Jump to ContentJump to Main Navigation
Bracing for Armageddon?The Science and Politics of Bioterrorism in America$

William R Clark

Print publication date: 2008

Print ISBN-13: 9780195336214

Published to Oxford Scholarship Online: September 2008

DOI: 10.1093/acprof:oso/9780195336214.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2019. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: null; date: 22 October 2019

The Ultimate Bioterrorist

The Ultimate Bioterrorist

Mother Nature!

(p.71) Chapter 5 The Ultimate Bioterrorist
Bracing for Armageddon?

William R. Clark

Oxford University Press

Abstract and Keywords

The greatest threat we face from biological agents of death today is not from humans – it is from nature itself. A bioterrorist attack that could kill hundreds, possibly thousands, of Americans is a possibility, but one with low probability. A pandemic caused by a naturally occurring biological pathogen that could kill tens of thousands, possibly millions, of Americans is an absolute certainty. Natural pandemics are a regularly occurring phenomenon throughout history. The tens of billions spent on bioterrorism defense have modestly improved our ability to resist a natural pandemic, but only modestly. This chapter examines the history of natural pandemics in America, and decribes what must be done if we are to avoid the tragedy of the 1918 flu pandemic.

Keywords:   1918 flu pandemic, SARS, avian flu, Asian flu, flu vaccines

THE AGENTS OF BIOTERRORISM WE DISCUSSED IN CHAPTER 3 have a common feature that distinguishes them from other agents of terror, such as explosives, chemicals, and nuclear devices: they are themselves not the product of human invention. They arose in nature. Most of them, at one time or another, have given rise to events—often repeatedly, throughout history—that closely resemble what we imagine a bioterrorist attack would be like. These events occurred without human causation, as a result of the constant attempts of various microbes to invade the human body and use it as a place to rear their young.

In fact, the greatest threat we face from biological agents of death today is not from humans—it is from nature itself. A bioterrorist attack that could kill hundreds, possibly thousands, of Americans is a possibility, but, as we will discuss in chapter 9, one with low probability. A pandemic caused by a naturally occurring biological pathogen that could kill tens of thousands, possibly millions of Americans is an absolute certainty. Natural pandemics1 are a regularly occurring phenomenon throughout history.

Historically, the most problematic pathogens for humans have been Yersinia pestis (plague) and the smallpox and influenza viruses. Our understanding of the natural ecology of Y. pestis in human and animal hosts makes it pretty unlikely we will ever (p.72)

                   The Ultimate BioterroristMother Nature!

Figure 5.1 Deaths from seasonal flu in the United States by year.

again face a natural plague pandemic. The smallpox virus no longer exists in nature. But we remain, after decades of study, terribly vulnerable to killer flu pandemics. Even seasonal flu still kills on average between 35,000 and 40,000 people in the United States each year (Figure 5.12). Three flu pandemics in the twentieth century—one major and two minor—claimed well over half a million lives in the United States alone, and almost certainly over 100,000,000 worldwide.

The uncertainty about exactly when a lethal flu pandemic could strike and the resulting social and economic disruption, in addition to the large numbers of deaths and illnesses even a moderate pandemic would cause, are in every way comparable to the uncertainty and consequences of a successful bioterrorist attack. Yet a reasonable wager would be that nine out of ten Americans fear the latter much more than the former. Why that might be will be a thread running through much of the rest of this book.

Our societal responses to a bioterrorist attack with the pathogens on the CDC lists are for all practical purposes the same as our responses would be should any of these agents, or any other pathogenic microbes, known or unknown to us now, make their way into the human population and create a natural epidemic or pandemic. The major difference between how such an event (p.73) would affect us today and the situation even a hundred years ago is the presence in many countries of strong public health systems. These same public health systems will also provide our major defense against the results of a bioterrorist attack.

To give some impression of what a natural pandemic might look like, let's take a look at several situations of relatively recent history: the three influenza pandemics of the twentieth century, the SARS pandemic of 2003, and the still uncertain health crisis that could be caused by the H5N1 variant of the avian flu virus.


We don't know exactly where the strain of influenza virus causing this pandemic arose. Most likely it was somewhere in Asia, probably China. It may have emerged as early as January of that year, or even earlier. Influenza outbreaks came in several waves during 1918. The first washed ashore in February in Spain, thus giving rise to the popular term “Spanish flu.” This flu proved to be highly contagious, but the resulting disease was relatively mild. The one that would kill so many people didn't arrive until September.

Although the 1918 pandemic coincided with the latter months of World War I, emergence of the virus was not caused by the war itself in the sense that it was developed or used as a weapon. Certainly crowded, dirty trenches and a general lack of hygiene everywhere—especially military camps—hastened the spread of natural outbreaks in areas where the war still raged. Half of American fighting men in this war died of the flu—not from artillery shells, bullets, or poison gas. Mysteriously, the pandemic ended very shortly after the war.

When it was over, barely a year after it began, close to a third of the world's population may have been infected. We don't really know how many people died worldwide. Estimates have ranged from twenty-five million to fifty or even a hundred million—possibly as much as several percent or more of the world's population at that time. Precise medical record keeping was still less than perfect in the United States and Europe, and practically nonexistent in many parts of the world. An estimated twenty (p.74) million people in the United States were infected, and between six and seven hundred thousand people perished—slightly more than have died after twenty-five years of AIDS, and more than have died in all U.S. wars through Vietnam.

All evidence suggests that two or three pandemic-level flu events per century is the historical norm. The death rate in even the more serious seasonal flu outbreaks—the kind we see almost every year—is around a tenth of one percent of those infected; the death rate during the 1918 pandemic was twenty-five times that. Virulence at this level was not recorded before 1918 (although it may well have occurred) and has not been seen since. One of the striking features of this influenza outbreak is that so many of its victims were between twenty and forty years of age, in the prime of life (Figure 5.2); the flu is normally most deadly for the

                   The Ultimate BioterroristMother Nature!

Figure 5.2 Proportional distributions by age. (A) Rate of infections in the 1918 pandemic; the disease struck across all age groups, although the young were particularly susceptible. (B) Rate of deaths in the 1918 pandemic; surprisingly, the death rate in individuals 1–40 years of age was extremely high, by comparison with (C), the rate of deaths seen in a standard outbreak of influenza. (Based on data presented in Jeffery K. Taubenberger and David M. Morens, “1918 Influenza: The Mother of All Pandemics,” Emerging Infectious Diseases 12(2006):15–22.)

(p.75) very young and the very old. The 1918 flu was also active during the spring and summer, whereas flu is usually a problem only in winter months.

The disease symptoms were like those of the flu generally—fever, achiness in joints and muscles, dizziness and weakness—but they were unusually harsh. Most people in fact recovered, but for too many others, death could come within just a few days of the onset of symptoms. Hemorrhaging in the lungs was common, causing victims to spit up quantities of blood-laced froth. As breathing became more difficult, many patients turned blue from lack of oxygen. Pneumonia could set in after a few days, was essentially untreatable, and was the most common cause of death.

All influenza viruses that have been involved in human epidemics or pandemics are of the influenza A viral group (Box 5.1). Influenza B and C viruses cause relatively mild cold- or flu-like symptoms in humans. Influenza A viruses are thought to have originated in aquatic fowl, and find a natural reservoir in many (p.76) types of birds, where they live in the gut, causing little or no damage.

Like HIV, flu viruses use RNA rather than DNA to build their genetic blueprint. Unlike DNA viruses, RNA viruses do not “proofread” their blueprint when it is copied from one generation to the next, resulting in a very high mutation rate—a characteristic of both HIV and flu viruses. That is what has made it difficult to develop effective vaccines for both of these pathogens.

Current strains of avian flu viruses have come into balance with their bird hosts. They do not kill them, and can enjoy a long and productive lifetime within their hosts. Thus, most mutations of their genetic blueprint are harmful, disrupting the balance between virus and host, and are selected against. Nonetheless, occasional mutations do arise in avian flu viruses that result in highly pathogenic forms that result in destruction of large numbers of birds before the viral mutants eventually disappear.

Influenza A also infects pigs as well as humans. Humans are not ordinarily infected directly by A-type viruses that pass back and forth among birds, but occasional mutants arising in birds or in pigs can produce a variant that is able to infect human cells. In both cases, the virus settles into the lungs, where, being completely foreign, it is vigorously attacked by the immune system. The rapid mutation rate of RNA viruses is now an advantage, as it changes rapidly in a thousand different directions, trying to escape immune destruction and live at peace with its new host.

Within the A group, viruses can be further classified based on two types of molecules they all display on their surface: the H (hemagglutinin) molecule, and the N (neuraminidase) molecule. These are viral proteins involved in the initial infection of target cells (H) and the escape of newly formed viruses from infected cells (N). The H proteins interact with receptors on host cells and promote entry of the viruses into them. That is one reason true avian viruses rarely cross into humans: they are selected to recognize bird cells, not human cells.

Influenza A viruses have sixteen possible different H molecules they can display, and nine different N molecules. Each strain of influenza A is named based on the combination of these two molecules that it uses: H1N1, H5N1, H7N7, etc. H1, H2, and H3 are most commonly found in viruses that infect humans, usually in (p.77) combination with N1 and N2. Analysis of the recovered samples of the 1918 virus discussed in the last chapter showed it was of the H1N1 strain, and it is thought to have migrated directly from birds to humans.

But at the time of the 1918 pandemic, scientists hadn't even identified viruses as pathogenic agents distinct from bacteria—that would not happen until the 1930s.3 The lungs of many who succumbed to the 1918 flu were full of bacteria, so it was assumed at the time this was yet another bacterial disease. We now know that heavy viral infections, particularly of the airways, are often accompanied by secondary bacterial infections, including the pneumococcal infections frequently associated with the pneumonia seen in late-stage flu. At any rate, since neither drugs that could slow the internal spread of the flu virus nor antibiotics to treat bacteria were available in 1918, it would have mattered little to have correctly diagnosed the causative agent. And of course there was no influenza vaccine available, or ventilators to assist with breathing.

The lack of any effective treatment also meant that public health measures were limited more or less to those that were present already in medieval times. Public health as a recognized medical discipline was still in its infancy in 1918, at least in the mind of the public, and compliance with official public health recommendations was irregular at best. It was understood that the disease was spread through aerosols generated by sneezing and coughing, and affected individuals were isolated at home or in segregated areas of hospitals. Entire groups of people and physical locations were often quarantined—military bases, prisons, and asylums, for example.

Hospitals overflowed with victims, and the scarcity of doctors and nurses caused by their recruitment into the war was compounded by the unwillingness of many who might have helped minister to the sick to even enter the hospitals. Medical and nursing students with little experience were often thrust into the fore of battles against the pandemic. Patients were confined to their beds, usually surrounded by bedsheets strung up to limit the spread of contagious aerosols. Both health care workers and patients wore masks. Isolation and quarantining of victims led to hostile interactions between the public and health and public (p.78) safety officers. Disposal of infected bodies became problematic—some communities even resorted to open trench mass burials, which again met with strong public resistance.

In the hardest-hit communities, such as Philadelphia, there was severe social and economic disruption. In that city, 11,000 people died in the first month alone. The lack of any national uniformity of public health standards meant that each state, city, and county basically made up its own rules. Schools and businesses were closed in many cities, as were bars and theaters of all sorts. Public gatherings were banned. Churches generally remained open, but congregants were urged to sit as far apart as possible. In some cities, even large weddings and funerals were forbidden, and people were not permitted in government buildings or on public transportation without a face mask. It was a time of enormous chaos.4

Although the story of the 1918 influenza pandemic lacks the stridency and militaristic drama of exercises like Dark Winter and Atlantic Storm, it certainly stands alongside them in terms of damage done in terms of human lives lost, social collapse, and economic disaster. But one very big difference leaps out in this comparison: the 1918 pandemic actually happened. And tens of millions of people died.


After the 1918 pandemic ended, flu reverted to its usual pattern of seasonal appearances. Most authorities think the pandemic H1N1 strain passed from humans into pigs, where it clearly was not as virulent. Milder versions of H1N1 must have arisen in pigs and passed back to humans, or mutated within humans themselves; at any rate, relatively benign H1N1 became the dominant FluA variant in humans for the next forty years.

Of course no flu variant or flu season is ever trivial; even today, with both vaccines and drugs to control the virus and antibiotics to manage secondary bacterial infections, tens of thousands of people in the United States (probably a million worldwide) still die each year from complications of seasonal flu. But the next flu outbreak after 1918 to rise to the status of a worldwide pandemic (p.79) came in 1957, triggered by a variant of the flu virus originating in Guizhou province in China, where it first infected humans in 1956. It was commonly referred to as the “Asian flu.”

Like most flu pandemics, including the one in 1918, this one was triggered by a viral variant that few living at the time had ever encountered. The prevalent H1N1 strain, probably while in pigs, appears to have interacted with an avian flu virus from ducks to produce an H2N2 variant. (The H and N genes and one other viral gene were from the bird virus; the rest of the genes were from the human virus.) This may be the first time a flu virus variant containing N2 had entered the human population in at least two generations, since there was no immunity to it among humans.

This new form of the virus was able to cause pneumonia in humans entirely on its own, without help from secondary bacterial infections of the lungs, although such infections did occur. Children were especially vulnerable to infection, and schools became major venues for spreading the disease. Many schools closed at least briefly, although relatively few children died. On the other hand, this virus was particularly deadly for persons with lung or heart disease, and took a high toll among the elderly. It was also very dangerous for women in the third trimester of pregnancy.

It has been suggested that this virus may have infected as many or more people as the 1918 variant, but the availability of a vaccine by May of that year,5 as well as antibiotics and improved public health services, contributed greatly to a reduction in overall mortality. Still, the toll it exacted—70,000 deaths in the United States and 2,000,000 to 4,000,000 worldwide—was certainly horrendous.

The H2N2 Asian flu virus largely displaced the H1N1 variant circulating among humans after 1918 and continued to cause flu outbreaks for several years after 1957, with a particularly serious eruption in 1958 involving disproportionately the elderly. But as increasing numbers of people built up immunity to it, through immunization or natural exposure, it gradually became less of a problem. It essentially disappeared after 1968, when it was largely supplanted by the infamous Hong Kong flu variant.

The 1957 version of the H2N2 virus almost made it back into the human population in 2004–05. A U.S. laboratory accidentally (p.80) included it in influenza test kits sent to other laboratories throughout the world. Fortunately, after an alert laboratory in Canada recognized the mistake, the kits were immediately recalled, and no one appears to have been infected.

Eleven years after the Asian flu, a third pandemic emerged from the Chinese mainland, becoming known as the Hong Kong flu.6 This time the culprit was an H3N2 variant of the influenza A virus. (The H3 and one other gene were of bird origin; the rest of the genes were human.) It first came ashore on the West Coast in the United States, probably with troops returning from Vietnam. From there it spread eastward, although not all states were ultimately affected except for sporadic cases. Although about 20 percent of people in affected countries became infected (an estimated 50,000,000 in the United States), this was a relatively mild pandemic, with only about 34,000 excess deaths7 in the United States and 700,000 to 1,000,000 worldwide. But once again, the elderly were hardest hit. Drugs that we have now to lessen the impact of flu—amantadine, rimantadine, ostelamivir (Tamiflu), and zanamivir (Relenza)—were not yet available. Less deadly forms of the H3N2 Hong Kong flu virus variant evolved over the next few years and are still the dominant form causing seasonal flu in humans to this day.

There are a number of reasons for the relative mildness of the Asian and Hong Kong flu pandemics. Neither variant in these latter outbreaks appear to have induced the same kind of violent response by the immune system, with collateral damage to normal tissues, thought to have been responsible for much of the damage seen with the 1918 H1N1 virus. Moreover, there had been continued improvements in flu vaccines used to control spread of flu viruses and in the antibiotics used to treat bacterial complications. The H3N2 variant of 1968 was immunologically cross-reactive with the H2N2 variant from 1957, and so people exposed to H2N2 over the intervening years may have had some degree of immunity to the Hong Kong virus. Also, the Hong Kong virus struck many American cities right at the December-January school break, reducing transmission among students. Some schools and colleges closed slightly early or delayed re- entry until the flu had subsided to reduce spreading to the larger population.

(p.81) So the twentieth century saw early on one of the deadliest flu pandemics in recorded history, and later gave rise to two more pandemics of decreasing intensity. Does this mean that flu pandemics are on their way out as a threat to humans? Would that it were so! We are now staring down the barrel at a new flu variant coming to us from birds that has the potential to equal or exceed the devastation wrought by the 1918 virus. We first saw this virus—this time an H5N1 variant—in 1997. But between that time and the present, we went through another pandemic with a completely different virus—SARS. Before we look more closely at H5N1 flu, let's take a moment to explore what happened in the SARS pandemic of 2003–04.


The first cases of what would come to be known as SARS were detected in China at the end of 2002. Several people in Guangdong province showed up at clinics and hospitals with severe flu-like symptoms. At the time, there was nothing to mark this as the emergence of a new and deadly disease that would eventually affect over 8,000 people in thirty countries, killing 774 of them. These numbers, from official World Health Organization records, may not tell the whole story. It is likely that more people in China were involved, and may have died, but either went unnoticed and unrecorded or were incorrectly diagnosed. There is evidence that SARS was present in China prior to 2002. For example, a subsequent analysis of 938 blood samples collected in Hong Kong in 2001 for unrelated purposes showed evidence that 17 of these individuals had been exposed to SARS.

The first few patients subsequently identified as having SARS apparently recovered, and may not have passed the disease on to others. But within a month or two many other such cases showed up in China and began to worry authorities. It takes time to figure out, in a situation like this, that what one is seeing in the clinic is not just another outbreak of a slightly nastier flu, but a new, different, and more deadly disease. China's public health system, while adequate, is not yet quite on a par with most Western (p.82) countries. That fact, coupled with the penchant of Chinese authorities for trying to keep disturbing news out of the press,8 slowed the development of awareness in the rest of the world that a major new health threat was arising. China did not alert the WHO until February 2003, by which time several dozen cases had been detected. China subsequently issued a public apology for its slowness in dealing with the SARS crisis.

Hong Kong was particularly hard hit early in the developing pandemic, as was Singapore. SARS had minimal impact in the United States. Only eight people were diagnosed with SARS; all of these had recently been in countries with verified SARS outbreaks. None died. However an American businessman traveling in Asia in early 2003 died of SARS in a Hanoi hospital. There was, moreover, a more serious importation of SARS into Canada, particularly Toronto. Of 251 cases officially diagnosed, 44 died. Nearly half of these were hospital or other medical personnel, and some hospitals in that city had to be quarantined. A high percentage of the 300 deaths in Hong Kong were also among health care workers. The public health systems in both cities were all but brought to their knees.9 Both cities also experienced serious economic disruptions and restrictions on travel.

These events, together with a better picture of what was happening in China, contributed to the sounding of a global health alert by the WHO. Public health agencies in countries throughout the world then issued alerts of their own, and began the serious work of preventing further spread of this new disease. The SARS pandemic peaked in May 2003, rapidly subsiding as the resulting containment measures took effect. The epidemic was essentially over by mid-2004 after a brief recurrence in China.

SARS is spread either through the air by sneezing and coughing aerosols or by direct contact with bodily fluids. As with many other contagious respiratory diseases, symptoms usually appear within two to three days of exposure to the triggering pathogen. The causative agent in the case of SARS is a newly emerged variant of the coronavirus, one of many viruses that can induce human colds. This never before seen variant, now called SARS-CoV, was first identified in March 2003 as a possible cause of SARS. An extraordinarily intense WHO-financed investigative campaign in laboratories around the world quickly confirmed (p.83) this, and on April 16, a month before the pandemic peaked, the WHO announced SARS-CoV as the official causative agent in SARS.

SARS-CoV was found a short time later in bats and civets in Guangdong province. It seems likely that SARS-CoV jumped to humans either through bat bites or through eating the meat of civets or other small mammals. Meat from civets, a carnivorous cat-like animal common in Asia, can be found in numerous meat shops throughout Guangdong. Whether bats or civets are a natural reservoir for SARS-CoV is unclear.

The symptoms of SARS are similar to the flu: fever, generalized achiness, lethargy, abdominal discomfort. There is usually a dry cough early on, and there may be shortness of breath, both of which would be unusual for the flu. Fever usually peaks in most patients four days after onset of symptoms, lung abnormalities are revealed by X-ray at day six, and oxygen sufficiency may be critically low at day 8. This latter may be particularly crucial in people over sixty, where fatality rates often approach 50 percent. In children and young adults, fatalities rarely exceeded 10 percent of those infected.

There was no vaccine available for this previously unknown viral variant, and flu drugs were ineffective against it. So the only interventions available at the time were traditional public health measures: making people aware of the symptoms and encouraging early self-reporting; identifying and isolating infected individuals and their first-degree contacts; urging the public to increase personal hygiene, and to wear face masks where appropriate; encouraging “social distancing” (avoiding mass gatherings) and closing schools where necessary; increasing surveillance at ports of entry for symptomatic individuals. Travel advisories warning people away from infected areas such as Toronto and Hong Kong were also issued by most governments.

Some of these measures seem to have been effective in places like Hong Kong, Singapore, and China proper. Government authorities in these cities tend to be a bit more heavy-handed in enforcing public health edicts, and this may have been effective in limiting the contagion emerging from these areas. It appears to have been less effective in Toronto. Only a few of the milder recommendations were issued by U.S. officials. An analysis of the (p.84) effectiveness of these various measures in containing the SARS pandemic has been published.10

SARS is a much slower moving infection within a population than the flu, offering more opportunities for intervention. But there is in fact no rapid, clear test that could be used in a doctor's office or small clinic to determine whether someone with what appears to be a bad case of the flu may actually have SARS. As the 2002–03 pandemic progressed, the major factor in determining whether people presenting themselves at a clinic or doctor's office with serious flu-like symptoms might have SARS was whether they had recently been in contact with someone who did have SARS or had traveled in a location where SARS was prevalent.

Laboratory testing could now conclusively identify SARS-CoV as the causative agent, but anyone even suspected of having SARS would likely be immediately placed in isolation, closely monitored, and given intense supportive therapy as needed until testing was completed. SARS cases can proceed rapidly into severe breathing difficulties and an insufficiency of oxygen, leading to respiratory collapse (acute respiratory distress) and a requirement for ventilator-assisted breathing.

As with most diseases caused by viruses, there is at present no specific treatment to control SARS-CoV. Indomethacin and interferon have been reported to be effective in the laboratory, but there is as yet no supporting evidence for the efficacy of these drugs in the clinic. There is also no vaccine presently available that protects against SARS-CoV, although several promising vaccines are under development, and one is currently being tested in human clinical trials being supervised by the U.S. National Institutes of Health. Researchers throughout the world are working steadily to develop a defensive armamentarium should the SARS virus ever reappear.


As noted earlier, influenza A viruses live most of the time in the gut of birds without sickening or killing them. But occasionally, random mutations arise in these viruses that can trigger a severe (p.85) form of avian flu, killing large numbers of birds. In the wild, these mutations can wreak havoc in isolated flocks, but the resulting explosions are usually self-extinguishing given the large spacing between flocks and their constant movement from place to place. Once the infected birds or even whole flocks die out, the lethal variant of the virus can disappear.

But modern methods of rearing poultry for commercial purposes result in huge concentrations of birds in very constricted spaces. Spread of mutant viruses can be extremely rapid in such populations, and the only solution to containing an outbreak is the immediate, compulsory destruction of every bird in the infected compound, whether symptomatic or not. Even then, shipment of infected live birds to other locations prior to confirmation of an outbreak (or even afterwards, in the interest of limiting financial loss), as well as movement of contaminated equipment, truckers and other workers between farms, can result in entire regions having to destroy enormous numbers of birds.

Such was the case in Hong Kong in 1997, with the emergence of a deadly avian flu variant of the type H5N1. This variant did not sit benignly in the gut of birds it infected, but penetrated into every organ and tissue of the body. This is thought to be due to changes in the H molecule, which determines which cell types the virus can invade. The result was rapid physiological collapse and death. In former times such outbreaks were referred to as “fowl plague.” Only in the 1950s were they recognized as a form of avian flu. Since that time there have been a dozen or so outbreaks, usually involving viruses bearing the H5 or H7 forms of hemagglutinin.

The 1997 Hong Kong outbreak, which involved several poultry farms, was finally quashed after the destruction of tens of thousands of birds, and it seemed that H5N1 would likely fade into the sorry history of fowl plague. But in May of that year, a three-year-old boy in Hong Kong was admitted to a hospital with a respiratory infection that quickly progressed into pneumonia. But as with birds, the H5N1 that had infected him spread far beyond the lungs. He went on to develop Reye's syndrome, acute respiratory distress, and kidney and liver failure. He died a few days later. The medical staff at the hospital, stunned by the violence of his disease, were determined to find out what had (p.86) caused it. Throat washings taken from the boy had been saved, and were analyzed for a wide range of viruses and bacteria. None of the tests detected anything. Samples were sent to WHO labs in London and Rotterdam and to the CDC in Atlanta. The Rotterdam lab was the first to come back with an answer: the throat washings were positive for the H5N1 variant of the avian influenza virus.

Since H5N1 had never been seen in humans before, extensive tests were carried out to see if a mistake had been made or the throat washings had been secondarily contaminated. Several labs around the world joined this effort, and it was soon absolutely clear that the young boy had indeed succumbed to a primary infection with the H5N1 virus. Further analysis of the boy's virus showed that it was virtually identical to the H5N1 flu virus involved in the recent local poultry outbreak, and that it had passed to him without modification in an intermediate host such as pigs. The clinical description of his illness and death were hauntingly familiar to those who had studied the 1918 flu pandemic.

Public health authorities in Hong Kong immediately tested other members of the boy's family, as well as medical staff that had attended him in the hospital. None of his family members showed any signs of having been in contact with the virus,11 but one of his nurses and some of his playmates did. Wider testing picked up a number of poultry workers who also showed signs of having harbored the virus. None of these had showed any signs of a flu infection. But when samples of the virus isolated from the boy's throat washings were tested on an experimental poultry flock in Georgia, the entire flock underwent an immediate and violent death from catastrophic influenza. From that point on, it was agreed that all work with the Hong Kong H5N1 virus must be carried out in high-security biocontainment laboratories, such as those designed to work with soil samples brought back from the moon or with CDC A-list pathogens.

In the months after the boy's death, no further cases emerged, and public health authorities began to hope his death might have been a fluke. It was unclear how or even whether he had passed the virus directly to others around him. It could not be ruled out that those who had contact with him and showed positive for (p.87) H5N1 had not independently picked it up from poultry. And he was the only one to have become ill.

But the message to infectious disease specialists was clear. Here was a virus that could—did—kill a human being, and it had passed directly from birds into humans. What would happen if this new H5N1 virus infected a human during the normal flu season, and that person was also carrying one of the relatively benign seasonal flu viruses such as the current H3N2 variant? The viruses could well recombine, producing a hybrid variant with H5N1's lethality and the ready human-to-human transmissibility of a seasonal flu virus.

A few months later, any hope that the death of the three-year-old child had been a one-shot occurrence disappeared forever. Over a period of several weeks, a total of seventeen more children and adults were admitted to hospitals with signs of a violent pneumonia and were found to be infected with H5N1. Five died. As in the 1918 pandemic, it appeared that most of those who died were previously healthy adults. That was the only sense in which the initial three-year-old victim may have been a fluke. Most of those who became seriously ill in this round had had contact with poultry; there was still no reason to suspect human-human transmission of H5N1.

And then, almost immediately afterward, poultry began dying again on Hong Kong farms, and even in live-chicken markets in the middle of the city. Public health authorities moved swiftly. As soon as H5N1 was confirmed as the cause and it became apparent that up to a quarter of the territory's poultry were infected, they decreed that all of Hong Kong's poultry must be destroyed—over a million and a half birds. The economic cost would be staggering, but there was no hesitation. The government ordered it done immediately.

Since its initial detection in Hong Kong in 1997, the H5N1 variant has spread to birds in over fifty countries (Figure 5.3). As of April 2007, the WHO had confirmed 291 cases of transmission to humans, with 172 deaths. A partial history of the spread of H5N1 and its interaction with humans is shown in Table 5.1. The mortality rate in humans known to be H5N1-infected has hovered at about 60 percent for the past decade, making it at least twenty-five times more deadly than the 1918 H1N1 flu virus. (p.88)

                   The Ultimate BioterroristMother Nature!

Figure 5.3 Cumulative spread and human deaths from H5N1 avian flu.

Victims so far have been predominantly women and children, but this could simply reflect their representation among individuals handling poultry.

The first (and so far only) confirmed case of human-to-human transmission occurred in Indonesia in 2006. A child contracted H5N1 from handling poultry and subsequently passed the virus to six other family members. No further spread outside the family was documented.12 It may be that this family had a genetic alteration that made them more susceptible to human-human transmission. But it is this possibility—a genetic alteration in the virus itself, particularly in the H protein which determines initial entrance of the virus into cells—that gives world public health officials nightmares. It is what happened in 1918, and it could happen again.13

How would we handle an outbreak of transmissible H5N1 influenza in the human population? The ideal solution would be a vaccine that is at least as effective as the current vaccines for seasonal flu. We don't have such a vaccine yet, but one that may offer partial protection has recently received FDA approval. The vaccine used an inactivated H5N1 virus isolated from humans infected through contact with birds. In healthy individuals, antibody levels judged sufficient for protection were induced in about (p.89)

Table 5.1 A Partial Summary of H5N1 Avian Flu Outbreaks Affecting Humans





Hong Kong

Epizootic in poultry. Eighteen humans infected, probably from poultry. Six died.


China, Hong Kong

Two Hong Kong family members returning from China developed H5N1 pneumonia. One died. Method of infection unknown.


Thailand, Vietnam

Epizootic in poultry; first case of possible human-human transmission reported.


China, Cambodia, Thailand, Vietnam, Indonesia

Epizootics in poultry; first instance of possible infection of humans from ducks.


Azerbaijan, China, Cambodia, Egypt, Djibouti, Thailand, Iraq, Turkey

Epizootics in poultry; swans may have been source in some cases.



First confirmed instance of human-human transmission. A child who probably picked up the virus from poultry passes it on to six other family members.



Two separate poultry farms infected.



Togo becomes seventh African country with H5N1 outbreak.


Great Britain

Over 100,00 H5N1-infected birds culled

For a complete timeline of avian influenza, see the WHO website http://www.who.int/csr/disease/avian_influenza/timeline_2007_04_20.pdf.

As of April, 2007, 291 conK rmed cases, 172 deaths (see http://www.who.int/csr/disease/avian_influenza/ country/cases_table_2007_04_11/en/index.html for updates).

(p.90) half of those immunized. Current seasonal flu vaccines protect 90 percent or more of recipients.

And to achieve even this level of protection required two injections of a rather large amount of the vaccine, spaced one month apart. So this vaccine is far from ideal, but it is all we have at the moment, and it is being stockpiled as a stopgap measure by the federal government while research into a more effective vaccine proceeds at a rapid pace. Enough of the recently approved vaccine for twenty million people is currently available. University and drug company researchers are working furiously, supported by federal funds, to develop a more effective vaccine. The government has already drawn up a prioritization list for who will receive H5N1 vaccines during an epidemic crisis.14

There are four drugs currently approved by the FDA for mitigating the impact of influenza A infections (Box 5.2). Amantadine and rimantadine stop the virus from replicating after it enters the cell; Relenza and Tamiflu block the ability of newly made viruses to exit from the infected cell. There is already evidence that amantidine and rimantidine are ineffective against the H5N1 virus as it presently exists. Relenza and Tamiflu are effective against current H5N1 strains, and are being stockpiled by both the federal government and individual states.

But the problem for both vaccines and antiviral drugs is that the H5N1 variant that presently exists, even in those cases where it has migrated into humans, is not the one we have to worry (p.91) about. As a minimum, some sort of mutation will have to occur in one or more genes of existing forms of H5N1 to allow it to spread easily among humans. A second mutation, allowing it to jump more readily from birds to humans, would really put us in deep trouble. It is possible, some think even likely, that these mutations could alter the ability of current H5N1 vaccines to block the virus. The effect of these mutations on the sensitivity of the new mutant(s) to antiviral drugs will also have to be assessed.

So we won't really know what we are dealing with, what it is we will have to protect ourselves against, until such mutations actually occur. The new mutants will have to be isolated and studied in the laboratory, and plans devised for the most effective vaccines and drugs. Laboratories and manufacturers around the world are poised to do this, and to do it on a 24/7 basis. But it will still take time. It is entirely possible that when such a mutant arises, we will be on our own for as much as six months before drugs or vaccines are available.

How bad could an H5N1 pandemic be? We won't know that either until we have the miscreant mutant in hand. It could be more lethal than current H5N1 strains, or less lethal. If it maintained its present lethality, a third of the people in the world became infected, and half of those died—the math is pretty straightforward. Remember, H5N1 as it exists now, when it infects humans, is about twenty times more lethal than the 1918 virus. Dr. Anthony Fauci, Director of the National Institute of Allergy and Infectious Diseases, considers the threat of an H5N1 flu pandemic greater than that of bioterrorism.15

We have the experience of those who managed the 1918 outbreak with nonpharmaceutical interventions to guide us.16 And it is possible that with our improved ability to manage influenza-like diseases—we now have mechanical ventilators to assist with breathing17; we know better how to manage secondary respiratory bacterial infections and pneumonia; we will (hopefully) have anti-flu drugs—we could reduce the lethality of infections considerably. Maybe we could cut it in half. But that could still mean a billion people dead worldwide, fifty million in the United States. That would pretty much bring the world as we know it to a standstill.

There is reason to be concerned. (p.92)


(1.) Epidemic is usually defined as a sudden, rampant spread of an infectious disease among humans within a single country, or possibly adjoining countries. When the disease involves a large number of countries or more than one continent, we use the term pandemic. The corresponding terms for movements of infectious diseases among animal populations are epizootic and panzootic.

(3.) In the first decades of the twentieth century, scientists did realize that some infectious agent existed that was much smaller in size than a bacterium, capable of passing through an extremely fine filter that trapped bacteria and incapable of being seen in a microscope, but they had no idea what it was. They referred to these agents as viruses. The influenza virus itself was discovered in 1933.

(4.) For an excellent and detailed account of the havoc wrought in the United States and elsewhere by the 1918 flu pandemic, see Gina Kolata, The Story of the Great Influenza Pandemic of 1918 (New York: Simon & Schuster, 1999).

(5.) An influenza vaccine was developed shortly after World War II, and by the mid-1950s most doctors and public health officials were familiar with its use.

(6.) Although the earliest cases were reported from the relatively open Hong Kong, officials there insisted this flu originated in the more secretive People's Republic of China, probably in the adjacent province of Guangdong.

(7.) Over those that would be expected from normal seasonal flu.

(8.) In fact, for quite some time information about SARS in China was issued by the propaganda arm of the Chinese Communist Party rather than the government's Health Ministry.

(9.) For a detailed analysis of the pandemic as it played out in Hong Kong and Toronto, see C. David Naylor, Cyril Chantler, and Sian Griffiths, “Learning From SARS in Hong Kong and Toronto,” Journal of the American Medical Association 291:2483–87.

(10.) David M. Bell, “Public Health Interventions and SARS Spread,” Emerging Infectious Diseases 10(2004):1900–06.

(11.) Anyone who has recently been infected by an influenza virus will have antibodies in their blood specific for that particular virus, and these can be readily detected in a simple laboratory test.

(12.) In addition, it has now been shown that a pregnant H5N1-infected woman had transmitted the virus to her fetus. See J. Gu et al., (p.195) “H5N1 Infection of the Respiratory Tract and Beyond: A Molecular Pathology Study.” Lancet (2007) 370:1106.

(13.) For a Dark Winter–like description of what an H5N1-based pandemic could look like in a major metropolitan area like New York, see Irwin Redlener, Americans At Risk (New York: Knopf, 2006), 29–36.

(15.) H. Markel et al., “Nonpharmaceutical Interventions Implemented by U.S. Cities During the 1918–1919 Influenza Pandemic,” Journal of the American Medical Association (2007) 298:644.

(16.) Quoted in Scott Shane, “U.S. Germ-Research Policy Is Protested By 758 Scientists,” New York Times, March 1, 2005.

(17.) Although not nearly enough to deal with a flu pandemic. There are roughly 100,000 ventilators in the United States; during a typical seasonal flu outbreak, the vast majority of these are in use. A major pandemic could easily require double the present number.