Twitter opinions linked to vaccine uptake

Imagine using Twitter to pinpoint the potential location of a disease outbreak. Tweets then become more than short messages flashing on the computer screen. And Twitter can do more--the sentiments shared on the social network site could drive us to get flu shots or avoid them. Indeed, Penn State University researchers say Twitter chatter about an H1N1 vaccination influenced users' decisions about getting the flu shots.

A researcher took 477,768 tweets from the micro-blogging site that included words related to vaccines, according to Penn State's account of the study. Later, he examined tweets from August 2009 through January 2010, the 6-month period during which the H1N1 virus was a big news story. With student-rated data on whether certain tweets were positive, negative or neutral about H1N1 vaccines, a Penn State computer programmer created an algorithm that corralled thousands of more messages from Twitter into those three categories.

By identifying the tone and geographic locations of the tweets about the vaccines, and comparing those data with CDC statistics on regional vaccination rates, the research team found patterns. In New England, for instance, there was a high rate of positive tweets about the vaccines that correlated with actual vaccination rates in that region.

"These results could be used strategically to develop public-health initiatives," Marcel Salathe, an assistant professor of biology and lead researcher of the study, said in the item from Penn State. "For example, targeted campaigns could be designed according to which region needs more prevention education. Such data also could be used to predict how many doses of a vaccine will be required in a particular area."

Of course, there's nothing new about the study of social networks in health behaviors and how even our friends' friends can, say, make us fat. For more on this phenomenon, check out Nicholas Christakis and James Fowler's 2009 book Connected. Yet, according to Penn State's piece, Salathe's work is the first case study of how a social media site impacts disease networks.

- check out Penn State's article
- and the article in PLoS Computational Biology