4
Review of the USAMRIID Environmental Impact Statement

The committee was charged with evaluating the Environmental Impact Statement (EIS) supporting the construction and operation of a new U.S. Army Medical Research Institute of Infectious Diseases (USAMRIID) facility in terms of the scientific adequacy and credibility of the analyses of health and safety risks associated with pathogen research. Most of the information on health and safety risks is contained in Appendix I (Hazard Assessment) of the EIS; other supporting information is distributed throughout the document. The EIS considers seven scenarios whereby pathogenic agents might escape from the containment laboratory. They include analyses of 1) biological aerosol releases from biosafety level (BSL)-3 and BSL-4 laboratories, 2) an escape of an infected animal, 3) biological material shipment, 4) terrorist acts, 5) external acts (such as natural disaster or mechanical failures), 6) exposure from a worker, and 7) cumulative impacts.

POTENTIAL SCENARIOS IN THE CONTEXT OF MAXIMUM CREDIBLE EVENTS AND RISKS TO THE COMMUNITY

The EIS estimates that under maximum credible event (MCE) scenarios community exposure to laboratory pathogens would be insignificant. In the sections below, the committee evaluates whether the scenarios are reasonably foreseeable and considers realistic exposure conditions, and whether the exposure estimations are justified.

Routes for Infectious Agent Release

There is a remote possibility that etiologic agents studied in the proposed USAMRIID facilities could exceed their intended containment by any one of a number of routes previously described. Except for animal escape, managed



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 42
4 Review of the USAMRIID Environmental Impact Statement The committee was charged with evaluating the Environmental Impact Statement (EIS) supporting the construction and operation of a new U.S. Army Medical Research Institute of Infectious Diseases (USAMRIID) facility in terms of the scientific adequacy and credibility of the analyses of health and safety risks associated with pathogen research. Most of the information on health and safety risks is contained in Appendix I (Hazard Assessment) of the EIS; other supporting information is distributed throughout the document. The EIS consid- ers seven scenarios whereby pathogenic agents might escape from the contain- ment laboratory. They include analyses of 1) biological aerosol releases from biosafety level (BSL)-3 and BSL-4 laboratories, 2) an escape of an infected animal, 3) biological material shipment, 4) terrorist acts, 5) external acts (such as natural disaster or mechanical failures), 6) exposure from a worker, and 7) cumulative impacts. POTENTIAL SCENARIOS IN THE CONTEXT OF MAXIMUM CREDIBLE EVENTS AND RISKS TO THE COMMUNITY The EIS estimates that under maximum credible event (MCE) scenarios community exposure to laboratory pathogens would be insignificant. In the sec- tions below, the committee evaluates whether the scenarios are reasonably fore- seeable and considers realistic exposure conditions, and whether the exposure estimations are justified. Routes for Infectious Agent Release There is a remote possibility that etiologic agents studied in the proposed USAMRIID facilities could exceed their intended containment by any one of a number of routes previously described. Except for animal escape, managed 42

OCR for page 42
43 Review of the USAMRIID Environmental Impact Statement transport, or terrorist acts, agents must move with indoor aerosols, wastewater, or solid waste streams before they potentially impact internal occupational re- ceptors and/or progress to the outdoor environment at large. Following media transport paradigms commonly employed by industrial hygienists and environ- mental engineers, each route—airborne, wastewater, and solids—is isolated and analyzed by the EIS, from a routine, but limited facilities design perspective. With respect to protecting biological indoor air quality as well as the im- mediate outdoor air quality, the committee finds the following key design pa- rameters and mechanical heat, ventilation, and air conditioning redundancies in accordance with or exceeding guidelines of the Centers for Disease Control and Prevention (CDC), the National Institutes of Health (NIH), the American Na- tional Standards Institute and the American Industrial Hygiene Association (ANSI/AIHA 2003), and the American Conference of Governmental Industrial Hygienists (ACGIH 1999, 2001) developed expressly for such high-exposure environments, specifically regarding: air exchange rates, vestive flow regimes, differential pressure systems, staged filtration, and protected heat transfer equipment. While special indoor air quality engineering features were presented in this context, the performance, reliability, and security concerning the opera- tions and maintenance of indoor air quality systems were not addressed in the EIS or its associated hazard assessment. Such an analysis would include, but would not be limited to: special contextual confirmation of mixing and flow regimes using widely accepted tracer tests under multi-season heating/cooling scenarios; the confirmation of filter blow-by under both clean and ripe condi- tions (in addition to smoke testing); and periodic stress challenges of critical pressure-sensitive infrastructure (in addition to Magnehelic calibration). Unlike the maintenance of indoor air quality, reliable inactivation of the capricious wastewater flows that contain potentially high concentrations of pathogenic agents relies on satellite in-situ laboratory pre-treatment prior to its transport and treatment in the building-centralized systems proposed under the USAMRIID expansion. With respect to treating comingled wastewater flows from the facility (apart from in-situ chemical pre-treatment), the committee finds key containment, transport, and design parameters in accordance with or exceed- ing the standards of the American Society of Microbiology (ASM 2007), CDC/NIH (2007), the American Public Health Association (APHA), the Ameri- can Society of Civil Engineers (ASCE), the American Water Works Association (AWWA), and the Water Environment Federation (WEF) (APHA/AWWA/ WEF 2005; ASCE/WEF 1992; WEF 1990, 1992, 1997), and guidelines written expressly for handling and treating combined wastewater flows containing dis- carded culture media and the biological fluids and sanitary sewage generated by USAMRIID’s animal testing. While special engineering features were presented in this context, the performance, reliability, and security concerning the opera- tions and maintenance of USAMRIID’s wastewater treatment and conveyance systems were not addressed in the EIS. Like its wastewater counterpart, reliable inactivation of pathogenic agents entrained in or on solid materials leaving the USAMRIID laboratories (exclud-

OCR for page 42
44 Health and Safety Risks of New USAMRIID High-Containment Facilities ing feces) relies on satellite in-situ treatment (autoclaves) prior to its transport and treatment in a centralized incineration system. With respect to treating com- ingled solid wastes from the facilities (apart from autoclave), the committee finds key containment, transport, and design parameters in accordance with, or exceeding the standards of the ASCE, ASM, and CDC/NIH, and guidelines writ- ten expressly for the handling and treatment of combined solid waste streams consisting predominantly of discarded culturing/assay supplies, sharps, plastics, and glassware, as well as selected animal parts. With exception of the extensive autoclave infrastructure built into the BSL-2, -3, and -4 suites, the performance of post-autoclave solid waste treatment systems or its reliability and security were not addressed in the EIS or its associated hazard assessment. Scenarios The EIS for the new USAMRIID facilities uses the MCE methodology to identify examples of events that may provide upper bounds on the risk posed to the public. The main MCE analysis in the EIS involved simulation of large aerosol releases from BSL-3 and BSL-4 laboratories as a result of centrifuge mishaps. Other events described include escapes of laboratory animals, an airplane flown into the laboratory, accidents during pathogen transport to the laboratory, and ex- posure through an infected laboratory worker. There is limited discussion of ac- tions taken by a laboratory employee that may circumvent biosecurity measures and maliciously expose members of the community to infectious agents. Transparency and Verification of Modeling The EIS documents its approach to risk analysis but is not transparent. An analysis is transparent when sufficient data are provided so that an independent observer with expert knowledge can replicate the results. The information and documentation provided in the EIS are insufficient for an independent assess- ment of the risks to the community posed by biological agents (see discussion below). Transparency of the EIS is an issue not just for this committee, but also for the public. One role played by the EIS has been to inform the public about the risks associated with the proposed project and how those risks are mitigated now and in the future. The committee believes that much greater elaboration would be necessary on the part of the EIS to adequately fulfill this informational role on behalf of the general public. For further discussion of the role of com- munications, see Chapter 5. The calculations of risk from the release scenarios appear to be incomplete and potentially incorrect. No attempts were made to systematically account for the biological characteristics of the organisms, and the analyses did not assess the potential for a local epidemic within the community or the efficacy of miti- gation plans for minimizing the effects of such an epidemic. Other scenarios lacked rigorous accounting and were supported with minimal data. The EIS does

OCR for page 42
45 Review of the USAMRIID Environmental Impact Statement not exhaustively consider the various possible routes through which members of the general public may be exposed to pathogens studied at USAMRIID. One possible route of exposure not considered in the EIS is the potential establish- ment of an infectious agent within a local animal or vector reservoir, and the subsequent low-level/long-term risks this may pose to the public. Risk of an Aerosol Release to the Community The MCE analyses in the EIS involved simulation of large aerosol releases from BSL-3 and BSL-4 laboratories as a result of centrifuge mishaps. In the scenarios, Coxiella burnetii (requiring BSL-3) and Ebola Zaire virus (requiring BSL-4) were released to the surrounding environment from an exhaust stack after vials in a centrifuge leaked and air filters failed to filter the pathogens. For these MCE scenarios, multiple mishaps must occur. First, a laboratory worker fails to use rubber O-ring gaskets to seal all six centrifuge tubes, so the safety caps leak during centrifugation. Second, the agent is leaked into the rotor com- partment, which is not sealed, and it is aerosolized into the laboratory. Third, the HEPA filter is inoperable (or one of two filters is inoperable in the BSL-4 sce- nario), which allows the agents to be released from the exhaust stack. The committee does not consider these MCE scenarios to be “reasonably foreseeable” accidents. They require simultaneous disregard of procedures, mul- tilevel engineering systems failures, and lack of administrative controls. Al- though there are reports documenting single-point failures—such as lapses in following proper procedures or administrative controls, engineering sub-system failure, and problems associated with personal protective equipment (PPE) and primary barriers—there are no documented reports in the literature of numerous simultaneous combined failures occurring in laboratories in the United States. Biosafety measures (outlined in Chapter 2) are designed to minimize the risk of such events. Laboratory workers are trained to use rubber O-ring gaskets to seal centrifuge tubes to prevent leakage of safety caps during centrifugation. Centri- fuges used in containment laboratories are themselves sealed to prevent such possible leakage. Rendering high-efficiency particulate air (HEPA) filters inef- fective in this context would mean they had not been properly installed and/or had been physically compromised without detection. Such compromise would cause pressure differences that would easily be recognized by building instru- mentation and alarm systems (and the laboratory would not be used until neces- sary repairs were made). The EIS used Gaussian plume calculations to estimate the maximum credible risk posed to the general public by the aerosol releases described above. The plume calculations were performed using the Hazard Prediction and As- sessment Capability (HPAC) software package developed by the Defense Threat Reduction Agency (DTRA). Gaussian plume calculations are a standard method for estimating risks from the atmospheric release of pollutants. In this EIS, plume calculations were used to estimate infectious doses after the accidental

OCR for page 42
46 Health and Safety Risks of New USAMRIID High-Containment Facilities release of a pathogen from BSL-3 and BSL-4 laboratories from the new USAMRIID facilities. The analysis of risk from an aerosolized pathogen release depends on local weather conditions affecting dispersal, environmental inactivation rates of pathogen after release, the geospatial density of susceptible human and animal hosts, and the biological characteristics of the pathogen itself. Future weather conditions for each month of the year were inferred based on meteorological records from the nearby Hagerstown Airport. Simulations were conducted under the conservative assumption that there was no environmental decay of the pathogen. There is no documentation on whether susceptible host concentrations were incorporated into the analysis of risk. The choice of C. burnetii and Ebola virus is puzzling. C. burnetii, a rickettsial agent that causes Q fever, is found in livestock reservoirs around the world, and is very capable of aerosol dispersal. In the United States, 169 cases of Q fever were reported in 2006 (CDC 2008). It can be isolated from pastureland in Maryland used for rearing sheep. Person-to- person transmission has not been documented. Thus, there is a low pre-existing risk from C. burnetii to all Maryland residents; the described release may not significantly enhance this risk. For Ebola virus, the only documented transmis- sion route from human-to-human is through direct contact with body fluids (Pourrut et al. 2007). Using the HPAC software package, puff-releases of aerosolized pathogen following large reportable accidents in the laboratory were simulated and the concentrations of pathogen, measured in human infectious doses, were calcu- lated. Considering a range of common meteorological conditions over the course of a year, it was determined that pathogen concentrations were insignificant at ground level beyond 300 meters from the points of release in all scenarios, the shortest distance to the Fort Detrick fence line. These results were considered conservative because of the large puff size and an infinite environmental half- life of selected pathogens, which ignores natural decay and environmental fac- tors that can significantly affect viability and infectious potential. The EIS con- cludes that there is no significant risk to the public from such an event based on these low concentrations. This analysis of aerosol dispersal risk is incomplete, and was not repro- ducible. The committee’s attempts to independently verify the calculations of doses of infectious agents delivered off site following puff releases found sig- nificantly higher concentrations than described in the report. However, the EIS’s published results are difficult to verify directly because specific data and param- eterizations of the simulation scenarios are not provided and the HPAC software is a closed-source system not available for independent review. In addition, the presentation of the risk analysis was incomplete. The most appropriate measure of per-person risk would be total infectious agent dose, which is calculated by the integration of concentration over time, with appropriate physiological pa- rameters. The EIS presents no documentation of an individual’s risk of infection (e.g., based on dose-response parameters) under the prescribed conditions or any description of the effect of population density and population size on the number

OCR for page 42
47 Review of the USAMRIID Environmental Impact Statement of cases expected. For most BSAT and other pathogens, human infectious dose- response relationships are poorly characterized. Despite the criticisms presented here, the aerosol release scenario is significantly more transparent than the other scenarios discussed in the EIS. Because of the issues raised here, the committee does not have a high degree of confidence in the conclusions of the Hazard As- sessment about this or the other MCE analyses. Risk of Secondary Transmission to the Community In addition to the shortfalls above, no modeling was performed on trans- mission of disease from an infected laboratory worker to family or community members. A review by the committee of the agents to be handled at the new USAMRIID facility indicates that very few are easily transmitted among hu- mans. But members of the Frederick County community have identified labora- tory-acquired infections as a concern, both because of the risk to facility em- ployees but also because they believe that employees could spread unusual and difficult-to-treat diseases to their families and the community at large. The EIS does not undertake any quantitative estimation of the risks and consequences of secondary infections that may occur subsequent to index cases. The primary reason for this appears to be that human-to-human transmission of most agents under study is rare. It is further noted that rapid detection of a bio- safety failure and the quick diagnosis of index cases will minimize further risk to the general public. Impacts from secondary infections following an initial infection will not vary among the three alternatives under consideration, as their spread in the community will not depend on the construction of the new USARMIID facility. However, secondary transmission risks are important com- ponents in the totaling of the consequences of individual biosafety failures. In cases involving communicable agents, a single index case of infection can have disproportionately large adverse public health consequences if effec- tive control measures are not in place. Two laboratory-acquired infections from the past decade have demonstrated that laboratory personnel can mediate sig- nificant public exposure (see Chapter 2). When accounting for risks posed by laboratory-acquired infections or the aerosol release of a pathogen, it is crucial to account not just for index cases of infection caused by direct exposure to pathogen release, but also secondary cases of infection that may result from the transmission of infection from an index case to members of the general public. An established set of tools exists within the academic literature on infectious diseases to assess such risks. Kermack-McKendrick and Reed-Frost models can be specified with relatively limited data in terms of basic reproductive numbers and serial intervals. They can be used to assess worst-case scenarios for the like- lihood of an epidemic occurring, predict speeds of epidemic spread within a community, and estimate the total number of people infected. Alternatively, large agent-based simulation models can incorporate detailed geospatial, demo- graphic, and socio-economic data to provide fine-grained projections about

OCR for page 42
48 Health and Safety Risks of New USAMRIID High-Containment Facilities transmission risks and the efficacies of mitigation strategies. These methods have been used in other contexts, including the MIDAS project’s pandemic pre- paredness efforts (NIGMS 2010) and the Government Accountability Office (GAO 2009) evaluation of risks of relocating research on foot and mouth disease at Plum Island Animal Disease Center to the mainland. Evaluation of mitigation strategies in all infectious disease models will re- quire detailed knowledge of the ecology and lifecycles of the pathogen. To iden- tify an appropriate MCE, a detailed enumeration of all laboratory pathogens and their characteristics would be needed. This is probably not possible for many pathogens because of incomplete scientific knowledge. Many of the characteris- tics are widely available, although some remain unknown. For example, a com- plete enumeration of all characteristics is not even known for influenza, a virus that has been studied for decades. A list of pathogens under study at USARMIID is provided by the EIS, but the epidemiologic characteristics, including transmission pathways, natural res- ervoirs, geographic distributions, and clinical outcomes of the pathogens, are not systematically documented. While the pathogen characteristics are crucial com- ponents of the risk assessment, the EIS does not appear to systematically stratify the risks of different pathogens to the general public. For instance, diseases that rely on arthropod vectors for transmission can be studied without significant risk of secondary transmissions so long as they are incompatible with the native ar- thropod populations. Such analyses would identify maximum credible risks, and would also provide guidance for the on-going mitigation of risk as emerging pathogens are added to the list of research candidates. Other Community Risks The EIS Hazard Assessment addresses the risks posed by laboratory ro- dents, rabbits, and primates to the public. The EIS summarizes design features and practices that make the escape of a laboratory rodent or rabbit very unlikely. A similar summary indicates that primate escapes are also highly unlikely under best practices. No calculations or comparative analyses are provided, but histo- ries of BSL-3 and BSL-4 laboratory incidents are consistent with these conclu- sions (DHS 2008). If a small mammal infected with a biological agent did escape from the laboratory, the EIS states that there is only a small chance the agent would be transmitted to a compatible human host or animal reservoir. No data, analysis, or calculations are provided to support this conclusion. Events subsequent to the escape of a non-human primate are not addressed, as primate escape is consid- ered significantly less likely and easier to mitigate. The hazard analysis does not address risks from arthropod escape. Several biological agents that may be studied at the new USAMRIID facility are trans- mitted by arthropod vectors (fleas, mosquitoes, and ticks), and the vectors may be employed in the course of research. The EIS should have addressed potential

OCR for page 42
49 Review of the USAMRIID Environmental Impact Statement concerns that an arthropod escape leading to the establishment of pathogen in a native animal or vector reservoir could result in long-term elevation in disease risk to the general public. POTENTIAL SCENARIOS IN THE CONTEXT OF MAXIMUM CREDIBLE EVENTS AND RISKS TO PERSONNEL The EIS does not provide scenarios describing potential exposure risks in- volving pathogens to USAMRIID laboratory personnel, but does cite a brief history of laboratory-acquired infections between 1989 and 2002. Review of these cases illustrates both means of transmission and procedures in place to address identification and treatment of affected laboratory workers (Section 2.3.4.3 of the EIS). Common risks to workers are needle- or sharps-stick acci- dents, inadvertent aerosol generation that leads to inhalation or ocular/mucosal exposure, and contact with infected laboratory animals. In the latter case, animal caretakers or laboratory workers may be exposed to zoonotic agents, such as Herpes B virus, from contact with laboratory animals; infectious aerosols shed in urine, blood, and other body secretions; or pathogens from bites and scratches (either from caged or escaped animals). When assessing risks to laboratory workers, reasonable scenarios can be developed using two simple and direct methodologies. One is to consider case studies of past events. The EIS does this in its discussion of a case of glanders acquired in 2000 (discussed in Chapter 2). The other methodology would be to consider the most credible exposure routes for laboratory workers. These should include needle sticks and aerosol events. Standard operating procedures are key in understanding, assessing, and mitigating the risks from credible events. Laboratory-acquired infections happen but are rare when compared to the thousands of person-years worked and number of laboratory-acquired infections resulting from exposure to pathogens. Based on a review of 234 recognized po- tential exposures and illnesses over a 14-year period (Rusnak et al. 2004b,c), the EIS cites only five recognized infections. All illnesses were caused by agents requiring BSL-3 containment. No infections have been reported for workers in U.S. laboratories under BSL-4 conditions. Exposures in BSL-3 laboratories that result in infection are infrequent. The EIS (Section 2.3.4.3) describes programs (i.e., training, special immunizations and medical monitoring, and reporting re- quirements) and regulations to identify and report potential exposures. Such reports initiate a monitoring process appropriate for the exposure circumstances and the infectious agent involved. One of the cited laboratory-acquired infec- tions involved a worker who in 2000 became infected with Burkholderia mallei, the causative agent of glanders. This case illustrates a failure at the user level. The individual did not follow standard operating procedures requiring glove use (resulting in inadequate personal protective equipment [PPE]) and requiring the prompt reporting of illness to USAMRIID’s Special Immunization Program clinic.

OCR for page 42
50 Health and Safety Risks of New USAMRIID High-Containment Facilities EISs recently developed for two other BSL-3/BSL-4 laboratories, the Na- tional Bio- and Agro-Defense Facility (DHS 2008) and Rocky Mountain Labo- ratories (NIH 2004) provide insights to laboratory-acquired infections based on person-hours worked in USAMRIID, CDC’s Special Pathogens Branch, and the National Institute for Communicable Diseases (NICD), a branch of the South Africa National Health Laboratory Service. Tables 4-1 and 4-2 point to the ex- tremely low occurrence of laboratory-acquired infections based on personnel hours worked in BSL-3 and BSL-4 in three different institutes (DHS 2008). TABLE 4-1 Personnel Hours Worked and Outcomes of Accidental Exposures to Infectious Agents: Intramural National Institute of Allergy and Infectious Diseases 1982-2003 Hours at Risk Bench Animal Total BSL-3 553,000 81,500 634,500 BSL-2/3 Pa 2,235,500 360,200 2,555,200 Total 2,788,500 441,700 3,189,700 Outcomes of Accidental Exposures Other Exposures, Clinical Infections Silent Infections No Infections BSL-3 1 2 9 BSL-2/3 Pa 0 2 15 Total 1 4 24 a P refers to partial, which was used in past practices preceding Biosafety in Microbiologi- cal and Biomedical Laboratories requirements. Partial laboratory status refers to the ab- sence of single-pass directional air flow through a HEPA filter within the laboratory; however, all bench work was conducted within a biosafety cabinet (BSC), which incorpo- rated HEPA filtration within the BSC before air was exhausted from the BSC to the facil- ity heating, ventilation, and air conditioning system. Source: DHS 2008. TABLE 4-2 Personnel Hours Worked and Outcomes of Accidental Exposures to BSL-4 Agents 1972-2003 Hours at Risk Incidents Infections USAMRIID 343,980 2 0 CDC Special Pathogens 120,560 2 0 NICD 40,000 2 0 Total 504,540 6 0 Source: DHS 2008.

OCR for page 42
51 Review of the USAMRIID Environmental Impact Statement Aerosol exposure is frequently associated with laboratory-acquired infec- tions, often because only a small proportion of such infections are associated with a recognized accident (Pike1979). There are, however, ample examples of laboratory-acquired infections transmission via the aerosol route, such as events involving Brucella (Olle-Goig and Canela-Soler 1987; Staszkiewicz et al. 1991), C. burnetii (Meiklejohn et al. 1981), and Venezuelan equine encephalitis (VEE) virus (AVMA 2006). Additional examples are also provided in Chapter 2, Table 2-1. A reasonable and credible scenario for an aerosol exposure in a BSL-3 laboratory might involve a laboratorian ignoring proper practice and creating a respirable aerosol while working at a biological safety cabinet. Assuming that the laboratorian was not wearing respiratory protection at the time (respirators are not universally required for work in BSL-3 labs) an unrecognized exposure and subsequent infection could occur. Depending on the agent (C. burnetii, for example) the subject would experience the sudden onset of characteristic symp- toms. A properly trained employee should recognize these symptoms and report to medical authorities. As noted above, this patient would pose no threat to the community because this agent is not transmitted from person to person. For the MCE in a BSL-4 setting, the EIS considered an unrealistic centri- fuge failure scenario similar to the one for the BSL-3 scenario, with Ebola virus as the infectious agent. Except for the Reston Ebola virus, which primarily in- fects non-human primates, Ebola is spread by direct contact with infectious blood or other bodily secretions and is not known to be transmissible by the aerosol route (Pourrut et al. 2007). A large aerosol release in a BSL-4 setting would have little effect on laboratory workers unless there was a simultaneous break in the one-piece, positive pressure suit. If the face piece of a suit is dam- aged, the positive pressure within the suit would limit exposure to aerosols, and the incident would be reportable and an aerosol exposure would be considered in the hazard assessment that would follow. A more likely event would be expo- sure by a needle stick, animal bite, or broken glass. Unlike aerosol exposures, sharps injuries in a laboratory are obvious and elicit a quick response on the part of the injured laboratorian. USAMRIID has adopted the requirements of the Occupational Safety and Health Administration (CFR 1910/1030; also in Bio- safety in Microbiological and Biomedical Laboratories [CDC/NIH 2007]) for preventing sharps injuries in its suite-specific safety manuals. Such injuries must be reported and evaluated by a medical panel to assess the risk of exposure. Thus, while a sharps injury in a laboratory at USAMRIID is possible, it is unlikely to result in a risk to the community. The risk to the individual may be high, but recognition of the event limits the unwitting exposure to the commu- nity because of procedures and policies in place to evaluate such injuries and promptly institute medical care or isolation as required. As documented in Chapter 2, USAMRIID has a rigorous biosafety pro- gram to prevent or reduce exposure risk for its personnel. The EIS (Section 2.3.4.3) discusses the occupational risks posed by pathogens. A discussion in the EIS regarding PPE, available immunizations, and post-exposure medical evalua-

OCR for page 42
52 Health and Safety Risks of New USAMRIID High-Containment Facilities tion and treatment of workers is needed to provide a comprehensive and bal- anced assessment of exposure and risk mitigation associated with exposure sce- narios and to demonstrate the high level of rigor of the existing biosafety pro- gram at USAMRIID. A better scenario would focus not only on the deficiencies and breaches of containment to the environment, but also on the risk posed to the workers in the room and individuals in adjacent work spaces (that is, rooms sharing a common entry corridor). Reduction in exposures of laboratory personnel to pathogens, and mitigat- ing laboratory-acquired infections through improved biosafety techniques and technology is achievable and can be demonstrated by comparing the frequency of laboratory-acquired infections from the 1950s through 2000 (Sulkin and Pike1951; Sewell 1995; Harding and Byers 2000). Today biological safety is based on a series of controls to include engineering controls, administrative con- trols, primary barriers, PPE, and workplace practices to achieve safety for the worker and the environment. These controls work in concert and each is de- signed to reinforce the others. If one control is weak (such as old facilities with aging engineering controls) additional PPE and enhanced work practices and perhaps additional training can be utilized to negate potential deficiencies in the weaker control. (See Chapter 2 for a review of operational procedures and his- tory of laboratory-acquired infections at USAMRIID.) While not part of the EIS’s Hazard Assessment, Section 2.3.4 describes measures to be used to protect both laboratory staff (and by extension, their families and community contacts) and the environment. These measures, based on the controls enumerated above and in Chapter 2, come from the CDC/NIH (2007) and are codified as Army Regulation 385-69 and the Department of the Army Pamphlet 385-69. Governing regulations incorporate by reference a host of federal and state regulations pertaining to laboratory and general safety. The practical application of the broader regulations is detailed in suite-specific safety manuals that are developed for each laboratory unit as required by 42 CFR 73 (2007) (Possession, Use, and Transfer of Select Agents and Toxins). CONSIDERATION OF ALTERNATIVE CONSTRUCTION SITES The EIS provided an ample and compelling rationale for why personnel in the new facility would not be able to effectively perform their mission—which is in part dependent on leveraging complementary capabilities of the National Biodefense Analysis and Countermeasures Center and the National Institute of Allergy and Infectious Diseases’ Integrated Research Facility—if sited remotely from its current location. However, the National Environmental Policy Act re- quires consideration of all reasonable alternatives, including reasonable alterna- tives not within the jurisdiction of the lead agency. In this case, the Army did not analyze any geographic alternative sites. Such an exercise would have al- lowed for a comparison of the advantages and disadvantages specific to each

OCR for page 42
53 Review of the USAMRIID Environmental Impact Statement location, and help guide strategies and mitigation efforts if differential risks are found. Furthermore, it would have distinguished risks and factors that are de- pendent on siting location (for example, the potential for disease transmission to livestock and wildlife in rural settings that could result in zoonotic outbreaks, or the availability of medical and emergency response personnel) and those that are independent of site (for example, risks of a malicious insider). FINDINGS  The EIS posed several problems: o The analyses of the risks and the mitigation measures to address them were not comprehensive and there was insufficient documenta- tion for an independent assessment of the risks to the community posed by biological agents. The problem was compounded by the fact that the MCE scenarios were not reasonably foreseeable acci- dents. o The epidemiologic characteristics, including transmission pathways, natural reservoirs, geographic distributions, and clinical outcomes of the pathogens, were not systematically documented. o There was incomplete consideration of some of the possible routes through which the general public might be exposed to pathogens. o Although the congressional mandate placing the National Inter- agency Biodefense Campus at Fort Detrick precludes siting the new USAMRIID facility elsewhere, it would have been appropriate for the EIS to include consideration of an alternative location, such as one in a less populated area. Such an exercise could have provided a comparison that identified advantages and disadvantages specific to each location, and guided preventive strategies and mitigation ef- forts if differential risks were found.  Although the EIS hazard assessment failed to provide adequate and credible technical analyses, it was determined in Chapter 2 that current proce- dures, regulations, physical security, and biosurety guidelines at USAMRIID meet or exceed accepted standards and practices. Thus, the committee has a high degree of confidence that polices and procedures are in place to provide appro- priate protections for workers and the public. Furthermore, the review in Chap- ter 3 indicates that the Army and Frederick County have the resources and the partnerships in place to address medical and emergency situations at the con- tainment laboratories.  Despite the problems identified with the EIS, the committee judged that it would not be useful to propose specific revisions to the EIS or supplementary analyses. The Record of Decision to construct the new USAMRIID was issued and construction has begun on the project.

OCR for page 42
54 Health and Safety Risks of New USAMRIID High-Containment Facilities RECOMMENDATION The committee recommends that the Army consider developing detailed and practical guidance for conducting hazard assessments of infectious agents for inclusion in the Army’s guidance for implementing the National Environ- mental Policy Act to improve future EIS processes and products.