Twitter vs. Zika—The role of social media in epidemic outbreaks surveillance

Objective The risk of global epidemic outbreaks led to the development of epidemic tracking services dedicated to track and warn against them. Despite the usefulness of these services, they suffer from shortcomings that impact their efficiency. This study examines the efficacy of social media in epidemic outbreak surveillance by studying Twitter users’ behavior related to the Zika virus outbreak in 2015–2016 and how this behavior can be used to track and predict Zika virus. Methods We collected 67,000 tweets in English and Spanish under the hashtag #Zikavirus and #Zika in the period from October 1st, 2015 to February 25th, 2016. We examined the tweets using text analytics techniques and extracted the important concepts. We analyzed the differences in using these concepts from one month to another. Results There are significant differences between the numbers of tweets during the Zika outbreak as well as between the concepts used in English and Spanish tweets. Discussion The differences in Zika epidemic related tweets evolved with the epidemic outbreak and reflected the different stages of the epidemic. However, those differences also reflected a digital divide between developed and developing communities. The number of tweets was related to the threatened community rather than the severity of the threat. Conclusion While Twitter can be used to augment current epidemic tracking systems, it cannot replace them. We identified digital divide and threat of misleading information as two factors that limit the dependence on Twitter as an epidemic tracking system.