Quotes of the Day

Desolate but beautiful, tucked in a northwest corner of China so distant from populated areas that it lies near the spot where Beijing developed its first atomic bomb, Bird Island could be the last place on Earth. But for migratory birds, the island—actually a small peninsula protruding into Qinghai Lake, China's largest saltwater lake—is the avian equivalent of a busy international airport. Each year in the late spring and summer, hundreds of thousands of wild birds of almost 200 different species land here to lay eggs and hatch their young, before departing for wintering grounds that range from Europe to South Asia. On May 4, Li Yinghua, a ranger at the Qinghai Nature Reserve's Bird Island Station, was making his daily rounds near an area popular with bar-headed geese when he spotted something he'd never seen in his two decades at the reserve. "One of the geese had wandered away from the group," he recalls. "It was walking so strangely, wobbling from side to side as if it were drunk." His voice drops to a whisper. "This goose seemed to be shivering." Li called his superiors who took the animal away for tests. He heard that it died soon after.

Over the next six weeks, thousands of other birds, chiefly bar-headed geese, would perish in the same way. In Hong Kong, animal virologist Guan Yi heard about Qinghai and immediately guessed that he knew the cause: H5N1, a deadly avian-flu virus that has ravaged Southeast Asia's poultry flocks and infected over 100 people since the end of 2003. Although the Chinese government initially rebuffed requests by the Geneva-based World Health Organization (WHO) to visit the site of the Qinghai outbreak—they wouldn't be given clearance until the end of June—Guan circumvented the red tape, using his network of mainland contacts to gather nearly 100 samples from dead birds. The results, which he and his team published on July 7 in the journal Nature, confirmed his suspicion: wild birds, generally thought to be immune to the effects of H5N1, were succumbing to the virus. That was important news. If H5N1 became established among migratory species, it would raise the risk that the virus could be spread across oceans and continents. Understanding just how the viral transmission belt operates "is urgent work for the entire world," Guan told Time recently, speaking in his office at the University of Hong Kong (HKU). "The virus could go to Europe, and from there to India. These questions are critical not just for science, but for humankind."

If that sounds like an alarmist's hype, it's not. For some time, health experts have warned of a worldwide bird-flu pandemic that could kill millions of people and wreck the global economy. "The most serious known health threat facing the world is avian flu," said WHO director-general Lee Jong-wook earlier this year. And the threat is growing all the time, as nature keeps dropping hints that the links in a chain of events leading to a deadly pandemic continue to be forged. This summer, H5N1 spread west—perhaps in migrating birds—to new territory, including Mongolia, Tibet, Siberia and Kazakhstan. European countries are taking precautions by tightening surveillance of flocks within their borders; in the Netherlands, officials in late August ordered farmers to move the nation's 90 million poultry indoors to prevent any contact with itinerant fowl. Meanwhile, in Southeast Asia, where at least 58 people have died and 150 million poultry have died or been culled because of avian flu since the end of 2003, the virus is still active; a Jakarta woman died of the disease on Sept. 10. The H5N1 virus has already shown it can be deadly to people who come into direct contact with infected birds or eat uncooked poultry. But bird-to-human transmission is relatively controllable because diseased flocks can be isolated or, usually, eliminated. The sum of all fears is that H5N1 could mutate into a strain with the ability to jump easily from person to person, as ordinary flu does. That could trigger a once-in-a-century catastrophe. How many would die? Nobody knows, or can know. But in the great flu pandemic at the end of World War I, up to 100 million lost their lives (see sidebar).

Another global flu pandemic that horrific is not inevitable. "The scientists at the WHO and elsewhere who have been beating the drum about the coming flu pandemic are overstating the evidence," says Paul Ewald, an evolutionary biologist at the University of Louisville. Standards of public health in much of the world are far better than they were in 1918. There is better understanding than ever before of how a disease propagates, and how it kills. New antiviral drugs are available—though there are not enough to go around in the event of a worldwide outbreak—and H5N1 vaccines are being developed by French drugmaker Sanofi-Aventis and U.S. biotechnology company Chiron. In the wake of Hurricane Katrina, which demonstrated the high cost of complacency, the world is beginning to heed WHO warnings—last week, U.S. President George W. Bush proposed an international partnership to fight the disease, and rich countries including Britain, France and South Korea have stepped up orders for antiviral drugs that could help to protect their citizens in an outbreak. But none of these factors, alone or together, are enough to let the world sleep easy. As Time reporters worked on this story—talking to a virologist hunting for genetic clues in the muddy backyard chicken farms of Asia, a doctor in Vietnam who attends some of those who have already been infected by H5N1, and an epidemiologist in the U.S. who tracks the movement of the virus around the globe—they were left with one abiding impression. Though we have the ability to prevent or mitigate a flu pandemic, those in the frontline of the battle against H5N1 are preparing for the worst. This is their story.

At his HKU office overlooking the south China Sea, Guan fishes a cigarette from his pack of Mild Sevens and sighs, rubbing his forehead. A research paper he has been working on for years, a document he says will explain where the H5N1 virus came from and how it has become so entrenched in much of Asia, has just been rejected by a major scientific journal. That's part of life for any research scientist, and Guan is confident that the paper will soon be picked up by another journal. Still, he is taking the blow personally. "A paper is just a paper, but this is something that's important for the whole world," says the researcher, who was born in mainland China. "If a pandemic occurs, no matter how developed this world is, it would suddenly shut down," he says. "And I know it's coming."

Messianic complexes aren't unusual among èlite medical researchers, though few are as vocal about it as Guan, who will tell listeners that his work will help save us all. But then, few scientists have Guan's credentials. It was Guan who identified civets as the major reservoir of the SARS virus, convincing Chinese officials in Guangzhou to cull the animals in January 2004. That helped to prevent a possible second SARS outbreak. Guan has been just as key with H5N1: he heads the Joint Influenza Research Center, a project between HKU and Shantou University in China's southern Guangdong province, that has helped to give scientists the most accurate picture yet of how H5N1 has evolved and spread. "Guan Yi is a very bright guy," says Robert Webster, a renowned flu expert at St. Jude Children's Research Hospital in Memphis, Tennessee. "He talks a million miles a minute, and a lot of the stuff you don't even understand, but he is extremely dedicated."

Guan is important not just because of who he is but where he is. With its densely packed populations of animals and humans, southern China is the traditional birthplace of influenza. The last two pandemics, of 1968 and 1957, originated there as avian-flu viruses that mutated and eventually passed into people. That's what happened in Hong Kong in 1997 with H5N1, as the virus first spread directly from sick poultry into human beings, infecting 18 people and killing six—an outbreak that was halted only when officials culled all 1.5 million of the territory's chickens. Later, Guan and his HKU colleagues—including Malik Peiris, who discovered the SARS coronavirus—helped to trace the origin of the new, more dangerous strain of H5N1 back to a single Guangdong goose.

Getting accurate intelligence about H5N1 in southern China—where it was concentrated until bird flu exploded throughout Southeast Asia at the end of 2003—is vital to understanding the virus. But Beijing has traditionally been reluctant to share information about infectious diseases with the rest of the world. Nevertheless, with his mainland connections and his base in Shantou, Guan has been able to keep HKU supplied with a steady stream of bird-flu viruses. That has enabled his lab to monitor how H5N1 is changing and how close it may be to mutating into more virulent and communicable strains. Using technicians who roam markets and poultry farms, Guan's team has taken more than 100,000 samples from birds throughout the country, from which they've sequenced over 250 different strains of H5N1. "Guan is our connection to China," says Yuen Kwok-yung, the head of the microbiology department at HKU. "By tracing the virus in surveillance, we can know what the best vaccine would be, if it did trigger a pandemic."

Acting outside Beijing's regimented official disease-response system, Guan and his team occasionally clash with China's Ministry of Agriculture, which maintains the country has been able to control avian flu through mass poultry vaccinations and strict government surveillance. When Guan's paper in Nature concluded that the genetic markers of the viruses found in the Qinghai outbreak pointed to southern China as the likely source, the Ministry's chief veterinary officer, Jia Youling, criticized Guan's conclusions and the quality of his research in an interview with the Asian Wall Street Journal. A spokesperson for the ministry told Time that the Shantou lab had failed safety standards under new rules established for mainland labs, a charge Guan denies. "This fighting has hurt me a lot, and damaged my progress in China," he says. "But in my mind I feel comfortable, because I do nothing wrong."

Ministry officials briefly investigated the Shantou lab, although HKU's Yuen says the Center is clarifying its status with Beijing, and bird-flu work there is continuing. But the minor spat underscores how flawed disease surveillance remains in China, where local officials treat epidemic data as state secrets unless Beijing orders disclosure. While WHO officials are quick to praise Beijing for its overall improvement in dealing with epidemics since the SARS outbreak more than two years ago, there continue to be frustrations. In Qinghai, for example, Beijing was slow to share samples taken from the outbreak site and turned down offers by international experts to help to coordinate research—especially surveillance work that would have looked for the virus in healthy birds, the prime candidates to spread flu via migration. (Birds sensitive to H5N1 would likely be too sick to migrate.) Yet when a much smaller outbreak of avian flu in wild birds was detected in Mongolia last month, the government welcomed a surveillance team from the Wildlife Conservation Society and the U.N.'s Food and Agriculture Organization (FAO). The team's preliminary results showed little evidence that large numbers of healthy birds were carrying the virus.

That's good news, but researchers still need a full account of what happened at Qinghai Lake. Some of those answers may come soon when China's National Avian Influenza Reference Laboratory, which did manage to conduct some research at the site, publishes its work. Juan Lubroth of the FAO, who has seen their data, says Beijing's slow reaction was due to bureaucratic red tape rather than deliberate deception, and believes the government will respond more comprehensively to the next outbreak. That might make Guan Yi's brand of maverick virology a little less necesary, but he insists he won't stop hunting the virus. "If we face this disease, and help each other, we have a chance to control it," he says. "But if we avoid it, and miss the chance to control it, we'll end up far worse than [we were] with SARS."

In a small room adjoining his office in Hanoi, Dr. Nguyen Hong Ha takes a chest X-ray image from a female bird-flu patient and displays it on a light board. Her lungs are covered with scattered clouds that indicate infection. "And this is the next day," he says, slapping another X-ray photo on the board; by now, her lungs are utterly obscured. "This patient died last year," says Ha.

As the director of the National Institute for Tropical Medicine in Hanoi's Bach Mai Hospital, Ha likely has more experience treating human cases of bird flu than anyone. While the rest of the world worries about a future H5N1 pandemic, Vietnamese are suffering right now—40 of the 58 people killed by bird flu since the end of 2003 have died in Vietnam. "During the last eight months, it's been a steady trickle," says Ha. "There's no season for this disease anymore."

H5N1 is unusually lethal. Roughly half of those who have contracted the disease have died, despite the use of artificial respirators and antiviral drugs. That compares with the 10% mortality rate among those who contracted SARS. Ha has watched the virus ravage the lungs of healthy young patients in a matter of days. He says the key to treatment is applying just the right amount of breathing assistance—too much, and an H5N1 patient's weakened lungs can burst. But survival ultimately comes down to "the patient's immune system and the will to fight."

It's a fight that can take everything from survivors but life itself. Before he caught H5N1 this past February, 21-year-old Nguyen Sy Tuan was already slim, carrying just 50 kg on his 1.65-m frame. On his release from Bach Mai Hospital in May, he weighed merely 32 kg. Today, just below his skeletal ribs, his body bears two angry red scars where doctors inserted tubes that pumped fluid out of his lungs; another scar on his throat marks the entry wound for a ventilation tube. Because he passed in and out of consciousness during his 82-day hospital stay, Tuan doesn't remember the tubes piercing his skin, or the local press mistakenly reporting his death. What he held onto was his will to endure. "I remembered that some people had already survived this disease, so I knew I could live," says Tuan, whose younger sister Nguyen Thi Ngoan also beat bird flu. "I always believed in my resistance."

Cases like Tuan's remain rare, because for all its lethality, H5N1 is still an avian virus adapted to killing birds, not people. A flu virus infects a cell by linking its hemagglutinin gene—the H in H5N1—with chemical receptors on the surface of the cell, like a key fitting a lock. Right now H5N1 remains a poor fit for most human cells, and even when it does cause an infection, the virus is concentrated in the lower respiratory system, making it harder to spread to someone else. (Common flu invades the upper respiratory system, and can be spread easily by coughing or sneezing.) But influenza is a master lockpicker, albeit by accident; flu viruses are so mistake-prone in their genetic replication process that they mutate constantly. Flu viruses can also mutate by swapping genes with other strains through reassortment, a process that's believed to have generated the last two pandemic flus. Although the vast majority of mutants prove unfit for survival, there's always a small chance that a new strain could arise that can be transmitted easily from human to human—either through a one-shot reassortment, or through the accumulation of minor mutations. "It's like a frequent flyer program," says Michael T. Osterholm, the head of the Center for Infectious Disease Research and Policy at the University of Minnesota. "Take enough trips and you can go anywhere." And every instance of a person catching H5N1 increases the odds that a deadlier strain will emerge—one with a genetic makeup that is better suited to human hosts. "You're playing Russian roulette every time you have a human infection," says Earl Brown, a virologist at the University of Ottawa.

If H5N1 turns nastier, Ha knows he will not be able to hold back the tide of dead patients. With its eight respirators, only six of which currently work, Bach Mai is as well-equipped as any hospital in the country to deal with a pandemic—which means it's barely equipped at all. Per-capita health expenditure in Vietnam is just $8, enough to buy a single rapid test for influenza. A pandemic, says Ha, "would be a disaster."

The lab tests confirmed the Vietnamese boy had died of H5N1, but his parents were sure the 13-year-old had never been in contact with sick poultry. For epidemiologist Tim Uyeki at the U.S. Centers for Disease Control (CDC) in Atlanta, part of a team responding to a bird-flu outbreak in Vietnam in January 2004, that was a red light, a possible sign of a human-to-human transmission. Uyeki traced the victim's contacts, investigating his family and friends, before neighbors finally provided the answer: the boy was an avid cockfighter, and held the roosters close when preparing them for a bout. "Influenza virus is excreted in the feces. Chickens and ducks have fecal matter all over them," Uyeki says in his office at cdc headquarters in Atlanta. "This boy was holding roosters with the rear end pointing toward his face, and that's direct contact." One more case closed, another lucky outcome on H5N1's genetic roulette wheel.

If the next spin comes bad, Uyeki will be among the first to know. His international colleagues at the WHO are devising a plan of last resort that they hope would smother a budding pandemic. Here's how it would work: sustained human-to-human transmission is confirmed and reconfirmed; the WHO tries to box in the emerging pandemic virus by treating all contacts of all known victims with the antiviral drug Tamiflu, which can prevent infection; and strict quarantines are imposed in the outbreak area, so unknown carriers are less likely to travel and spread disease. According to mathematical models, if the virus remains relatively hard to spread; if the affected country is willing to report itself as the potential source of a pandemic; and if officials are able to respond within about three weeks of the first cases—then the disease might be contained. "No one has tested this in the past, but we all know how devastating the consequences of a pandemic would be," says Margaret Chan, the WHO's director of communicable disease. "If this solution has even a small chance of success, we need to try it."

Another option for fighting a pandemic: develop a vaccine and inoculate as many people as possible so their immune systems recognize and attack the H5N1 virus. Such vaccines are in the final testing stages, but this approach is no panacea. Viruses can mutate unexpectedly, making it difficult for drug companies to keep up. Researchers can engineer a vaccine tailored to a new flu strain within weeks, but production in high volumes still depends on 1950s-era techniques. Months could go by before the first batch would be ready. As St Jude's Webster says: "How many people are going to die in the meantime?" Moreover, because only a few companies are capable of mass producing a vaccine, the world would be hard-pressed to boost output beyond the current 300 million annual doses. "Basically, we haven't prepared for a pandemic," says Hitoshi Oshitani, an avian-flu expert with the WHO's Western Pacific office. "You need huge resources to be able to respond, and I don't think any country in the world is well prepared." Osterholm is even more pessimistic: "We have as much chance of stopping a pandemic as we would of putting a curtain around Minnesota and keeping out winter."

Still, some scientists argue that the threat of a devastating global outbreak is overstated, and that the next pandemic is more likely to result in the lower death rates of the 1957 Asian flu pandemic or the 1968 Hong Kong flu outbreak, which killed about 2 million and 1 million people respectively. That would be a far cry from the 100 million death toll in 1918. But even if the mortality rate is relatively low, the economic impact could be severe. The cost of the SARS outbreak was estimated to be $40 billion in the Asia-Pacific region alone. "If we really had evidence that the genie was out of the bottle, the world would go into outright panic," says Osterholm, who believes that even a weak pandemic could trigger the kind of quarantines and travel restrictions that could bring the global economy to a standstill. "We're much more vulnerable now because in the past we were more self-sufficient."

For Guan, there is an awareness that a potential disaster is brewing and that an urgent opportunity to prepare is being squandered. "Our job is to make the world better," he says. "But like an ostrich, we hide our face in the sand." It might be tomorrow, it might be five years from now, or it might be longer, but a pandemic is coming. In the mutable, maddening world of influenza, that much is certain. The uncertainty is our response.

• Bryan Walsh
• As Hurricane Katrina showed, hoping for the best isn't enough when it comes to preparing for potential catastrophe. Here's how three top scientists are racing to avert an avian-flu pandemic