Thursday, January 31, 2013

Your doc is following you on Twitter

The doctor is now following you on Twitter


Scientists say social media can track diseases better than the CDC


January 30, 2013|Jen Wieczner



Savvy analysts and investors have been using Twitter to predict election results and identify new hot stocks. (See The wishdom of crowds)
Now a group of researchers at Johns Hopkins University is using the social media site to track the movement and severity of the flu — and their work may be used to address other public-health issues. (See U.S. researchers tracking flu through Twitter)
As people tweet personal details about their health (and illnesses), the scientists screen the social network to gather data on influenza cases in the U.S. While health experts have in the past had trouble monitoring the virus, since people with the flu often stay home in bed, social media sites like Twitter may provide a unique real-time window into national epidemics even as they play out behind closed doors.
“When an influx of mentions of the flu start appearing on Twitter, it can be used to see where the strain started and how quickly it is spreading just by following the conversations about it,” says Sheldon Levine, community manager for Sysomos, a social media monitoring and analytics company.


Inspired by the power of the tweets to reveal trends faster than traditional surveys and polling, the researchers began monitoring Twitter for health issues about two years ago. They saw potential, for example, in the social media site’s ability to pick up an earthquake in Washington, D.C., moments before people felt it in Manhattan — faster than seismic monitoring equipment could even have detected it, says Michael Paul, a doctoral student in computer science collaborating on the studies. Twitter, the researchers say, might yield flu intel faster than the Centers for Disease Control and Prevention.
(A 2010 analysis of the “Twitter mood” by researchers at Indiana University predicted whether the Dow Jones Industrial Average would rise with nearly 90% accuracy — up to a week ahead of time. (See Analyzing almost 10 million tweets, research finds public mood can predict Dow days in advance)
“Because people are sharing so much on Twitter in their everyday lives, we thought we could leverage that to look at people’s health,” Paul says.
He and his colleagues have since been able to map out the danger zones for allergies in real time, with Twitter users inadvertently identifying certain cities and states that are allergen hotbeds. The researchers are also looking at other health issues to pick up on trends, including obesity, hospital errors and self-medication, as well as drug misuse and abuse.

No comments:

Post a Comment