Overall, a majority of youth younger than 18 years expressed support for T21 (Table 1). Younger students aged 11–14 years were more supportive of T21 than older students aged 15–17 years (72.5 vs. 58.5%, respectively). Less than half (42.9%) of students aged 18 years and older supported T21. Greater support for T21 was observed among females (66.5% vs. 61.8% for males), tobacco nonusers (70.9% vs. 30.3% for users of tobacco in any form [data not shown]), and students without a tobacco user in their household (68.4% vs. 57.8% who live with a tobacco user). Over two-thirds (68.8%) of students who believed tobacco products are dangerous supported T21 compared to 35.1% who held opposite risk conceptions.

Among current nonusers, support for T21 varied inversely by receptivity to protobacco peer influence. Over three-fourths (76.5%) of nonusers who reported not being receptive to using tobacco products if offered by a friend supported the proposed age restriction. Among nonusers with low receptivity, 70.4% supported T21, while 51.9% of those most vulnerable to peer influence expressed support.

Among users, support varied by product type used, from 18.8% among multiproduct users to 49.5% among exclusive users of e-cigarettes. Among users who recently made a direct tobacco purchase, 1 in 10 (12.1%) supported T21. Among users who obtained tobacco through social sources, 1 in 4 (25.0%) supported T21. Similar support was observed among users who reported taking tobacco products or getting them some other way (24.0%).

Among nonusers, adjusted results (Table 2) revealed that the odds of supporting T21 differed based on receptivity to peer influence. Compared with highly receptive students, those with low or no receptivity had two-fold higher odds (low receptivity, aOR = 1.9; nonreceptivity, aOR = 2.5) of supporting T21. In contrast, users who bought tobacco through a direct purchase had 70% lower odds (aOR = .3) of supporting T21 (Table 3). Similarly, users who reported recently accessing tobacco through social sources or obtaining tobacco products through alternate means had 30% (aOR = .7) and 40% (aOR = .6) lower odds, respectively, of supporting T21.

This study demonstrates that the majority of middle and high school students support T21. Given adults’ majority support of T21 [21–25], findings reinforce that T21 favorability spans generations. Among tobacco nonusers, the strongest support for T21 was found among students with lower receptivity to pro-tobacco peer influence. Among tobacco users, those who accessed tobacco products through direct purchase, social sources, or other means (e.g., theft) were less supportive of T21 compared with those who did not report accessing tobacco in these ways. Exclusive users of e-cigarettes had greater odds of supporting T21 compared with users of other types of tobacco products, possibly reflecting confusion about e-cigarettes’ categorization as a tobacco product [5] and an assumption that e-cigarettes would not be included in T21 policies. Those who were younger (i.e., further away from proposed age restriction) and those who believed tobacco products are dangerous had greater odds of supporting T21. These findings indicate that individual, interpersonal, and community factors are associated with T21 support among youth, which could serve as a useful evidence base for states and communities considering T21.

The Family Smoking Prevention and Tobacco Control Act of 2009 (“Tobacco Control Act”) gave the U.S. Food and Drug Administration (FDA) broad authority over the manufacturing, marketing, and sale of tobacco products. This authority granted by the U.S. Congress allows FDA, among other regulatory powers, the right to restrict advertising and cigarette and smokeless tobacco sales to youth [7]; as of 2016, the FDA’s authority further extends to a wider range of tobacco products, including e-cigarettes [5]. The Tobacco Control Act, however, prohibits the FDA from setting a national minimum legal age above 18 years for selling tobacco products. Importantly, the Tobacco Control Act does not preempt state and local efforts with regard to many tobacco prevention and control strategies, including T21. Therefore, efforts to increase the minimum legal age above 18 years are left to states and local jurisdictions. Presently, 25% of the U.S. population is covered by a T21 policy at the state or local level (http://tobacco21.org).

Under the assumption of a national T21 policy, simulation studies have been conducted to estimate the public health implications of an increased minimum legal age for tobacco product sales. Over time, tobacco use initiation among adolescents aged 15‒17 years would be expected to decline 25% following implementation of a national T21 policy. Researchers predicted that, by 2100, smoking prevalence among adults would drop 12% beyond expected declines, resulting from current tobacco control policies. Furthermore, 250,000 smoking-related deaths would be averted, including 36% premature deaths to persons < 65 years of age [7]. Absent a uniform minimum legal age of 21 years for tobacco product sales, these predicted outcomes depend on collective state and local action to restrict legal access to tobacco. In real-world conditions, change will depend on specific T21 policy elements, adequate policy implementation and enforcement, and sufficient time for shifts in social norms to occur [32,33].

To offer the full scope of public health protection across the numerous types of tobacco products being used in the U.S. marketplace, model T21 policies address the diversity of tobacco products (specifically including e-cigarettes and other electronic products [9]), emphasize penalties and enforcement for under-age sales (not possession), carry no exemptions for special populations (e.g., military), and disallow pre-emption of stronger, local ordinances [34]. To protect youth and young adults from the harms of nicotine addiction, the U.S. Surgeon General outlined a multipronged, evidence-based strategy that includes increasing and enforcing minimum age laws for selling tobacco products, including e-cigarettes [5]. Given e-cigarettes’ potential to renormalize smoking, and the growing body of science suggesting that e-cigarette use among youth may lead to subsequent conventional cigarette smoking [35,36], continued examination of attitudes toward T21 by product use is warranted. To curtail underage access, enforcement of age restrictions is a critical component of T21 laws aimed primarily at the retail environment [7]. Frequently cited best practices for implementing T21 include intentional planning of enforcement strategies, compliance inspections, effective penalties for violations (including criminal prosecution), and merchant education [32].

T21 policies can also help shape societal norms and reinforce the denormalization of tobacco use behaviors. At a group or organizational level (e.g., school, sports team), compliance with T21 policies (and, otherwise, smoke-free campus policies) may impact actual and perceived tobacco use prevalence norms, both of which are associated with tobacco product susceptibility and/or use [28,37]. In family systems, parental controls (e.g., home punishments) possibly in response to legal and social risks posed by violations of T21 laws that include penalties for individuals’ possessing, using, and furnishing tobacco to under-age persons may further denormalize youth tobacco use [7]. Over time, a decline in tobacco prevalence may also be accelerated by younger students’ behavioral modeling of their older peers who may increasingly adapt to shifting social norms unfavorable to tobacco use [32]. Support for T21 is associated with lower intentions to initiate cigarette smoking and increased intentions to quit [26]. Thus, as an indicator of social norms, periodic assessment of youth attitudes toward T21 can be useful to examine associated changes in tobacco use susceptibility, initiation, and prevalence.

Our study is subject to some limitations. First, given the study’s cross-sectional design, we are unable to ascertain whether factors had a temporal influence on support for T21. Absent longitudinal data, we can only detect that certain associations exist. Second, the receptivity to protobacco peer influence scale was newly developed for this study; its psycho-metric properties are unknown without further validation. Third, because students self-reported their attitudes and tobacco use behavior, estimates are subject to response and recall biases; however, research supports the validity of self-reporting smoking behavior among youth [38]. Finally, the findings might not be generalizable to all youth; those who are home-schooled, have dropped out of school, or are in detention centers are not included in this survey. However, despite these limitations, overall trends are generally similar to other nationally representative surveys of tobacco use among youth [39,40].

T21 policies reflect one of the several strategies that comprise a comprehensive approach to population-based tobacco prevention and control [5]. As a strategy to reduce tobacco use initiation, T21 garners support that spans generations. Results demonstrate that individual, interpersonal, and community factors influence support of T21 among youth. For states and communities considering T21, results underscore the importance of taking a social-ecological approach to T21 implementation.