Abstract
Social media platforms are home to large volumes of ambiguous hashtag-based claims about the health, modified-risk, and cessation benefits of electronic cigarette products (e.g., #Vapingsavedmylife). The objective of this study was to qualitatively explore how young adults interpret these hashtags on the popular platform Instagram. Specifically, we sought to identify if they view these hashtags as making health-related claims, and if they find these claims to be credible and valid. We conducted 12 focus groups in 2018 with non–tobacco users, smokers, dual users, and vapers between the ages of 18 and 24 (n = 69). Using real Instagram posts to guide discussion, participants reflected on the meaning of potentially claims-making hashtags. Participants interpreted the majority of the hashtags as making health-related claims. However, many participants felt that the claims were too exaggerated to be entirely valid. Some participants, including dual users and vapers, argued that smoking and vaping were largely equivalent. Smokers were particularly skeptical of claims. Findings suggest that the U. S. Food and Drug Administration should consider hashtag-based claims in their regulatory efforts. However, further research is needed on how to pragmatically address claims taking the form of hashtags given legal and practical constraints.
Marketing for electronic cigarettes (e-cigarettes) is now widespread on social media, forcing tobacco control researchers and advocates to contend with new avenues through which youth and young adults may be exposed to content that promotes nicotine use. For example, many e-liquids used to refill e-cigarettes use cartoons in their promotions on visual social media platforms (Allem et al., 2018). E-cigarettes are also frequently promoted by individual social media users who are sponsored by brands and retailers, allowing posts to reach new audiences and blurring the lines between commercial and organic content with regard to First Amendment protections (Laestadius, Wahl, et al., 2019). There have also been new developments in the use of unauthorized health, modified-risk, and cessation claims as more marketing moves onto social media.
At the time of writing, there are no U.S. Food and Drug Administration (FDA) authorized claims that can be made about modified risk or smoking cessation with regard to e-cigarettes. Despite this, explicit modified risk and cessation claims were a common feature of early marketing for e-cigarette products (Grana & Ling 2014; Klein et al. 2016). These types of claims appear to be at least partially effective at increasing interest in trying e-cigarettes (El-Toukhy et al., 2018; Pepper et al., 2014; Yang et al., 2019). Claims are problematic in the context of nonsmoking populations or smokers who become dual users but are potentially useful with regard to educating smokers who are otherwise unwilling to quit smoking (Kozlowski & Sweanor, 2018).
However, on social media, and on Instagram in particular, explicit claims appear to be relatively rare (Chu et al., 2016; Laestadius, Wahl, et al., 2019; Lee et al., 2017; Majmundar et al., 2019). Instagram has large volumes of e-cigarette and e-liquid content (Chu et al., 2016; Czaplicki et al., 2019; Laestadius, Wahl, et al., 2019; Majmundar et al., 2019), but most claims take the form of ambiguous hashtags, such as #Notblowingsmoke, #Quitsmoking, and #Vapeordie (Laestadius, Wahl, et al., 2019; Chu et al., 2016; Laestadius et al., 2016). These practices raise questions about how potentially claims-making hashtags are interpreted. If they are viewed as claims-making by the public despite their nontraditional format, they should also be considered claims for regulatory purposes. The presence of these hashtags on Instagram is of particular concern since Instagram is home to a large volume of adolescent and young adult social media users. For example, Instagram is used by 72% of U.S. teens and 75% of 18- to 24-year-olds (Anderson & Jiang, 2018; Perrin & Anderson, 2019).
Prior research suggests that adults and adolescents are at least partially able to infer claims from more implicit communications about risk and cessation (El-Toukhy et al., 2018; Kim et al., 2019). While hashtags may seem unlikely to ever take on the role of an officially approved modified-risk or cessation statement, they reflect the current reality of how e-cigarettes and e-liquids are being promoted on social media (Laestadius, Wahl, et al., 2019). The volume of the content in question makes hashtags a priority for tobacco regulatory science. For example, #Vapingsavedmylife has been used on over 700,000 posts as of November 2019. Hashtags are also unique from traditional written claims in other mediums in that they have connectivity to other posts made with the same hashtag, allowing people to use them to both make a post easier to find and to signal membership in a community (Highfield & Leaver, 2014). These promotional strategies are critical as greater exposure to e-cigarette content—e-cigarette-related posts by lay people or e-cigarette marketing by manufacturers or retailers—on social media is associated with current e-cigarette use (Pokhrel et al., 2018). An understanding of these hashtags is essential for an understanding of, and regulatory response to, tobacco product promotions in the social media era.
To date, there has been no consideration of the role of hashtags in claims-making around tobacco products of any type. Our study sought to address two key questions through qualitative focus groups: (1) Do young adult Instagram users view hashtags inferring health, modified-risk, and cessation benefits about e-liquids as actual health-related claims? and (2) Do they feel that these types of claims are valid and credible? A better understanding of how these hashtags are interpreted will help shed on light on how they should be handled by both regulatory agencies and platforms themselves, as well as contribute to broader public health understanding of the role of hashtags in health communication.
Method
We recruited participants in the Milwaukee, Wisconsin, metropolitan area, which has a population of over 1.5 million residents (Metropolitan Milwaukee Association of Commerce, 2019), through social media ads and flyers at local vape and tobacco shops, coffee shops, libraries, and one private and one public university. Participants were screened to ensure they were active Instagram users (viewing their Instagram account at least once per week), between the ages of 18 and 24, and English speakers. The institutional review board at the University of Wisconsin-Milwaukee approved the research protocol.
Between September and December 2018, we conducted three focus groups with vapers (n = 20), three with dual users (n = 18), three with smokers (n = 12), and three with non–tobacco users (n = 19). Prior research suggests that three focus groups are sufficient to reach saturation of themes in relatively homogenous groups (Guest et al., 2017). Participants were provided with a meal and a $50 gift card. Each focus group lasted about 90 minutes and had four to eight participants. After providing informed consent, participants completed a short questionnaire on demographics, tobacco use, and perception of promotional Instagram posts. Focus group discussion was organized around actual Instagram posts that were chosen to represent common themes and hashtags identified in prior content analysis of Instagram posts promoting e-liquids (Laestadius, Wahl, et al., 2019). Specifically, we sought to capture posts with possible cessation, harm reduction, and smokefree claims. All posts were from Instagram accounts owned by brands, stores, or individual Instagram users with explicit commercial ties (i.e., sponsored by brands/stores or owners of brands/stores).
In each focus group, a trained moderator led discussion around five of eight possible Instagram posts. Posts were rotated in order to generate perceptions of a broader range of content. Written descriptions of posts have been published in prior research that considered the appeal of Instagram marketing for e-liquid (Laestadius, Penndorf, et al., 2019). In addition to displaying posts on a projector, participants were provided with printed copies of each post and account user information. Discussion around hashtags for each post started with open ended discussion around “What do you think the hashtags are saying about this e-liquid?” and led into probes about how they interpreted and viewed the credibility of specific hashtag claims (e.g., “What does the ‘vaping saved my life’ hashtag mean to you?”). The hashtags included in posts and chosen for probing were #Vapingsavedmylife, #Quitsmoking (used adjacent to #Startvaping in the post caption), #Stopsmoking, #Vapeordie, and #Notblowingsmoke. The final hashtag is of note because it originated with a provaping campaign designed to counter a California tobacco control effort called Still Blowing Smoke (Allem et al., 2017). Accordingly, this hashtag may be interpreted differently depending on prior awareness of this effort. As hashtags are never viewed in isolation on Instagram, hashtags were displayed in conjunction with the posts they were used on. While this may complicate the ability to isolate the inherent meaning of hashtags, it is also critical to note that hashtags are part of a holistic unit (image, caption, hashtags, comments, and likes) rather than free-standing elements that are evaluated independently (Laestadius, 2017). See Table 1 for Instagram post volume associated with each hashtag. As part of a larger focus group study, discussion also covered broader questions about the appeal of posts. All focus groups were audio-recorded.
Hashtags and Associated Post Volume as of November 21, 2019.
Following professional verbatim transcription, we used a framework analysis approach to analyze data (Gale et al., 2013). To allow for discussion and codebook creation, the first transcript from each smoking status was independently coded line by line by the lead, second, and third authors. The codebook included codes capturing the perceived credibility of hashtag claims (e.g., valid, invalid, unclear/unspecified, and not seen as a claim) and interpretations of the benefits and risks of e-liquid products and vaping (e.g., harm reduction, cessation, smokeless tobacco, addiction, no benefits over smoking, etc.). The first author and second author then applied the codebook to all remaining transcripts using MAXQDA 2018, with each transcript analyzed by both coders with reference to frequency, specificity, emotion, and extensiveness of comments (Krueger & Casey, 2015). Any coding disagreements were discussed and resolved, allowing for the codebook and coding process to continue to be refined (Barbour, 2001). Coded transcript segments were then charted by smoking status and claims related code (e.g., invalid claim) into a matrix in Microsoft Excel and memos were written to facilitate further analysis and reflection on data across smoking statuses. Since not all focus group participants responded to each question and some participants responded at greater length than others, we present findings using qualitative descriptors and quotes rather than precise numerical counts that may be misinterpreted (Krueger & Casey, 2015).
Results
Of the 69 participants who participated across 12 focus groups, 58.0% were male, 62.3% were White and 15.9% were African American, and 69.1% had completed some college or technical training. The average age of participants was 21, with vapers on average 3 years younger than smokers. See Table 2 for further participant demographics. Overall, participants were fairly positive toward Instagram promotions, with 40.6% indicating that they trust promotional posts on Instagram and 58% indicating that they pay attention to them. Prior exposure to e-cigarette and e-liquid promotional content on Instagram was relatively high, with 58.8% of participants indicating that they had seen such posts in the past.
Participant Characteristics by Smoking Status.
Hashtags Can Convey Multiple Meanings at Once
In several instances, participants expressed the idea that hashtags could contain multiple meanings simultaneously. For example, #Notblowingsmoke, which serves as a popular idiom for not misleading someone, was interpreted both metaphorically and as a literal health claim. None of the participants indicated that they were aware that the hashtag was originally developed as part of an advocacy campaign rather than as a claim. One smoker explained, “I think it’s just a play on words like you’re not lying, you’re not smoking. It’s just kind of supposed to be like a cheeky way of saying like ‘start vaping’ I guess” (male, age 24). Although participants were mixed as to how they felt about the accuracy of the claim, it was broadly seen as making implicit claims that e-cigarettes are a form of smokeless tobacco, with a lower risk profile than cigarettes. Another current smoker described it as follows, “I understand why the not blowing smoke [hashtag] because you’re not blowing smoke, it’s water vapor that you’re really blowing out so- I guess that’s more the healthier aspect that they’re trying to throw out” (female, age 24).
By contrast, #Vapeordie was seen as distinct from the other hashtags in its explicit reference to a well-known phrasing used in popular culture to signal a strong commitment to something. As a result, participants of all smoking statuses primarily viewed this hashtag as more of a reference to the vaping lifestyle and subculture than a suggestion that vaping was less harmful than smoking. One vaper explained, “To me that’s like ‘ride or die’; like you’re either in or you’re out. You know like ‘you do this hardcore or you’re not one of us’” (female, age 22).
Perceived Intent and Impact Are Not Always Aligned
#Quitsmoking, #Stopsmoking, and #Vapingsavedmylife were frequently viewed as attempting to convey positive health-related claims about risk and cessation. One vaper disagreed with the accuracy of the claims but still noted that “they’re definitely trying to say it’s better [than smoking]” (male, age 21). A non–tobacco user similarly noted, “I would say it’s trying to promote [vaping] as an alternative, saying it’s better, it’s better for you, it’s healthier in a way” (male, age 18) With regard to cessation, one smoker explained that he “could see the hashtag carrying some meaning for certain individuals that probably take in less nicotine now or don’t smoke a pack a day anymore” (male, age 24).
However, some participants felt torn on if the intent of the hashtags was truly to convey claims, even if they personally read them as claims. This was largely due to their understanding of hashtag conventions on Instagram, where hashtags are often used more for searchability and expressions of community affiliation than for conveying meaning. A dual user noted, “People literally just copy and paste [hashtags] just to try to get views, it doesn’t necessarily reflect what they like” (female, age 23). A non–tobacco user similarly explained, Well I think “not blowing smoke” is a vape-specific thing, it’s like “oh we’re not blowing smoke, we’re blowing vapor,” and it just kind of caught on as a really badly done pun. . . That’s probably how that tag started, I’m guessing that this guy is using it just to get into a community, like people would post with the “not blowing smoke” and then they would post like pictures of themselves vaping, so, he’s probably posting in that tag to try to get people to notice him, same with all the other tags. (Male, age 19)
Skepticism of Claims Is High, Particularly for Smokers
While the hashtags were most often seen as making health-related claims, there was significantly less support for the idea that these were fully convincing claims. Credibility was diminished in instances when participants reflected on the fact that the Instagram users were making the claims for promotional purposes. One vaper explained, “[#Vapingsavedmylife] could just be being used to grab smokers and just get them switched over because they’re looking at profits” (male, age 20). With regard to validity, most participants struggled to reconcile some level of belief in the benefits of vaping with a sense that many claims were “overblown,” “very dramatic,” or “hyperbole.” For example, a dual user expressed her concerns about #Vapingsavedmylife and #Quitsmoking: It’s really cringey, and I understand that vaping could help, could possible help you quit smoking because that’s how I started, and I think it’s a good thing. But overselling it is just not a good marketing strategy, you know, it’s just too much and too in your face. (Female, age 24)
Overall, vapers and some of the dual users expressed more support for the validity of the hashtag claims than smokers and non–tobacco users. This was largely grounded in perceptions of vapor being healthier than cigarette smoke and positive personal and familial experiences with using e-cigarettes, including the benefits of e-cigarettes for oral fixation. Some non–tobacco users also felt that cessation related hashtags truthfully indicated that the products would be helpful for smokers.
However, there was also a strong sense of equivalency between smoking and vaping among many participants, which undercut the validity of the hashtag claims. One non–tobacco user exclaimed, “[#Quitsmoking #Startvaping] doesn’t make sense- because it’s just as bad! (male, age 23).” While many non–tobacco users also doubted #Vapingsavedmylife and #Notblowingsmoke because they felt e-cigarettes still produced harmful smoke, smokers and dual users were particularly defensive on this point. One smoker noted, “If you’re vaping you’re still smoking” (male, age 20), and a dual explained, “My thing is why does it say ‘stop smoking’? ‘Hashtag stop smoking’? Like are you saying that vaping is like better than smoking? Like it doesn’t make it any better than smoking” (female, age 22). Dual user groups appeared to have strongly mixed views on the relative benefits of vaping to smoking.
Equivalency concerns also undercut #Stopsmoking and #Quitsmoking, with many arguing that addiction to nicotine is undesirable regardless of delivery mechanism. Smokers also raised concerns about e-cigarettes posing unknown risks and potentially being more dangerous and more addictive than smoking. Accordingly, switching to e-cigarettes was seen as potentially risky, not true cessation, and in some ways pointless, expressed by comments like “I feel like this is just a crutch, it’s not really a fix” (male, age 22). Vapers themselves also frequently recognized that vaping was still problematic. One vaper elaborated, Like it says “hashtag quit smoking,” “hashtag start vaping,” if you’re going to quit smoking to get rid of the nicotine addiction but you start vaping, that seems counterintuitive to me. Not that I’m one to talk, because I’m you know, a vaper but . . . It’s just trading one addiction for another, it’s not actually helping you so, I don’t think it should be advertised towards trying to help anybody because I personally don’t think that vaping is any better than smoking, even though I do it. (Female, age 22)
Claim Credibility Depends on Product Traits
Finally, the validity of cessation claim hashtags was partially shaped by the traits of the e-liquid featured in the Instagram post. For nicotine users, 0 mg nicotine e-liquids effectively undercut the validity of the cessation claim hashtags because many felt that complete withdrawal from nicotine would prevent a successful switch away from cigarettes. One dual user felt that the claim made no sense in this context and noted, “If they are trying to advertise as a quitting thing, why do they have the zero-milligram nicotine thing? I mean what the shit? That doesn’t—like that’s not going to help anyone . . .” (male, age 23). By contrast, the presence of a 0 mg nicotine e-liquid in the post instead amplified the validity of cessation hashtags for non–tobacco users, who envisioned smokers wanting to switch to nicotine-free products.
Discussion
Our study qualitatively examined perceptions of health-related vaping hashtags among young adults, finding that ambiguous claims-making hashtags are broadly interpreted as health-related claims. Even participants who questioned the intent of the hashtags still felt that they conveyed health-related claims. This suggests that the FDA should consider hashtag-based claims in their regulatory efforts. Based on the e-cigarette deeming rule issued by the FDA in 2016, it would seem that this type of claims-making would be covered (FDA, 2016). Furthermore, when considering whether or not a claim is made, the FDA may take into account “any claim or statement made by or on behalf of a manufacturer that explicitly or implicitly promotes a product for a particular use” (FDA, 2017). When establishing a product’s intended use, the FDA is not bound by the manufacturer or distributor’s subjective claims of intent but rather “can consider objective evidence, which may include a variety of direct and circumstantial evidence” (FDA, 2017).
However, any effort to regulate hashtags is complicated by several factors. First, not all of the hashtags were seen as claims-making by the participants. For example, #Vapeordie was broadly seen as a statement about subculture membership. Accordingly, a blanket ban on all hashtags that could be seen as claims-making would overstep regulatory authority. Government pressure on platforms to ban content would also face First Amendment challenges from private citizens expressing personal views. That said, platforms themselves appear willing to engage despite potential public backlash. For example, current Instagram searches for #vaccine compel viewers to read information suggesting that they visit the Centers for Disease Control and Prevention website before they view search results (Instagram, 2019). Instagram has also banned searches for more explicitly misleading claims such as #VaccinesCauseAIDS (Yurieff, 2019).
This suggests that platforms have the capacity to restrict content. More broadly, however, the issue of claims around vaping must grapple with the discrepancies between the U.S. regulatory approach to vaping and that of the United Kingdom. Unlike the FDA, the National Health Service holds that e-cigarettes offer cessation and harm reduction benefits (NHS Smokefree, 2019). As a platform with an international presence, Instagram would need to decide if something qualifies as misinformation in a more global context on the basis of it being unauthorized by the U.S. government. Similar to Instagram’s practice for #vaccine, the provision of links with information from different health agencies when someone performs a search for vaping content may serve as an appropriate middle ground approach.
Second, any list of claims-making hashtags would need to be continuously updated, as online slang and platform vernaculars shift over time. Hashtags are part of a folksonomy (in contrast to a taxonomy), in which hashtags are ever emerging and evolving through social and collective processes (Highfield & Leaver, 2014). In short, the hashtags prevalent on e-liquid posts in 2019 may be quite different from the ones in the years to come. For example, at the time of writing, backlash against an American Heart Association campaign led to the use of #quitlying as a hashtag for conveying the benefits of vaping relative to smoking (Bentley, 2019). As with #Notblowingsmoke, this example indicates the dual meanings that can be present in hashtags since #quitlying was initially developed by the American Heart Association as a part of its own antivaping campaign. To understand how these hashtags emerge and evolve over time, it is critical for future research to also engage with the social media users who choose to apply these hashtags to their posts. This will shed light on how hashtags are chosen, as well as when and why they shift.
Finally, the targeting of specific hashtags by platforms or regulators speeds up the emergence of new hashtags designed to avoid detection. Efforts by Instagram to ban self-injury hashtags merely resulted in multiple new variants of the hashtags emerging (Moreno et al., 2016). Further research is needed on how to approach this issue given regulatory and practical constraints. In light of the difficulty of fully addressing this issue through regulatory approaches, it is also important to consider ways to ensure that health-promoting information flows parallel to pro–substance use information on social media (Laestadius, Guidry, et al., 2019). One approach may be to build on recent work on strategies for the identification of social media users who hold ambiguous views about tobacco products, enabling them to receive more targeted health information (Chu et al., 2019).
With regard to perceptions on the validity and credibility of claims, most participants found elements of truth in the hashtags but felt that they overstated the benefits of vaping. Higher levels of skepticism and defensiveness among smokers have been documented in prior studies, with some suggestion that this may be a means of justifying continued combustible tobacco use (Ma et al., 2019; Owusu et al., 2019). Prior research also suggests that smokers may view e-cigarettes as a symbol of their addiction to nicotine and the smoking-related stigma they are seeking to avoid (Tokle & Pedersen, 2019). However, views grounded in the equivalency of vaping and smoking were not limited to smokers. This is in line with recent surveys indicating that over half of U.S. adults now perceive e-cigarettes to be at least as harmful as cigarettes, potentially attributable to the media failing to fully clarify the nuances between absolute and relative harm (Huang, 2019). Given growing evidence that e-cigarettes may have smoking cessation uses (Hajek et al., 2019; Kalkhoran et al., 2020), the ways in which smokers view claims about e-cigarette products warrants further attention.
Our study benefited from the inclusion of individuals across four different smoking statuses and has strong ecological validity from the use of real Instagram posts. However, our study is not without limitations. Qualitative sampling approaches were used, and given the focus group format, not all participants engaged with each question equally. Accordingly, we presented findings purely qualitatively so as to not give the impression of broader prevalence rates. Second, despite best efforts, our sample size differed across smoking status groups. Because of the imbalance, there is some chance that smokers’ perspectives were not as well represented as the perspectives of others. In addition, the use of real posts limited the number of distinct hashtags that could be considered, and other post elements may have helped shape hashtag interpretations. Finally, transferability of qualitative results to other settings should also be considered in light of the fact that this study was conducted with young adults in a Midwestern state.
As new forms of health-related claims about tobacco emerge on social media, there is a strong need to revisit our understanding of what counts as a claim. This study offers the first exploration of hashtag-based claims for tobacco products, finding both that young adults do view many ambiguous claims-making hashtags as actual claims and that the interpretation of those claims is largely consistent with recent studies considering more traditional claim-formats. As claims continue to evolve, further research will be needed to follow their usage and the ways in which they influence different populations. Experimental studies considering how different smoking status groups respond to claims would be particularly valuable for products that may hold harm reduction benefits.
Footnotes
Acknowledgements
The authors would like to thank Alana Schneider, Alex Fairburn, Holly Milaeger, and Cameron Macdonald for their contributions to recruitment, planning, and facilitation.
Authors’ Note
The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health (NIH) or the U.S. Food and Drug Administration (FDA). The NIH and FDA had no role in the study design; collection, analysis, or interpretation of the data; the writing of the manuscript; or the decision to submit the paper for publication.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the National Cancer Institute at the National Institutes of Health and the U.S. Food and Drug Administration Center for Tobacco Products (Grant No. R03CA216528).
