In Drill, CDC Practices For Influenza Pandemic
Agency Uses Mock Outbreak to Prepare for Disaster

By David Brown
Washington Post Staff Writer
Sunday, February 4, 2007

ATLANTA -- A 22-year-old Georgetown University swim team member just back from Indonesia eats dinner with his teammates but then develops a fever and doesn't accompany them to a meet in New York.

That is how a flu pandemic in the United States started.

A winter storm bears down on saltless and plowless Atlanta, closing schools and scaring commuters.

That is how it ended.

The Centers for Disease Control and Prevention held a pandemic "war game" here last week but aborted it at midnight Wednesday, halfway through its planned 24-hour run. The reason -- ironically, but perhaps not surprisingly -- was concern about public safety. CDC did not want a hundred of its most valuable employees rushing on Thursday morning over ice-slick roads to a mock disaster.

A real pandemic, which is considered inevitable, won't be stopped by weather, which in this case turned out to be mostly rain. In fact, it probably won't be stopped by anything. But public health experts believe it could be made less disastrous with practice and preparation.

For that reason, CDC's director, Julie L. Gerberding, who spent Wednesday facing nail-biting pseudo-decisions -- and made the real one to call things off -- was happy with half a pandemic.

"The long-frame view is that this was a tremendous success. We exercised, we learned and we are definitely committed to this process," she said when it was over.

The federal government is hard at work trying to ready the country for a global outbreak of a new, highly transmissible strain of influenza -- a pandemic. Such events occur at unpredictable intervals. There were three in the last century, in 1918-19, 1957 and 1968.

The H5N1 strain of avian influenza, or bird flu, has killed millions of birds and 164 people, mostly in Asia, since 2003. It does not pass from person to person efficiently. But it is highly lethal and still evolving; many experts believe it has the potential to cause a pandemic.

Last week's exercise was conducted by MPRI, an Alexandria-based consultant with a $7 million contract to help CDC plan for pandemic flu. Founded in 1987 as Military Professionals Resources Inc., the company is run principally by retired officers.

Horace G. "Pete" Taylor, a retired Army lieutenant general, led a team of 24 people who wrote the outbreak scenario, "injected" new information as the exercise ran, played the roles of non-CDC decision makers, and are now helping the agency analyze its performance. Taylor, 69, used to run war games at the Pentagon.

The exercise began Jan. 12 with e-mail traffic to CDC officials describing H5N1 flu activity in Indonesia. With six cases and five deaths in 2007, that country is currently the most active place for bird flu.

A week ago a fictional element appeared in the e-mails in the form of the Georgetown student's story. Two days before the exercise began, his virus was identified as H5N1. When the game began at 7:30 Wednesday morning, he was a few hours from dying and two of his housemates were down with the flu, its type not yet identified.

About a hundred people -- many in the Navy-like uniforms of the U.S. Public Health Service -- filled the CDC's emergency operations center, typing and clicking at benchlike workstations, milling and meeting as maps, presentations and video feeds filled screens at the front of the room.

The exercise sought to capture the elements that make outbreaks hazardous and decisions difficult.

For example:

The ill person came from a country with less-than-optimal disease surveillance and laboratory testing. Nobody was certain what the facts of the Indonesian outbreak were. Was there casual spread of the illness in Jakarta? Were there outbreaks in other parts of the country? These are not minor details. They help determine whether the World Health Organization -- with lots of CDC help -- will try to mount a mass distribution of antiviral drugs in a one-shot effort to snuff out the virus at its source.

A passenger on one of the student's flights had died in Chicago and another was sick in Baltimore. Was this just a coincidence? It was flu season, and regular influenza virus makes some people very sick -- killing some of them.

The number of people who might be incubating flu grew by the hour. Should the whole swim team be quarantined? How about all the other swim teams at the meet?

The federal government has 20 million doses of oseltamivir (Tamiflu). When should this stockpile be distributed to the states? At the first hint of a pandemic, when many doses might prevent cases but also be wasted? Or later, when the outbreak is confirmed and the drugs will go only to the ill, possibly saving their lives?

There were hints that the Georgetown student might not be the only source. A passenger on a flight from Singapore to Hawaii developed fever and "respiratory distress." Did he have some link to Indonesia? Should all flights from Asia be diverted to the 20-odd airports that can screen and quarantine passengers? Should CDC push to have people leaving certain countries be medically evaluated?

And then there were the small annoyances.

The scenario said the student was "living in an on-campus dorm." But this was wrong. He lived in a rented house in Northwest with 10 people -- an error that gave headaches to the two cartographers who were marking satellite photos of the District -- all the places the "index patient" had been.

The exercise included a mock news conference. CDC media officers questioned Gerberding while four real reporters (from The Washington Post, the New York Times, the Associated Press and the Canadian Press) watched.

Gerberding ran two high-level meetings like an experienced college professor in a freshman seminar. She regularly summarized the avalanche of facts, cleared the air with concrete questions and demanded unambiguous answers from her advisers.

"We probably have to assume the worst," she said at one point, noting that even with woefully incomplete data from Indonesia it appeared that one-third of the people there with the flu were dying. Later, she asked: "So, do we agree there is strongly supporting evidence that we have a pandemic?" The senior staff agreed there was.

In the news conference, however, she avoided eye contact with reporters, emphasized uncertainty and concentrated on the paucity of facts. A news release quoted her as saying: "We have not yet seen sustained or repeated human-to-human transmission of H5N1. It's important for people to remember that one case does not mean that a pandemic has begun." She later admitted she hadn't had time to read the quotations attributed to her.

The game will pick up in April with an exercise in which pandemic flu has spread to many states. In a final round in May, the virus gets to Atlanta and takes out 40 percent of CDC's workforce.

It's unlikely there will be any snow to save them.

View all comments that have been posted about this article.

© 2007 The Washington Post Company