As we know, social media can be used to keep in touch with friends who are thousands of miles away, grow a business, get feedback about anything under the sun, meet up with locals for a drink after work, interact with celebrities, or any number of things that, simply put, are galvanized by people communicating with other people. Now, Penn State’s Marcel Salathé, an assistant professor of biology, has devised a unique way to use data gathered from social media service Twitter: tracking H1N1 vaccination rates and attitudes of positivity or negativity related to the procedure.
Believed to be the first case study in how social media can be used to interact with disease networks, its success seems to indicate that it won’t be the last. But how did Salathé arrive at the decision to use Twitter for this very specific purpose? He explains that “People tweet because they want other members of the public to hear what they have to say. Tweets are very short — a maximum of 140 characters — so users have to express their opinions and beliefs about a particular subject very concisely.”
Salathé’s research began by collecting tweets that included vaccination terms, keywords, and phrases. From these resulting 477,768 tweets, he was able to assess sentiments related to statements made in these tweets. As he was trying to specifically gather data about popular opinion toward the H1N1 vaccine, he took a random sample of 10 percent of these tweets and had students evaluate their positivity, negativity, neutrality, or irrelevance. Statements that were made concerning irrelevant vaccinations were discarded from the study. Statements that expressed the Twitter user’s desire to get an H1N1 vaccination were considered positive, whereas statements that derided the H1N1 vaccine as dangerous and to be avoided were considered negative. From the assessments made to this 10 percent, he was able to design an algorithm that would evaluate the remainder.
As locations are often disclosed by Twitter users, Salathé was able to correlate the expressed attitudes by region and compare them with Centers for Disease Control and Prevention (CDC) vaccination rate estimates. New England, for instance, had the highest level of tweets that displayed an overall positive attitude toward the H1N1 vaccine, and CDC records showed it as also being the region with the highest percentage of vaccinations.
“These results could be used strategically to develop public-health initiatives,” Salathé explains. “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.”
Another interesting observation made by Salathé was a confirmation of the notion that people who communicate with one another online tend to do so with people who think like them in a “birds of a feather flock together” sort of a mentality. Predictably, this kind of interaction results in an environment where dissenting opinions and ideas never enter the picture — which not only has the potential to keep important facts at bay, but serves to affirm whatever cockamamie convictions people dream up as facts.
“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. By definition, herd immunity only works if unvaccinated, unprotected individuals are distributed sparsely throughout the population, buffered from the disease by vaccinated individuals. Unfortunately, the data from Twitter seem to indicate that the buffer of protection cannot be counted on if these clusters exist in real, geographical space.”
Using the lessons learned from this experiment in social media and disease control correlation, Salathé believes they’ll also apply to diseases that have more to do with choice than dumb luck — that is, lifestyle diseases vs. contagious diseases.
“Behavior-influenced diseases always have existed, but, until recently, they were masked: People died of infectious diseases relatively early in their life cycles. So behavior-influenced diseases weren’t really on anyone’s radar. Now that heart disease — a malady caused, at least in part, by lifestyle — 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,” Salathé says.
