A lot of sexual tweets in your area? Local syphilis rates could be on the rise. Oversharing on social media may be annoying, but it could predict the next outbreak of sexually transmitted disease.
Sean Young at the University of California, Los Angeles, and his colleagues analysed tweets posted in 2012. They narrowed down millions of posts to 8538 tweets that could be geolocated to US counties and included keywords such as 鈥渟ex鈥� or 鈥渟uck鈥� used in a sexual context. Areas where sexual tweets were posted saw a 2.7 per cent increase in syphilis rates the following year.
One way to predict an increase in syphilis rates is to look at the current number of cases 鈥� for example, each county that had a higher than average number of cases in 2012 saw a 0.6 per cent increase the following year. Tweets corresponded to a bigger change in syphilis rates, so they could provide a more effective way of planning where best to allocate resources.
Advertisement
鈥淭here鈥檚 potentially as much or more information in what people say online as there is in where people live, what education they have, or how much they earn,鈥� says Young.
Even though using social media to predict sexually transmitted disease will never be perfect, if the analysis is carefully carried out it 鈥渃an provide a lot of information,鈥� says Alessandro Vespignani at Northeastern University, Boston, who was one of the authors of a prominent paper on Google Flu Trends, which used a similar method in an attempt to predict flu outbreaks.
Preventive Medicine
This article appears in print under the headline 鈥淭racking syphilis with Twitter鈥�
