COVID-19 Vaccines on TikTok: A Big-Data Analysis of Entangled Discourses

The Normalization of Vaping on TikTok Using Computer Vision, Natural Language Processing, and Qualitative Thematic Analysis: Mixed Methods Study

BACKGROUND

Social media posts that portray vaping in positive social contexts shape people's perceptions and serve to normalize vaping. Despite restrictions on depicting or promoting controlled substances, vape-related content is easily accessible on TikTok. There is a need to understand strategies used in promoting vaping on TikTok, especially among susceptible youth audiences.

OBJECTIVE

This study seeks to comprehensively describe direct (ie, explicit promotional efforts) and indirect (ie, subtler strategies) themes promoting vaping on TikTok using a mixture of computational and qualitative thematic analyses of social media posts. In addition, we aim to describe how these themes might play a role in normalizing vaping behavior on TikTok for youth audiences, thereby informing public health communication and regulatory policies regarding vaping endorsements on TikTok.

METHODS

We collected 14,002 unique TikTok posts using 50 vape-related hashtags (eg, #vapetok and #boxmod). Using the k-means unsupervised machine learning algorithm, we identified clusters and then categorized posts qualitatively based on themes. Next, we organized all videos from the posts thematically and extracted the visual features of each theme using 3 machine learning-based model architectures: residual network (ResNet) with 50 layers (ResNet50), Visual Geometry Group model with 16 layers, and vision transformer. We chose the best-performing model, ResNet50, to thoroughly analyze the image clustering output. To assess clustering accuracy, we examined 4.01% (441/10,990) of the samples from each video cluster. Finally, we randomly selected 50 videos (5% of the total videos) from each theme, which were qualitatively coded and compared with the machine-derived classification for validation.

RESULTS

We successfully identified 5 major themes from the TikTok posts. Vape product marketing (1160/10,990, 8.28%) reflected direct marketing, while the other 4 themes reflected indirect marketing: TikTok influencer (3775/14,002, 26.96%), general vape (2741/14,002, 19.58%), vape brands (2042/14,002, 14.58%), and vaping cessation (1272/14,002, 9.08%). The ResNet50 model successfully classified clusters based on image features, achieving an average F1-score of 0.97, the highest among the 3 models. Qualitative content analyses indicated that vaping was depicted as a normal, routine part of daily life, with TikTok influencers subtly incorporating vaping into popular culture (eg, gaming, skateboarding, and tattooing) and social practices (eg, shopping sprees, driving, and grocery shopping).

CONCLUSIONS

The results from both computational and qualitative analyses of text and visual data reveal that vaping is normalized on TikTok. Our identified themes underscore how everyday conversations, promotional content, and the influence of popular figures collectively contribute to depicting vaping as a normal and accepted aspect of daily life on TikTok. Our study provides valuable insights for regulatory policies and public health initiatives aimed at tackling the normalization of vaping on social media platforms.

Surveillance of Gun-Related Conversations on Twitter

Gun violence in the USA is a documented public health crisis. Publicly accessible data from Twitter posts can be used to rapidly capture and describe the public's recent conversations about guns. Because these gun-related conversations change rapidly, it is important to provide regularly updated information on them. Twitter posts containing gun-related terms were obtained from January 1, 2022 to June 30, 2022. To understand topics of gun-related tweets (N = 449,492), topic modeling was performed with Top2Vec. Gun ownership control, concern about gun safety and its impact on children and schools, and the Second Amendment were major areas of the gun-related discourse on Twitter. Several identified topics in this study were a consequence of the study period, including "Discourse on Capitol Riots," and "Wartime and Military Use of Guns," with the latter topic containing conversations about the Russia-Ukraine War. Conversations around the influence of the National Rifle Association (NRA) on gun policies and pro-gun ownership perspectives were also part of the public discourse. The intersection between alcohol, substance use, and gun use was infrequently observed. Findings suggest that gun-related conversations in social media such as Twitter can inform public health researchers.

Anti-stigma narratives and emotional comfort against health crisis: a context analysis of UGC short videos from patients with COVID-19 infections

Patients narratives are being recorded increasingly frequently and spontaneously in short user produced content (UGC) films, which may have an impact on the vlogger's health as well as the public's comprehension of the relevant health concerns. This paper addressed three research questions regarding the population characteristics of UGC video publishers, the narrative theme of the videos, and the emotional orientation of the commenters. This study aimed to deepen our understanding of COVID-19 patients' narrative intentions and emotional needs through the theoretical frameworks of theory of planned behavior (TPB) and negative dominance theory (NDT). We collected 335 videos from 28 COVID-19 patients and 572,052 comments as samples on Douyin platform, the largest short-video website in China. Using Latent Semantic Analysis, we analyzed the descriptive information of the video blogs, the narrative textual information of the videos, and the emotional orientation of the comments. Our findings revealled seven categories of narrative themes, with 52.1% of video comments exhibiting a positive emotional orientation. Within a framework integrating TPB and NDT theories, we analyzed the behavioral intentions of vloggers and viewers during COVID-19 epidemic, and interpreted the persistent posting of videos and the active posting of comments as positive actions that counteracted the multiple effects of negative messages. This study contributes to the understanding of individual narratives in macro-risk communication, both theoretically and empirically, and offers policy recommendations in relevant fields.

COVID-19 and Public Health: Analysis of Opinions in Social Media

The article presents the results of research of public opinion during the third wave of the COVID-19 pandemic in Russia. The study touches on the attitude of citizens to public health, as well as the reaction of social media users to government measures in a crisis situation during a pandemic. Special attention is paid to the phenomenon of infodemic and methods of detecting cases of the spread of false and unverified information about diseases. The article demonstrates the application of an interdisciplinary approach using network analysis of texts and sociological research. A model for detecting social stress in the textual communication of social network users using a specially trained neural network and linguistic analysis methods is presented. The validity and validity of the results of the analysis of social network data were verified using a sociological survey. This approach allows us to identify points of tension in matters of public health promotion, during crisis situations to improve interaction between the government and society, and to timely adjust government plans and actions to ensure resilience in emergency situations for public health purposes.

Digital Media Exposure and Health Beliefs Influencing Influenza Vaccination Intentions: An Empirical Research in China

The purpose of this study was to investigate whether/how digital media exposure influences people's intention to influenza vaccination. Through an anonymous online survey, we collected data on Chinese people's exposure to influenza and influenza vaccine information on digital media platforms and their attitudes toward influenza vaccines (N = 600). The structural equation model analysis results strongly support to the research hypotheses and the proposed model. The findings reveal three major themes: (1) digital media exposure significantly influence the susceptibility and severity of influenza. (2) After exposure to digital media, it is helpful to understand the vaccine's benefits, reduce the barriers to vaccination, and finally improve the intention to vaccination. (3) Users receive cues to action from digital media, and their vaccination intention tends to be positive. These findings explore how digital media exposure influences influenza vaccination intention and may provide insights into vaccine promotion efforts in countries. Research has shown that digital media exposure contributes to getting vaccinated against influenza.