The Washington Post
Figuring out where and when the next flu outbreak will occur may soon be less of a guessing game, thanks to technology from the meteorological world.
To predict when and where the flu may hit, scientists at Columbia University's Mailman School of Public Health and the National Center for Atmospheric Research have collaborated on a model that operates like a weather modeling system.
"This system can potentially be used to estimate the probability of regional outbreaks of the flu several weeks in advance," said Alicia Karspeck, a scientist at NCAR and a co-author of a study published in October in the Proceedings of the National Academy of Sciences.
After developing their model, Karspeck and Columbia University colleague Jeffrey Shaman ran it for the 2003-2008 flu seasons in New York. The results, they wrote, showed that "real-time skillful predictions of peak timing [of a flu outbreak] can be made more than 7 weeks in advance of the actual peak."
Shaman and Karspeck's system uses near-real-time data from Google Flu Trends, which tracks flu activity. The data can help overcome incomplete understanding about how weather and other factors (such as the exchange of germs indoors) may affect flu transmission.
While the authors caution that their study represents just a first step in influenza forecasting, they are hopeful about its application: "We will soon reach an era when reliable forecasts of some infectious pathogens are as commonplace as weather predictions."