NPR logo

Deaths In Early Virus Outbreaks Can Be Misleading

  • Download
  • <iframe src="" width="100%" height="290" frameborder="0" scrolling="no" title="NPR embedded audio player">
  • Transcript
Deaths In Early Virus Outbreaks Can Be Misleading

Global Health

Deaths In Early Virus Outbreaks Can Be Misleading

Deaths In Early Virus Outbreaks Can Be Misleading

  • Download
  • <iframe src="" width="100%" height="290" frameborder="0" scrolling="no" title="NPR embedded audio player">
  • Transcript

'Flu Shots'

Get the latest updates on the swine flu outbreak.

The first reports about swine flu in Mexico made the disease sound highly lethal. But now, public health officials are saying the new H1N1 strain may be no more deadly than plain old seasonal flu.

Disease experts say a downgrade isn't surprising. It's a product of the way most disease outbreaks are detected. At first, testing tends to focus on the sickest people, and misses those who have the virus but have milder or no symptoms.

Take SARS, for example. In 2003, the virus struck the city of Toronto with what appeared to be terrifying force.

Dr. Rob Fowler was on call for the critical care unit at Sunnybrook Hospital. He remembers getting a call about a man with severe breathing problems. But what sticks in Fowler's mind is what he learned about the man's family.

"A brother had been admitted to hospital earlier in the day and had suffered a respiratory arrest," Fowler says. "Another first-degree relative of the patient was in an ambulance en route to another hospital's intensive care unit in the city." And the man's mother had died before she made it to the hospital.

Fowler says that was just the beginning.

Early Numbers Can Be Misleading

Over the next few months, he says, "We saw hundreds of patients with SARS, and it really shut down not just the medical infrastructure, but really the financial and social structure of the city."

One reason the effect was so profound, Fowler says, was that at first it appeared that SARS was killing nearly 50 percent of the people it infected. But as the outbreak progressed, it became clear that the true death rate was closer to 5 percent.

Fowler says the early figures were misleading because they didn't include people who only had mild illnesses.

"The ones that you notice are the folks that are different because they've gotten very sick," he says.

Same Pattern With West Nile Virus

Several years earlier, something similar had happened with West Nile virus in New York. There, the first confirmed cases were people with a rare brain inflammation.

One reason was that the test for West Nile had just been developed, so labs limited testing to critically ill patients, says Dr. Annie Fine of the New York City health department.

But before long, labs had the capacity to do many more tests, and health workers began to knock on doors to collect blood samples from people who weren't sick. Those samples showed that thousands of people had been exposed to the virus, but had no symptoms or just a mild fever.

Fine says the experience taught her that it's "very, very difficult" to figure out how many people have been infected with any particular virus, including swine flu.

Severe Cases Only Represent Small Fraction Of Illness

With swine flu, as with SARS and West Nile, the first identified cases in Mexico were people in hospitals who'd become very ill.

Initially, health officials had detected a surprising number of otherwise healthy young people hospitalized with pneumonia. Many of them died. And the first tests for swine flu were done in this group.

That made sense, says Dr. Frederick Hayden, a flu expert at the University of Virginia, but was bound to make the virus look more dangerous than it actually is.

"The hospitalized patients really represent only a fraction of all those affected," he says.

And mild cases of swine flu could have gone undetected for weeks, Hayden says. But this time, flu experts got a break. They found out that the virus had infected dozens of students from a high school in New York after some of them had visited Mexico.

Hayden says that allowed scientists to study a large group of flu patients who weren't in a hospital. And in this group, he says, "There were no complications or fatalities."

Hayden says new cases around the globe could still reveal a more deadly virus. But he says most people's experience with swine flu is more likely to be like that of the students than of the people identified in Mexican hospitals.