Twitter can predict rates of heart disease

Researchers from the University of Pennsylvania have found that they can use Twitter to accurately predict rates of heart disease in specific areas.

Coronary heart disease (where arteries become clogged with fatty plaque, making it harder for blood to move around the body) is the leading cause of death worldwide. Risk factors include stress, a low income, and smoking. Now it also looks like how positive people are in a particular community is linked to the spread of the disease.

The researchers looked at tweets posted from 2009-10 by residents from 1300 counties across the U.S. A linguistic analysis showed a correlation between swearing and negative words like ‘hate’ and a higher rate of death from heart disease, while the opposite was true for people used words like ‘wonderful’ and ‘friends’ and tweeted about positive experiences.

This doesn’t necessarily mean that being an online Pollyanna protects your heart: negative word choices are likely to reflect negative emotions, which are linked to poor health choices. Still, if only for your followers’ sake, it can’t hurt to try to look on the bright side on social media.

Then again, the important thing to remember about this research is that the results reflect specific areas rather than accounts: if you’re living in a town of negative tweeters, you could be at risk even if your hilarious #tbt is the highlight of everyone’s week. (Just what we needed: something new to worry about when we need to move house. Although don’t panic, because that could make heart disease more likely, too.)

Whether you want the language of your tweets analysed in this kind of detail or not, the social network’s becoming a popular source of information for health researchers, with a UCLA scientist using it last year to find areas most at risk of HIV outbreaks. As in that case, the Pennsylvania researchers hope that their study might provide the basis for an outreach programme aimed at lowering the risk and incidence of disease in the worst-affected areas. Its success would be measured by Twitter, naturally.

Image via Andreas Eldh’s Flickr.