IT, Friends Predict Flu Outbreaks
By
John Pulley
Using electronic health records and predictive models based on the dynamics of social networks, Harvard researchers were able to identify a group of college students who came down with the flu two weeks earlier than did a randomly selected control group. Monitoring the health of individuals whose social connections make them more vulnerable to infections diseases could serve as an effective early warning system for outbreaks, the researchers said.
Using electronic health records and predictive models based on the dynamics of social networks, Harvard researchers were able to identify a group of college students who came down with the flu two weeks earlier than did a randomly selected control group. Monitoring the health of individuals whose social connections make them more vulnerable to infections diseases could serve as an effective early warning system for outbreaks, the researchers said.
The experiment built on previous work by Dr. Nicholas A. Christakis and James H. Fowler, who have found that social networks influence individuals' health, according to the New York Times, which reported on the Harvard experiment. The researchers made a splash a few years ago with a paper that found a 57 percent increase in obesity rates among people with an obese friend. Moreover, friends of friends who put on pounds also influenced one's vulnerability to weight gain, reported Christakis and Fowler. Further study found that other conditions and behaviors, from smoking to happiness, tended to spread through networks, as well.
The researchers posited that they could predict the spread of an infectious disease on campus by identifying students who have a lot of social connections, they wrote in a paper published last week in PLos One. In theory, students who have more friends will have more exposure to the influenza virus, resulting in their becoming sick earlier than less-connected peers.
Researchers compared infections among members of a randomly selected control group and a group of social connectors identified as friends by members of the control group. Social network theory holds that the people we identify as friends tend to have more friends than we do.
"You're more likely to be friends with popular people than with loners," the Times Reported. "And those popular people tend to be closer to the core of a social network."
Researchers used electronic medical records to track students who complained of flu-like symptoms to university health-care providers. Without electronic records, monitoring the clinic visits of 744 students in the study would have been difficult.
As predicted, the "friends group" became ill almost two weeks earlier than the control group. (To see a video illustration of the virus's spread through the social network, click here.) Using "friend monitoring," health officials could identify flu trends "faster than methods now used by the Centers for Disease Control and Prevention," the Times reported. Other researchers have found that social networks can have a positive effect on the adoption and maintenance of healthful habits, noted the article.