Abstract
Context: The first imported US Ebola hemorrhagic fever case during the 2014 West Africa Ebola outbreak triggered an increase in online activity through various social media platforms, including Twitter.
Objectives: The purpose of our study was to examine characteristics of local health departments (LHDs) tweeting about Ebola, in addition to how and when LHDs were communicating Ebola-related messages.
Design: All tweets sent by 287 LHDs known to be using Twitter were collected from September 3 to November 2, 2014. Twitter data were merged with the 2013 National Association of County & City Health Officials Profile study to assess LHD characteristics associated with sending Ebola-related tweets. To examine the content of Ebola tweets, we reviewed all such tweets and developed a codebook including 4 major message categories: information giving, news update, event promotion, and preparedness. A time line tracking the trends in Ebola tweets was created by aligning daily tweets with major Ebola news events posted on the Centers for Disease Control and Prevention Ebola Web site.
Results: Approximately 60% (n = 174) of all LHDs using Twitter sent a total of 1648 Ebola-related tweets during the study period. Sending more tweets in general (odds ratio: 2.42; 95% confidence interval, 1.00-5.84) and employing at least 1 public information specialist (odds ratio: 2.61; 95% confidence interval, 1.14-5.95) significantly increased the odds that an LHD tweeted about Ebola. Of all the Ebola tweets collected, 78.6% were information giving, 22.5% were on preparedness, 20.8% were news updates, and 10.3% were event promotion tweets. A temporal analysis of Ebola tweets indicated 5 distinct waves, each corresponding with major Ebola news events.
Conclusions: Twitter has become a communication tool frequently used by many LHDs to respond to novel outbreaks, but messaging strategies vary widely across LHDs. We recommend that LHDs increase tweet frequency during public health emergencies in order to ensure timely dissemination of critical information.