Twitter Tracks Outbreaks Of Disease
According to new research by Brigham Young University, such tweets could prove beneficial for health officials who are trying to stop outbreaks.
The study analyzed 24 million tweets from 10 million unique users. They discovered that accurate location information is available for about 15 percent of tweets (gathered from profiles of users and tweets that contain GPS information). That's likely a critical number for an early-warning system that could identify terms such as "fever," "flu" and "coughing" in the city or state.
"One of the things this paper shows is that the distribution of tweets is about the same as the distribution of the population , so we get a good representation of the country," said BYU professor Christophe Giraud-Carrier. "That's an additional good reason to be able to prove that it's valid, particularly if you're planning examine things like diseases spreading."
Professor Giraud-Carrier (@ChristopheGC) and his computer science students from BYU report their findings in a recent issue of the Journal of Medical Internet Research.
Researchers discovered that Twitter's location-tagging feature, which allows tweets to be tagged with a location did not provide the information they had hoped for. They found that just 2 percent of tweets had the GPS info. This is a lower rate than the results of a survey conducted by Twitter users.
Giraud-Carrier has stated that there is a gap between what you think you know and what you actually do.
User profiles are typically used to find and parse the location information. Of course, there are some who use this location field to create a joke, i.e. "Somewhere in my imagination" or "a cube world in Minecraft." However, the researchers found that the information provided by users was correct 88 percent of the time. A portion of the inaccuracies result from people tweeting when traveling.
Health officials from public health could collect state-level information , or even better, for 15% of tweets. This indicates the viability and feasibility of a system based on Twitter for monitoring diseases to complement the data that has been confirmed from sentinel hospitals.
Scott Burton, a graduate student who was the primary author of the study, said that "the first step is to search for symptoms that are linked to real-world location indicators" and then plot them on the map. "You can also check to find out if people are discussing actual diagnosis or self-reported symptoms such as 'The doctor said I'm suffering from the flu.'
Two BYU health science professors worked on the project with computer scientists. Professor Josh West says speed is the main advantage Twitter provides health officials.
"If people from a particular region are reporting similar symptoms on Twitter public health officials could issue a notice to providers to gear up for something," West said. I love It could prove very useful in these situations.
Kesler Tanner, a BYU student, is the coauthor of the study. He created the code to get the data from Twitter. He will be graduating in April and heading off to graduate school to pursue an Ph.D.
