What can Twitter tell us about H1N1 vaccinations?

After sorting through nearly 480,000 tweets, researchers show that people against the vaccine tend to influence those in their circles, creating under-vaccinated communities.
Written by Janet Fang, Contributor

Social media can affect the spread of diseases… that is, according to a new study on attitudes toward the H1N1 vaccine.

Marcel Salathé and Shashank Khandelwal from Penn State University studied how Twitter users expressed their sentiments about the relatively new swine flu vaccine.

In particular, they tracked how the attitudes correlated with vaccination rates and how people with the same for or against feelings influenced each other in their social circles.

"It is very likely that negative opinions of vaccination are contagious on online social networks," Salathé says.

  1. They amassed 477,768 tweets with vaccination-related keywords and phrases.
  2. Then they tracked sentiments about the H1N1 vaccine between August 2009, when news of the vaccine was made public, up through January 2010. The first 10% were sorted by hand, the rest were done using an algorithm.
  3. A tweet expressing a desire to get the H1N1 vaccine is considered positive; a tweet saying the vaccine causes harm is negative. (A tweet concerning the hepatitis B vaccine is considered irrelevant.)
  4. Sentiments were categorized geographically. (The highest positive-sentiment users were from New England.)
  5. Using data from the Centers for Disease Control and Prevention (CDC), they correlated vaccination attitudes with vaccination rates. (New England also had the highest H1N1 vaccination rate.)

"These results could be used strategically to develop public-health initiatives," Salathé explains. Campaigns could be targeted at regions that need more prevention education, or the data could help predict doses required in particular areas.

They also found, not surprisingly, that users followed like-minded people. "The public-health message here is obvious," Salathé says. "If anti-vaccination communities cluster in real, geographical space, as well, then this is likely to lead to under-vaccinated communities that are at great risk of local outbreaks."

AND, negative tweets spiked when the vaccine was first announced and during periods of vaccine recall. More-positive sentiments popped up when the vaccine was first shipped across the US.

"People tweet because they want other members of the public to hear what they have to say," adds Salathé, who thinks the analysis could be used for noninfectious diseases like obesity and hypertension. Except, as he puts it, behaviors such as smoking or unhealthy eating habits can be infectious.

"Now that heart disease,” he says, “is moving to the top of the list of killers, it might be wise to focus on how social media influences behaviors such as poor diet and infrequent exercise.”

The study was published last week in PLoS Computation Biology.

Image by alkoga via Flickr

This post was originally published on Smartplanet.com

Editorial standards