The Effects of Tobacco Coverage in the Public Communication Environment on Young People's Decisions to Smoke Combustible Cigarettes

Accuracy of a large language model in distinguishing anti- and pro-vaccination messages on social media: The case of human papillomavirus vaccination

Objective

Vaccination has engendered a spectrum of public opinions, with social media acting as a crucial platform for health-related discussions. The emergence of artificial intelligence technologies, such as large language models (LLMs), offers a novel opportunity to efficiently investigate public discourses. This research assesses the accuracy of ChatGPT, a widely used and freely available service built upon an LLM, for sentiment analysis to discern different stances toward Human Papillomavirus (HPV) vaccination.

Methods

Messages related to HPV vaccination were collected from social media supporting different message formats: Facebook (long format) and Twitter (short format). A selection of 1,000 human-evaluated messages was input into the LLM, which generated multiple response instances containing its classification results. Accuracy was measured for each message as the level of concurrence between human and machine decisions, ranging between 0 and 1.

Results

Average accuracy was notably high when 20 response instances were used to determine the machine decision of each message: .882 (SE = .021) and .750 (SE = .029) for anti- and pro-vaccination long-form; .773 (SE = .027) and .723 (SE = .029) for anti- and pro-vaccination short-form, respectively. Using only three or even one instance did not lead to a severe decrease in accuracy. However, for long-form messages, the language model exhibited significantly lower accuracy in categorizing pro-vaccination messages than anti-vaccination ones.

Conclusions

ChatGPT shows potential in analyzing public opinions on HPV vaccination using social media content. However, understanding the characteristics and limitations of a language model within specific public health contexts remains imperative.

Using Exogenous Social Media Exposure Measures to Assess the Effects of Smokeless Tobacco-Related Social Media Content on Smokeless Tobacco Sales in the United States

INTRODUCTION

Prior research on the effects of social media promotion of tobacco products has predominantly relied on survey-based self-report measures of marketing exposure, which potentially introduce endogeneity, recall, and selection biases. New approaches can enhance measurement and help better understand the effects of exposure to tobacco-related messages in a dynamic social media marketing environment. We used geolocation-specific tweet rate as an exogenous indicator of exposure to smokeless tobacco (ST)-related content and employed this measure to examine the influence of social media marketing on ST sales.

AIMS AND METHODS

Autoregressive error models were used to analyze the association between the ST-relevant tweet rate (aggregated by 4-week period from February 12, 2017 to June 26, 2021 and scaled by population density) and logarithmic ST unit sales across time by product type (newer, snus, conventional) in the United States, accounting for autocorrelated errors. Interrupted time series approach was used to control for policy change effects.

RESULTS

ST product category-related tweet rates were associated with ST unit sales of newer and conventional products, controlling for price, relevant policy events, and the coronavirus disease 2019 (COVID-19) pandemic. On average, 100-unit increase in the number of newer ST-related tweets was associated with 14% increase in unit sales (RR = 1.14; p = .01); 100-unit increase in conventional ST tweets was associated with ~1% increase in unit sales (p = .04). Average price was negatively associated with the unit sales.

CONCLUSIONS

Study findings reveal that ST social media tweet rate was related to increased ST consumption and illustrate the utility of exogenous measures in conceptualizing and assessing effects in the complex media environment.

IMPLICATIONS

Tobacco control initiatives should include efforts to monitor the role of social media in promoting tobacco use. Surveillance of social media platforms is critical to monitor emerging tobacco product-related marketing strategies and promotional content reach. Exogenous measures of potential exposure to social media messages can supplement survey data to study media effects on tobacco consumption.

Effects of a nicotine warning label and vaping cessation resources on young adults' perceptions of pro-vaping instagram influencer posts

BACKGROUND

Exposure to social media content promoting e-cigarette use ("vaping") is associated with subsequent tobacco use among young adults. Adding features to pro-vaping Instagram influencer posts, such as a nicotine warning label and vaping cessation resources, could help counteract posts' negative influence.

METHODS

Young adults (N = 2,179; Mage = 22.6 [SD = 0.4]; 53.0 % cisgender women, 45.1 % Hispanic) completed an online experiment in 2021-2022 through an ongoing prospective cohort study. Participants viewed three simulated pro-vaping Instagram influencer posts in a four-group, between-subjects design. Post features differed by experimental condition: "label-only" (nicotine warning label on post), "link-only" (link to vaping cessation resources under post), "L&L" (label and link), or "control" (neither). Participants rated each influencer's traits (honest, trustworthy, informed, smart, attractive, popular; 0-100 %). After viewing all three posts, participants reported use intentions, susceptibility, positive and negative expectancies, and harm perceptions around the fictitious advertised vaping product. Past-month vapers additionally reported their desire and self-efficacy for quitting.

RESULTS

L&L (versus control and link-only) participants viewed influencers as more honest, trustworthy, and informed. L&L (versus control) participants had lower odds of susceptibility to using the advertised product, lower positive expectancies, and greater negative expectancies. The label and link did not significantly affect participants' intentions to use the product, perceived harm of the product, or desire or self-efficacy for quitting vaping.

CONCLUSIONS

Providing a nicotine warning label and link to vaping cessation resources on influencers' Instagram posts may have the unintended effect of increasing positive perceptions of the influencer. However, they may reduce susceptibility to product use.

Scanned information exposure and support for tobacco regulations among US youth and young adult tobacco product users and non-users

The influences of information exposure on youth and young adults' (YYA) support for smoking/vaping regulations have been understudied. This study examines (i) the relationships between routine exposure to (i.e. scanning) anti-smoking/pro-vaping information and YYA support for anti-smoking/vaping regulations and (ii) whether these relationships differ across YYA users and non-users of tobacco products. We analyzed the data from a nationally representative two-wave rolling cross-sectional survey of YYA in the United States, collected from 2014 to 2017 (baseline n = 10 642; follow-up n = 4001). Less than 5% of the participants ever scanned pro-smoking and anti-vaping information. Scanning anti-smoking information had significant positive relationships with support for all anti-smoking policies cross-sectionally, and this pattern was longitudinally significant in two anti-smoking policy contexts. Scanning pro-vaping information had significant negative associations with support for anti-vaping policies cross-sectionally, but not longitudinally. The lagged positive relationships between scanning anti-smoking information and support for anti-smoking regulations were stronger among YYA smokers than among YYA non-smokers, whereas evidence from adult data suggested the opposite pattern. The findings suggest that scanning information can affect YYA support for tobacco regulations. Future efforts are required to investigate mechanisms underlying the influences of scanned information on YYA support for tobacco regulations.

An Analysis of Twitter Posts About the U.S. Food and Drug Administration's Menthol Ban

INTRODUCTION

Although the U.S. Food and Drug Administration (FDA) banned characterizing flavors in cigarettes in 2009, this initial ban exempted menthol. After examining numerous reports on the adverse health effects of menthol cigarettes, the FDA proposed a menthol ban in April 2022. This study analyzed Twitter data to describe public reaction to this announcement.

AIMS AND METHODS

Posts containing the word "menthol" and/or "#menthol" were collected from April 21, 2022 to May 5, 2022 from Twitter's Streaming Application Programming Interface (API). A random sampling procedure supplied 1041 tweets for analysis. Following an inductive approach to content analysis, posts were classified into one or more of 11 themes.

RESULTS

Posts discussed the FDA announcement (n = 153, 14.7%), racial discrimination (n = 101, 9.7%), distrust in government (n = 67, 6.4%), inconsistencies between policies (n = 52, 5.0%), public health benefits (n = 42, 4%), freedom of choice (n = 22, 2.1%), and health equity (n = 21, 2.0%). Posts contained misinformation (n = 20, 1.9%), and discussed the potential for illicit markets (n = 18, 1.7%) and the need for cessation support (n = 4, 0.4%). 541 (52.0%) tweets did not fit into any of the prescribed themes.

CONCLUSIONS

Twitter posts with the word "menthol" commonly discussed distrust in government and mentioned racial discrimination. Findings demonstrated the possibility of near real-time Twitter monitoring of public opinion on a menthol ban. These data may be valuable for designing tobacco control health communication campaigns in the future.

IMPLICATIONS

The U.S. FDA proposed a ban on menthol cigarettes in April 2022. This study's content analyzed Twitter posts over a 2-week period to understand the public's response to the proposed menthol ban. Twitter posts with the word "menthol" often discussed distrust in government and mentioned racial discrimination. Findings demonstrated the possibility of near real-time Twitter monitoring of public opinion of regulatory action. Findings underscore the need to educate the public about the potential health benefits of banning menthol from cigarettes, particularly for populations that experience tobacco-related health disparities.

The impact of COVID-19 misinformation and trust in institutions on preventive behaviors

Misinformation related to coronavirus disease 2019 (COVID-19) has the potential to suppress preventive behaviors that mitigate the spread of COVID-19. Early research on the behavioral consequences of COVID-19 misinformation is mixed, and most rely on cross-sectional data. We examined whether believing in COVID-19 misinformation at one time point influences engaging in preventive behaviors later. In addition, we investigated the role of trust in institutions. We conducted a two-wave survey in South Korea and examined the association between belief in COVID-19 misinformation at Wave 1 and preventive behaviors at Wave 2 controlling for preventive behaviors at Wave 1. We also analyzed whether there is an interaction between belief in COVID-19 misinformation and trust in institutions. Belief in COVID-19 misinformation at Wave 1 significantly increased avoidance of preventive behaviors at Wave 2, but after accounting for trust in institutions, this effect disappeared. Rather, trust in institutions significantly decreased avoidance of preventive behaviors. In addition, misinformation increased avoidance of preventive behaviors among those who trusted institutions the most. Results suggest that building trust in institutions is essential in promoting COVID-19 preventive behaviors. Belief in COVID-19 misinformation may have harmful effects, but these effects were pronounced for those who highly trust institutions.

Search Term Identification Methods for Computational Health Communication: Word Embedding and Network Approach for Health Content on YouTube

Background

Common methods for extracting content in health communication research typically involve using a set of well-established queries, often names of medical procedures or diseases, that are often technical or rarely used in the public discussion of health topics. Although these methods produce high recall (ie, retrieve highly relevant content), they tend to overlook health messages that feature colloquial language and layperson vocabularies on social media. Given how such messages could contain misinformation or obscure content that circumvents official medical concepts, correctly identifying (and analyzing) them is crucial to the study of user-generated health content on social media platforms.

Objective

Health communication scholars would benefit from a retrieval process that goes beyond the use of standard terminologies as search queries. Motivated by this, this study aims to put forward a search term identification method to improve the retrieval of user-generated health content on social media. We focused on cancer screening tests as a subject and YouTube as a platform case study.

Methods

We retrieved YouTube videos using cancer screening procedures (colonoscopy, fecal occult blood test, mammogram, and pap test) as seed queries. We then trained word embedding models using text features from these videos to identify the nearest neighbor terms that are semantically similar to cancer screening tests in colloquial language. Retrieving more YouTube videos from the top neighbor terms, we coded a sample of 150 random videos from each term for relevance. We then used text mining to examine the new content retrieved from these videos and network analysis to inspect the relations between the newly retrieved videos and videos from the seed queries.

Results

The top terms with semantic similarities to cancer screening tests were identified via word embedding models. Text mining analysis showed that the 5 nearest neighbor terms retrieved content that was novel and contextually diverse, beyond the content retrieved from cancer screening concepts alone. Results from network analysis showed that the newly retrieved videos had at least one total degree of connection (sum of indegree and outdegree) with seed videos according to YouTube relatedness measures.

Conclusions

We demonstrated a retrieval technique to improve recall and minimize precision loss, which can be extended to various health topics on YouTube, a popular video-sharing social media platform. We discussed how health communication scholars can apply the technique to inspect the performance of the retrieval strategy before investing human coding resources and outlined suggestions on how such a technique can be extended to other health contexts.