Abstract
BACKGROUND: The recent outbreak of bubonic plague in Madagascar reminds us of the continuing public health challenges posed by such deadly diseases in various parts of the world years after their eradication. This study examines the role of Twitter in public health disease surveillance with special focus on how Twitter intensity, time, and location issues explain Twitter plague message delay.
METHOD: We retrospectively analyzed the Twitter feeds of the 2014 bubonic plague outbreak in Madagascar. The analyses are based on the plague-related data available in the public domain between November 19th and 27th 2014. The data were compiled in March 2015. We calculated the time differential between the tweets and retweets, and analyzed various characteristics of the Tweets including Twitter intensity of the users.
RESULTS: A total of 6873 Twitter users were included in the study, of which 52% tweeted plague-related information during the morning hours (before mid-day), and 87% of the tweets came from the west of the epicenter of the plague. More importantly, while session of tweet lease and relative location had effect on message lapse, absolute location did not. Additionally, we found no evidence of differential effect of location on message lapse based on relative location i.e. tweets from west or east nor number of following. However, there is evidence that more intense Twitter use appears to have significant effect on message lapse such that as the number of tweets became more intense, time differential between the tweets and retweets increased while higher number of retweets diminished message lapse.
CONCLUSION: This study affirms that Twitter can play an important role in ongoing disease surveillance and the timely dissemination of information during public health emergencies independent of the time and space restrictions. Further ways should be explored to embed social media channels in routine public health practice. Copyright (C) 2017 King Saud Bin Abdulaziz University for Health Sciences. Published by Elsevier Ltd. All rights reserved.