Tobacco use among youth remains a major public health concern. Tobacco use is often initiated during adolescence (U.S. Department of Health and Human Services, 2014); and those who smoke in adulthood are more likely to have tried smoking before the age of 18 years (U.S. Department of Health and Human Services, 2014). In 2023, an estimated 15% of middle school and 28% of high school students in the United States (U.S.) reported ever using tobacco (Birdsey et al., 2023). A quarter of youth who currently use tobacco report polytobacco use (using ≥ 2 products) (Gentzke et al., 2022). Polytobacco use in adolescence increases the risk of nicotine dependence in adulthood (U.S. Department of Health and Human Services, 2012).

Social media (e.g., Instagram, Snapchat) is a critical environment for socializing and information seeking among adolescents (Best et al., 2014). Over 90% of 13–17 year olds use at least one social media platform. (Anderson et al., 2018) Increased time on social media platforms increases opportunities for exposure to tobacco-related content. One study documented over 200 Facebook fan pages and 100,000 pro-tobacco videos on YouTube (Liang et al., 2015). Such exposures may influence smoking attitudes, intentions, and behaviors among youth (Huang et al., 2014, Vogel et al., 2021). Prior work has found associations between social media use frequency and initiation of e-cigarettes (Sampasa-Kanyinga and Hamilton, 2018).

Tobacco companies have a long history of marketing targeted at youth (Cummings, 2002, Perry, 1999, Pollay, 2000). As individuals have started to spend more time online, companies have adapted and relied increasingly on online strategies (Pierce et al., 2018, Shan and Azagba, 2022, Soneji et al., 2018, Li et al., 2021, Padon et al., 2018, Donaldson et al., 2022). The U.S. Food and Drug Administration (FDA) has regulatory authority over the marketing and advertising of tobacco products. This authority was expanded in 2016 when congress passed the “Deeming rule,” giving the FDA regulatory authority over the manufacturing, distribution, and marketing of new and emerging tobacco products (Food et al., 2016). In spite of these regulations—which theoretically extend to online marketing practices—youth encounter content generated by tobacco brands (hereon called promotions) on social media (Liang et al., 2015, Huang et al., 2019, Campaign for Tobacco-Free Kids, 2018). To-date, the majority of the enforcement of and restrictions on social media-based tobacco promotions are undertaken at the discretion of social media platforms (Pierce and Gilpin, 2004). However, platform policies typically extend only to promotions paid directly via the site (i.e., sponsored posts), leaving loop holes for certain promotional strategies such as branded accounts, influencer collaborations, and fan pages (Kong et al., 2022). Brands popular among youth, such as JUUL, have been accused of predatory social media marketing practices including ads designed to appeal to youth, posts offering giveaways, and content highlighting products that are easily concealed and available in flavors appealing to a younger demographic (Huang et al., 2019, Kim et al., 2019, Lee et al., 2023).

Tobacco exposures on social media (both brand- and user-generated) may influence tobacco expectancies and behaviors via social learning, which suggests behaviors are learned by observing and modeling the behaviors of others (Mead et al., 2014). The often visual and attention grabbing nature of social media posts and positive feedback mechanisms, such as likes and comments, are foundational to social networks, making them optimized for social learning. Positive feedback (e.g., likes, followers, comments) from sharing or engaging with tobacco-related content on social media may directly enhance positive outcome expectancies among youth (Bandura, 1977). Effects of content exposure could also be more indirect wherein general exposure to tobacco content as well as observing the positive feedback others receive from sharing tobacco content increases the salience of the topic and reduces perceived harms (Zhang et al., 2024). Youth may be particularly vulnerable to tobacco content exposure on these platforms as prior content analyses of tobacco promotions on social media have found themes and imagery that may appeal to youth including: use of young-looking models, inclusion of smoking cues, depictions of socializing, and flavor indicators (e.g., fruit imagery) (Lee et al., 2023, Kostygina et al., 2023, Buente et al., 2020, Liu et al., 2018). Exposure to such content is further exacerbated by social media algorithms, which emphasize repetition and reinforcement. Repeated tobacco content on these platforms can therefore foster and reinforce misperceptions regarding the prevalence, severity, and acceptability of tobacco use (Elmore et al., 2017); and has been shown to alter purchasing intentions and tobacco use (Vogel et al., 2021, Pokhrel et al., 2018, Lee et al., 2021).

Prior work has found associations between exposure to promotional tobacco content and tobacco use among youth varying in magnitude from moderate (odds ratios of 1.1–1.6) to large effects (odds ratios of 2.0 or higher) (Pierce et al., 2018, Shan and Azagba, 2022, Soneji et al., 2018, Li et al., 2021, Padon et al., 2018, Donaldson et al., 2022, Zheng et al., 2021). Yet, the effect of social media use frequency and subsequent promotional exposures on tobacco initiation among previously tobacco naïve youth has not been studied longitudinally. It is critical to explore the role of engagement with promotional content among tobacco naïve youth as prior work suggests exposure to tobacco advertisements increases curiosity (Margolis et al., 2018, Donaldson et al., 2022) and such susceptibility is linked to experimentation (Strong et al., 2019). In addition, focus on a longitudinal sample with individuals with no prior tobacco use enhances research rigor by addressing temporality concerns in prior research that is cross-sectional or focused on those with a history of tobacco use (Donaldson et al., 2022).

Our study aims were to examine associations between 1) social media use frequency and 2) a measure of social media tobacco engagement (liking/following tobacco brands) with risk of tobacco initiation over approximately one year as a primary outcome within a tobacco naïve youth sample. Given concerns about polytobacco use in youth (U.S. Department of Health and Human Services, 2012); we also assessed polytobacco initiation. We used data from the Population Assessment for Tobacco and Health (PATH) study, a nationally representative sample of U.S. youth. We hypothesized (H1) that youth who report daily or more frequent social media use at analytic baseline (compared to less frequent) are at higher risk of initiating any tobacco use at follow-up. We hypothesized results would be similar for polytobacco initiation (H2). In addition, (H3) youth who report liking/following tobacco brands on social media (compared to those who do not), are at higher risk for using any tobacco and polytobacco initiation at follow-up.