Figuring out the flu

Arthur Allen is the author of "Vaccine: the Controversial Story of Medicine's Greatest Lifesaver."

On July 6, 1968, the National Communicable Disease Center in Atlanta announced that it “did not expect any widespread outbreaks of influenza” in the 1968-69 flu season. But the NCDC -- the predecessor of the Centers for Disease Control and Prevention -- soon had to correct itself.

By late September, the so-called Hong Kong flu had stricken Marines returning from Vietnam. By mid-December, the flu was knocking thousands of Americans off their feet. The 1968 pandemic claimed the lives of actress Tallulah Bankhead and master spy Allen Dulles, and President Lyndon Johnson spent one of his last, lonely White House weeks in bed with the flu.

The cause of the pandemic was a major shift in the type of hemagglutinin protein on the surface of the virus. Flu viruses that have undergone this shift often spread quickly, but they aren’t necessarily any more deadly than seasonal strains. So it was with the 1968 virus. Only a careful reader of the New York Times would have noticed the flu pandemic. Stories about it tended to land on Page 65 and go on for all of three paragraphs. The flu season of 1951 -- not a pandemic year -- was, in fact, more lethal in many parts of the world.

Contrast that with the blare of publicity, alarm and school closings that followed the CDC’s announcement in late April that a potential pandemic was in the offing. The world’s swift reaction to the new swine flu virus displayed the superb power of our technology when it comes to detecting and reporting on things that are very small or very far away. That’s mainly what the biotech and information revolutions have done: increase our powers of detection. But detection is only the first step to understanding, and detected facts can be deceptive if they are incomplete.


For example, we’re very good at finding tumor antigen levels in elderly men’s blood but not so good at figuring out whether this sign of a prostate cancer is a cause for alarm or action. We now can find out whether genes predispose us to breast cancer or stroke, but evaluating the risk and deciding what to do about it are far from obvious. Do we remove a young woman’s breasts if she carries a certain gene for breast cancer? Do we choose invasive, but preventive, brain surgery to place an arterial clamp?

How we categorize what happens around us can profoundly affect our perceptions of risk. In 1968, Americans had not been exposed to a steady stream of disturbing news about a bird flu virus in Asia. Pandemic was not a household word associated with terror and globalism, the way it is now. Although many noticed that 1968 was a bad flu year, most of us lacked a doom-laden category in which to place that information. We went about our business, free of excessive virus fright. We’ve all heard, now, about how fearsome pandemics can be, and thanks to the remarkable advances of molecular biology and computing, we have an easier time identifying them. But it turns out that putting a name to something is not only a way of taming it. It can also be a way of spreading needless fear.

In the case of the swine flu investigation, word of the virus’ novel type preceded any ability to determine how serious it was. Initial reports from Mexico suggested a high mortality rate. But that information may have been misleading for a number of reasons, one being that Mexicans often generally don’t seek care until they are very sick.

Our initial impressions of the pathogen’s virulence were based on the number of people who showed up at the hospital, not the total number of infected people. By considering the fatality rate of those hospitalized, we got a distorted picture of the overall case/fatality ratio. Assuming this virus doesn’t undergo major mutations -- and of course, it still might -- it seems more likely to produce an outcome similar to the 1968 pandemic, which had a mortality rate of .01%, rather than the 1918 outbreak, which killed 2% of those who got sick.

Health authorities have cautioned from the start that there are too many unknowns about this virus to predict its course. They simply called attention to the potential threat of the pandemic and began preparing for it. This is what we pay them for, and why we spent several billion dollars on pandemic flu preparedness over the last eight years.

But the deceptive nature of partial information is less well understood by laypeople. We are battered by the latest headlines and hysteria from cable news programs that thrive on alarm, whether a crisis is real or potential. Public authorities, it seems to me, have handled the swine flu outbreak well. The news media, and the rest of us, need to learn how to deal calmly with incomplete scientific information.