TY  - JOUR
AU  - Mendez, Samuel R
AU  - Munoz-Najar, Sebastian
AU  - Emmons, Karen M
AU  - Viswanath, Kasisomayajula
PY  - 2025
DA  - 2025/1/3
TI  - US State Public Health Agencies' Use of Twitter From 2012 to 2022: Observational Study
JO  - J Med Internet Res
SP  - e59786
VL  - 27
KW  - social media
KW  - health communication
KW  - Twitter
KW  - tweet
KW  - public health
KW  - state government
KW  - government agencies
KW  - information technology
KW  - data science
KW  - communication tool
KW  - COVID-19 pandemic
KW  - data collection
KW  - theoretical framework
KW  - message
KW  - interaction
AB  - Background: Twitter (subsequently rebranded as X) is acknowledged by US health agencies, including the US Centers for Disease Control and Prevention (CDC), as an important public health communication tool. However, there is a lack of data describing its use by state health agencies over time. This knowledge is important amid a changing social media landscape in the wake of the COVID-19 pandemic. Objective: The study aimed to describe US state health agencies’ use of Twitter from 2012 through 2022. Furthermore, we organized our data collection and analysis around the theoretical framework of the networked public to contribute to the broader literature on health communication beyond a single platform. Methods: We used Twitter application programming interface data as indicators of state health agencies’ engagement with the 4 key qualities of communication in a networked public: scalability, persistence, replicability, and searchability. To assess scalability, we calculated tweet volume and audience engagement metrics per tweet. To assess persistence, we calculated the portion of tweets that were manual retweets or included an account mention. To assess replicability, we calculated the portion of tweets that were retweets or quote tweets. To assess searchability, we calculated the portion of tweets using at least 1 hashtag. Results: We observed a COVID-19 pandemic–era shift in state health agency engagement with scalability. The overall volume of tweets increased suddenly from less than 50,000 tweets in 2019 to over 94,000 in 2020, resulting in an average of 5.3 per day. Though mean tweets per day fell in 2021 and 2022, this COVID-19 pandemic–era low was still higher than the pre–COVID-19 pandemic peak. We also observed a more fragmented approach to searchability aligning with the start of the COVID-19 pandemic. More state-specific hashtags were among the top 10 during the COVID-19 pandemic, compared with more general hashtags related to disease outbreaks and natural disasters in years before. We did not observe such a clear COVID-19 pandemic–era shift in engagement with replicability. The portion of tweets mentioning a CDC account gradually rose and fell around a peak of 7.0% in 2018. Similarly, the rate of retweets of a CDC account rose and fell gradually around a peak of 5.4% in 2018. We did not observe a clear COVID-19 pandemic–era shift in persistence. The portion of tweets mentioning any account reached a maximum of 21% in 2013. It oscillated for much of the study period before dropping off in 2021 and reaching a minimum of 10% in 2022. Before 2018, the top 10 mentioned accounts included at least 2 non-CDC or corporate accounts. From 2018 onward, state agencies were much more prominent. Conclusions: Overall, we observed a more fragmented approach to state health agency communication on Twitter during the pandemic, prioritizing volume over searchability, formally replicating existing messages, and leaving traces of interactions with other accounts. 
SN  - 1438-8871
UR  - https://www.jmir.org/2025/1/e59786
UR  - https://doi.org/10.2196/59786
UR  - http://www.ncbi.nlm.nih.gov/pubmed/39752190
DO  - 10.2196/59786
ID  - info:doi/10.2196/59786
ER  -