Select your localized edition:

Close ×

More Ways to Connect

Discover one of our 28 local entrepreneurial communities »

Be the first to know as we launch in new countries and markets around the globe.

Interested in bringing MIT Technology Review to your local market?

MIT Technology ReviewMIT Technology Review - logo

 

Unsupported browser: Your browser does not meet modern web standards. See how it scores »

Epidemics spread quickly and erratically, and researchers have been hunting for better ways to predict outbreaks for some time. But although widespread technology is providing innovative ways to pinpoint emerging outbreaks–from social media to internet trackers–government organizations still can’t get as great a jump on them as they’d like.

James Fowler, who studies genetics and social networks at the University of California at San Diego, has a novel proposal for them: Rather than trying to get a grasp on what everyone is doing by studying vast networks of data, focus on what the popular people do. In research published today in the open-access journal PLoS ONE, Fowler and his Harvard colleague Nicholas Christakis followed two groups of people during the 2009 H1N1 or so-called “bird flu” pandemic–one group chosen at random from the Harvard undergrad population (the control), and one group comprised of friends of the first group.

Granted, college campuses are insular and may not be representative of larger networks, such as those in large cities; but they provide a perfect petri dish (so to speak) for experiments like Fowler’s. By the end of campus outbreak, Fowler’s data clearly showed that people in the friends group were, for better or worse, ahead of the epidemic curve. On average, these students came down with the flu 13.9 days before the control group.

Being dubbed a “friend” meant that someone was more likely to be widely connected than a randomly selected person in the control group. And because such people were more deeply entwined in various campus social groups (real, face-to-face social networks), they were more likely to be exposed to the H1N1 virus as it was just entering circulation. In other words, these people were early detectors of the H1N1 virus before it peaked on campus or across the country.

“The best the CDC can do right now is to lag a couple of days behind an epidemic,” Fowler says. Internet search data can do a bit better, providing data that reflects the here and now. But his method, he says, “is a crystal ball for looking to see what will happen to the whole population. These are the people to look at if you want to see what will happen in the future.”

Fowler wants to combine these human sensors with other methods in development, such as those that monitor internet search terms. “We could potentially follow the online behavior of the friend group. And if the friend group is looking up information about flu symptoms and what kind of cough syrup works best, we suspect that will give you advance information,” he says.

0 comments about this story. Start the discussion »

Tagged: Biomedicine, social networks, disease, H1N1, influenza

Reprints and Permissions | Send feedback to the editor

From the Archives

Close

Introducing MIT Technology Review Insider.

Already a Magazine subscriber?

You're automatically an Insider. It's easy to activate or upgrade your account.

Activate Your Account

Become an Insider

It's the new way to subscribe. Get even more of the tech news, research, and discoveries you crave.

Sign Up

Learn More

Find out why MIT Technology Review Insider is for you and explore your options.

Show Me