Imagine you’re standing around the office water cooler as your coworkers drift by. One day in late fall, you start hearing the same things over and over, things like: “Oooh, my stomach isn’t great today,” or “I’ve had this cough for three days and it just won’t quit.” You might figure that was a smart time to get your flu shot.
But disease surveillance organizations, the groups that track and predict disease outbreaks to make sure we have enough hospital beds, vaccines, medicines, and other supplies, can’t stand by every water cooler in the world waiting for up-to-the-minute information. Or they couldn’t–not until now.
The World’s Biggest Water Cooler
Social media, especially short-post “microblogging” sites like Twitter and Facebook, can act like the world’s biggest water cooler, bringing together millions of people’s posts on their daily activities, their thoughts and opinions–and their health. Scanning such data for keywords like “flu” and “cough,” as well as references to time (such as how long a poster reports having had an illness) can give disease surveillance organizations an amazingly accurate picture of a growing epidemic or the probable length of this year’s flu season.
Twitter in particular was the subject of a 2012 paper that found that the popular social media site can improve predictions about seasonal influenza. Tracking keywords from people’s posts about themselves could actually outperform the CDC in terms of accurately predicting flu locations and symptoms. Such real-time data helps cities and counties, as well as hospitals and other providers, prepare for flu season and ensure that sufficient resources are available.
You Can Play, Too
One of the most exciting and interactive social-media disease surveillance sites is crowdbreaks. Created by the Center for Infectious Disease Dynamics at Penn State, the site uses a vast number of individual inputs to more accurately “read” social media sites. In other words, you can help out: all you have to do is go to the link above and answer some questions.
The questions ask, basically, “Is this social media nugget we found really about what our computer thinks it’s about?” So for example, maybe the computer found a Facebook post that says, “I’ve had this cold for a week, aarrg!” Crowdbreaks might show you this post and ask, “Is this a statement about the common cold?” When you answer yes, the site’s algorithms are tweaked to recognize similar posts in the future. And you’ve helped improve disease prediction with one simple click.
Bring it Home
If you’d rather find out what diseases might be coming your way, there’s HealthMap, a site created by a team at Boston Children’s Hospital that actually provides an interactive map of ongoing infections. You can choose to view the global map or the local one, or just look at the most recent news-worthy items, meaning the experience is customizable to your needs.
The site is particularly useful for parents who want to know whether infections diseases, including flu, might be coming to their towns (and possibly schools) in the near future.
Check it Out
For more information on how social media is changing disease surveillance and prediction, check out NIH’s report on the subject, which is chock full of useful information about the wide variety of new projects in this area.
But most of all, keep tweeting, posting to Facebook, and publishing your blogs; you might just be saving a life.