Twitter and other real-time social media outlets can be utilised to detect HIV outbreaks as well as drug behaviour, according to a group of researchers from UCLA.
The study, which was published in the Preventive Medicine journal, was conducted via UCLA’s Center for Digital Behavior. It showed that there is a relation between the geographical outbreaks in the US and tweets containing phrases indicative of risky sexually practices and drug-related behaviour.
The group gathered more than 550 million tweets from 26 May to 9 December 2012. They then developed an algorithm that could locate phrases including words such as “get high” and “sex”.
They plotted the tweets on a map and ran statistical models to verify if the locations matched areas with reported HIV cases.
However, the researchers revealed that they used 2009 public data. Public health experts need frequently updated data in order to predict future outbreaks.
Meanwhile, other tools and studies have already proven the power of new media in tracking numerous health outbreaks, such as Google Flu Trends.
However, this is the first study “to suggest that Twitter can be used to predict people’s health-related behaviours and as a method for monitoring HIV risk behaviours and drug use,” said Sean Young, Center for Digital Behavior’s co-director.
The study listed Texas, New York, California and Florida as the states with the largest proportion of HIV-related tweets, while Delaware, Washington, D.C., South Carolina and Louisiana posted the highest per capita rate of HIV risk-related tweets.