Christophe Fraser et al, Science

Going viral. A look at how air travel helped H1N1 spread.

Early Lessons From Mexico's Swine Flu Outbreak

Jon is a staff writer for Science.

The first quick and dirty analysis of Mexico's swine flu outbreak suggests that the H1N1 virus is about as dangerous as the virus behind a 1957 pandemic that killed 2 million people worldwide. But it's not nearly as lethal as the bug that caused the 1918 Spanish flu pandemic.

The world first became aware of this latest outbreak of swine flu in late April. Researchers recognized that the new virus was also causing disease in Mexico. As of today, there have been more than 5000 confirmed cases worldwide and fewer than 50 confirmed deaths.

In an effort to understand how the virus spread, an international team led by epidemiologist and disease modeler Neil Ferguson of Imperial College London closely examined the Mexican outbreak with the most current data available. The researchers calculated that each infected person transmitted the virus to about 1 1/2 other people and that it had been around long enough to copy and spread between groups of people between 14 to 73 times.

The team also showed a strong link between air travel out of Mexico and confirmed cases in other countries (see picture). The case fatality rate--the percentage of infected people who died--was estimated to be 0.4%, with a range between 0.3% and 1.5%. That's far less than the devastating pandemic of 1918 but still substantial, the team reports online today in Science.

Epidemiologist Arnold Monto, a flu specialist at the University of Michigan School of Public Health in Ann Arbor, praises the speedy collaboration among researchers from the United Kingdom, Mexico, and the World Health Organization to produce the study. He notes that in other outbreaks, many heavily affected countries have hesitated to share data.

Yet Monto cautions that the findings are provisional and that the wide ranges in the estimates indicate limited confidence in the conclusions. "They have been dealing with what they have to work with in terms of the data," says Monto.

Epidemiologist Lone Simonsen at George Washington University in Washington, D.C., also notes that the number of confirmed cases in Mexico probably vastly underestimate the situation, as the sickest cases are most likely to be tested for the virus. As more mild cases are detected, the denominator could rise, which would lower the fatality rate. "It's quite possible that the confirmed cases are the tip of the iceberg," she says. "The uncertainty in the denominator means they could be two or three orders of magnitude off."

Still, both Simonsen and Monto say that they welcome fast and dirty analyses right now. "We need to very speedily learn about the infections in the first affected countries," says Simonsen. "We have burning questions to answer."

You can find Science's full coverage of swine flu here.

Posted in Health, Biology