Analyzing Social Media to Adapt Communication Strategy During Public Health Emergencies

Background: The use of social media during emergencies and disasters has increasingly put the public at the forefront of communication. Through platforms like microblogs (e.g., Twitter) and social networking sites (e.g., Facebook), individuals can distribute information about the event to a broad array of people and organizations in minutes or even seconds.  The communication that happens in social media affects the response context, including how event characteristics and responder organizations are portrayed.  The public’s use of social media may not intend to help or hinder communication during emergencies but what is shared can have an enormous impact. Social media exposes participants to the concerns of others who may be bearing public health risks. Capturing social media conversations during emergencies provides critical insights into what people in actual or perceived risk are thinking, feeling, doing, and sharing about the event.  Social media can help identify affected populations, find rumors or misinformation, and engage influential potential allies.  In addition, by tracking specific concerns through social media – which moves much more quickly than traditional news media – agency decision-makers have an opportunity to address relevant issues while they are still evolving. 

Program background: CDC's Emergency Communication System (ECS) is dedicated to ensuring that timely, accurate, credible, targeted and actionable information, messages and guidance reach the public and stakeholders during public health emergencies. This presentation will describe how CDC’s Emergency Communication System (ECS) has engaged in systematic social media monitoring during recent public health emergencies, including the 2011 Japan earthquake and the global polio eradication initiative.

Evaluation Methods and Results: We will show how our team conducted daily, real-time surveillance of social media using a variety of media monitoring tools to analyze the communication environment and identify key themes, information gaps, miscommunication, rumors, and emerging issues.  We will discuss how this information was reported to key agency decision-makers for communication strategy development and implementation of valuable course-corrections that improved the effectiveness of information sharing during a public health response. Finally, we discuss the unique challenges encountered in the use of current social media monitoring tools for communication surveillance, data collection, and analysis for public health emergency response. These challenges include: 1) the accelerated pace of new social media technologies and their use by the public, 2) the need to quickly analyze and report findings from a large volume of data with limited time and resources, 3) the potential amplification of less-relevant secondary and tertiary issues, and 4) the limitations of social media tools to provide the information, facilitate analysis, and be flexible to the needs of a public health emergency response.

Conclusions: Knowing what is in the communication “environment” of social media enables public health agencies to address public questions and concern and respond more rapidly and effectively to ongoing disasters.

Implications for research and/or practice: It is important to have a systematic process of evaluating tools and methods for analyzing large amounts of social media data to inform communication strategy during public health emergencies.