Temporal analysis of users’ engagements in CDC public health information on Twitter/X

Purpose The objective of this paper is to investigate the temporal patterns, trends and valuable insights regarding user engagements in public health information diffusion from the Centers for Disease Control and Prevention (CDC) channel on Twitter/X. Design/methodology/approach The health information regression models on the 10 CDC channels on Twitter/X were developed by applying regression analysis methods. The user engagement velocities were explored, and the half-life of user engagement and response times was examined. Findings The findings show that logarithmic regression models were the most suitable for predicting user engagement types. The most frequent user engagement types on Twitter/X were views, followed by likes and retweets. Regarding average user engagement velocity, views ranked first, followed by likes, retweets, replies, quotes and bookmarks in that order. The half-life values for various user engagement types ranged from 52.68 to 54.60 days. The study also revealed response times for different user engagement types: bookmarks (1.53 days), likes (1.05 days), quotes (2.22 days), replies (1.21 days), retweets (1.07 days) and views (0.71 days). Social implications The findings serve as invaluable tools for public health practitioners and administrators, facilitating scientific assessments and predictions regarding the impact of CDC public information on the public. Originality/value This study used scientific research methods to accurately describe user engagement changes over time for an authoritative public health agency channel on social media. The analysis on user engagements included views, likes, retweets, replies, quotes and bookmarks. Other important characteristics like velocity and half-life for user engagement types were addressed.