Google Flu Trends, a recently released free tool from the company’s philanthropic division, Google.org, may detect regional influenza outbreaks 7-10 days faster than traditional methods.
Google’s bio-surveillance tool relies on the fact that many people enter phrases like “do I have the flu?” into search engines long before calling their physicians. Working with the Centers for Disease Control, Google created a basket of search phrases that suggest influenza and then aggregated all the hits by location. Terms like thermometer, muscle aches, and flu symptoms made the list.
Flu Trends utilizes terms like minimal, moderate, and intense to describe flu-related search activity and displays its results on a map (shown), along with flu prevention tips, vaccination site locators and links to helpful Web sites.
This process is light-years ahead of normal tracking systems which rely on people to visit a doctor and get tested for the flu, and then for positive results to be relayed to and analyzed by the CDC.
Last February for example, the CDC reported an influenza outbreak in mid-Atlantic states, but in retrospect Google’s search data had revealed a regional spike in flu-related search activity 2 weeks earlier.
Superior lead times can save lives by enabling officials to target educational campaigns and resources where they are needed, and by motivating people to ramp up personal hygiene practices like hand washing .
As compelling and wonderfully modern as this sounds, Google Flu Trends is still an emerging technology that must undergo rigorous prospective testing. The tool will inevitably be associated with false positive reports for example, because common but less deadly viral illnesses present with the same symptoms as influenza. The inclusion or exclusion of certain phrases into the above-mentioned basket of terms is likely to impact the tool’s accuracy.
A research paper outlining Google Flu Trends methodology will soon be published in Nature. That’s a good start.