Article Text

Download PDFPDF

Social media use by leading US e-cigarette, cigarette, smokeless tobacco, cigar and hookah brands
Free
  1. Erin Keely O'Brien1,
  2. Leah Hoffman2,
  3. Mario Antonio Navarro2,
  4. Ollie Ganz2
  1. 1 Office of Science, Food and Drug Administration Center for Tobacco Products, Silver Spring, Maryland, USA
  2. 2 Office of Health Communication and Education, Food and Drug Administration Center for Tobacco Products, Silver Spring, Maryland, USA
  1. Correspondence to Dr Erin Keely O'Brien, Office of Science, Food and Drug Administration Center for Tobacco Products, Silver Spring, MD 20993, USA; erin.obrien{at}fda.hhs.gov

Abstract

Background Youth and young adults frequently use social media and are susceptible to tobacco use. This study is the first to provide a systematic overview of how leading tobacco product brands use popular social media platforms.

Methods We identified 112 leading brands of e-cigarettes, hookah, cigars, cigarettes and smokeless tobacco based on sales and self-report user data. We searched for each brand on six platforms: Instagram, Facebook, Twitter, YouTube, Pinterest and Tumblr. In early 2019, we conducted a content analysis of each page, coding: age restrictions, warning display, page characteristics and post characteristics.

Results Cigarette brands were generally not present. Most e-cigarettes, hookah and cigar brands had pages on at least two platforms. One-third of smokeless brands had pages on at least one platform. Few brands had pages on Pinterest and Tumblr. Most pages had existed for 3–5 years. Overall, brand pages rarely used age gating, did not display health warnings, generally posted images of a product alone and often used hashtags unrelated to tobacco. Brands commonly used special features like ephemeral posts on Instagram and pop-up chat windows on Facebook. Many pages displayed images of young people and mentioned flavour. Median followers per brand ranged from about 1 000–10 000, and total followers summed across brands reached over 5 million on Facebook and Instagram alone.

Conclusions Leading brands of most tobacco product types use social media extensively. Several findings identify issues related to youth exposure to and appeal of tobacco social media marketing. Findings can inform tobacco education efforts and regulation.

  • advertising and promotion
  • electronic nicotine delivery devices
  • non-cigarette tobacco products
  • tobacco industry
  • media

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Tobacco companies spend billions on marketing1 2 and promote their products on digital channels such as smartphone-optimised websites,3 apps4 and social media.5 Studying how tobacco companies use changing technology to reach consumers is important because exposure to digital tobacco marketing,6 including social media marketing,7 predicts later tobacco use. Furthermore, nearly all youth and young adults—two populations susceptible to tobacco use8—use multiple social media platforms several times per day.9 10 Although research has documented that some tobacco brands have a social media presence,3 5 11 no research has provided a comprehensive overview of the use of social media marketing by leading brands of e-cigarettes, cigarettes, smokeless tobacco, cigars and hookah tobacco. The current research describes what social media platforms brands use, how they use them, whether they restrict youth access or display health warnings and how they reach and engage their audience.

Tobacco social media marketing

Social media platforms allow individuals and companies to connect and create, share and interact with content including text, videos, photos and links. Social media can be a more powerful marketing tool than print media, because of its ability to precisely target consumers’ specific demographic profiles, engage them in two-way conversations, learn about who interacts with their content and provide real-time data on effectiveness.12 13 Social media marketing includes promotions via paid digital advertising, via compensated ‘influencers’ with large networks, and on brand pages that companies can create for free and use to share content (eg, links and videos).

Many tobacco companies have brand-sponsored social media pages on platforms such as Facebook, Twitter, Instagram, YouTube and Pinterest.3 14–22 Based on previous research, e-cigarette and hookah brands may have a greater social media presence than cigar and smokeless brands, and cigarette companies may not use social media.3 18 Next, we review what is known and unknown about tobacco social media marketing.

Age restrictions

Exposure to tobacco social media marketing is related to tobacco use cross-sectionally in youth23 and longitudinally in young adults.7 Few existing policies limit the extent to which tobacco brands can reach youth. The minimum age to start a social media account is typically 13 years old, and most adolescents use social media multiple times per day.9 10 While several platforms popular with youth have policies that restrict paid direct digital tobacco product advertising (online supplementary table 1), these policies may not be well enforced,18 and they do not appear to address branded social media pages or paid influencers.24 Furthermore, only some platforms offer functions to page owners to restrict the ages of viewers (online supplementary table 1); no platforms require the use of age restrictions for all tobacco-related content; and tobacco branded pages may not voluntarily elect to employ these restrictions.15 25 Given youth’s frequent use of social media, the lack of comprehensive restrictions on tobacco content, and broad reach of this content,14 16 22 there is a high potential for youth exposure.26 27 However, little research has systematically examined youth exposure or voluntary use of age restrictions in tobacco social media marketing, though two studies estimate that a large proportion of JUUL Twitter followers are under age 18 years.28 29

Warnings

Warnings regarding the health effects of tobacco use can serve as a source of health information, discourage initiation and promote cessation.30 Several federal laws require warnings to be displayed on packaging and advertising for cigarettes, smokeless tobacco products, e-cigarettes and hookah tobacco sold or distributed in the USA.31–35 Although the law also requires warnings for cigars (and pipe tobacco), the US District Court for the District of Columbia issued an order enjoining Food and Drug Administration (FDA) from enforcing this requirement until after a related court case is settled.36 Several cigar companies must still display health warnings because of an existing consent agreement with the Federal Trade Commission (FTC).37

We only identified one study that examined warning display for social media.38 This study was limited to examining JUUL on Instagram and found that required nicotine warnings were included at the beginning of testimonial videos. We also note that the FDA and the FTC have issued warning letters to e-liquid manufacturers for failure to display the required nicotine warning in social media posts of influencers made on the manufacturer’s behalf.39

Page and post characteristics

Many studies have described public social media posts about tobacco products on social media19 40–45; however, only several focus specifically on posts originating from tobacco brands. One study of 2012–2015 posts on Facebook, Instagram and Pinterest posts found that Blu and NJOY e-cigarette brands focused on lifestyle-based promotions, while Logic and Metro brands mainly promoted device features.22 A study of 2015 Blu Twitter posts found that many hashtags were related to a specific music festival rather than e-cigarettes.46 A study of JUUL posts on Instagram, Twitter and Facebook from 2015 to 2018 found that JUUL primarily used youth and young adult models, heavily promoted flavours and used hashtags both related and unrelated to the brand and vaping.38 Only one study described content of a non-e-cigarette brand and found that Swisher Sweets used Instagram and YouTube to promote their ‘Artist Project’, which featured musicians and promoted sponsored concerts.16

While these studies provide general qualitative description of post content, they are generally limited to e-cigarettes, and they do not systematically describe some specific content of interest. No studies have assessed the general prevalence of content that may particularly appeal to youth (eg, young models,47 flavours48 49) or the use of hashtags that could lead to youth exposure. Furthermore, these studies do not describe the general types of posts made by tobacco brands (eg, whether they depict people using the product) or general features of brand pages (eg, how long they have been established, post frequency/recency and use of special features such as ephemeral Instagram ‘stories’).

Reach and engagement

Tobacco posts from influencers24 and posts mentioning e-cigarettes41 reach millions of people. Studies specifically focused on posts originating from tobacco brands suggest that they also have extensive reach.14 16 22 One study of four e-cigarette brands’ 2012–2015 posts on Facebook, Instagram and Pinterest found 5 022 posts were shared almost 3 000 times.22 Another study of Blu e-cigarette’s 2014 Twitter posts found they reached over 2 000 individuals, including ‘members of the general public who were not necessarily e-cigarette advocates or users’.14 Another study found YouTube videos from Swisher Sweet’s brand page alone had over 200 000 views.16 Only one study reported indicators of post engagement and found that on average each post was liked a handful of times and had 0–2 comments.22

Gaps in the literature

As described, there are many gaps in terms of the content examined of tobacco company social media marketing (eg, little research has examined age restrictions, health warnings, the featuring of young people, hashtags and followers). Other research gaps are derived from common methodological limitations in past studies. First, past studies have typically focused on one platform,18 27 43 50 one platform and one product16 46 or a small sample of brands.5 22 Most studies focused on e-cigarettes and did not use a systematic method. Additionally, past studies examining the prevalence of tobacco companies on social media searched tobacco brands’ websites for social media links, rather than searching social media platforms themselves,3 18 thus missing pages not promoted on websites.

The current study

This content analysis seeks to describe the overall landscape of tobacco brands’ use of social media by identifying and content analysing leading cigarette, e-cigarette, cigar, smokeless tobacco and hookah brands’ pages on six popular social media platforms. This study has four goals: to describe (1) brand presence across platforms, (2) brands’ use of age restrictions and display of health warnings, (3) pages and post content and (4) brand pages’ follower count and engagement.

Method

Brand identification

In November 2018,we used several sources to identify leading brands of e-cigarettes, hookah, cigars, cigarettes and smokeless tobacco. For all products, we used (A) Nielsen data on top 20 brands for each type based on July 2017–June 2018 dollar sales from Nielsen Scantrack data for total US convenience stores and all outlets combined, and (B) Euromonitor’s identified US market leaders in 2017 based on retail volume (except for hookah). For hookah, we also used self-report data from wave 4 of the Population Assessment of Tobacco and Health Study.51 For e-cigarettes, we collectively considered top brands in Nielsen’s eight e-cigarette categories (electronic cigarette disposable, electronic cigarettes kits, electronic cigarettes refills liquid, electronic cigarettes refills non-liquid, electronic cigars disposable, electronic hookahs disposable, electronic hookahs refills liquid and electronic pipes kits). We used multiple additional methods because the Nielsen data do not include online or vape shop sales. We accomplished this using a multistep process. First, we identified the top 16 online retailers that sold multiple brands in 2016–2017 as reported by ECigIntelligence.52 Then, we searched each retailer on Google Trends, looked at ‘related queries’ and identified three additional online retailers. We ranked each retailer by search score, identified the top five (elementvape.com, vapewild.com, myfreedomsmokes.com, breazy.com and vapordna.com) and identified 16 brands that at least two retailers characterised as bestselling. Finally, we identified three up-and-coming e-cigarette brands in ‘related topics’ and ‘related queries’ that appeared from searching Google Trends for ‘e-cigarette,’ ‘vape’ and ‘vaping’. There was overlap between several data sources; online supplementary table 2 lists the brands and notes the unique number of brands added by the method used.

Identifying brand-sponsored pages

In November 2018, we searched Instagram, Facebook, Twitter, YouTube, Pinterest and Tumblr. We used desktop computers to search for each brand and product category name (eg, “Camel cigarettes”) on each social media platform and reviewed the first page of results. We included public pages that appeared to be brand sponsored based on having at least one of the following characteristics: the page title used the term ‘official’; it linked to the brand’s website; the content solely promoted one brand; or, for Twitter and Instagram, were ‘verified’. We excluded pages that: were described as a ‘fan page’; were described as unofficial; had most posts in a language other than English; were described as exclusively representing a country other than the USA (eg, ‘Vapeccino Japan’); or had no posts. We attempted to include SnapChat in the study and searched it using the smartphone app; however, the accounts we identified as having the same name as a brand (eg, ‘Kangertech’) did not have any bio information or posts. Thus, we could not assess whether they appeared brand sponsored and do not report results. For the other platforms, where we did not identify a page for a brand, a second author double-checked this finding. In searching for the 112 brands across platforms, we identified 58 pages on Instagram, 53 on Facebook, 49 in Twitter, 42 on YouTube, 26 on Pinterest and 17 on Tumblr.

Coding pages

In January–March 2019, using desktop computers, we coded characteristics as described below. First, all four authors dual coded 10 randomly selected pages. Because we had high agreement (91%; range across variables: 89%–100%); we did not dual code the remaining pages; they were each coded by one author. Before proceeding, we reconciled differences and added detail to the coding scheme.

Most of our coding scheme was deductive (ie, predetermined). However, several characteristics we coded were inductive (ie, derived from the data; eg, post content and hashtags). To develop inductive codes, we first recorded a general text description during the dual coding phase, then met and discussed the main themes. Then, we added codes for these themes to the coding scheme and coded the text originally recorded.

Age restrictions and warning display

We coded whether the account used age gating by attempting to access the web address of each page without logging into the social media platform. We coded the page as having age gating if we could not see the page content without logging in (Instagram and Facebook) or if we received a notice that we had to be logged in to verify age and view the content (YouTube). We note that to restrict viewer age on Twitter, page owners must both implement age gating on who can follow the Twitter account and limit all posts to only followers. Among pages that had no age gating, we also coded whether there were any statements about age restrictions in the company bio (eg, ‘About’ section on Facebook) or in the three most recent posts. We coded whether any warnings about the health effects of using the product were displayed anywhere on the page, including the homepage, bio or three recent posts.

Page characteristics

We coded several basic characteristics of each social media page: page founding date; activity level (days since last post and number of posts within the past 7 days); and bio characteristics (inclusion of a website link; inclusion of hashtags). We also coded whether each page included specific features, including whether it: had any ephemeral posts (that disappeared after a certain time; eg, Instagram stories); included a direct link to purchase products (eg, ‘Shop Now’ Facebook button); and had a chat window automatically appear on Facebook. Finally, three authors (EKO, MAN and LH) coded whether anyone appearing younger than 25 years old was visible on the home page or three most recent posts; because this is subjective, we only coded this as ‘yes’ if we all agreed.

Post content

We examined the past three most recent posts and coded whether any: included a link to buy a specific product; mentioned a flavour other than tobacco, menthol or mint; were videos (and if so, number of views); featured a product image by itself; featured a close-up image of a product in a hand (‘hand check’)46; and featured a person using the product. For YouTube, we did not code whether videos included an image of a product or an image of a product in a hand because we were not confident we did not miss this after viewing videos several times. Finally, we coded whether any of the past three posts used hashtags, which function as keywords attached to a post to help people search for it. Among those that did, we coded whether brands included hashtags unrelated to tobacco (eg, #superbowl2019), or related to brand name (eg, #myblu) or product category (eg, #vapelife).

Follower count and engagement

We recorded page statistics, including number of followers, number of page views (YouTube only) and number of monthly viewers (Pinterest only). Note that while number of followers is a proxy for post reach, it is subject to some error because not all followers view posts and non-followers can also view posts. For each page’s three most recent posts, we recorded number of likes, comments and shares as measures of viewer engagement. We also recorded two measures of brands engaging viewers: whether at least one of the past three posts (A) was reposted content and (B) included brand responses to commenters.

Results

Search results

Leading tobacco brands had an extensive presence on social media, as the majority of leading e-cigarettes, hookah and cigar brands had pages on at least two social media platforms, and one-third of smokeless tobacco brands had pages on at least one platform. Instagram, Facebook and Twitter were the most commonly used platforms. Only one cigarette brand had one page; they were excluded from further analyses. All pages were public except for Nat Sherman’s Twitter page, which had content limited to followers only; they did not accept our follow request so we excluded this page from results (table 1).

Table 1

Per cent and number of leading brands having pages on each social media platform by tobacco product category

Age restrictions and warning display

The ability to restrict visible content by age was available on Instagram, Facebook, YouTube and Twitter (ie, viewers must be logged into an account where their birthdate was at least 18 or 21 years to view page content). Few pages used any age restrictions, and thus their content was fully visible to youth; across products, no Twitter pages, very few YouTube and Instagram pages (2%–5%) and one-third of Facebook pages were age restricted. On Facebook, hookah websites were the most likely to be age restricted (60.0%) and e-cigarettes least likely (16.7%). We also noted that some pages that chose not to use age restrictions had statements about minimum age requirements in the bio and/or recent posts (eg, ‘Must be 18 to follow’, Al Fakher) (table 2).

Table 2

Social media pages’ age restrictions and warning display

Overall, few pages displayed any health warnings anywhere on the page, including the homepage, bio or three most recent posts. No Tumblr pages, few Pinterest pages (7.7%) and about a third of Instagram and Facebook pages contained warnings. A variety of methods were used to display warnings, including displaying them: in banner photos in either large clear text (eg, General Snus on Facebook) or very small text (eg, Punch Cigars on Twitter); at the end of post descriptions after several blank lines of text, requiring viewers to scroll down to see it (eg, UWell on Instagram); as text within an image that is too small to read on a desktop computer screen (eg, Flair Vapor on Instagram); and as a band clearly visible across the top or bottom of an image (eg, Social Smoke on Facebook). YouTube videos with warnings generally displayed them at the beginning or the end of the video, as a full screen (eg, Stoker’s) or as small text (eg, Tonic Shisha).

Page characteristics

Hookah and cigar Facebook and Twitter pages had a slightly older median ages (7–8 years) compared with e-cigarettes and smokeless pages (3–5 years). Facebook allows page owners to select the ‘category’ that the page belongs to (data not shown in table). Of the 53 pages we coded, only four pages (7.5%) were categorised as tobacco/e-cigarette stores/companies. The most common categories were ‘product/service’ (35.8%), ‘electronics’ (26.4%, all e-cigarettes brands) and ‘company’ (11.3%) (table 3).

Table 3

Social media page general characteristics

Across products, Instagram and Facebook pages were most active in terms of post recency and frequency—median time since the last post was approximately 1 week, and there was a median of two posts per week. For the other platforms, the most recent post was from a median of 29 days ago (Twitter), over 3 months ago (YouTube) or over a year ago (Pinterest and Tumblr). Examining activity level by product brand category, e-cigarettes brands were the most active (many posting daily), followed by cigar brands then hookah brands. Smokeless brands were virtually inactive.

Across tobacco products and platforms, brand pages typically had one or two sentences in their bio that described and promoted the brand. Examples include ‘Making minds tingle with taste since 1997 (Fumari)’ and ‘Established in 1912, and home of the finest cigars in the world (Arturo Fuente)’. Most bios linked to the branded websites except on Tumblr. While about half of Instagram bios included branded hashtags (eg, #mymontecristo), few other pages did so. On Instagram, five e-cigarette brands and one smokeless brand (Copenhagen) included contact information for those interested in being an influencer or promoter (eg, Eleaf: ‘Click the link to be a promoter!’; Lost Vape: ‘Welcome photographers and influencers to join our team’). On YouTube, two bios solicited product reviewers (VandyVape and Joyetech) and one advertised an app (SMOK).

Most e-cigarettes pages (75.9%) and nearly half of hookah (45.4%) and cigar (40%) pages featured ephemeral Instagram ‘stories’. There were no ephemeral posts on Facebook or YouTube. On Facebook, about half of e-cigarettes pages and one of the two smokeless pages included a chat window that popped up automatically to engage page visitors. On Facebook, a ‘Shop Now’ button linking directly to an online retailer was common on pages of all products type brands except hookah; direct links to purchase products were also present for some e-cigarettes brands on Instagram and YouTube.

Post content

We examined the past three most recent posts of each page and reported the proportion that had at least one post with specific characteristics. Overall, most posts were still images rather than videos. Most images featured a specific product by itself, and ‘hand checks’ of a hand holding a product were common for e-cigarette and cigar posts. About a quarter of Instagram and Facebook posts featured a person using a tobacco product, as did more than half of YouTube videos. Across platforms, e-cigarettes and hookah videos were commonly demonstration videos. While at least a quarter of e-cigarettes pages had at least one post with direct links to buy specific products promoted in posts, this was rare for posts from other products. Hookah pages were the most likely to have at least one post mentioning flavour; more than half did so on Instagram, Facebook, YouTube and Tumblr. A smaller proportion of e-cigarettes page posts (7%–21%) did so, as did several cigar posts (table 4).

Table 4

Content of leading tobacco brand’s social media posts

Across brand types, hashtags were generally were used in at least one of the past three posts on almost all Instagram and Tumblr pages (88%–90%), more than half of Facebook and Twitter pages (58%–59%) and few Pinterest and YouTube pages (7%–19%). Generally, posts included multiple hashtags. Almost all had had a hashtag specific to the product brand (72%–100%; eg, #myblu), and most had a hashtag related to the product category (58%–100%; eg, #vapelife). Across platforms, most hookah and cigar posts also included hashtags unrelated to tobacco (eg, #superbowl2019); this was also the case for a proportion of e-cigarette posts.

On homepages and the three most recent posts, we found at least one photo of a person who appeared younger than age 25 years in a significant proportion of pages across platforms for all types of product pages except smokeless; the largest proportion was on Instagram (43.1%). Examining results by product type, we found this for at least half of hookah pages on Instagram, Facebook, YouTube and Tumblr; at least 40% of e-cigarettes and cigar pages for Instagram; about a quarter of e-cigarettes pages on Facebook, YouTube and Pinterest; and for about a quarter of cigar pages on Facebook and YouTube (table 4).

Follower count and engagement

Tobacco brand pages had the highest median number of followers on Instagram and Facebook (about 8–10 000), fewer on Twitter and YouTube (about 1–2 000) and the fewest on Pinterest (90). Across platforms, e-cigarettes pages had the highest median followers, followed by hookah, cigar and smokeless pages. Looking at the total number of followers by platform, brand pages were followed by over 5 million people on Facebook and Instagram alone; most followers were following e-cigarettes pages (table 5).

Table 5

Reach and engagement

We also coded each page’s three most recent posts for indicators of engagement. Based on these measures, Instagram, Facebook and YouTube had the most audience engagement. Median number of likes ranged from about 10 to 20 across platforms, except for Instagram, where e-cigarette brands had a much higher median (429). Overall, posts generally had 0–2 shares and comments. Brands were the most responsive to commenters on Instagram, Facebook and YouTube (12.5%–21.4%). Brands reposted content most often on Instagram, Twitter and Tumblr (17.6%–50.0%).

Discussion

Our systematic content analysis described the overall landscape of tobacco social media marketing. We found that leading tobacco brand use of social media is widespread: the majority of e-cigarette, hookah and cigar brands had active pages on at least two social media platforms, and one-third of smokeless tobacco brands had mostly inactive pages on at least one platform. Consistent with prior research,3 5 we found that cigarette brands generally did not have pages. These findings are significant, as tobacco social media marketing7 and digital tobacco marketing in general6 predict youth and young adults’ subsequent tobacco use.

Comparing results by platform, Instagram, Facebook, Twitter and YouTube were the most commonly and actively used platforms, and most brand pages were well-established, having existed for at least several years. Most pages on Pinterest and Tumblr were inactive. Brands commonly used special features of specific platforms, such as ephemeral Instagram ‘stories’ and pop-up chat windows on Facebook. Based on the number of likes and comments, viewers engaged most with posts on Facebook, Instagram and YouTube; brands also most commonly responded to comments on these platforms. Brand pages have many followers, with median number of followers per brand ranging from about 1 000–10 000 and total followers summed across brands reaching over 5 million between Facebook and Instagram alone.

Brand pages for nearly all product types shared several characteristics. Most brand pages: rarely used age gating; did not display warnings; linked to branded websites; and commonly posted images of products alone. Our finding that posts often used multiple hashtags that were about the brand name, product type and unrelated to tobacco were similar to one study of JUUL.38 However, there were also differences by product type. Hookah brand posts were unique in that most mentioned flavour. Smokeless brand pages were unique in that they generally did not: mention flavour, show people using products or show people appearing under age 25 years. E-cigarette brand pages were unique in that they commonly featured ‘hand check’ images and linked directly to online stores. Comparing our results to past studies that examined links to online stores, we found that more e-cigarette Twitter posts linked to stores compared with a 2008–2013 analysis,42 and fewer Facebook pages for e-cigarettes, hookah and cigar brands had ‘Shop Now’ buttons on Facebook compared with a 2017 analysis.18

A combination of specific findings suggest concerns related to tobacco social media marketing and youth. First, consistent with past research,3 25 the vast majority of tobacco brand social media pages are fully visible to youth, because few elected to use stringent methods of restricting youth access (ie, restricting based on the birthdate of the users’ profile). Furthermore, few elected to state the content was only for adults of a certain age (a less stringent method). This visibility is especially concerning for several reasons. First, a large proportion of e-cigarettes, hookah and cigar brands appeared to feature an image of at least one person under age 25 years, which may particularly appeal to youth.47 Most hookah posts and a smaller proportion of e-cigarettes and cigar posts mention flavours; this may also particularly appeal to youth.48 49 Most pages did not include health warnings, which could dissuade youth product interest.30 Posts often included hashtags unrelated to tobacco; this expands post reach so youth who have not searched for tobacco content but are following a particular hashtag (eg, ‘#superbowl2019’) can encounter it. Many pages and posts also provided direct links to buy products, which is a concern because past research has found that some websites do not verify age when selling tobacco products online.3 53 Considering that millions of people follow these tobacco brands, that most youth access multiple social media platforms multiple times per day9 10 and that exposure to this marketing is related to use,7 23 this combination of findings indicates a significant public health concern.

Implications

Findings have implications for researchers, policymakers and social media companies. First, findings can inform what platforms and product types researchers prioritise in future studies (eg, prioritise e-cigarette and hookah brands, and Instagram and Facebook). Findings also identify potential challenges to capturing and studying some tobacco social media posts. For example, we found that brands often use ephemeral Instagram ‘stories’ that disappear in as little as 24 hours. Additionally, we found that e-cigarette brands use social media to recruit influencers, whose posts often do not disclose sponsorship.24 26 54 55 Furthermore, results related to specific platforms used for different product types are informative to public health practitioners who design public education campaigns that address tobacco industry marketing efforts.

Study results also have implications for policymakers and regulators. Our research suggests that policies to reduce youth exposure to tobacco marketing could include requiring marketers to: use age-gating tools to prevent youth exposure to posts; use independent age verification on any linked websites selling tobacco products; display health warnings prominently on posts; and limit hashtags to those related to tobacco to avoid posts being found by people who are tobacco naïve. Our findings may be relevant to the UK’s Advertising Standards Authority, which regulates tobacco advertising and has rules banning online advertising of e-cigarettes but allows factual statements about such products on company websites. Recently, they ruled that Vype e-cigarette’s Instagram page was advertising and thus not permitted as it did not fall under the website exception56; our research indicates there are many other public tobacco brand social media pages that could be subject to this rule. In the USA, results can provide context for FDA’s review of applications to market tobacco products (eg, Premarket Tobacco Applications57; Modified Risk Tobacco Product Applications).58 For example, they may inform postmarket reporting requirements related to social media marketing, such as those described in the iQOS marketing orders.59

Finally, findings may have implications for social media companies. Most social media companies ban paid tobacco advertising (online supplementary table 1), and recently, Facebook and Instagram banned tobacco brands from paying to promote influencer posts.60 However, public tobacco brand pages—the subject of this study—do not appear to be banned or restricted. Our finding that most tobacco brands have pages on social media and do not use age restrictions may inform future social media company decisions about regulating these pages.

Limitations and future directions

Our study had several limitations. First, we only had one round of practice coding. Although we had high reliability, this could be improved in future research. Second, we did not confirm page ownership with tobacco companies and thus may have included pages that appeared brand-sponsored but were not. Third, our study focused on describing how leading tobacco brands explicitly use social media to market their products publicly; describing tobacco brand use of social media influencers was beyond this scope and remains an area of future research. Fourth, while large tobacco companies with multinational brands often use similar marketing strategies worldwide,61 this study was limited to English-language pages and excluded accounts that were for countries other than the USA. Therefore, results may not be generalisable to tobacco social media marketing worldwide. Fifth, findings are time limited, as this study analysed a cross-section of social media pages from early 2019. Tobacco brand use of social media and social media platforms themselves are rapidly changing; future research can consider updating our findings. Finally, while our study shines a light on the opportunities for youth to be exposed to and engage with tobacco social media marketing, we did not actually measure this in our study. Future research can examine topics such as youth exposure to and engagement with this marketing, how exposure relates to later use and potential sociodemographic differences.

Conclusions

This systematic content analysis is the first to describe the overall landscape of tobacco brands use of social media: it describes how leading brands of e-cigarette, hookah, cigar and smokeless tobacco use Instagram, Facebook, Twitter, YouTube, Pinterest and Tumblr. We found that most of these brands have pages that: have existed for years on multiple platforms, do not use age restriction tools or display health warnings and have a sum total of over 5 million followers. We also describe key characteristics of brand pages and posts, such as featuring young people and mentioning flavour, and note that a combination of these characteristics suggest concern for youth. Findings provide perspective for regulators and can inform future research and tobacco education efforts.

What this paper adds

  • Understanding how tobacco companies market products on social media is important, as the vast majority of youth and young adults—who are vulnerable to tobacco initiation—use several social media platforms multiple times per day, and exposure to tobacco social media marketing is related to tobacco use.

  • Past studies have noted that tobacco brands market their products on social media but are often limited to studying one or several brands on a single platform. Furthermore, past studies have examined limited aspects of tobacco brand use of social media and have not systematically assessed key features such as use of age restrictions, display of warnings, page and post content, follower count and post engagement.

  • The current study fills this gap by systematically content analysing these features in leading e-cigarette, hookah, cigar and smokeless tobacco brand pages on Instagram, Facebook, Twitter, YouTube, Pinterest and Tumblr.

  • Findings provide an overall description of tobacco company use of popular social media platforms. While cigarette brands are not present on social media, most other tobacco product brands had pages on multiple platforms. Key results include that most brand pages do not use age restriction tools and thus are visible to youth, that most pages do not display health warnings and that Instagram and Facebook pages alone have a sum total of over 5 million followers.

References

Footnotes

  • Contributors All authors contributed to the idea for the research and completed coding. EKO led drafting the coding scheme and manuscript and analysed results. LH led identifying leading brands of products and contributed to the introduction. MAN led drafting the results. OG contributed to the discussion. All authors provided feedback on the draft manuscript and approved the final version.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Patient consent for publication Not required.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Data availability statement No data are available.