Examining the Network of Local Health Departments and Their Information Sources On Twitter

Theoretical Background and research questions/hypothesis: Implementing evidence-based public health programs results in a greater likelihood of success for local health departments (LHDs) and is among the recently-developed LHD accreditation standards. A recent study found that LHD leaders use a variety of information sources when they are making decisions related to programming; one of the sources reported by a small number of LHD leaders was social media.

Methods: To learn more about the information sources that LHDs are following on social media, we collected and examined Twitter users being followed by LHDs. We identified those followed by more than 10% of health departments and grouped them by type of organization. Network data comprised of the LHDs and the commonly followed Twitter users was collected and examined using standard and statistical network methods.

Results: As of mid-2012, 216 LHDs nationwide had adopted Twitter. In January 2013 these LHDs were following 19,494 total unique Twitter users altogether; 116 Twitter users were followed by more than 10% of the 216 LHDs using Twitter. Among these commonly followed information sources were 64 government organizations including local (n=13) and state health departments (n=3) along with national programs (e.g., MyPlate) and divisions of national organizations (e.g., USDAFoodSafety). Twenty-six different media outlets were represented, including newspapers, television, and web-based organizations. LHDs were also following 14 non-profit organizations and programs, including the American Heart Association, American Cancer Association, and American Lung Association. LHDs also followed 10 professional organizations (e.g., NACCHO), including four affiliated with the American Public Health Association. Finally, the list of followers included five educational institutions: two schools of public health and three medical schools. The network comprised of the local health departments and these organizations included 317 nodes connected by 13,874 links with two isolates. The overall network density was .14, indicating that 14% of possible ties existed. Initial statistical network models indicated that, after controlling for network density, LHDs were significantly more likely than chance to follow government (b=.37; p<.001), non-profit (b=.36; p<.001), and professional organizations (b=.54; p<.001), while they were significantly less likely than chance to follow other LHDs (b=-1.94; p<.001), media (b=-.24; p<.001), and educational sources (b=-.15; p=.02).

Conclusions: Sixty-four government sources and 26 media outlets were among the most followed Twitter users for LHDs. However, while LHDs were more likely to follow government organizations (among others), they were less likely to follow media outlets.

Implications for research and/or practice: Future studies should examine the content of the Twitter feeds LHDs follow in order to determine what information LHDs are gaining. In addition, public health journals are a major source of evidence on effective public health programming, however, not a single journal was among the top LHD information sources. In addition, LHDs were less likely than chance to follow educational sources, where researchers might be highlighted by their institutions when new findings are published. Researchers may wish to consider other venues for dissemination such as partnering with professional associations like the National Association of County and City Health Officials (NACCHO) in order to reach practitioners at the local level.