Characteristics of Adolescents’ and Young Adults’ Exposure to and Engagement with Nicotine and Tobacco Product Content on Social Media

Abstract

To inform policy and messaging, this study examined characteristics of adolescents’ and young adults’ (AYAs’) exposure to and engagement with nicotine and tobacco product (NTP) social media (SM) content. In this cross-sectional survey study, AYAs aged 13–26 (N=1,163) reported current NTP use, SM use frequency, and exposure to and engagement with SM content promoting and opposing NTP use (i.e. frequency, source[s], format[s], platform[s]). Participants who used NTPs (vs. did not use) were more likely to report having seen NTP content (p-values<.001). Prevalent sources were companies/brands (46.6%) and influencers (44.4%); prevalent formats were video (65.4%) and image (50.7%). Exposure to content promoting NTP use was prevalent on several popular platforms (e.g. TikTok, Instagram, Snapchat); exposure to content opposing NTP use was most prevalent on YouTube (75.8%). Among those reporting content engagement (i.e. liking, commenting on, or sharing NTP content; 34.6%), 57.2% engaged with influencer content. Participants reported engaging with content promoting and opposing NTP use on popular platforms (e.g. TikTok, Instagram, YouTube). Participants with (versus without) current NTP use were significantly more likely to use most SM platforms and to report NTP content exposure and engagement (p-values<.05). Results suggest that NTP education messaging and enforcement of platforms’ content restrictions are needed.

Nicotine and tobacco product (NTP) use among adolescents and young adults (AYAs) is a persistent public health challenge. In 2021, 2,060,000 United States (U.S.) high school students (Gentzke et al., 2022) and 5,614,000 U.S. YAs aged 18–25 (Substance Abuse and Mental Health Services Administration, 2023) reported current NTP use. Psychosocial factors, such as exposure to digital marketing (Venrick et al., 2023) and social norms (East, McNeill, Thrasher, & Hitchman, 2021), are known to influence AYAs’ NTP use. Further, social media is a nearly ubiquitous feature of AYAs’ social environments, with 91.5% of U.S. middle and high school students (Gentzke et al., 2022) and 84% of U.S. YAs (Auxier & Anderson, 2021) reporting social media use in 2021. Meta-analytic evidence suggests that exposure to and engagement with NTP social media content are consistently associated with NTP use (Donaldson, Dormanesh, Perez, Majmundar, & Allem, 2022), underscoring the need to understand and address AYAs’ exposure to and engagement with such content.

Content analyses have uncovered a preponderance of social media content promoting NTP use, spanning traditional marketing, influencer marketing (i.e., product promotion by individuals with large social media followings thought to influence trends (de Veirman, Cauberghe, & Hudders, 2017)), and people who post on social media uncompensated. Companies promoting NTPs (e.g., e-cigarettes, cigars) have used youth-friendly advertising tactics, including cartoons (Dormanesh, Kirkpatrick, & Allem, 2020), linking NTP use to partying (Kong et al., 2023), and product giveaways (Kong, Kuguru, Bhatti, Sen, & Morean, 2021). As of 2021, most major social media platforms prohibit paid ads for NTPs; however, few platforms explicitly prohibit sponsored content from influencers (Kong et al., 2022). The U.S. Food and Drug Administration (FDA) requires NTP advertisements with visual components, including social media posts, to include a conspicuous warning label regarding the addictiveness of nicotine (U.S. FDA, 2020). Recent estimates suggest that only 14–15% of e-cigarette and e-liquid promotional posts on Instagram carry FDA-compliant warning labels (Laestadius, Wahl, Vassey, & Cho, 2020; Silver, Bertrand, Kucherlapaty, & Schillo, 2022), indicating a need for stronger enforcement. Despite platforms’ restrictions on NTP content, Facebook and Instagram posts promoting NTPs often use the platforms’ branded content tools, which can be used to optimize product promotion (Laestadius et al., 2023).

NTP-related social media content from individuals can proliferate quickly via hashtags, even when a brand’s official social media accounts are no longer accessible (Valdez & Unger, 2021). Individual users may normalize use by depicting themselves using NTPs in everyday life (Lim & Lee, 2023; Vogel et al., 2023) and sharing their experiences with NTPs (Shao et al., 2022; Smith, Buckton, Patterson, & Hilton, 2023). NTP-related social media content may glamorize and facilitate NTP use through aesthetic appeal (Kong, LaVallee, Rams, Ramamurthi, & Krishnan-Sarin, 2019; Sun et al., 2021), direct product sales (Jancey et al., 2023; Liu et al., 2020; Vogel et al., 2023), and instructional content (Massey et al., 2020). Although most research has focused on e-cigarette social media content, products such as premium cigars (Kong et al., 2023), little cigars and cigarillos (LCCs) (Wu et al., 2023), and heated tobacco products (Abroms et al., 2023; Chen, Xue, Xie, & Li, 2023) are also promoted on social media, with desirable lifestyle themes (Chen, Xue, Xie, & Li, 2023), poor adherence to warning label requirements (Chen, Xue, Xie, & Li, 2023), and little age-gating to prevent underage access to promotional content (Kong et al., 2023). Social media content that portrays NTPs positively may encourage AYAs’ NTP use (Litt & Stock, 2011; Vogel et al., 2021). According to the Prototype Willingness Model, AYAs form a mental image, or “prototype,” of a person who engages in a certain behavior (e.g., NTP use). If the prototype is positive, AYAs may be more willing to engage in the behavior when presented with the opportunity (e.g., in a social situation) even if they do not explicitly intend to (Gibbons & Gerrard, 1995; Gibbons, Gerrard, Blanton, & Russell, 1998).

Aligning with the centrality of social media in AYAs’ social environments, public health campaigns, such as The Real Cost (U.S. FDA, 2023), Truth (Truth Initiative, 2023), and campaigns targeting specific populations at increased risk for tobacco use (e.g., This Free Life (U.S. FDA, 2022b), Fresh Empire (U.S. FDA, 2022a)), established a social media presence. In a nationally representative study, adolescents with greater (versus lesser) social media use were more likely to recall seeing The Real Cost e-cigarette use prevention ads (Vereen et al., 2022). A majority of adolescents (15–17%) report visiting the three most popular social media platforms (i.e., YouTube, TikTok, Snapchat) daily, with 14–17% visiting these platforms “almost constantly” (Anderson, Faverio, & Gottfried, 2023). Social media may be an effective way to reach many adolescents with NTP educational messaging. Exposure to NTP education messaging is associated with more negative attitudes toward NTP use and greater odds of a quit attempt among those using NTPs (Farrelly et al., 2002; MacMonegle et al., 2022; Oyapero et al., 2021). Among media outlets, social media is unique in its potential for engagement with the target audience. Monitoring AYAs’ engagement with NTP-related social media content is imperative, as engagement with content promoting NTP use may be associated with greater odds of subsequent NTP use, and engagement with content opposing NTP use may be associated with lower odds (Yang, Clendennen, & Loukas, 2023).

Content analyses have described NTP content that is publicly available on social media platforms. However, AYAs are also exposed to private content that cannot be scraped from social media (e.g., through peers’ private accounts). Documentation of the content publicly available on social media does not necessarily reflect the content exposure that is most memorable and influential to AYAs. Engagement data mined from social media are typically not linkable to individuals’ characteristics and may include engagement from older adults and organizations as well as AYAs. AYAs who use (vs. do not use) NTPs may have more exposure to and engagement with NTP-related social media content. Social media algorithms display content based on the data they have gathered about the user’s interests (O’Brien, 2023); thus, using NTPs may increase odds of exposure to NTP-related content. According to the Elaboration Likelihood Model of persuasion, individuals process persuasive messages (e.g., social media posts about NTPs) more deeply, with more attention to the message itself rather than peripheral features (e.g., attractiveness of the spokesperson), when the message is personally relevant (Petty & Cacioppo, 1986). NTP content is more personally relevant to AYAs who use (vs. do not use) NTPs, and may be more engaging for those who use. Understanding the nuances of NTP-related social media content to which AYAs are exposed, and with which they engage, can inform social media platforms’ enforcement priorities and help guide public health campaigns toward optimal social media messaging.

This cross-sectional, observational survey study aimed to examine AYAs’ perceptions of their exposure to and engagement with NTP content, and to compare exposure and engagement between AYAs who do and do not currently use NTPs. AYAs aged 13–26 were surveyed about their exposure to and engagement with NTP-related social media content. Aims were to: 1) describe characteristics of NTP-related social media content exposure (e.g., frequency, content valence, sources, platforms), 2) describe characteristics of NTP-related content engagement, and 3) compare prevalence and characteristics of NTP-related content exposure and engagement between AYAs with and without recent NTP use. Aims 1 and 2 were descriptive and exploratory. Due to social media algorithms that target users’ specific interests, we hypothesized that reports of NTP-related content exposure would be more prevalent among AYAs with (versus without) NTP use (Aim 3). Due to the personal relevance of NTP-related content among AYAs who use NTPs, we hypothesized that NTP-related content engagement would also be more prevalent among AYAs with (versus without) NTP use (Aim 3). We also explored differences in NTP-related content exposure and engagement characteristics (e.g., content valence, sources, platforms) by NTP use status.

Methods

Participants and Procedures

U.S. AYAs aged 13–26 who reported using at least one social media platform at least 5 days per week were recruited for a survey study through Qualtrics Research Services, which recruits respondents who have consented to being contacted with research participation opportunities through online panels (Qualtrics Research Services, 2024). We oversampled for past-month NTP use, aiming for 30% of respondents. Potential participants received an invitation via e-mail from their panel provider and provided electronic assent (participants aged 13–17) or electronic consent (aged 18–26) to complete the eligibility screener and survey (measures described below) online in July 2023. Participants were compensated by their panel providers. The protocol was approved by the Westat Institutional Review Board. The full text of the survey is available as Online Supplementary Material.

Measures

Social Media Use, NTP Use, and Sociodemographics. Participants reported use frequency (never, every few months, every few weeks, 1–4 days a week, 5–6 days a week, once or twice a day, many times a day) of 12 social media platforms: Facebook, Twitter, Instagram, TikTok, Snapchat, Discord, Reddit, YouTube, Pinterest, Twitch, Slack, and BeReal. We chose these platforms based on prior research that reviewed social media platforms’ NTP content policies in 2021 (Kong et al., 2022). Studying these platforms will add information on the NTP content that AYAs are seeing in the context of existing NTP social media policies. Participants also reported ever-use and past-month (current) use (yes/no) of 13 NTPs (cigarettes, traditional cigars, cigarillos, little cigars, pipe, chewing tobacco, snuff, snus, hookah/shisha, e-cigarette or vaping device containing nicotine, nicotine pouches, nicotine gummies, and IQOS) with a “select all that apply” checklist. Sociodemographic characteristics measured were age, race (American Indian or Alaska Native, Asian, Black or African American, Native Hawaiian or other Pacific Islander, White, another race), Hispanic/Latinx ethnicity (yes/no), sex assigned at birth (male/female), gender identity (boy/man, girl/woman, non-binary, another gender), and sexual orientation (asexual, bisexual, gay, lesbian, pansexual, queer, questioning or unsure, straight/heterosexual, another identity).

NTP Content Exposure. We first measured frequency of NTP content exposure, defined as “any content related to nicotine/tobacco on social media” (never, only once or twice, every few months, every few weeks, 1–2 days a week, 3–6 days a week, at least once a day). Those with any exposure reported the balance of “anti-use” (i.e., opposing NTP use) and “pro-use” (i.e., promoting NTP use) content (only anti-use, more anti-use than pro-use, about equal, more pro-use than anti-use, only pro-use, I don’t remember), content source(s) (companies/brands, influencers, friends, acquaintances, other), and content format(s) (posts with images, posts with only text, videos or “reels,” transient content [“content that goes away after a period of time”], other). Checklists (yes/no) measured exposure to “pro-NTP” and “anti-NTP” use content on 12 social media platforms (Facebook, Twitter, Instagram, TikTok, Snapchat, Discord, Reddit, YouTube, Pinterest, Twitch, Slack, and BeReal).

NTP Content Engagement. We first measured whether participants had “ever liked, commented on, or shared a post that included NTPs” (no, yes, don’t remember). Those responding “yes” reported frequency (only once or twice, every few months, every few weeks, 1–2 days a week, 3–6 days a week, at least once a day), content source(s) (companies/brands, influencers, friends, acquaintances, other), and content format(s) (posts with images, posts with only text, videos or “reels,” transient content [“content that goes away after a period of time”], other). We then queried whether participants had ever posted something on social media that included NTPs (no, yes, don’t remember). Those responding “yes” reported frequency, content source(s), and content format(s). Participants who had ever liked, commented on, shared, or posted NTP content used checklists to report on which platform(s) they had engaged with NTP content (measured separately for content promoting and opposing NTP use).

Supplemental Measures: Characteristics of Most Recent NTP Content Seen. Participants who reported having ever been exposed to NTP content reported characteristics of their most recent exposure to content promoting and opposing NTP use: content source (as listed previously), social media platform (as listed previously), timeframe of exposure (within the past week, past month, past 6 months, past year, more than a year ago, I don’t remember), products featured (as listed previously). Participants who reported exposure to content promoting NTP use also reported flavor(s) featured (tobacco-flavored or unflavored, menthol, mint, fruit, dessert, other, I don’t remember).

Statistical Analysis

Descriptive statistics were examined for all social media use, NTP use, sociodemographic, NTP content exposure, and NTP content engagement variables. Response options with low frequencies were aggregated. Participants with and without current NTP use were compared on social media use, sociodemographic, NTP content exposure, and NTP content engagement variables using chi-square tests of independence for categorical variables and an independent-samples t-test for age (measured continuously). Statistical significance was defined as p < .05. Analyses were conducted with SPSS 29.0.

Results

Participant Characteristics

Participant characteristics are presented in Table 1. The sample (N = 1,163; M_age = 19.5 [SD = 3.6], 79.3% girls/women) was racially and ethnically diverse (38.8% non-Hispanic White, 30.4% Hispanic or Latinx, 18.7% non-Hispanic Black, 12.1% non-Hispanic another race) and 70% heterosexual. Social media use and NTP use characteristics are presented in Table 2. With oversampling, 25.4% reported any current NTP use (57.6% of whom used e-cigarettes with nicotine, 35.6% cigarettes, 34.2% other combustible NTP[s], 14.9% other noncombustible NTP[s]), and 44.1% reported ever-use (67.3% of whom used e-cigarettes with nicotine, 48.3% cigarettes, 51.5% other combustible NTP[s], 24.0% other noncombustible NTP[s]). The most commonly used social media platforms were YouTube (96.9% any use, 60.6% daily use), TikTok (87.9% any use, 72.7% daily use), Instagram (86.3% any use, 60.8% daily use), and Snapchat (81.4% any use, 57.7% daily use). Participants with (versus without) current NTP use were significantly older (p < .001), more likely to identify with a non-heterosexual identity (p < .001), more likely to report use of most social media platforms (p-values < .05) with the exception of YouTube and Pinterest, and more likely to report daily use of Snapchat, Facebook, Twitter, Discord, Reddit, Twitch, BeReal, and Slack (p-values < .05).

Table 1. Participant sociodemographic characteristics by nicotine or tobacco product (NTP) use status in a sample of adolescents and young adults surveyed in 2023

	No current NTP use (n = 872)	Current NTP use (n = 291)	Full sample (N = 1,163)
	% (n)
Age (M/SD)***	18.9 (3.5)	21.2 (3.1)	19.5 (3.6)
Race and ethnicity (n = 1,130)
NH White	37.9 (317)	41.3 (121)	38.8 (438)
Hispanic or Latinx	30.5 (255)	30.4 (89)	30.4 (344)
NH Black or African American	18.8 (157)	18.4 (54)	18.7 (211)
NH another race^a	12.9 (108)	9.9 (29)	12.1 (137)
Sex assigned at birth (n = 1,126)
Female	83.8 (702)	79.5 (229)	82.7 (931)
Male	16.2 (136)	20.5 (59)	17.3 (195)
Gender (n = 1,133)
Girl/woman	80.3 (677)	76.6 (222)	79.3 (899)
Boy/man	16.4 (138)	18.6 (54)	16.9 (192)
Non-binary or another gender	3.3 (28)	4.8 (14)	3.7 (42)
Sexual orientation (n = 1,088)***
Straight or heterosexual	72.8 (581)	62.4 (181)	70.0 (762)
Another orientation^b	27.2 (217)	37.6 (109)	30.0 (326)

Note: NH = Non-Hispanic. Differences by NTP use status were examined using a t-test (age) and chi-square tests of independence (remaining variables) and are designated with: *p < .05 **p < .01 ***p < .0001

^aIncludes American Indian or Alaska Native, Asian, Native Hawaiian or Pacific Islander, multiracial, or another race

^bIncludes asexual, bisexual, gay, lesbian, pansexual, queer, questioning or unsure, or another orientation

Table 2. Social media and nicotine or tobacco product (NTP) use characteristics in a sample of adolescents and young adults surveyed in 2023

	No current NTP use (n = 872)	Current NTP use (n = 291)	Full sample (N = 1,163)
	% (n)
Any social media platform use
YouTube	97.2 (844)	95.9 (283)	96.9 (1,127)
TikTok*	86.8 (753)	91.2 (269)	87.9 (1,022)
Instagram*	85.0 (738)	90.2 (266)	86.3 (1,004)
Snapchat***	79.0 (686)	88.5 (261)	81.4 (947)
Pinterest	75.9 (659)	77.3 (228)	76.3 (887)
Facebook***	60.7 (527)	83.4 (246)	66.5 (773)
Twitter***	55.3 (480)	69.8 (206)	59.0 (686)
Discord*	40.9 (355)	48.1 (142)	42.7 (497)
Reddit***	35.0 (304)	55.9 (165)	40.3 (469)
Twitch***	30.5 (265)	45.8 (135)	34.4 (400)
BeReal*	23.5 (204)	30.2 (89)	25.2 (293)
Slack***	9.8 (85)	20.0 (59)	12.4 (144)
Social media daily use frequency
YouTube	60.6 (526)	60.7 (179)	60.6 (705)
TikTok	71.9 (624)	75.3 (222)	72.7 (846)
Instagram	60.0 (521)	63.1 (186)	60.8 (707)
Snapchat**	56.1 (487)	62.4 (184)	57.7 (671)
Pinterest	26.8 (233)	27.1 (80)	26.9 (313)
Facebook***	28.6 (248)	48.1 (142)	33.5 (390)
Twitter***	17.3 (150)	26.4 (78)	19.6 (228)
Discord**	14.7 (128)	12.5 (37)	14.2 (165)
Reddit***	7.4 (64)	14.9 (44)	9.3 (108)
Twitch***	6.8 (59)	10.5 (31)	7.7 (90)
BeReal*	10.1 (88)	10.2 (30)	10.1 (118)
Slack***	3.0 (26)	4.1 (12)	3.3 (38)
Ever NTP use
Never NTP use			55.9 (650)
Any ever NTP use			44.1 (513)
E-cigarettes with nicotine			67.3 (345)
Cigarettes			48.3 (248)
Other combustible NTP^a			51.5 (264)
Other noncombustible NTP^b			24.0 (123)
Current (i.e., past-month) NTP use
No current NTP use			74.6 (868)
Any current NTP use			25.4 (295)
E-cigarettes with nicotine			57.6 (170)
Cigarettes			35.6 (105)
Other combustible NTP^a			34.2 (101)
Other noncombustible NTP^b			14.9 (44)

Note: Differences in any (versus no) social media platform use and daily (versus non-daily or never) social media platform use by NTP use status were examined using chi-square tests of independence and are designated with: *p < .05 **p < .01 ***p < .001

^aIncludes cigars, cigarillos, little cigars, pipe filled with tobacco, hookah/shisha

^bIncludes chewing tobacco, snuff, snus, nicotine pouches, nicotine gummies, IQOS

Aim 1: Describing NTP Content Exposure

Reports of NTP content exposure (Table 3) were prevalent (72.7% reporting any exposure) and included both content promoting (65.2%) and content opposing (82.7%) NTP use. Most commonly reported content sources were companies/brands (46.6%) and influencers (44.4%), rather than friends (31.7%) or acquaintances (23.6%). Most commonly reported content formats were videos or “reels” (65.4%) and posts with pictures (50.7%). Exposure to content promoting NTP use was most prevalent on TikTok (49.3%), Instagram (44.7%), and Snapchat (41.8%); exposure to content opposing NTP use was most prevalent on YouTube (75.8%), TikTok (63.9%), and Instagram (63.0%; Supplemental Table 1).

Table 3. Reported nicotine and tobacco product (NTP) content exposure by NTP use status in a sample of adolescents and young adults surveyed in 2023

	No current NTP use (n = 872)	Current NTP use (n = 291)	Full sample (N = 1,163)
	% (n)
NTP content exposure frequency***
Never	32.7 (284)	11.5 (34)	27.3 (318)
Only once or twice	19.1 (166)	19.3 (57)	19.2 (223)
Every few months to every few weeks	29.3 (254)	31.2 (92)	29.8 (346)
Daily or more	18.9 (164)	38.0 (112)	23.7 (276)
Any exposure	67.3 (584)	88.5 (261)	72.7 (845)
	No current NTP use (n = 584)	Current NTP use	Full sample (n = 845)
		(n = 261)
Content valence (among those with any content exposure, n = 845)
Any “anti-NTP use” content***	79.1 (462)	90.8 (236)	82.7 (698)
Any “pro-NTP use” content***	59.9 (350)	76.9 (200)	65.2 (550)
Content sources (among those with any content exposure, n = 845)
Companies/brands***	42.5 (248)	55.9 (146)	46.6 (394)
Influencers**	40.8 (238)	52.5 (137)	44.4 (375)
Friends***	24.5 (143)	47.9 (125)	31.7 (268)
Acquaintances*	21.1 (123)	29.1 (76)	23.6 (199)
Content format (among those with any content exposure, n = 845)
Posts with pictures	49.5 (289)	53.3 (139)	50.7 (428)
Posts with only text***	18.3 (107)	34.1 (89)	23.2 (196)
Videos or “reels”**	62.2 (584)	72.8 (190)	65.4 (553)
Transient content^a ***	24.8 (145)	38.3 (100)	29.0 (245)

Note: Differences by NTP use status were examined using chi-square tests of independence and are designated with: *p < .05 **p < .01 ***p < .001

^a“Content that goes away after a period of time (e.g., stories, snaps)”

Aim 2: Describing NTP Content Engagement

NTP content engagement (Table 4) was reported by 34.6% of participants who had ever been exposed to NTP content (n = 845), 22.6% of whom reported engaging with NTP content at least weekly (42.1% every few months to weeks, 35.3% only once or twice). Participants most commonly engaged with content from influencers (57.2%) and friends (47.9%), and less commonly engaged with content from companies/brands (35.6%) and acquaintances (30.5%). Videos or “reels” (58.9%) and posts with pictures (54.1%) were the most common content formats for engagement. Among participants with any NTP content exposure, 43.2% reported having posted about NTPs themselves (20.8% at least weekly). Engagement with content promoting NTP use was most prevalent on TikTok (21.1%), Instagram (20.5%), and Snapchat (20.4%); engagement with content opposing NTP use was most prevalent on YouTube (22.6%), TikTok (22.3%), and Instagram (22.2%; Supplemental Table 1).

Table 4. Reported nicotine or tobacco product (NTP) content engagement by NTP use status, among those with any NTP content exposure (N = 845) in a sample of adolescents and young adults surveyed in 2023

	No current NTP use (n = 584)	Current NTP use (n = 261)	Full sample (N = 845)
	% (n)
Ever liked, commented on, or shared NTP post***	24.1 (141)	57.9 (151)	34.6 (292)
Only once or twice	40.4 (57)	30.5 (46)	35.3 (103)
Every few months to weeks	41.8 (59)	42.4 (64)	42.1 (123)
At least weekly	17.7 (25)	27.2 (41)	22.6 (66)
Content sources^a
Companies/brands**	27.0 (38)	43.7 (66)	35.6 (104)
Influencers	53.9 (76)	60.3 (91)	57.2 (167)
Friends**	39.0 (55)	56.3 (85)	47.9 (140)
Acquaintances*	24.1 (34)	36.4 (55)	30.5 (89)
Content format^a
Posts with pictures	48.2 (68)	59.6 (90)	54.1 (158)
Posts with only text***	19.1 (27)	41.1 (62)	30.5 (89)
Videos or “reels”	56.0 (79)	61.6 (93)	58.9 (172)
Transient content^b	22.0 (31)	29.1 (44)	25.7 (75)
Ever posted about NTPs***	8.9 (52)	48.5 (126)	21.1 (178)
Only once or twice	55.8 (29)	38.1 (48)	43.3 (77)
Every few months to weeks	30.8 (16)	38.1 (48)	36.0 (64)
At least weekly	13.5 (7)	23.8 (30)	20.8 (37)

Note: Differences by NTP use status were examined using chi-square tests of independence and are designated with: *p < .05 **p < .01 ***p < .001

^aAmong those who ever liked, commented on, or shared NTP content (no current NTP use: n = 141; current NTP use: n = 151; full sample: N = 292)

^b“Content that goes away after a period of time (e.g., stories, snaps)”

Aim 3: Comparing Exposure and Engagement by NTP Use Status

NTP content exposure. Differences in NTP content exposure by NTP use status are reported in Table 3. NTP content exposure was significantly more prevalent and frequent among those with (versus without) current NTP use (88.5% vs. 67.3% reporting any exposure, p < .001). Among those reporting any content exposure, 82.7% were exposed to content opposing NTP use (90.8% with vs. 79.1% without current NTP use, p < .001), and 65.2% were exposed to content promoting NTP use (76.9% with vs. 59.9% without current NTP use, p < .001). Participants differed by NTP use status in prevalence of content exposure from companies/brands (55.9% with vs. 42.5% without current NTP use, p < .001), influencers (52.5% with vs. 40.8% without current NTP use, p = .002), friends (47.9% with vs. 24.5% without current NTP use, p < .001), and acquaintances (29.1% with vs. 21.1% without current NTP use, p = .011). NTP use status was not associated with seeing posts with pictures (p = .311); however, those with versus without NTP use were significantly more likely to have seen posts with only text (34.1% vs. 18.3%, p < .001), videos or “reels” (72.8% vs. 62.2%, p = .003), and transient content (e.g., “stories,” snaps; 38.3% vs. 24.8%, p < .001).

NTP content engagement. Among those with any NTP content exposure (n = 845), participants with (versus without) current NTP use were significantly more likely to report liking, commenting on, or sharing an NTP post (57.9% vs. 24.1%; p < .001); to engage with content from companies/brands (43.7% vs. 27.0%; p = .003), friends (56.3% vs. 39.0%; p = .003), and acquaintances (36.4% vs. 24.1%; p = .022); to engage with text-only posts (41.1% vs. 19.1%; p < .001); and to post about NTPs (48.5% vs. 8.9%; p < .001). NTP use status was not associated with engaging with content from influencers or with engaging with posts with pictures, videos or “reels,” or transient content (p-values > .05). Significantly greater proportions of participants with (vs. without) current NTP use engaged with both pro- and anti-NTP use content on nearly all platforms (p-values < .05; Supplemental Table 1).

Supplemental Analysis: Describing Characteristics of Most Recent NTP Content Exposure

Among participants reporting exposure to content promoting NTP use (n = 550) and exposure to content opposing NTP use (n = 698), most recently viewed content (Supplemental Table 2) most commonly came from an influencer or celebrity (promoting: 33.5%; opposing: 30.5%) or company/brand (promoting: 21.6%; opposing: 28.9%). Most common platforms were Instagram (promoting: 26.5%; opposing: 17.0%), TikTok (promoting: 23.3%; opposing: 26.2%) and YouTube (promoting: 9.5%; opposing: 21.8%). Most participants reported viewing NTP content recently (promoting: 21.3% within the past week, 31.3% within the past month; opposing: 23.5% within the past week, 33.2% within the past month). E-cigarettes (promoting: 67.5%; opposing: 70.6%) and cigarettes (promoting: 27.8%; opposing: 39.3%) were commonly featured. Among those with pro-NTP content exposure, the most commonly featured flavors were fruit (39.5%), mint (18.7%), and menthol (17.3%); however, a sizable proportion (28.9%) did not remember which flavor(s) were featured.

Discussion

In this cross-sectional survey, AYAs in the U.S. endorsed seeing NTP-related social media content that both promoted and opposed NTP use. NTP-related content came from a variety of sources (e.g., influencers, friends), and mostly included imagery and/or videos. A majority of AYAs with current NTP use reported liking, commenting on, or sharing an NTP-related post, especially from influencers and friends. More participants reported engaging with content that had imagery and/or videos than text-only content. Nearly half of AYAs who currently use NTPs and had ever been exposed to NTP content had also posted about NTPs themselves. Exposure to and engagement with content promoting NTP use were most common on Instagram, Snapchat, and TikTok, while content opposing NTP use was predominant on Instagram, TikTok, and YouTube. Among AYAs who had reported any exposure to NTP content, most had seen NTP content within the past month.

The observed frequency of exposure to NTP-related content (72.7%) is consistent with findings from the 2021 National Youth Tobacco Survey, a nationally representative sample of U.S. middle and high school students (Gentzke et al., 2022). The current study additionally found that, consistent with our hypothesis, a greater proportion of AYAs with (versus without) current NTP use were exposed to content promoting and/or content opposing NTP use. Among AYAs already using NTPs, exposure to content promoting NTP use—particularly imagery—could trigger urges to use NTPs (King et al., 2016). However, the high prevalence of exposure to content opposing NTP use (90.8% of those reporting any content exposure) suggests that NTP educational messaging is also reaching AYAs who use NTPs. In addition to educational messaging, content opposing NTP use was reportedly posted by friends and acquaintances. Social media may help establish social norms of abstaining from NTP use; however, promotion of NTPs on social media may further increase their popularity. Promotional content that participants reported viewing most recently commonly featured NTPs (i.e., cigarettes, e-cigarettes, cigars, little cigars or cigarillos) and flavors (i.e., fruit, mint, menthol) that are popular with AYAs (Birdsey et al., 2023; Gaiha, Lempert, McKelvey, & Halpern-Felsher, 2022; Patrick, Miech, Johnston, & O’Malley, 2023). NTP content exposure is associated with subsequent NTP use, and may play a causal role (Cavazos-Rehg et al., 2021; Clendennen, Loukas, Vandewater, Perry, & Wilkinson, 2020; Donaldson, Dormanesh, Perez, Majmundar, & Allem, 2022; Pérez, Spells, Bluestein, Harrell, & Hébert, 2022; Ranker et al., 2024; Vassey, Galimov, Kennedy, Vogel, & Unger, 2022; Vogel et al., 2021). The current study suggests that although exposure to beneficial content opposing NTP use is prevalent, exposure to potentially harmful content promoting NTP use (especially e-cigarette use) remains widespread.

Exposure to content promoting and opposing NTP use came from a variety of sources, including sponsored sources (i.e., companies/brands and influencers) and non-sponsored sources (e.g., friends, acquaintances). Among AYAs with and without current tobacco use, recalling exposure to sponsored content was more prevalent than recalling exposure to non-sponsored content. It is unclear whether exposure to sponsored content was truly more prevalent, or simply more memorable. Promotional content is designed to be impactful (i.e., to influence purchasing behavior), whereas NTP content posted by friends and acquaintances may be perceived as peers simply documenting their daily lives (Vogel et al., 2023). Promotional content and non-sponsored content may affect NTP use behavior through different mechanisms. Influencer marketing can be used to promote NTP use as part of an aspirational lifestyle (Jackler et al., 2019). In alignment with the Prototype Willingness Model (Gibbons & Gerrard, 1995; Gibbons, Gerrard, Blanton, & Russell, 1998), influencer marketing may portray purported benefits of NTP use (e.g., fitting in socially, looking attractive), leading AYAs to view the influencers positively and emulate their behavior. Concurrently, seeing peers in one’s own social network use NTPs on social media may normalize NTP use by making it seem as though such use is more common and socially acceptable among AYAs than it actually is (Nesi, Rothenberg, Hussong, & Jackson, 2017).

Engagement with NTP content, in the forms of “liking,” commenting on, sharing, or posting NTP content, was significantly more prevalent among those with (versus without) current NTP use. Among AYAs reporting any exposure to NTP content who did not currently use NTPs, 8.9% reported posting about NTPs themselves. While a small proportion, AYAs who engage with NTP content may be at high risk for initiating or resuming NTP use (Ranker et al., 2024; Vogel, Barrington-Trimis, Vassey, Soto, & Unger, 2024). Importantly, engagement with content opposing NTP use was more prevalent than engagement with content promoting NTP use on every social media platform, among those not currently using NTPs. AYAs who do not use NTPs may share NTP educational messaging with their own social networks, thereby amplifying NTP use prevention messages. Among those reporting both current NTP use and exposure to NTP content, a majority (57.9%) had liked, commented on, or shared NTP content, and nearly half (48.5%) had posted their own NTP content. Among those with and without current use, influencer content was more engaging than other content sources, and videos were the most engaging content format. In response to surveyed AYAs’ interest in having a social media-based support system for vaping cessation and seeing influencers quit vaping, Truth Initiative’s “Quit Together” campaign worked with influencers to produce content opposing NTP use in video form (Truth Initiative, 2021). Platforms with reportedly engaging content included Instagram, Snapchat, TikTok, and YouTube, all of which are entirely video-based or have a strong video component. The engaging nature of video content presents a challenge for regulation, but an opportunity for public health campaigns.

NTP use was associated with more frequent use of Snapchat—one of the top platforms where AYAs reported exposure to NTP content. Prior research has also found Snapchat use to be associated with NTP use among AYAs (Fuemmeler et al., 2024; Lee et al., 2021; Massey, Brockenberry, & Harrell, 2021). One of Snapchat’s defining features is that content is only visible for a limited time, and the platform alerts the person who posted the content when someone takes a screenshot of it (Antonelli, 2023). Underage AYAs may post about NTP use on Snapchat because the content is less likely to be found by parents or employers. Consistent with the Prototype Willingness Model (Gibbons & Gerrard, 1995; Gibbons, Gerrard, Blanton, & Russell, 1998), seeing glamorized NTP content from peers on Snapchat may increase AYAs’ willingness to use NTPs. Mounting evidence highlights that greater enforcement of Snapchat’s policies surrounding content that promotes NTP use (Kong et al., 2022) is needed. Recent research suggests that viewing NTP ads on Snapchat may explain associations between Snapchat use and NTP use (Fuemmeler et al., 2024). Examining other potential mechanisms of associations between Snapchat use and NTP use—including, but not limited to beliefs that NTP content on Snapchat will go undetected—may help further guide prevention and intervention efforts. Although participants in this study more frequently reported seeing content opposing NTP use on YouTube than on Snapchat, prior research pilot testing NTP educational messaging found greater reach on Snapchat than on YouTube (England et al., 2021).

Limitations and Future Directions

Results should be interpreted in the context of several limitations. First, the sample of AYAs was non-representative and we oversampled AYAs with current NTP use; therefore, prevalence estimates should be interpreted with caution. Although the sample was diverse in race, ethnicity, and sexual orientation, the majority of participants (79.3%) identified as girls or women. A prior study found that among AYAs who had previously vaped, females were more likely than males to report being influenced to vape by social media content (Davidson & Al-Hamdani, 2023). Greater proportions of men/boys and non-binary individuals are needed to achieve a more holistic understanding of AYAs’ NTP-related social media content exposure and engagement. However, gender was not significantly associated with current NTP use in this sample. Second, due to survey space limitations and concerns about participant burden, not all survey items were separated by content valence (i.e., promoting or opposing NTP use), and different levels of engagement (i.e., “liking” versus commenting versus posting) were not disaggregated. Prevalence of specific content sources and formats may differ between content promoting and content opposing NTP use. Qualitative research could explore specific characteristics of exposure to and engagement with content promoting and opposing NTP use, to capture details of AYAs’ experiences with NTP content that may have been missed in this survey study. Third, “public health campaign” or similar phrasing was not a provided response option. Participants may have classified these content sources as “companies/brands,” “influencers,” or neither. Future research should provide a response option for NTP educational messaging.

Fourth, all measures were self-reported and results may be subject to recall bias, such that participants did not accurately report their content exposure or engagement. However, for the current research question, participants’ perceptions of their content exposure and engagement were of greatest interest. Self-report measures enabled us to capture content that was most memorable to participants. One purpose of the study was to examine frequency of NTP content exposure and engagement, which we suspected may be infrequent experiences for many AYAs. Most survey items queried ever-exposure and ever-engagement (i.e., not necessarily recent) so as not to exclude NTP content exposure and engagement that were memorable to participants but may not have occurred recently. Most participants reported fairly recent content exposure (within the past week or the past month). Self-reported characteristics (i.e., source, platform) of participants’ most recently viewed NTP content were concordant with their reports of NTP content characteristics across time. Although all retrospective self-report measures are subject to recall bias, this concordance suggests participants may have been able to accurately recall the NTP content that was most salient to them across time. Future research could compare self-reported survey data with participant-provided records of their social media activity, which can be downloaded from most major social media platforms (e.g., Facebook, Instagram, Snapchat, TikTok), to examine concordance between measures at the individual level.

Implications and Conclusions

Exposure to social media content opposing NTP use was pervasive among AYAs both with and without current NTP use. However, content promoting NTP use, including influencer marketing, continues to reach AYAs on popular platforms such as Snapchat and TikTok. Despite most social media platforms restricting some NTP promotional content (Kong et al., 2022), this study suggests that AYAs, especially those who use NTPs, continued encountering social media content promoting NTPs from both advertisers and peers in 2023. These results support a need for stronger enforcement of comprehensive NTP content restrictions on popular social media platforms (Vassey & Unger, 2023). Methods such as computer vision analysis may be useful in detecting NTP content that violates platforms’ policies (Murthy et al., 2024; Wu et al., 2023). Video-based NTP educational messaging on platforms where promotional messaging is common (e.g., Instagram, TikTok) may help counteract promotional messaging by engaging AYAs. Results also support a need for transparency in social media algorithms, which may target AYAs who use or are susceptible to using NTPs. Legislation introduced by the U.S. Congress would require social media platforms to disclose how minors’ information is used, report harms to minors that result from their platforms, and refrain from advertising NTPs and other age-restricted products to minors (Congressional Research Service [CRS], 2023). Social media content that glamorizes NTP use may be especially persuasive to AYAs, who are highly susceptible to social influence. As the social media landscape evolves, content promoting NTP use should be restricted to protect AYAs.

Acknowledgments

This study is a cross-institution collaborative project from the Marketing Influences Special Interest Group supported, in part, by the Center for Coordination of Analytics, Science, Enhancement and Logistics (CASEL) in Tobacco Regulatory Science U54DA046060-01 (National Institute of Drug Abuse [NIDA] and the Food and Drug Administration’s Center for Tobacco Products [FDA CTP]). Additional support was provided by NIDA (K01 DA055073); the National Heart, Lung, and Blood Institute (NHLBI) and the FDA CTP (U54HL120163); the American Heart Association (20YVNR35500014); NIDA and the FDA CTP (U54 DA036151, R01 DA049878); the National Cancer Institute (NCI) and FDA (U54CA228110); NCI (P30CA225520) and the Oklahoma Tobacco Settlement Endowment Trust (R22-03).

Disclosure Statement

The authors have no conflicts of interest to declare. Content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health or the Food and Drug Administration.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author, EAV, upon reasonable request.

Supplementary Material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/10810730.2024.2355291

Additional information

Funding

The work was supported by the American Heart Association [20YVNR35500014]; National Cancer Institute [P30CA225520]; National Heart, Lung, and Blood Institute [U54HL120163]; National Institute on Drug Abuse [K01DA055073]; Tobacco Settlement Endowment Trust [R22-03].