Examining the influence of vape advocates on tobacco control policy discourse on Twitter

Theoretical Background and research questions/hypothesis: Twitter is an important avenue through which tobacco regulation evolves in the public consciousness. Prior research has identified the spread of mis- and dis-information regarding tobacco products on Twitter and a possibly disproportionate influence of self-identified vape advocates broadly voicing anti-regulatory views. The objective of this research was to examine the most influential users on Twitter contributing to the conversation around tobacco control policy.

Methods: We used a keyword filter (F1 = .91) to identify N=3,159,807 tobacco-policy tweets by n=58,369 users from the full corpus of tobacco-related content between the vaping-associated lung injury outbreak starting in August 2019, continuing through the period before the COVID-19 pandemic (period 1), during the pandemic (period 2), and after pandemic stay-at-home orders were lifted and vaccines were rolled out (period 3). These data continue through July of 2021. The most retweeted users and users with the highest calculated sustained influence (H index) were identified and their profiles were coded to identify vape advocates. Independent sample t-tests were used to assess for statistically significant differences in H index between vape advocates and non-vape advocates amongst top users in each period.

Results: After an initial peak in 2019, both the number of Twitter users and volume of tobacco policy-related tweets declined. Total posts and retweets received were heavily concentrated among the top users at each of the three time periods. The top 100 most retweeted users received 56%, 48%, and 70% of all retweets during periods 1, 2 and 3, respectively. The 100 users most influential users (top H index) received 23%, 38%, and 17% of all retweets during each period respectively. After coding user profiles, vape advocates comprised 44.5% (n=146) of the most influential and top retweeted users across all three time periods. Additionally, vape advocates had significantly higher measures of H index in each period compared to those who were not vape advocates.

Conclusions: E-cigarette advocates were disproportionately represented among the most retweeted and the most influential users revealing the strong influence of anti-tobacco regulation advocates on Twitter. The online environment for tobacco-policy related discourse on Twitter is becoming increasingly concentrated, with a vocal minority of vape advocates producing and disseminating a disproportionate plurality of content.

Implications for research and/or practice: Characterizing the voices driving this conversation demonstrates that tobacco policy-related discussion on Twitter is not representative of public opinion, but rather an interest group that has leveraged the platform to amplify anti-regulatory sentiment. These findings are informative for researchers and practitioners seeking to understand the role of Twitter on public understanding of tobacco control policy.