apart from the information on MRHC and communication with health providers, the SSRA does not state what the specific gaps are, where the deficiencies occur, what capabilities are missing, or what plans have been developed to fix the insufficiencies. In other words, the SSRA lacks detailed analysis of the risks related to the emergency management and response capacities in the area. For example, other hospitals and medical providers within 20 miles of the proposed NBAF location were not evaluated as part of the risk assessment. The other health centers could be critically important in the event of an external release or a case of laboratory- acquired infection, but these rural and small community hospitals would not be able to handle the kind of outbreak that would involve a BSL-4 pathogen. A comprehensive analysis of regional health and medical assets is not included in the SSRA. The SSRA does not discuss the capabilities of local, regional, and state public health agencies, nor does it discuss surveillance and laboratory diagnostic systems. Animal Health The SSRA acknowledges the need to provide increased training and funding to the local and regional emergency response communities so that they can meet their increased obligations for detection, prevention, response, and recovery in the event of a foreign animal or zoonotic disease outbreak due to the presence of the NBAF in Manhattan, Kansas. The rapid response to a foreign animal or zoonotic disease outbreak will provide the greatest challenge to local and regional responders. Of particular concern is that there is no active national surveillance system for FMD detection. Passive surveillance is based on a traditional system in which practicing veterinarians report suspected “vesicular diseases”, and this would not be applicable to the zoonotic pathogens that would be in the NBAF. To fulfill the SSRA’s recommendation of enhancing local diagnostic capability to support regional surveillance and traceback capability, it will be essential to provide funding and validated tests to enable the National Animal Health Laboratory Network (NAHLN) at a minimum to conduct routine active surveillance for the agents under investigation in the NBAF. For an active FMD surveillance system to become fully operational, the NAHLN will need to expand its repertoire of testing and will need diagnostic surge capacity for those agents in the event of an outbreak. The KSU College of Veterinary Medicine Diagnostic Laboratory is a member of the NAHLN. The SSRA recognizes that the Manhattan region serves as a hub of animal movement and could potentially allow infected animals to move great distances from an initial focus of infection, which would increase the likelihood of a widespread outbreak if FMDv escaped from the NBAF. In a widespread outbreak, USDA may implement an emergency vaccination policy for the outbreak areas as one of its response strategies. If such an approach is taken, then adequate supplies of vaccines, associated equipment, and trained vaccination teams would need to be available to respond to such an outbreak (as mandated for the National Veterinary Stockpile under Homeland Security Presidential Directive 9), and there will need to be a well- developed policy agreed to by all 50 states as to who would have priority for vaccine distribution. 

CONSIDERATION OF SITE-SPECIFIC RISK AND MITIGATION FACTORS 25 REFERENCES CFIA (Canadian Food Inspection Agency). 2008. Foot-and-mouth disease hazard specific plan: Control area activities (5.3 Zones within the control area). Available online at http://www.inspection.gc.ca/english/anima/heasan/man/fmdfie/fmdfie-5e.shtml [accessed September 30, 2010]. Haas, C.N., J.B. Rose, and C.P. Gerba. 1999. Quantitative Microbial Risk Assessment. New York: John Wiley & Sons, Inc. 464 pages. Lott, N., S. McCown, and T. Ross, 2000. 1998-1999 Tornadoes and a Long-Term U.S. Tornado Climatology. In: National Climatic Data Center Technical Report 99-02, NCDC/NOAA: Asheville, NC. NOAA (National Oceanic and Atmospheric Administration). 2010. Storm Prediction Center: Frequently Asked Questions. Available online at www.spc.noaa.gov/faq/tornado/#Climatology [accessed August 30, 2010]. NRC (National Research Council). 1983. Risk Assessment in the Federal Government: Managing the Process. Washington, DC: National Academy Press. NRC. 2010a. Review of the Department of Homeland Security’s Approach to Risk Analysis. Washington, DC: The National Academies Press. NRC. 2010b. Evaluation of the Health and Safety Risks of the New USAMRIID High- Containment Facilities at Fort Detrick, Maryland. Washington, DC: The National Academies Press. Riley County. 2010. Riley County Community GIS Website: Map of Manhattan, Kansas. Available online at http://gis.rileycountyks.gov/website/rileyco/viewer.htm [accessed October 13, 2010]. USDA-NASS (U.S. Department of Agriculture, National Agricultural Statistics Service). 2009. “Quick Stats -- Statistics by State, Livestock Inventory January 2009 Agricultural Statistical Annual.” Available online at nass.usda.gov/statistics by state [accessed October 20, 2009]. 

EVALUATION OF THE NBAF SITE-SPECIFIC RISK ASSESSMENT 26