Anthrax Decontamination After the 2001 Attacks: Social and Political Context
The question of when a facility is safe for reoccupation cannot be answered with physical data alone. Science provides highly sophisticated tools that help diminish uncertainties and, despite those uncertainties, help policy makers map out possible courses of action. However, the issue of safety goes beyond numerical calculations. The perception of what is safe ultimately depends on whether people believe what the technical experts and policy makers say about safety. The question, “How clean is safe?” is the same as, “What level of risk is acceptable?” Thus, if we are to build sensible policy recommendations, the physical and life sciences must work hand in hand with the social sciences.
This chapter discusses the social aspects of decontamination, using as case studies the major sites that were contaminated with Bacillus anthracis in fall of 2001. Four major cleanups resulted from the arrival of contaminated letters in the mail: the American Media, Inc. (AMI), building in Boca Raton, Florida; the National Broadcasting Company (NBC) offices in New York City; the U.S. Capitol Complex in Washington, DC; and two facilities of the U.S. Postal Service (USPS).1 Remediation of the AMI and NBC buildings was done by private companies; government agencies led the remediation effort in the Capitol and USPS buildings. Separate entities controlled each contamination epicenter, and top decision makers gained varying degrees of stakeholder confidence for their health and safety pronouncements. A review of the four case studies reveals the
ways the larger social, economic, and political context affects how human health risk is defined, remedied, or contested. Another perspective on the anthrax attacks of 2001 can be found in P.S. Brachman’s thorough commentary, Bioterrorism: An Update with a Focus on Anthrax (2002).
UNCERTAIN SCIENCE, CERTAIN SOCIAL DIVISION
A successful decontamination project will require the following:
Removal of the threatening agent to the greatest extent feasible.
Certification that the property is as safe as it was before contamination occurred.
Public or stakeholder acceptance of the credibility of those who have certified the safety of the property.
Those three elements are both technical and social, and they are all difficult to achieve. But without all three elements, in the case of a real-world decontamination effort, the answer to “How clean is safe?” would likely be, “Cleaner than you claim.”
Consider, for example, the hypothetical situation of an anthrax contamination of a major metropolitan airport, San Francisco International (SFO). Decades of practical work in decontaminating laboratories at Fort Detrick, Maryland, and elsewhere, demonstrate convincingly that cleanup and reoccupation of buildings are achievable. But even if those same techniques used at Fort Detrick were applied at SFO, at the end of the project, someone must say, “It is safe to go back into the airport.”
Were that announcement marred by uncertainty about residual contamination, or about the decision-making process regarding safety, it is unlikely that SFO would reopen. If decision makers stated that scientists think they may have decontaminated SFO, stakeholders would not likely be convinced that the airport is safe. And yet, officials would not be able to claim with certainty that no spores remained after cleanup. “Zero spores,” after all, is an undetectable quantity. Policymakers must be able to state, with credibility and defensibility, that they have used the most conservative science available and that they have used proven decontamination techniques (Hsu et al., 2002). Decision makers also will need the public to see their decisions as legitimate. Legitimacy can not be commanded, and it does not flow automatically from competent science.
No Universal Definition of Health Risk
There is no documented threshold for cleanup of B. anthracis—especially in its weaponized form—below which no health effects would occur. Raber and colleagues (2003) noted that uncertainties surrounding B. anthracis contamina-
tion, such as its persistence and LD50, present considerable difficulties in defining the amount of cleanup possible. And they note that “it is important to emphasize that decontamination and cleanup issues are not only agent-specific, but scenario-dependent and site-specific as well.” Similarly, an NRC report (1996) noted that “the appropriate level of effort for a risk characterization is situation specific.” The implication for this committee’s task is that the committee can provide general guidance about when a facility is safe for reoccupation, but it cannot provide a recipe for every eventuality.
Raber and colleagues (2004), summarizing a report by the U.S. Government Accounting Office (GAO), also brought special attention to the issue (p. 38):
First, there is now consensus among experts that even a few anthrax spores could be harmful to a susceptible individual. Second, according to officials from the US Army Medical Research Institute of Infectious Disease, what is most important is not the number of spores in a facility, but whether or not any spores are found. Finally, the unpredictability of the lethality of anthrax, the broad spectrum of population potentially at risk of exposure, and the inability to determine the route that contaminated mail might take as well as the extent of cross contamination make it “… extremely difficult to establish the health risks associated with a release of a biological agent, such as anthrax.”
Analytic-Deliberative Process to Identify Health Risk
A previous NRC report on risk assessment states “it is necessary to reconceive risk characterization in order to increase the likelihood of achieving sound and acceptable decisions” (NRC, 1996). That report conceived of risk characterization as more than a specification of technical inputs. It recommended “a combination of analysis and deliberation” which it called the analytic-deliberative process, as a way to create and provide information about risk that would
… describe a potentially hazardous situation in as accurate, thorough, and decision-relevant a manner as possible, addressing the significant concerns of the interested and affected parties, and to make this information understandable and accessible to public officials and to the parties.
That committee also noted that when the stakes are high but public trust in responsible organizations is low, “the organization may need to make special efforts to ensure” that risk decision processes are seen as legitimate by affected parties. “Adequate risk analysis and characterization thus depend on incorporating the perspectives and knowledge of the interested and affected parties from the earliest phases of the effort to understand the risks” (NRC, 1996). The report noted that the analytic-deliberative process should incorporate scientifically rigorous analysis with the perspectives and knowledge of the interested and affected parties. “The process must have an appropriate diverse participation or represen-
tation of the spectrum of interested and affected parties, of decision makers, and of specialists in risk analysis, at each step.”
In testimony before the U.S. House of Representatives, Myke Reid drew attention to problems of cost, noting that “huge costs and delays have sometimes resulted when a risk situation was inadequately diagnosed, a problem misformulated, key interested and affected parties did not participate, or analysis proceeded unintegrated with deliberation” (Reid, 2003). Such conditions would likely hold in a future biocontamination of a public facility. But it need not be so. Tools such as the decision-making framework outlined in Chapters 11 and 12 can lower the probability that such trouble would ensue.
CASE STUDY SELECTION
Case studies can be used to illuminate the social aspects of a decontamination project. Along with extant research on risk communication and disaster response, case studies provide lessons learned—both positive and negative. As this committee explicates the case studies, it is not passing judgment on decision makers or decisions. The choices were often difficult, and they occurred in fast-moving and highly uncertain technical and political environments. The cases provide a range of examples for leadership, risk communication, and risk dilemmas. The point is not to blame but to use available knowledge to address the important question of what constitutes an acceptable level of cleanup for safe reoccupation of a facility.
The anthrax attacks in the fall of 2001 (the major milestones are chronicled in Box 4-1) resulted in extensive contamination of several facilities. But even in locations where the contamination was relatively contained, substantial disruption resulted. The amount of B. anthracis in two letters—one addressed to Senator Patrick Leahy (D-Vermont) and one to Senator Thomas Daschle (D-South Dakota)—has been estimated at 1-2 grams (D. Canter, EPA, presentation to committee, November 24, 2003). The total cost of decontamination of the affected buildings on Capitol Hill in Washington, DC, apparently cannot be estimated. “Capitol Hill anthrax-related cleanup cost for all 30 sites is estimated to be about $27 million by EPA. A total cost estimate for the Brentwood and Trenton facilities of about $200 million is probably an underestimate. The estimate for stripping and fumigation alone at the State Department to date is about $10 million” (D. Canter, EPA, presentation to committee, November 24, 2003). Chapman and Leng (2004) report that “almost $1 billion [was spent] to test for, remediate, and prevent anthrax contamination.” Despite the lack of solid information about the total cost, two conclusions can be made confidently: The costs were high and, in some cases, unnecessarily so because there was not enough accurate information available about the buildings themselves or about how to proceed. Accurate floor plans, for example, and validated protocols for sampling were not available at the start of the process (Schaudies and Robinson, 2003).
October 2, 2001—An infectious disease physician recognized a possible case of inhalational anthrax in a man hospitalized in Palm Beach County, Florida. This physician contacted the local health officer in Palm Beach County, who immediately began a public health investigation. By October 2, there were already 7 persons with cutaneous anthrax in the northeastern United States, but none had yet been diagnosed.
October 4—The microbiologic diagnosis of B. anthracis was confirmed by the Florida Department of Health and the Centers for Disease Control and Prevention (CDC), and the diagnosis was made public. Epidemiologic and environmental investigations were launched to determine the source of the patient’s anthrax exposure. Evidence of contamination with B. anthracis was found at American Media Inc. (AMI) in Boca Raton, Florida, where this first victim worked as a photo editor.
October 5—The first victim of the anthrax attacks died. A second AMI employee, who had been hospitalized for pneumonia on September 30, was diagnosed with inhalational anthrax. He was an employee in the AMI mailroom.
October 6—The Palm Beach County Health Department began to obtain nasal swabs from those who had been in the AMI building in an attempt to define exposure groups. Because nasal swab testing was known to be an insensitive diagnostic test, the health department also recommended prophylactic antibiotics for all those people who had been in the AMI building for at least 1 hour since August 1 regardless of the results of their nasal swab tests. Environmental samples taken from the mailroom showed evidence of B. anthracis.
October 7—A nasal swab was positive on another employee. A swab from the first victim’s computer screen was positive. The AMI building was closed.
October 9—The New York City Department of Health notified CDC of a woman with a skin lesion consistent with cutaneous anthrax. The woman, an assistant to NBC anchor Tom Brokaw, had handled a powder-containing letter postmarked September 18 at her workplace.
October 13—Another cutaneous case of anthrax was recognized in a 7-month-old infant who had visited his mother’s workplace, the ABC office building on West 66th Street in Manhattan, on September 28.
October 13—Symptoms of cutaneous and inhalational anthrax in New Jersey postal workers began to be observed and reported by physicians to the New York City Health Department. Diagnoses of anthrax are confirmed by the CDC on October 18 and 19.
October 15—A staff member in the office of Senator Daschle in the Hart Senate Office Building opened a letter (postmarked October 9) which contained a powder and a note identifying the powder as anthrax. The powder tested positive for B. anthracis on October 16. Nasal swab testing of anthrax spores was performed on 340 Senate staff members and visitors to the building who potentially were exposed and to approximately 5,000 other people who self-referred for testing. This testing indicated exposure in 28 persons. Antimicrobial prophylaxis was administered on a broader scale and environmental testing was initiated.
October 19—CDC linked the four confirmed cases of anthrax to “intentional delivery of B. anthracis spores through mailed letters or packages.”
October 19-22—Four postal workers at the Brentwood Mail Processing and Distribution Center in the District of Columbia were hospitalized with inhalational anthrax. The Brentwood facility was closed on October 21. On October 22 two of these four postal workers died.
October 24—CDC sent an advisory to state health officials via the Health Alert Network recommending antibiotic prophylaxis to prevent anthrax for all people who had been in the non-public mail operations area at the USPS’s Brentwood Road Postal Distribution Center or who had worked in the non-public mail operations areas at postal facilities that had received mail directly from the Brentwood facility since October 11.
October 27—A CDC alert recommended antibiotic prophylaxis for workers in the mail facilities that supplied the CIA, the House office buildings, the Supreme Court, Walter Reed Army Institute of Research, the White House, and the Southwest Postal Station after preliminary environmental sampling revealed B. anthracis contamination in these mailrooms.
October 31—A 61-year-old female hospital stockroom worker in New York City died from inhalational anthrax after she had become ill with malaise and myalgias on October 25. The source of her exposure remains unknown despite extensive epidemiologic investigation.
November 16—A 94-year-old woman residing in Oxford, Connecticut, was hospitalized with fever, cough, and weakness. She died on November 19. Her diagnosis was confirmed as B. anthracis on November 20 by the Connecticut Department of Public Health Laboratory. Subsequent environmental and epidemiological testing indicated exposure from cross-contaminated letters.
Reprinted from Gursky et al., 2003.
Cost, however, might not be the major consideration in the decontamination of a public transportation facility. There is no economic justification that would overcome the perception that such a facility has been poisoned with weaponized anthrax. Chapman and Leng (2004), on considering a contamination event at SFO, say, “One year of cleanup (shutdown) would translate to 6-10 years for recovery of the airport…. With preplanning, it might be possible to reduce the time from years to months.” Given the significant capital investment an airport complex represents and its role in the overall national economy, decision makers might expend extremely large amounts of money to avoid the permanent loss of an airport.
American Media, Inc.
The AMI building in Boca Raton, Florida, was the first in which B. anthracis spores were detected in the fall of 2001. The contamination was extensive, and the building remained contaminated until July 2004. Details of the receipt of a letter containing a white powder were obtained in retrospective interviews the Federal Bureau of Investigation (FBI) conducted with AMI employees, but the letter was never found. In fact, during interviews with employees, it was speculated there might have been a second letter or a package that contained white powder. It was not until Robert Stevens, an AMI employee, became ill, was hospitalized and subsequently diagnosed, and died of anthrax that government officials began to test areas where Mr. Stevens lived and worked. Places he shopped and the areas he fished also were tested.
Stevens was hospitalized October 2, 2001. His diagnosis of inhalational anthrax was announced on October 4, 2001. The next day, U.S. Department of Health and Human Services Secretary Tommy Thompson announced that the case was “isolated.” New York City Mayor Rudolph Giuliani told New York residents that they should not be concerned. Thompson also said, “We do know that he drank water out of a stream when he was traveling to North Carolina last week” (CNN, 2001). Those overconfident pronouncements were unwarranted and are recognized as the kind of statements that lead to mistrust of officials and experts (Freudenburg, 1993). A team from CDC traveled to North Carolina to test areas Stevens had visited the previous week. In addition, in an effort to determine the possible source of the infection, those close to Stevens were interviewed (Cole, 2003).
Medical practitioners involved with Mr. Stevens’s care notified state health officials as soon as they suspected inhalational anthrax. Local health authorities quickly took action to mobilize a response team without waiting for confirmatory testing from CDC. The team alerted health care providers at area hospitals to signs and symptoms of anthrax, which possibly led to the diagnosis of a second case of inhalation anthrax in Florida. The response team set up a telephone hotline for persons who believed they had been exposed, and a website was
created to address questions from the public about anthrax. The AMI case also led to the initiation of a large-scale postexposure prophylaxis program (Heyman, 2002).
Initial sampling of the AMI building was carried out primarily by the FBI, which sought forensic evidence. A response team from CDC also was sent to quantify the building’s contamination. Two weeks after the initial investigation began, EPA took over the lead in collecting samples from the building. Sampling was performed to track contamination in the building so that remediation recommendations could be made to the building’s owner. A plan delineating sampling procedures in response to a biothreat may be similar to epidemiological sampling strategies, but those actions had not been performed on a large scale with B. anthracis until the AMI case. The CDC sampling team initially collected samples primarily from the mailroom and from Stevens’s office to quantify contamination in those areas (K. Martinez, CDC, presentation to committee, November 24, 2003).
By the time the U.S. Environmental Protection Agency (EPA) began to collect samples, it was obvious from the results of previous FBI and CDC testing that the AMI building was highly contaminated with B. anthracis spores. EPA used blueprints of the building to identify areas for collecting samples from various surfaces. Samples were taken primarily using wipe and HEPA vacuum sock methods. Those samples eventually helped to determine the method of decontamination to be used. During the summer of 2002, CDC and the FBI returned to the AMI building to collect additional samples (K. Martinez, CDC, presentation to committee, November 24, 2003).
In July 2004, the AMI building was declared to have been successfully decontaminated. There are lessons from the case that could be applied to a contaminated public transportation facility, but some of the specifics of the AMI example are unusual and would likely not apply to our hypothetical future case. The AMI case raises the possibility that a contaminated building could be abandoned and that nobody would take responsibility for its remediation. It is also an example of a remediated building no longer in use for its original purpose: AMI employees relocated in 2001. The AMI building is smaller than the other facilities that had to be decontaminated. Most important, decontamination experts had 3 years to plan their procedures. All of those factors limit the lessons from the AMI case, because none of them is likely to operate if a large public transportation facility were to be contaminated with B. anthracis.
There are, nonetheless, several interesting observations to make. The AMI building cost “significantly less than $5 million” to decontaminate (J. Mason, Sabre Technical Services, presentation to committee, October 13, 2004) and the decontamination effort was a technical success. Extensive sampling done throughout the building after decontamination showed that “the ‘no growth’ standard for all environmental samples was achieved as a result of the fumigation” (Sabre Technical Services, 2004). The standard was no growth to an 8 log kill, and the
company responsible for the cleanup demonstrated its results with about 2000 spore strips. The AMI building experience also demonstrates the utility of having clearly established lines of responsibility for decontaminating a facility. The vendor—anticipating the need to explain how a successful cleanup was validated—used three-dimensional graphic software to track the sampling database, making an otherwise complex endeavor readily understandable to a nontechnical audience.
The committee was told (J. Mason, Sabre Technical Services, presentation to committee, October 13, 2004) that although the compressors were turned off, the air conditioning was left running at the AMI building for 2 years, and that the air inside the building was probably exchanging with the outside air “four times a day.” Greater control of the building and its contaminants was instituted after a responsibility for decontamination was clearly established.
National Broadcasting Company
A letter tainted with B. anthracis arrived at NBC on September 19, 2001. Two cases of cutaneous anthrax were later identified there—one of them in an NBC employee who tested positive for cutaneous anthrax on October 12, 2001. There was extensive destruction of the physical plant at NBC. Contaminated areas included the Nightly News set, the mailroom, and a security office. A decision was reached early to evacuate the third floor, where the news operation is located. Extensive sampling was conducted above and below the third floor: 1200 employees were tested with nasal swabs; the test was administered to all employees who requested it (J. Eck, NBC, presentation to committee, March 29, 2004).
NBC officials concluded that one “can’t do enough communicating” and that “even if [they] found one spore” they would continue to decontaminate (J. Eck, NBC, presentation to committee, March 29, 2004). NBC management said it wanted to be able to “say with a straight face to our employees that we sampled until we found no more spores.” NBC also communicated to its employees through the behavior of upper management. Managers “made sure [they] went down and ate at the commissary” to demonstrate that it was safe to eat there. Indeed, managers said they “sought out people’s opinions not only by walking around but all email was responded to” and that was a major reason there was “no panic at any point.” Although NBC seems to have inspired considerable confidence in its employees, no research effort was conducted to assess employee concern.
Because there were no guidelines to follow regarding what constituted adequate cleanup, a “cross section of employees were involved in management of the crisis.” Personnel from different departments were involved in making critical decisions. Management believes that involvement helped build trust through-
out the organization. In all, 1200 employees were tested with nasal swabs, and all tests were negative. Of course nasal swabbing is a not diagnostic tool, so its utility in the NBC case was a way to communicate to employees that the company was competent to perform during the crisis.
New York City Department of Health and Mental Hygiene officials believed that NBC management went overboard (J. Prud’homme, NYCDOHMH, presentation to committee, January 28, 2004), “setting a standard not every company could live up to.” “When in doubt we gutted,” was the decontamination principle employed by NBC corporate leadership (J. Eck, NBC, presentation to committee, March 29, 2004). NBC set an explicit policy that it would continue sampling until no B. anthracis was detected. Yet management did not promise that every spore was gone. No unconditional guarantee was offered or demanded.
The NBC case is an apparent success story, both technically and socially, but there are limits to the conclusions it supports. The committee’s information on this case came exclusively from two NBC managers and from a city health official responsible for overseeing the restoration effort. Requests from committee staff and from a committee member for interviews with a wider range of NBC employees were not responded to positively. The committee asked NBC managers how they knew that employees were satisfied with the decontamination, and it was told that the lack of complaints was indicative of the acceptability of risk.
Nevertheless, the NBC response appears to have involved transparency in decision making, constant communication of information to stakeholders, involvement of affected parties in deciding policy, and a commitment to additional cleaning if contamination was discovered. The goal would be zero spores, the committee was told. That would seem important both as a technical goal and as a commitment to health and safety on the part of management.
There was an extensive search for B. anthracis contamination in more than two dozen government buildings in Washington, D.C. The uncertainties were much greater than those at NBC and AMI, although they were similar to those faced by the USPS. There were more stakeholders involved and there was more attention from the media. There also was a pressing need, or a perceived pressing need, to reopen the buildings quickly. According to the EPA’s Federal On-Scene Coordinator’s (FOSC) Report, “The Capitol Hill Site initially consisted of 26 buildings with suspected anthrax contamination. All 26 buildings were sampled; anthrax was detected in seven buildings, all of which were decontaminated and cleared for re-entry after confirmation sampling” (EPA, 2002). It was a massive effort, involving more than 50 organizations. “Trillions” of anthrax spores were removed in the decontamination (EPA, 2002). Those seven buildings were:
P Street Warehouse
Supreme Court Building
The committee did not review the decontamination effort for each building but the official decontamination standard was the same for each building: Extensive sampling should show that there was no spore growth. In each case there was no significant conflict among stakeholders about when or whether to return a building to operational status. To some degree, that represents a risk communication success. However, there were different procedures for certifying that buildings could be reoccupied, and there was no coherent, organized entity that oversaw decontamination throughout. That organizational failure likely added to the uncertainties, costs, and length of time that buildings were closed. A more coordinated response would have increased efficiency and effectiveness. The FOSC report notes that “no single entity” would accept responsibility for reoccupying the buildings after decontamination (EPA, 2002). It also noted that CDC “left the site after the first few weeks” and did not return until the end of the decontamination. There is obviously a need for clear lines of responsibility in any decontamination effort. It seems likely that the cleanups at the Capitol complex would have gone more smoothly had there been a broader understanding of responsibilities.
Some “20 to 130 initial samples collected in each building” were initially taken in the 26 buildings. Overall, EPA spent about $27 million “to clean up anthrax contamination on Capitol Hill, using funding from its Superfund program” (EPA, 2002). The steep expenses incurred seemed to result from several factors. The high-profile users of the buildings undoubtedly created pressure to reopen the buildings quickly, yet a conservative definition of “clean” was adopted by EPA. More important, there was a lack of a standard protocol to drive remediation, which in some cases led to repeated decontamination.
The B. anthracis crisis in Washington, D.C., started on October 15, 2001, on the 6th floor of the Hart Senate Office Building (HSOB), when a Senate staff member opened an envelope addressed to Senator Tom Daschle. Senate staff had been trained to be alert for B. anthracis, so they knew the procedure. The Capitol Police were on the scene within minutes, soon followed by the force’s hazardous device unit (Hsu et al., 2002). On-the-spot tests gave positive indicators for B. anthracis within 15 minutes. The ventilation system was shut off about 45 minutes after the initial discovery of the contamination. Medical staff immediately collected nasal swabs from those most likely to have been exposed, and they initiated antibiotic prophylaxis. Within 9 hours of the initial exposure, everyone
in the office suites of Senator Daschle and Senator Russell Feingold (D-Wisconsin) and official responders had been tested with a nasal swab (Hsu et al., 2002).
EPA was notified on October 16 and the building was closed on the evening of October 17. Over the next three days, nasal swab samples were collected from all HSOB employees and from anyone else on Capitol Hill who requested it (Hsu et al., 2002). One report in the Journal of the American Medical Association reported that more than 7000 nasal swabs were analyzed (Weis et al., 2002). Everyone tested was given antibiotic prophylaxis, pending test results. CDC arrived on October 16 and defined “the population at risk…as persons in the exposed area during or after the time the contaminated envelope was processed or opened” (Hsu et al., 2002). Other organizations, including EPA, the U.S. Coast Guard, FEMA, FBI, and the National Institute of Occupational Safety and Health, were also involved in the response: “The incident response involved coordination of more than 50 organizations” (Schaudies and Robinson, 2003). The situation at the Hart Building and other contaminated buildings was unprecedented and confusing. Initially, some expected the entire building to be decontaminated. Later it was decided that by following the mail trail, only those parts of the building that were sampled and found to be contaminated would be treated (D. Canter, EPA, presentation to committee, November 24, 2003).
EPA defined the acceptability of the Hart Building cleanup as zero B. anthracis growth on any samples taken. The standard was not zero B. anthracis, which is impossible to demonstrate. To ensure credibility, EPA took a large number of samples and, according to Raber and colleagues (2003) “verbally indicated that essentially all surfaces in the Hart Building were swabbed.” Faced with unprecedented problems at HSOB, the determination was made that there was no acceptable level of B. anthracis spores that could remain in the building (D. Canter, EPA, presentation to committee, November 24, 2003). “Cleanup” therefore was defined as no detectable growth of B anthracis spores in any sample. HSOB reopened on January 22, 2002.
In a report available from the Office of Pesticide Programs at EPA, Schaudies and Robinson (2003) note that numerous problems with the information available to facilitate the response in the Capitol complex—problems that included inaccurate floor plans, nonexistent protocols for sampling, long hours, and constant strain:
… the sampling and remediation activities were successful overall based on the fact that all clearance samples showed no growth in any areas previously contaminated with B. anthracis. In addition, no one has presented with symptoms of anthrax since buildings on Capitol Hill were remediated and cleared for reoccupancy. This is clearly the best measure of the success for the response and remediation activities … Over 9,000 samples were collected throughout the course of the response.
From available documents and presentations to the committee, it is apparent that many things went right on Capitol Hill in the fall of 2001. But there were problems, too, and any future decontamination effort can benefit from attention to them. The Capitol Hill response was uncoordinated, and it was marked by inconsistencies, especially concerning the closing of buildings. It does not appear that the Supreme Court building was closed at all, even though B. anthracis was found there. The contaminated letter was opened in the Hart Building on October 15, but the building was not closed until 2 days later on October 17. The Longworth Building was closed that day, and most of it was reopened three weeks later on November 5 (EPA, 2002). The Ford Building was closed October 20, as was the Dirksen Building. The P Street Warehouse, a mail facility, was never closed, even though B. anthracis was found in several places within it. The Russell Building was closed on October 20, and then again on November 17, and then decontamination occurred.
There were two sign-off procedures for the Capitol complex. The Dirksen Building and the Supreme Court were certified “informally,” in the words of the FOSC’s report (EPA, 2002). The certification was done by circulating a “sign-off” sheet between the Incident Commander, a contractor, and the attending physician. The same process was used for parts of the P Street Warehouse and the Russell Building. Later, a more thorough process developed that increased the number of agents to sign off on decontamination.
One result of the uncoordinated response was a haphazard decontamination standard. EPA initially had “set a criterion of zero spore growth for determining whether decontamination had been successful in each building. However, no frame of reference existed for such a criterion. The meeting between CDC and EPA concluded that best professional judgment should be used in reviewing the data to determine when the remaining buildings were ready for reoccupation” (EPA, 2002).
An effective response to hazardous situations requires decision-making transparency, coordinated decisions, and meaningful risk communication. Every action that officials and organizations take is fraught with communicative import. Consider nasal swabbing: Nasal swab sampling for B. anthracis is not diagnostically useful. It was known in the fall of 2001 that, indeed, some experts argue that it is a complete waste of resources to conduct massive nasal swab testing. But the extensive sampling likely conveyed the message that the hazard that building occupants might face was being taken seriously.
It is important to remember that all official actions have meaning beyond their instrumental utility. For example, on Capitol Hill extensive efforts were made to provide antibiotic prophylaxis. Such actions are medically useful—there likely would have been more discovered cases of anthrax absent such prescription. But the action also sent the message that people’s concerns were valid.
United States Postal Service
Because the B. anthracis traveled through the postal system, several USPS facilities—although the precise number is not clear—were contaminated with weaponized B. anthracis, some more extensively than others. The Hamilton Processing and Distribution Center in Trenton, New Jersey, and the Brentwood Processing and Distribution Center in Washington D.C. (later renamed Curseen-Morris Processing and Distribution Center to honor the men who died from anthrax) experienced extensive contamination. Another postal facility, in Wallingford, Connecticut, was the route through which contamination reached a citizen in Oxford, Connecticut. There were two cases of anthrax at Hamilton, four at Brentwood with two deaths, and one death in Connecticut. All of the deaths were from inhalational anthrax.
The contaminated envelopes addressed to Senators Tom Daschle and Patrick Leahy entered the mail stream at Hamilton on October 9, 2001 and Brentwood on October 11, 2001. The Wallingford facility was most likely cross-contaminated by mail sent from Hamilton. The doses to which people were exposed are not known, but the doses are presumed to have been higher among postal workers than among other people.
The technical problems of detection and decontamination at USPS facilities were similar to those at the Capitol complex, although the volume of space at Hamilton and Brentwood that required decontamination was considerably larger. According to the Morbidity and Mortality Weekly Report, on October 18, 2001, a “postal service contractor” took 29 samples from the mail sorting area at Brentwood (CDC, 2001). CDC initiated its own investigation there on October 20. The fate of the original 29 samples taken by the USPS contractor is not clear. Dewan and colleagues (2002) indicate that the first inhalational anthrax case at Brentwood was diagnosed on October 19 and confirmed on October 21. Also on October 21, postal worker Thomas Morris, Jr., was diagnosed with anthrax (he died later that night). Brentwood was closed that day. Nasal swab samples were taken from Brentwood employees and from people who visited the facility between October 10 and October 21, for a total of 3110 people (Dewan et al., 2002). Seventy-eight percent of Brentwood employees were given antibiotic prophylaxis (1870 of 2403) (Dewan et al., 2002).
There seems to have been difficulty identifying B. anthracis at the Wallingford postal facility. Counter to CDC recommendations, initial sampling by USPS was done with dry wipes; that effort yielded no positive samples. After four attempts, the last two using wet wipes and HEPA vacuums, mail-sorting machines were found to be contaminated (GAO, 2003). From the final samples, taken on November 28, two results were provided by the CDC-contracted laboratory. One was “about 3 million colony forming units (CFUs) of anthrax (that is, 5.5 million CFUs per gram of dust) in a sample collected from a heavily contaminated mail-sorting machine.” Decontamination of the machines began on Decem-
ber 2. According to GAO (2003), “when anthrax contamination was first identified [at Wallingford, on December 2, 2001]—USPS met with workers to inform them that ‘trace’ amounts of anthrax had been found in the samples collected on November 28.” The phrase “trace amount” was used apparently on advice of the chief epidemiologist for the state of Connecticut. On December 21, 2001, district managers told workers there was a “concentration” of spores in a sample. The more quantitative information was not conveyed to workers until 9 months later.
It is generally thought that USPS management could have communicated more effectively with employees about decisions and procedures for nasal swabbing, antibiotic prophylaxis, and decontamination. Comparing their experience with events at the Capitol complex, some USPS employees expressed the belief that their concerns were not taken seriously. The Hart Building had been closed quickly, thousands were immediately put on antibiotic prophylaxis, and nasal swabbing was used extensively. By contrast, USPS facilities were not closed until there were official diagnoses of anthrax, the recommended dosage of antibiotics was different for employees in the two places, and nasal swabbing was more limited at the postal facilities than it had been on Capitol Hill.
Such differences were taken by some USPS employees as symbols that their concerns were not as important as were those of Capitol complex employees. Additionally, all four anthrax cases at Brentwood were among African Americans, which contributed to a perception among some that race was important in attending to the crisis. The committee does not believe that race was a factor in the deliberations, but such perceptions clearly can be important in establishing trust among stakeholders.
It was quickly known that B. anthracis had contaminated the Daschle suite in the Hart Building because people saw the powder and it was analyzed immediately. This was not the case at any of the postal facilities. At the time, the prevailing assumption was that B. anthracis could not escape from a taped envelope. By the time officials realized that postal facilities were contaminated with B. anthracis it was generally too late to employ nasal swabs as a means of assessing the extent of contamination. It is clear that there were some behavioral and perceptual issues regarding the postal service contaminations. It is not this committee’s task to research those issues in great detail. It is, however, this committee’s task to glean lessons from available evidence regarding acceptable risk. Judging from that evidence, and projecting a B. anthracis attack on a public transportation facility, the committee has concluded that risk acceptability would be enhanced to the extent that trust is fostered between labor and management and to the extent that decision making about important issues includes those who might bear the brunt of decisions about risks.
There are two important limitations in this discussion about the B. anthracis contamination at USPS facilities. First, the committee heard no direct testimony from postal workers. Second, although the committee requested decontamination data from the USPS, those requests were not responded to positively.
The events of the fall of 2001 provide several valuable lessons. Before those incidents it was not known that cross-contamination with B. anthracis can be a significant risk. Although it was thought that 8000 to 10000 anthrax spores are needed to infect someone, this is an average figure and some people may be infected at much lower doses. The spread of B. anthracis was thought to be unlikely from a closed envelope. We now know that not only can B. anthracis escape from an envelope but that it can subsequently spread throughout a facility (Kournikakis et al., 2003).
The committee has heard from several sources that the needs of law enforcement and public health agencies sometimes do not coincide, and that the lack of concordance can hinder effective responses to a crisis. The committee concurs with the opening sentence of a recent agreement signed by law enforcement agencies and the New York City Department of Health: “In the event of a suspected or confirmed bioterrorist (“BT”) event, it is essential that public health and law enforcement agencies coordinate their investigations closely, so that shared objectives (e.g., determining where and when a release may have occurred) can be reached” (NYDOHMH, 2004).
The committee itself encountered organizational resistance to complete sharing of relevant data. Notable in this regard is a conclusion reached by the 9/11 Commission regarding the other terrorist attacks of 2001: “The culture of agencies feeling they own the information they gathered at taxpayer expense must be replaced by a culture in which the agencies instead feel they have a duty to inform the public” (NCTAUS, 2004).
The anthrax response and subsequent decontamination experiences provide several lessons. The acceptance of risk is likely to be enhanced if trust is fostered and preserved among authorities, subject matter experts, and affected parties; if decision making about key issues includes those who must bear the brunt of the consequences of those decisions; if affected populations see—because of transparent decision making—that their health and safety are given higher priority than material considerations, such as disrupted work schedules and cleanup costs; if the complex technical dimensions of the problem are translated into terms that are meaningful to nontechnical audiences; and if constant and open communications is maintained between responsible officials and directly and indirectly affected parties.
FINDINGS AND RECOMMENDATIONS
Acceptability is not a technical concept. It is a values concept. It is, therefore, best constructed through an analytical and deliberative process that involves key stakeholders in a potentially harmful situation. Without trust, acceptability is difficult
to achieve. Effective leadership in dangerous situations is based on openness and honesty, even when bad news must be conveyed. Transparency in decision making can contribute substantially to ensuring the acceptability of risk. Panic is rare in disasters, and it is an unhelpful idea for explaining how people respond to frightening situations and information. After the 2001 anthrax attacks, decision makers sometimes relied on assumptions that later proved unfounded; their subsequent actions resulted in significant problems with communicating the degree of risk involved to the stakeholders.
Risk managers who face potential contamination should assume that the problem could be worse than they initially think. In remediation projects, the public should be seen as an asset, not a liability, and information should be made available widely. Indeed, the public should participate actively in decision making in the aftermath of an attack. Following the lead of previous work by the National Academies, the committee recommends that an analytical deliberative process be used to determine appropriate approaches for cleanup.
Relevant data from the sites contaminated in 2001 were not shared with all necessary parties, partly because of the differing goals and objectives of law enforcement and public health agencies. Lack of data sharing can compromise health in the aftermath of a biological attack.
Agencies and organizations entrusted with data relevant to public health should make every effort to share this information. Cooperation is the key to decreasing public anxiety, and agreements, such as the one signed by the New York City Department of Health and relevant law enforcement agencies, should be in place to protect public health and safety by allowing the process of forensic evidence collection and decontamination to proceed unimpeded by one another.
Brachman, P.S. 2002. Bioterrorism: an update with a focus on anthrax. American Journal of Epidemiology 155(11): 981-987.
CDC (U.S. Centers for Disease Control and Prevention). 2001. Evaluation of Bacillus anthracis contamination inside the Brentwood Mail Processing and Distribution Center—District of Columbia, October 2001. MMWR 50(5): 1129-1133.
Chapman, R.E., and C.J. Leng. 2004. Cost-Effective Responses to Terrorist Risks in Constructed Facilities. Gaithersburg, Maryland: National Institute of Standards and Technology. NISTIR 7073.
CNN. 2001. Florida anthrax case ‘isolated’. [Online] Available at: http://archives.cnn.com/2001/HEALTH/10/05/florida.anthrax/.
Cole, L.A. 2003. The Anthrax Letters. Washington, DC: Joseph Henry Press.
Dewan, K., A.M. Fry, K. Laserson, B.C. Tierney, C.P. Quinn, J.A. Haslett, L.N. Broyles, A. Shane, K.L. Winthrop, I. Walks, L. Siegel, T. Hales, V.A. Semenova, S. Romero-Steiner, C. Elie, R. Khabbaz, A.S. Khan, R.A. Hajjeh, A. Schuchat, and the Washington, D.C. Anthrax Response Team. 2002. Inhalational anthrax outbreak among postal workers, Washington, D.C., 2001. Emerging Infectious Diseases 8(10):1066-1072.
EPA (U.S. Environmental Protection Agency). 2002. Federal On-scene Coordinator’s Report for the Capitol Hill Site, Washington, D.C. EPA Region 3, Philadelphia, PA: GS-10F-0076K.
Freudenburg, W.R. 1993. Risk and recreancy: Weber, the division of labor, and the rationality of risk perceptions. Social Forces 71(4): 909-932.
GAO (U.S. General Accounting Office). 2003. Capitol Hill Anthrax Incident. Report to the Chairman, Committee on Finance, U.S. Senate: GAO-03-686.
Gursky, E., T.V. Inglesby, and T. O’Toole. 2003. Anthrax 2001: observations on the medical and public health response. Biosecurity and Bioterrorism 1: 97-110.
Heyman, D. 2002. Lessons from the Anthrax Attacks, Implications for U.S. Bioterrorism Preparedness: A Report on a National Forum on Biodefense. Defense Threat Reduction Agency Report.
Hsu, V.P., S.L. Lukacs, T. Handzel, J. Hayslett, S. Harper, T. Hales, V.A. Semenova, S. Romero Steiner, C. Elie, C.P. Quinn, R. Khabbaz, A.S. Khan, G. Martin, J. Eisold, A. Schuchat, and R.A. Hajjeh. 2002. Opening a Bacillus anthracis-containing envelope, Capitol Hill, Washington, D.C.: the public health response. Emerging Infectious Diseases 8(10): 1039-1043.
Kournikakis, B., S.J. Armour, C.A. Boulet, M. Spence, and B. Parsons. 2001. Risk Assessment of Anthrax Threat Letters. Defence Research Establishment Suffield: Technical Report, DRES TR-2001-048.
NCTAUS (National Commission on Terrorist Attacks Upon the United States). 2004. The 9/11 Commission Report. Washington, DC: Government Printing Office.
NRC (National Research Council). 1996. Understanding Risks: Informed Decisions in a Democratic Society. Washington, DC: National Academy Press
NYCDOHMH (New York City Department of Health and Mental Hygiene), NYPD (New York City Police Department), and FBI (Federal Bureau of Investigation). 2004. Joint Field Investigation Following a Suspected Bioterrist Incident. [Online] Available at: http://www.phppo.cdc.gov/od/phlp/docs/Investigations.PDF.
Raber, E., T.M. Carlsen, K.J. Folks, R.D. Kirvel, J.I. Daniels, and K.T. Bogen. 2003. How Clean is Clean Enough? Recent Developments in Response to Threats Posed by Chemical and Biological Warfare Agents. Livermore, California: U.S. Department of Energy Lawrence Livermore National Laboratory.
Raber, E., T.M. Carlsen, K.J. Folks, R.D. Kirvel, J.I. Daniels, and K.T. Bogen. 2004. How clean is clean enough? Recent developments in response to threats posed by chemical and biological warfare agents. International Journal of Environmental Health Research 14(1): 31-41.
Reid, M. 2003. Re-opening of the US Postal Service’s Joseph Curseen, Jr. and Thomas Morris, Jr. Processing and Distribution Center. Testimony Before Committee on Government Reform, U.S. House of Representatives: October 23, 2003.
Sabre Technical Services. 2004. FIFRA Crisis Examption Final Report. Albany, New York: Sabre Technical Services, LLP.
Schaudies, R.P., and D.A. Robinson. 2003. Analysis of Chlorine Dioxide Remidiation of Washington, DC, Bacillus anthracis Contamination. McLean, Virginia: Science Applications International Corporation.
Weis, C.P., A.J. Intrpido, A.K. Miller, P.G. Cowin, M.A. Durno, J.S. Gebhardt, and R. Bull. 2002. Secondary aerosolization of viable Bacillus anthracis spores in a contaminated US Senate office. Journal of the American Medical Association 288(22): 2853-2858.