By LAWRENCE K. ALTMAN and GINA KOLATA, The New York Times
On Saturday, Sept. 8, employees of Inova Fairfax Hospital in Northern Virginia rehearsed their response to a mock terrorist attack that would overwhelm the medical system with hundreds or thousands of victims. The drill went well.
Meanwhile, at the Health and Human Services Department in Washington, officials were perplexed by Mr. Stevens's illness. Could he have become infected in some unorthodox way?
Government officials acknowledge that they found it hard to imagine a case in which a terrorist aimed airborne anthrax at just one person. "Everything we knew about the disease just did not fit with what was going on," Dr. Henderson said. "We were totally baffled."
As for health officials' initial response, Dr. Henderson said they often play down the seriousness of an outbreak to avoid frightening the public. But he added that he himself was critical of the practice. "It is not reassuring," he said.
Still, James Adams, a terrorism expert who is a senior associate at the Center for Strategic and International Studies in Washington, said he could sympathize with government officials. "It's the worst political problem," Mr. Adams said. The truth of the matter, he added, is: "We have no solution. Therefore, we can't bear to tell you about it."
But it was soon becoming impossible to play down the events. Reporters clamored for information. The requests "just really buried us in a way that we had not anticipated," Dr. Henderson said.
The Centers for Disease Control and Prevention, the federal agency responsible for investigating diseases, kept silent.
"Early on we were under the Federal Emergency Management Act and the decisions were made that C.D.C. should not be a locus of communications, in part because it was a criminal investigation and we were not really clear what the appropriate message was to put out," said Dr. Julie L. Gerberding, acting deputy at the agency's infectious disease center. "Soon thereafter it became clear that C.D.C. was desperately needed as a spokesperson for this outbreak, but by that time we were in a reactive state."
Mr. Thompson later held news conferences about the anthrax outbreak, assisted by government scientists. But his initial remarks left a lasting impression of overconfidence.
A basic principle of sound public health management is to have scientists inform and interpret what has happened promptly. But when communication was most critical early in the outbreak, no government scientist consistently delivered the clear message many people said was needed.
In describing the general problem, Dr. John F. Eisold, the Capitol physician, said that "the message was clearly a medical message, and you have got to have medical people talking about medical facts and not nonmedical people prescribing antibiotics."
Federal health officials were confused and disorganized. "We felt very strongly about the need to be available and to communicate, and there was just no way in the world you could," Dr. Henderson said. "We were just paralyzed." Meanwhile, the information vacuum was being filled by "experts" who came forward, some with questionable qualifications.
Another problem was the disease control agency's usual way of working. It "has traditionally been a very deliberative body, scientific in nature, that makes good policy decisions agonizingly over time," said Dr. Georges C. Benjamin, the secretary of the Maryland Health Department and president of the Association of State and Territorial Health Officials.
Ordinarily, once epidemiologists understand the nature of a particular microbe and the way it is transmitted, they can develop effective public health responses. But the rush of new information about anthrax and the postal system meant that the agency had little time to be deliberative. Dr. Benjamin said that meant "a new paradigm" for the agency; Dr. Gerberding of the disease control centers agreed and said the lesson had now been learned.
The Warnings: 'A Weird Disease' and Its Nuances
Anthrax "is a weird disease," Dr. Henderson said.
Before the attacks, he said, he did not appreciate many of its nuances. He and others writing a primer on anthrax in The Journal of the American Medical Association in 1999 relied primarily on reports of 18 patients who developed anthrax in the United States, a limited amount of information on a 1979 outbreak in Sverdlovsk, in the former Soviet Union, and a limited amount of research in the laboratory and with animals.
Scientists knew that anthrax spores could lie dormant in soil for decades and then cause disease. They also knew that inhalation anthrax occurred when spores entered the lungs and were swept into lymph nodes in the mediastinum, in the middle of the chest, where they germinated and pumped out toxins. But scientists still had much to learn.
Few scientists had ever considered how dangerous anthrax spores might be if they were sent through the mail. But in retrospect, it is clear that there were clues.
In Canada, military scientists investigated the question as a result of a hoax — a letter said to contain anthrax that forced the government to close a building in Ottawa last January.
That led Dr. Bill Kournikakis and his team at the Defense Research Establishment in Alberta to conduct an experiment. They used a harmless microbe to mimic how spores might disperse in an office or mailroom if an envelope containing anthrax were opened.
Their conclusion was chilling: a person could inhale a lethal dose of spores within seconds of opening an envelope. Those who remained in an affected room for more than 10 minutes could inhale far more than a lethal dose, depending on their location and the air flow.
In early October, when American epidemiologists linked the spread of anthrax to the postal system, Dr. Kournikakis said he sent the report to the disease control centers. But it went unread in the blizzard of e-mail messages that arrived at the agency, and it was not until three weeks later that officials learned of the study through other channels.
A similar warning had come in 1999, from William C. Patrick III, a government germ warfare expert, in a report for a government contractor exploring what might happen if an anthrax letter was opened. Finely milled spores, he wrote, could easily contaminate an office.
Medical experts also misjudged the difficulty doctors would have in diagnosing inhalation anthrax, assuming that a sophisticated surveillance system was needed to detect an attack. But Dr. Bush, the Florida infectious disease expert, says he knew immediately what was wrong with Mr. Stevens, the first patient with inhalation anthrax. He saw Mr. Stevens on the morning of Oct. 2. By 2 p.m., he was convinced.
"I had four pieces of information, all consistent with anthrax and not consistent with other organisms on my short list," Dr. Bush said. He called the local health department, telling officials there that he thought he had a victim of bioterrorism. And he sent samples of the bacteria to a state reference laboratory for further tests. By 8:30 the next morning, all three tests had come back positive.
Medical textbooks say that inhalation anthrax starts with mild, flulike symptoms that are hard to recognize, and that by the time it progresses to its severe phase, it is easy to diagnose but virtually impossible to cure. But the two postal workers who came to the emergency room at Inova Fairfax in October did not have textbook symptoms. The first patient did not even seem very ill, but a CT scan of his chest showed telltale signs of anthrax. The second patient complained of the worst headache of his life. But he did not have the classic signs of inhalation anthrax — bacteria in his spinal fluid and abnormalities in a chest scan. Doctors learned he had anthrax only when they examined his blood and saw the characteristic boxcar-shaped anthrax bacteria.
Both patients recovered with aggressive treatment — another surprise, considering how deadly the advanced stage of the disease was assumed to be. But that expectation was based on what scientists knew about the 1979 outbreak in Sverdlovsk, which was caused by a plume of spores accidentally released from a bioweapons factory.
Now, Dr. Henderson said, scientists realize they misread scientific papers, never appreciating that many more Soviets may have had the disease and survived. It is unclear how effective antibiotics were in Sverdlovsk, he said, because no one is sure how many people were given antibiotics and for how long they took them.
The anthrax attacks also pointed to another scientific mystery: how many spores does it take to infect someone? Could one spore cause a fatal disease? The two most recent deaths, of two women who were not postal workers — Kathy T. Nguyen in the Bronx and Ottilie W. Lundgren in Oxford, Conn. — raise the question, because no spores were found in their homes and the source of their infection is unknown.
Dr. Henderson and others now say that the outbreak illustrates an important lesson: the temptation to draw firm conclusions from a small database should be resisted, even if it is the only information available.
"There is a lot of feeling that we didn't know what we were doing as scientists in giving advice," he said. "But, sorry, we haven't had a lot of anthrax around to know just how it's going to behave."
The Collaboration: 'Layers and Levels' of Teamwork
Teamwork is essential in any epidemiologic investigation. But, Dr. Gerberding of the disease control agency said, "In retrospect, we were certainly not prepared for layers and levels of collaboration" among a vast array of government agencies and professional organizations "that would be required to be efficient and successful" in the anthrax outbreak.
The agency quickly deployed hundreds of workers and created an operations center, installing banks of telephones so epidemiologists could relay information from colleagues in the field to top officials at its headquarters in Atlanta and then on to the National Security Agency, the Central Intelligence Agency and the White House. Even so, calls from state and local health officials and departments and doctors across the country flooded the lines.
And responsibilities for health matters were fragmented. "We were very much aware that we had no jurisdiction over federal facilities whether it was the V.A. or the post office," said Dr. Matthew L. Cartter, an official of the Connecticut Health Department. He urged local, state and federal agencies to work out a memorandum of understanding to clarify lines of jurisdiction before another outbreak.
In New York City, the Health Department had prepared itself for inhalation anthrax in recent years, building liaisons between hospitals and specialists in infectious disease, pulmonary disease and emergency room care. But health officials overlooked dermatologists and surgeons, who treated the first anthrax cases — the skin form. "Very few dermatologists had ever heard of us or knew how to reach us," said Dr. Marcelle Layton, an assistant commissioner in the department's communicable disease bureau.
Anthrax also challenged health officials and law enforcement agencies to work together. And each group had something to learn. For F.B.I. agents, it was how to obtain evidence without contaminating the scene of a medical investigation. For epidemiologists, it was how to collect specimens without disturbing the chain of custody in a criminal case.
"It's a different mind-set, of using epidemiology to reconstruct the circumstances of the exposure that resulted in disease," said Dr. Bradley A. Perkins, a top anthrax investigator at the federal disease centers. Law enforcement officials "immediately recognized the value of that in prosecuting the criminal case," he said.
The Testing: Samples Deluge the Laboratories
On Oct. 15, a letter stuffed with anthrax spores was opened at the office of Senator Tom Daschle. The next day, 2,500 people who had potentially been exposed lined up for nasal swabs. Many were terrified. "People thought each spore was plutonium," said Dr. Eisold, the Capitol physician.
But although the nasal swabbing continued wherever anthrax spores showed up, epidemiologists soon discovered that it was of little use in detecting illness. Its main role was in helping determine where and how far spores had spread.
Soon officials in every state were hit with an avalanche of samples to test — from nasal swabs, from suspicious letters, from swabs of offices and rooms, from clothing, from soil. "You could never have prepared for the volume that you had to process," said Dr. Lou Turner, the director of the North Carolina Laboratory of Public Health.
The disease control agency regards environmental microbiology as one of its strengths, Dr. Gerberding said, but it soon learned that the discipline had a long way to go when it came to anthrax — in particular, sampling the air for spores, disinfecting an area and monitoring it for spores and particles that might escape when envelopes are put under mechanical pressure.
The agency believed that it was prepared for a real anthrax outbreak. It had created a network of laboratories to aid in rapidly detecting microbes. Although the network worked well, the assumption had been that the labs would mostly test specimens from sick patients. Instead, most tests were for spores in the environment — and for hoaxes. The agency had to expand lab space and open a new lab at its headquarters just to test more than 5,500 specimens for spores.
Some health officials complained about the data coming back from testing labs. Does a negative report mean that the laboratory used only a quick screening test, or that it also performed a culture? Such details are important, particularly for laboratory reports that will be evaluated by law enforcement officials and others who would not understand what tests were done, health officials said.
A new problem has emerged: how to return the variety of items — rugs, envelopes, china, even a 50-gallon drum — that were tested and found not to be contaminated.
"We have to figure out how to get rid of all this, which is still evidence and still in the chain of custody," said Dr. Elizabeth Franko, the director of the Georgia Public Health Laboratory. "Either law enforcement needs to come get it, or they need to sign off and say it is trash and they do not want it back."
The anthrax attack was much less horrific than it might have been. But medical and terrorism experts say that situation is due in large part to luck. Considering the size of the postal system, relatively few people were infected. And unlike smallpox, among other possible terrorist weapons, anthrax is not spread from person to person.
In deconstructing the response, Dr. Hunker, the terrorism expert, said, it will be important to investigate what role luck played, to avoid having to rely on it in the future.
The backbone of the Centers for Disease Control and Prevention's response to health emergencies is a corps of epidemiologists known as the Epidemic Intelligence Service. Forensic epidemiology has not been part of their training. But now it has to be, Dr. Gerberding said.
And experts caution that the anthrax outbreak may not be over.
"We still do not know who put anthrax in the mail, we still do not know if they used all they had, and we still do not know how to make all the mail safe," said Dr. John O. Agwunobi, the Florida secretary of health. "So the question becomes how quickly can we apply what we have learned so far to the next event."