Abstract
A BRIGADE OF 5ELF-MADE BIRD FLU EXPERTS IS TURNING THE OUTBREAK RUMOR MILL 1NTO AN ONLINE INFORMATION FACTORY.
“Unconfirmed reports that more than 120 people have died from the avian flu in western China increase concerns–and significantly raise the stakes–of a Chinese government cover-up. If true, the consequences could he grave.”
So began a June 2005 I e-mail dispatched by two I renowned Asia watchers at the Manhattan-based Eurasia Group, a private intelligence shop that monitors the globe for impending political crises and has a history of spotting disease outbreaks before the world press. More than 100 deaths was “grave” because such a high fatality rate suggested that avian flu had mutated out of its zoonotic state (spreading among birds and livestock) and was being transmitted human-to-human: the requisite stepping-stone to a global pandemic.
The analysts discovered the China rumor during their regular scans of non-English web sites, user groups, and blogs. A Chinese-language site called Boxun.com, or Abundant News, purportedly maintained by Chinese dissidents, had initially reported the deaths. Its authors claimed the Chinese government had prohibited the media and health monitors from traveling to the outbreak site, the rural Gangcha County in China's second poorest province, Qinghai. Boxun also claimed that authorities had quarantined 1,300 others who may have been infected by the flu strain, known as H5N1, which is the pathological descendent of the 1918 strain that killed more than 600,000 people in the United States.
Beijing denied any human deaths, just as it had during previous suspected outbreaks. But this rumor was worrisome, because it followed the confirmed deaths of more than 100 migratory ducks days earlier at a nearby lake. It didn't take long for the news to spread beyond Boxun. Packs of bug bloggers–who ingest and analyze every online scrap of information about avian flu news–pounced on the report. These self-made experts, whose devotion to their online lives competes with the demands of their day jobs, were shouting from the blogosphere's rooftops long before bird flu became the mass media story du jour. Going forward, they are positioning themselves as key players in preparing for a global outbreak.
Among those leading the charge was virologist Henry Niman, whose web site Recombinomics.com is a virtual situation room for bird flu news, boasting color-coded maps tracking confirmed and suspected outbreaks of H5N1 in Indonesia, China, and Europe. A newsfeed on the site, updated daily, offers commentary and background on reports culled from around the world. (In the space of a typical day, Recombinomics posts reports on countries as far afield as Turkey, Bulgaria, and Canada.)
Niman is one of the acknowledged grandfathers of the bug blogosphere. His site is a key hub for the dissemination of H5N1 rumors. He posted the Boxun reports after running them through an online translation service, which spit out butchered descriptions of the scene in Qinghai: “The infection birds and beasts flu causes the casualty already increased to 121 people….” Over at the Agonist.org–a world affairs blog frequented by a mix of amateur bug hunters, seasoned epidemiologists, and public health professionals–message board posters grabbed the Niman translations, ran some of their own, and scanned other Chinese media sites for any mention of the H5N1 outbreak. They dutifully posted everything for all to see, added their own views, and pointed readers to other bloggers, weaving threads of intelligence into an emerging tapestry.
Were the Boxun reports credible? The coveted essence of a blog–raw, often unfiltered information–is also its downfall. The blogosphere can turn into a polluted echo chamber in minutes, owing to flawed analysis, unsubstantiated rumors, and writers with axes to grind. The Agonist members considered that the Boxun reports could be a dissident attempt to embarrass Beijing. Yet, the bug bloggers also had good reason to be suspicious of the Chinese government, in light of its failure to alert the global community to the appearance of Severe Acute Respiratory Syndrome (SARS) two years earlier.
In fact, it was the SARS outbreak that revealed how effectively the online community could circumvent China's tight-lipped secrecy. In February 2003, an analyst at Intellibridge Corp.–whose assets Eurasia Group has since acquired–found a report in a Hong Kong newspaper about a mysterious respiratory virus in nearby Guangzhou, a city of more than 4 million people. The analyst sent the report to ProMED-mail, a web site and mailing list service for reports of disease outbreaks around the world, run by the International Society for Infectious Diseases. The service reaches thousands of members in more than 150 countries and is published in five languages, including Chinese. (The World Health Organization and other official sources “are constrained in their reporting by the need for bureaucratic clearance,” the service informs readers. “ProMED-mail … has no such constraints, and posts outbreak reports seven days a week.”)
Around the same time, a medical doctor and health consultant received an e-mail alluding to an “epidemic” breaking in China. The sender cited “an acquaintance of mine from a teacher's chat room” who lived in Guangzhou, and “reports that the hospitals there have been closed and people are dying.” The doctor sent the message to ProMED-mail. An employee with the Tennessee State Health Department also found news of the illness on Boxun, which cited a poster at RenAi Hospital in Guangzhou pushing doctors and employees to fight the disease, then thought to be a form of pneumonia. He too dispatched the news to ProMED. Within days, the Western world's medical establishment was alerted to the new virus. And not long after, one could scarcely turn on a television or walk through a grocery store checkout line without encountering news stories and headlines about SARS. The online community had the story first.
Two years later, in June 2005, the bloggers again took to their battle stations. As the reports of avian flu from Qinghai accumulated without any official rebuttal, one writer captured the dread-laden resignation as the rumor mill tilted into high gear: “We are all at the mercy of the Chinese at the moment. Let's hope they get it right. And hope the migratory birds don't carry the virus anywhere else…. And hope … and hope….”
Ultimately, the rumors of H5N1 human infections proved false. But in early June 2005, no one knew that. Every morsel in the bloggers' ravenous Qinghai data diet was a rumor. The bug bloggers themselves are the first to acknowledge the challenges and limitations of tracking a viral infection in cyberspace. “In the hours we daily spend gathering information, most of the world is getting on with its life,” wrote Crawford Kilian in an online column. Kilian, a Canadian professor and science fiction writer, publishes H5N1 (crofsblogs.typepad.com/h5nl), a flu blog mainstay. “The pandemic that draws us together is technically nonexistent: A lot of chickens and wildfowl are dying, but very few people.”
Only World Health Organization (WHO) investigators could have credibly tested the alleged victims in Qinghai to confirm or refute H5N1 as the cause of death. The Chinese government hadn't yet invited them in. So, where were the WHO's H5N1 experts? In all likelihood, they were sitting at computers, just like the bloggers.
Back up a year and a half. In January 2004, 14 people in Vietnam showed up in provincial hospitals in severe respiratory distress. Lab tests confirmed at least three H5N1 infections. WHO officials based in Manila rushed to the scene. (The Vietnamese government has not resisted WHO intervention.) What they found scared them. The infections were sporadic, and no one knew whether these three were the only victims. Had they contracted the flu from livestock, or was H5N1 now passing person-to-person? The Manila team, knowing they were about to set the health world hyperventilating, issued an international public health alert on January 13. And then they listened.
Duck and cover: Chinese health workers prepare to kill a duckling in the vicinity of a suspected bird flu outbreak.
As feared, news of the H5N1 infections spread and fanned anxiety worldwide. Unconfirmed outbreak reports, or “disease rumors,” started proliferating. Left unchecked, those anecdotes could inspire extreme measures–nations closing borders; airlines abandoning flight routes; and the wholesale extermination of poultry, which is a primary source of income throughout Southeast Asia. In order to prevent such an outcome, the WHO implemented “enhanced rumor surveillance” of unofficial disease reports, an intense process of using expert analysis and human networks to determine the veracity of every claim. The internet and media produced many of the leads.
“Rumor surveillance is a passive process,” wrote Gina Samaan, a disease researcher who coauthored a study of the Vietnam surveillance in the March 2005 issue of Emerging Infectious Diseases, published by the U.S. Centers for Disease Control and Prevention (CDC). “In an enhanced system, rumor surveillance is intensified by actively seeking out rumors and undertaking more rigorous follow-up. This surveillance includes analyzing more media sources and regularly requesting information” from members of WHO's global network of health practitioners.
The Vietnam rumor-watchers swapped stories, tossed around theories, and prodded officials for confirmation. What did people hear on the street? What did they see in villages? Were people coughing? Sneezing? The Manila office became rumor central and the home base of a rumor surveillance officer, who “actively assessed media sources [mostly in English, but some in Japanese and Arabic] and e-mail-based public health discussion,” Samaan wrote. The officer also tapped the Global Public Health Intelligence Network, a WHO electronic system that continuously checks more than 600 media sources in several languages, including Chinese.
In essence, and in practice, this is no different from how the bug bloggers operate. “Rumor surveillance is basically having people poised and paying attention to the environment around them,” says Howard Hill, a private consultant who worked for the CDC in Atlanta for 35 years. Most recently, Hill was the senior adviser to the director of the viral diseases division, which manages CDC's flu program. “When they hear someone has an infection or a condition that's deadly, they know those people need to be treated.” That's how CDC operated its sexually transmitted diseases program, in which health workers diagnosed sexually transmitted disease infection by spotting secondary characteristics, such as lesions or unusual hair loss, Hill says. “I found many cases of secondary syphilis just by being observant.”
But the bloggers and WHO officials, with their rumor-watcher-in-chief, differ in one important respect. The rumor officer followed up every report with a request to the appropriate WHO country officer to check out the story on the ground. Bloggers, strapped to their keyboards and their day jobs, can't take that critical step. But, just like their comrades in Manila, in 2004 they sat at their desks, read e-mails and articles, and fretted with their friends and colleagues. And their efforts rivaled those of their WHO counterparts. From January 20 to February 26, the WHO identified 40 rumors, about half from the media, according to Samaan. In the same time period, 80 of 97 topics posted at the Agonist–more than 80 percent–concerned reports of bird flu, and not just in Vietnam. The bloggers found reports of strains in Pakistan, Russia, China, and a “mild strain” in Texas. Each posted report contained multiple comments, forming a running dialogue. In many of them, the steadiness of investigation, the refusal to give in to panic without the full facts being known, was omnipresent.
The blogs also offer something frequently missing in the governmental health world–candor. “There is a time for blunt speaking, and the blog offers that,” says “Revere,” a prominent public health scientist and academic who maintains the highly regarded web site Effect Measure (effectmeasure.blogspot.com). (The blogger's pseudonym is in honor of Paul Revere, who was a member of the first local board of health in the United States.) “In public, I say, ‘CDC has [management] problems,’” says Revere. “On the blog I say, ‘[CDC director Julie] Gerberding is spineless.’” Indeed, confronted with the possibility of an imminent pandemic, Effect Measure has little tolerance for the posturing of public officials. Revere refers to the WHO as the “World Reassurance Organization,” accusing it of downplaying evidence of bird flu outbreaks in Indonesia so as not to offend member governments and disrupt trade and tourism in Asia. Likewise, Effect Measure had harsh words for the European Union, which convened a 25-nation emergency summit on the bird flu in October 2005, only to have the European health commissioner declare, “The fact [that] we have avian flu in Europe does not affect the possibility of a human influenza pandemic.” (An irate Revere responded: “I kinda thought that the more infected animals and the more widespread the virus, that would also, kinda, like, you know, increase the risk a little? … I don't know about you, but the one thing that makes me want to panic is when the leaders of 25 countries meet in emergency session and tell me not to panic”)
Some bloggers believe they're doing a job that the medical professionals are ignoring. “I'm an enthusiastic amateur epidemiologist,” says Melanie Mattson, who writes the political blog Just a Bump in the Beltway (node707.com). Mattson chooses her stories using a simple rubric. She wants “the stuff that ordinary people don't read. Quite frankly the stuff that doctors don't read,” including the international media, scientific journals, and fellow bug bloggers. Like many of her colleagues, Mattson worries that the professional health class isn't ready for a flu pandemic because doctors and specialists haven't been following the disease closely. “Docs are as unprepared for this as laypeople are,” she argues.
Revere thinks Effect Measure adds value to the information health workers can get from other sources. “I'm not just a news filter,” the blogger says. “I take certain things, highlight certain things, then add a commentary.” Anonymity hasn't diminished Revere's credibility, but it does complicate the logistics of blogging. “It's sort of a pain in the ass actually,” says Revere, who has to maintain a secret identity by using a special e-mail address to correspond with readers and sources.
MEET THE BUG BLOGGERS
Recombinomics' Henry Niman.
Public health workers have educated themselves by paying attention and [sharing what they learn. And so have the bloggers. In the early days of blogs–about three years ago–the few authors out there started reading one another. In the process, they learned how to blog, but also about how the avian flu develops and spreads.
“We [became] like a little Stone Age tribe of hunter-gatherers … getting stuff, sharing it… learning what was good and not good to eat,” says H5Nl's Kilian. In only a few years, that online tribe has branched, with flu blogs emerging in Japan, Israel, France, and Germany. Bloggers report hits to their sites from all over the world, including from CDC and WHO addresses.
Bug bloggers have evolved into a “kind of human syndication service,” Kilian says. They pull together and analyze information that interests them, often with a precision and institutional memory that escapes the mainstream media. While the bloggers aren't all public health experts, they certainly know more about bird flu than most of the population. And if anyone else wants to become that sophisticated, all he or she has to do is read.
To that end, prominent bloggers such as Mattson and Revere have launched the ambitious Flu Wiki (fluwikie.com). Published in four languages, Flu Wiki acts as a sort of homeland security guide for local communities on how to prepare for and cope with potential flu outbreaks. Its contributors include biologists, epidemiologists, and virologists from all over the world. The site was born out of the apparently unanimous conviction among bug bloggers that the federal government would be of little help during a pandemic–a perspective recently reinforced by Pat Libbey, the executive director of the National Association of County and City Health Officials. “Communities, in large part, will be on their own,” he predicted in a recent interview.
Since the flu is contagious before symptoms develop, once victims begin showing up, epidemiologists would have to assume that the flu had already spread over a vast geographic area. At that point, the federal government “can't come in and take over,” Libbey observed. “The math alone just doesn't work.” (Or, as Revere wryly wrote, “The cavalry isn't coming.”) Mattson says Flu Wiki receives at least a few hundred hits per hour. And the readers, Mattson and her colleagues believe, are spending more time than average web surfers reading material–up to 20 minutes per visit.
Still, the number of readers is small compared to the web traffic of mainstream media outlets, with whom the bug bloggers (like bloggers everywhere) have a symbiotic, love-hate relationship. In a column he wrote for an online publication, Kilian described a mysterious “pig fever” that broke out in China in June 2005. Was this avian flu mutating through pigs, much as the Spanish flu did in 1918? “Chinese authorities said it was a bacterial infection, Streptococcus suis
Playing chicken: Doctors in South Korea seek to calm public fears by eating chicken in November 2005.
Yet, the circular dance of information exchange can benefit both the bloggers and the media. Some bug bloggers are emerging as expert sources: Recombinomics' Niman, for instance, has been quoted by CNN, United Press International, and the Sydney Morning Herald. Establishment journalists also serve the blogs: Mattson, the Flu Wiki co-creator, says Declan Butler, the European correspondent for Nature, contributes to the blog “and sends me things on a regular basis.” And when stories of avian flu show up in the mainstream press, it can drive traffic toward the blogs. After ABC News aired a segment on bird flu earlier this year, Kilian saw his traffic grow dramatically. As he described it in a post: “This was clearly due to viewers of one program, who promptly booted their computers, Googled ‘H5N1,’ and found my site as No. 2 out of close to two million pages.”
And this may be the bug bloggers' biggest contribution to public health. While the WHO and CDC and other official bodies can go that extra mile and deploy experts to the hot zone to scientifically verify a rumor, the bloggers, who have now become as informed as most health professionals, can help the public make sense of what they're hearing and seeing. They help turn the rumor mill into an information factory. Considering that confusion and mass panic in the wake of a global pandemic of H5N1 are to be feared almost as much as the disease itself, bug bloggers are now a valuable addition to public health–the frontline correspondents in the flu war.
BEST OF THE BULLETIN ARCHIVE: Infectious disease
↗
↗
↗
For these articles and more, visit the online Bulletin Archive at www.bulletinarchive.org.