In the fall of 2001 there were several incidents of bioterrorism in which preparations of Bacillus anthracis were mailed to public and private institutions. The acts led to 5 deaths from inhalational anthrax and to more than 20 cases of inhalational or cutaneous anthrax. More than 30,000 people were given prophylactic antibiotic therapy (Lane and Fauci, 2001). People were infected not only at locations where the contaminated envelopes had been opened but also at sites through which the unopened mail had passed, including postal distribution centers and local mail rooms. Decontamination was required at many locations.
Decontamination was an extensive undertaking, both for cleanup and for communication involving building managers, government agencies, a host of private- and public-sector experts, and affected building occupants and users. Decisions had to be made about which sites required cleanup, what method to use, how to determine the effectiveness of the cleanup, and how “clean” the building had to be for reoccupation. The responses of people who occupied the building in the affected areas and in other locations and those of other stakeholders had to be considered in the overall remediation effort. Responses ran the gamut from apparent confidence in decision makers to outright distrust and hostility. The stakes were high, and there was considerable stress attached to the uncertainties concerning the successful completion of the project. For these reasons, the significance of the social aspect of recovery from a bioterrorism attack cannot be overstated.
The cost of decontamination was significant. Remediation of U.S. Postal Service facilities alone cost more than $200 million. Clearly, it is desirable to control expenses, so it is of national interest to learn how to respond to and
recover from a bioterrorist attack in a manner that effectively reduces the risk posed by exposure to the biological agent to an acceptable degree of assurance and without incurring unnecessary expense. Factors to consider for effective response and recovery include the geographic extent of contamination, the timing and duration, and the method of decontamination, all of which are influenced by characteristics of the agent released. Underlying all of this is a question: “How clean is clean enough?” The original question placed before the Committee on Standards and Policies for Decontaminating Public Facilities Affected by Exposure to Harmful Biological Agents was “How clean is safe?” Is there a standard that we should anticipate, beyond which additional decontamination efforts would yield insubstantial benefit?
In response to the attacks of 2001 and their subsequent cleanup, the Department of Homeland Security funded a project called the Restoration and Domestic Demonstration and Application Program, which was run by the Lawrence Livermore National Laboratory (LLNL) and Sandia National Laboratories. As part of the study, the LLNL subcontracted to the National Research Council (NRC) to convene a committee of experts to consider the criteria that must be met for a cleanup to be declared successful, allowing the reoccupation of a facility. The committee specifically was asked to consider a scenario in which decontamination of a facility approaches completion, but it was not asked to review all issues that should be considered in the aftermath of a biological attack. Therefore, this report does not address in any detail the risk that such an attack would occur, the emergency response to an attack, the identification of the appropriate allocation of resources for research or response, or broader public health issues related to transmissible diseases. It does not recommend specific decontamination technologies, although such technologies are discussed to provide information for those who need to understand them to make informed decisions about reoccupation. Rather, the report reviews the key factors that influence decision making and lays the foundation for establishing standards and policies for relevant aspects of biological decontamination. Because the sponsors are seeking additional information for their demonstration project, this committee focused its effort on indoor facilities, and it uses an airport as a model.
The committee considered the issues outlined above and concluded that remediation must meet appropriate technical considerations from several perspectives, and it must be convincing to the stakeholders, especially the users of a facility. There is no single standard to apply to all situations. Thus, based on its scientific analysis of the available information, the committee outlined steps that would help achieve a socially acceptable standard for cleanup.
This report has 12 chapters, each containing information relevant to the decision about what constitutes “acceptable cleanup.” It considers the history of biological weapons, the biology of 3 microorganisms that are considered threats for use in biological warfare, the nature of the response to infection, the issues associated with the determination of infectious dose and quantitative microbio-
logical risk analysis, the social issues involved in making decisions, reagents and techniques used in decontamination, approaches to detection and surveillance of microbiological contamination, and environmental factors in buildings that affect the distribution and accessibility of the agent to decontamination. Chapter 3 and Appendix D contain case studies on some of the buildings that were decontaminated after the 2001 anthrax attacks and on other facilities that have been cleaned after biological or chemical contamination. Because of the particular concern of LLNL, the final chapter specifically addresses the considerations that apply to a major airport.
What has become clear is that there are three levels of consideration in the approach to decontamination: The first is the fact that decontamination within a “reasonable” period is necessary and must proceed in a manner that is consistent with available knowledge and current regulations. The social aspect of decontamination and safe reoccupation—for example, stakeholder and occupant concerns—is another consideration. The third involves the recognition that there is a lack of information that could influence both the effectiveness and the cost of decontamination, a state of affairs that should be remedied. Findings and associated recommendations appear at the end of each chapter. Selected key findings and recommendations are summarized here.
The charge to the committee is addressed by discussions in five areas: infectious dose, natural background of microbiological flora of interest, risk assessment, past cleanup efforts, and residual contamination. Key messages of the report for each area are indicated here.
The 2001 anthrax attacks called into question the state of knowledge on what constitutes the infectious dose for B. anthracis. Infectious dose is the term often used to denote the number of organisms it is believed are necessary to overwhelm host defense mechanisms and establish an infection that can lead to disease. The committee concluded that standard infectious doses for harmful biological agents that could be used as weapons cannot be determined with confidence because the infectivity and virulence of harmful agents can vary by strain, within species, and by type of preparation into weapons. Currently available data on dose–response relationships are not as extensive as demanded by modern scientific standards, and, in most cases, the human data cover only the exposure of healthy young adults.
The committee acknowledged that natural environmental background concentrations of various microorganisms have been assessed in some areas and that most people in those locations tolerate exposure without adverse effects, perhaps
because those people have developed immunity gradually. The concept of natural background is difficult to apply in evaluating acts of bioterrorism in indoor public facilities because it is not likely that a detectable natural background of harmful agents, such as those under consideration in this report, would exist in indoor public facilities. Moreover, the agent used in an act of bioterrorism could be different from its natural form if it has been weaponized. (Weaponized microorganisms are processed to enhance stability, infectivity, environmental half-life, or ease of dissemination.)
Quantitative risk assessment models often are used to evaluate complex situations. The models have four steps: hazard identification, exposure assessment, dose–response assessment, and risk characterization. Although such models can be useful in assessing the risk of exposure to harmful biological agents after cleanup, the essential data to support thorough analysis via quantitative risk assessment are lacking for some agents that might be used as biological weapons.
PAST CLEANUP EFFORTS
A review of the B. anthracis cleanup after the events of 2001 provides insight about the approaches that should be used in the event of a future attack.
Some biological agents in their natural forms would likely degrade rapidly enough that extensive cleanup would not be necessary after an initial decontamination. However, a preliminary analysis of the agent might not reveal alterations that could influence its viability. Therefore, a full characterization would be necessary to evaluate the effect of genetic or physical modifications on its viability. After cleanup, continuous medical monitoring might be useful to ensure the safety of those who would occupy the decontaminated space.
A contaminated facility cannot be guaranteed to be agent-free even after cleanup because it is impossible to prove the complete absence of an agent. The committee was asked to consider whether there is a “safe” amount of residual contamination. It concluded that there is insufficient information to quantify a “safe” amount of residual biological agent in a decontaminated facility. Further research could provide information on host response to specific doses that would decrease the uncertainty and make a quantitative approach more useful. However, the risk different people or groups of people are willing to tolerate will always vary. Therefore, the issues related to decision making raised in this report will continue to be relevant. The report considers lessons from the response to the 2001 anthrax attacks and from other situations involving chemical or radiological
decontamination; the idea of a risk assessment framework, including current knowledge of dose–response relationships; the role of indoor air movement; the various approaches to sampling for biological agents; and the technologies available for decontamination. All of those issues would be important for decision makers to consider in the event a facility requires decontamination.
Based on its analysis of the issue areas listed above, the committee made 26 recommendations. Those described in this Executive Summary are organized into four areas: risk assessment, health, sampling and decontamination standards, and decision making.
Quantitative microbial risk assessment, a discipline developed over the past 20 years, has been used to inform decision making about microbial hazards in food safety, drinking-water quality, and in hospital isolation rooms. The potential variations in agents of the same species and among potential human hosts (immune status) strongly suggest that the limited information currently available should be interpreted cautiously. Although the nonthreshold model implies that there is no amount below which an agent poses zero risk, it indicates that the probability of infection is extremely low. A threshold model, in contrast, implies a definitive threshold below which no infection would occur. Therefore, nonthreshold dose–response models offer a more cautious approach that is appropriate for describing the human response to exposure to a diversity of infectious agents by ingestion, inhalation, and other routes. Full characterization (including screening for known threat agents, genetically modified agents, and emerging-threat organisms) of a suspected biological pathogen is required for proper analysis and to inform decision making. Identification and characterization of the properties of an organism, and the amount and extent of its concentration when cleanup begins, are critical to making decisions about response options (Findings 5-1, 6-1, 6-2, 8-1).
The threat posed by naturally occurring infectious disease is well characterized, and the information available is useful for planning the response to an attack. But weaponized biological agents could pose distinct threats, especially when it comes to decontamination. Dose–response data for most of the pathogens of concern are either incomplete or have not been analyzed critically in the open literature. A complete risk analysis depends on the availability of information about each variable, and the information on agents that might be used in a biological attack is weak for some variables. Because publicly available data on which to base human dose–response assessments for the critical pathogens are minimal, we often must rely on animal data. However, our understanding of interspecies extrapolation from animals to humans remains poor (Findings 2-1, 5-2, 8-2).
The committee recommends that a risk assessment approach be adopted as one component of decision making for determining the adequacy of decontami-
nation efforts after a release or suspected release of a biological contaminant. More data on dose–response relationships are needed to conduct a practical, as opposed to a theoretical, risk analysis for any given biological attack. Available dose–response data for pathogens of concern should be analyzed using nonthreshold dose–response models. Targeted research will help inform decision making on extrapolation for the pathogens of concern. That work might use multiple species of organisms or study animal and human tissues to provide information that is relevant for human exposures. With the increasing difficulty of performing nonhuman primate studies, it will become more important to develop in vitro techniques that can be used to develop dose–response information (Recommendations 5-1, 5-2, 8-1, 8-2).
A characterization system should be developed to inexpensively identify, or partially characterize, all potential threat agents, including genetically modified and emerging-threat agents. Decontamination decisions and plans should account for the natural characteristics of a specific pathogen and for the weaponization characteristics of the agent. Weaponized agents (such as weaponized B. anthracis) could vary from crude to sophisticated preparations, formulated to enhance dispersal, increase suspension-in-air time, extend viability, or increase their ability to penetrate a target organism. Given the potential deviation from the natural form, it is not possible to say with complete certainty that a particular agent would pose zero risk at a given exposure. For that reason, the contaminating agent or agents should be characterized before the approach for large-scale remediation is chosen. The remediation approach should ensure adequate destruction or removal of the amount of the agent present at the start of the procedure (Recommendations 2-1, 6-1, 6-2).
People and microorganisms cohabit the world; sometimes their interactions result in human disease. Where people face a greater risk of exposure to pathogens (for example, in laboratories or hospitals), biological safety policies protect against human disease. Decontamination is not a discreet activity, but it is part of a larger set of controls over dangerous microorganisms and their potential to affect health. The earlier contamination is detected the easier it will be to restrict the area of contamination and the number of people exposed. Different monitoring systems—environmental (Biowatch) and medical (syndromic surveillance)—have been put in place with the hope of obtaining the earliest indicator regarding the release of a biological agent. Some type of postevent medical monitoring of the health of people exposed to a contaminant is critical to ensuring confidence in a facility’s safety. The purpose and outcome of medical monitoring should be made transparent to affected parties. Because of different objectives for law enforcement agencies and public health agencies, data from the sites contaminated in 2001 were not shared with all relevant parties. Lack of data sharing can
compromise human health in the aftermath of a biological attack (Findings 3-3, 4-2, 6-3).
Existing environmental monitoring systems and syndromic surveillance systems should be evaluated for the ability to provide information that can be used cost effectively to detect and limit the spread of dangerous biological agents. Such programs should be able to detect an agent or to identify unusual clusters of symptoms or disease within a period that ensures the earliest possible public health intervention. If the systems prove effective and affordable, they might be deployed at public facilities that are likely to be targets of an attack or whose removal from service because of contamination would produce catastrophic economic effects. Those systems also could be useful for postevent monitoring. Planning for future incidents should consider mechanisms for establishing a centralized and sustained effort to track the health of people who are exposed, or potentially exposed, to pathogens. Such a program should be evaluated for effectiveness and practicality. Agencies and organizations entrusted with data relevant to public health should make every effort to share the information. To achieve the primary goal of protecting public health and safety, federal agencies should prepare memoranda of understanding to increase cooperation and decrease anxiety (Recommendations 3-3, 4-2, and 6-3).
SAMPLING AND DECONTAMINATION STANDARDS
Biological agents can spread beyond their point of initial release in air-handling systems, by the reaerosolization of contaminants from floors and other surfaces as a result of foot traffic or air currents, through adhesion to people or their clothing, and by transmission from one person to another. The result could be widespread dispersal of biological contaminants within a building, into transportation and transit vehicles, and into homes or other sites. Indoor air-handling systems can redistribute biological agents by carrying airborne contaminants throughout buildings and then outdoors. If appropriate actions are taken, however, air-handling systems also can be used to confine contaminants and reduce the effects of contamination. General guidance from the Centers for Disease Control and Prevention (CDC) directs sampling of B. anthracis spores, but there is no official guidance for the collection of vegetative B. anthracis, plague bacteria, or smallpox virions. Different threat substances require different sampling protocols. The wide variety of collection approaches currently in use results in widely varying efficiencies, which impede quantification of the extent of initial contamination. Adequate training of decontamination teams is essential for effective remediation and validation. The federal sterilization “metric” of using test strips to verify a “6-logarithm kill” (the reduction of the amount of live contaminant by 6 orders of magnitude) was used as a standard for the remediation of the Hart Senate Office Building in Washington, D.C. However, the amount of contaminant present before decontamination can be many orders of magnitude higher:
One gram of dried B. anthracis can produce 1011 to 1012 active spores. The current standard could leave large amounts of viable organisms (Findings 7-1, 7-2, 9-1, 9-3, 10-3, 10-4).
In the aftermath of an attack, an extensive survey should be done to determine the extent to which biological contamination has spread. Building operators should act now to gain a thorough understanding of air flow under normal operating conditions and the potential adverse or beneficial consequences of a shutdown for the spread of airborne contaminants. Appropriate actions therefore could be taken to minimize the dispersal of contaminants once a release has been identified.
Sampling protocols must be appropriate to the threat. B. anthracis sampling should follow published guidance from CDC, including protocols published by the National Institute for Occupational Safety and Health. With input from the Department of Defense, CDC and the American Society for Microbiology should develop sampling and analysis guidelines for the other threat agents. Sampling and analysis methods must be standardized and incorporated into a general sampling plan. Research should be conducted to assess the efficiency of sampling collection and analysis procedures for each type of biological threat substance. Unless the sampling efficiency is known, the amount of contaminant present when cleanup begins cannot be estimated with confidence. A general sampling plan should result from consensus among facility stakeholders, medical and public health personnel, environmental experts, decontamination technologists, laboratory analysts, and worker safety representatives. It should encompass three phases: (1) confirmation and contamination baseline, (2) assessment and characterization, and (3) decontamination effectiveness. The Environmental Protection Agency (EPA) and CDC should establish standards for remediation and validation of contaminated buildings and evaluate current and emerging decontamination techniques to determine efficiencies (Recommendations 7-1, 7-2, 9-1, 9-3, 10-3, 10-4).
Risk is a complex issue, and willingness to accept risk varies among people and circumstances. There are divergent ideas about how much responsibility the government or the owners and operators of public facilities and lands should take to limit public exposure to risk. Those issues have been addressed in various situations, and many policy-making lessons can be learned, for example, from Superfund1 and the Department of Energy cleanup experiences. But if safety-related standards and protocols are devised behind closed doors, without the
advice or consent of affected and interested parties, those standards are likely to be questioned or rejected outright.
Lack of transparency for policy decisions that directly affect public health—even in the context of a proclaimed national security interest—can severely erode public confidence. Initiating a planning procedure that involves relevant stakeholders before an event occurs would expedite decontamination and improve the acceptability of decisions made during and after decontamination. Effective response to and recovery from a biological attack requires expertise and input from scientists, building engineers, and stakeholders. Response and recovery alike can be accomplished promptly if there is planning that involves all stakeholders and those with appropriate scientific expertise. Although building owners and managers could begin the planning that involves the building structures and operations, technical and scientific planning involves expertise that is scattered across government agencies. In the event of an actual biological attack, the availability of a soundly drawn plan will certainly hasten the reopening of a facility (Findings 3-1, 3-2, 11-1, 12-2).
Authorities who contemplate how to respond to biological attacks should base their plans on lessons from experiences with decontamination in the broadest sense; they should not consider their charge a completely novel task. Affected parties should be involved in risk management decision making and should participate in technical discussions. Planning should identify those interested parties, form them into a working group, and have them interact regularly in anticipation of coming together to guide an actual recovery effort. The view that a facility is once again “safe” for use will be accepted by the interested parties only if they have had an active, meaningful role in reaching that conclusion (Recommendations 3-1, 3-2).
Building owners and managers should begin to plan immediately. The committee recommends that the National Response Plan (specifically the Biological Incident Annex) or some other suitable federal document be expanded to provide more scientific and technical information on biological weapons, decontamination, sampling and surveying, and epidemiology. The document should describe how a team would operate to collect information pertinent to a response to and recovery from a biological attack. That document also should identify who would be responsible for convening the team. The committee recognizes that formation of such a team might take time, and it therefore outlines the following immediate, short-term, and long-term goals for building managers and the government to consider.
Building managers and owners should convene an Operations Working Group that includes representatives of all relevant stakeholder groups to devise a biological attack response and recovery plan. Because the group would not have
all the necessary scientific and technical expertise, it should identify the appropriate government agencies and officials to be contacted in the event of an attack.
The government should identify a group with the appropriate technical and scientific expertise to assist building owners and managers in the event of a biological attack. That group should work with the Operations Working Group to devise the best courses of action for response and recovery.
The federal government should devise a mechanism for keeping government and other interested parties abreast of developments and new technology in surveillance, sampling, and decontamination and revise the standards and policies for decontamination iteratively. The mechanism should ensure that building managers and owners are kept informed (Recommendation 11-1).
Lane, C.L., and A.S. Fauci. 2001. Bioterrorism on the home front, a new challenge for American medicine. Journal of the American Medical Association 286: 2595-2596.