@Article{info:doi/10.2196/53666, author="Zhou, Runtao and Xie, Zidian and Tang, Qihang and Li, Dongmei", title="Social Network Analysis of e-Cigarette--Related Social Media Influencers on Twitter/X: Observational Study", journal="JMIR Form Res", year="2024", month="Apr", day="1", volume="8", pages="e53666", keywords="social network; social media; influencer; electronic cigarettes; e-cigarette; vaping; vape; Twitter; observational study; aerosol; consumer; influencers; social network analysis; antivaping; campaigns", abstract="Background: An e-cigarette uses a battery to heat a liquid that generates an aerosol for consumers to inhale. e-Cigarette use (vaping) has been associated with respiratory disease, cardiovascular disease, and cognitive functions. Recently, vaping has become increasingly popular, especially among youth and young adults. Objective: The aim of this study was to understand the social networks of Twitter (now rebranded as X) influencers related to e-cigarettes through social network analysis. Methods: Through the Twitter streaming application programming interface, we identified 3,617,766 unique Twitter accounts posting e-cigarette--related tweets from May 3, 2021, to June 10, 2022. Among these, we identified 33 e-cigarette influencers. The followers of these influencers were grouped according to whether or not they post about e-cigarettes themselves; specifically, the former group was defined as having posted at least five e-cigarette--related tweets in the past year, whereas the latter group was defined as followers that had not posted any e-cigarette--related tweets in the past 3 years. We randomly sampled 100 user accounts among each group of e-cigarette influencer followers and created corresponding social networks for each e-cigarette influencer. We compared various network measures (eg, clustering coefficient) between the networks of the two follower groups. Results: Major topics from e-cigarette--related tweets posted by the 33 e-cigarette influencers included advocating against vaping policy (48.0{\%}), vaping as a method to quit smoking (28.0{\%}), and vaping product promotion (24.0{\%}). The follower networks of these 33 influencers showed more connections for those who also post about e-cigarettes than for followers who do not post about e-cigarettes, with significantly higher clustering coefficients for the former group (0.398 vs 0.098; P=.005). Further, networks of followers who post about e-cigarettes exhibited substantially more incoming and outgoing connections than those of followers who do not post about e-cigarettes, with significantly higher in-degree (0.273 vs 0.084; P=.02), closeness (0.452 vs 0.137; P=.04), betweenness (0.036 vs 0.008; P=.001), and out-of-degree (0.097 vs 0.014; P=.02) centrality values. The followers who post about e-cigarettes also had a significantly (P<.001) higher number of followers (n=322) than that of followers who do not post about e-cigarettes (n=201). The number of tweets in the networks of followers who post about e-cigarettes was significantly higher than that in the networks of followers who do not post about e-cigarettes (93 vs 43; P<.001). Two major topics discussed in the networks of followers who post about e-cigarettes included promoting e-cigarette products or vaping activity (55.7{\%}) and vaping being a help for smoking cessation and harm reduction (44.3{\%}). Conclusions: Followers of e-cigarette influencers who also post about e-cigarettes have more closely connected networks than those of followers who do not themselves post about e-cigarettes. These findings provide a potentially practical intervention approach for future antivaping campaigns. ", issn="2561-326X", doi="10.2196/53666", url="https://formative.jmir.org/2024/1/e53666", url="https://doi.org/10.2196/53666", url="http://www.ncbi.nlm.nih.gov/pubmed/38557555" }