Prohibiting juvenile access to tobacco: Violation rates, cigarette sales, and youth smoking

Highlights

• The efficacy of juvenile tobacco sales prohibitions can be empirically tested.

• States test retailer violation rates of laws prohibiting juvenile tobacco sales.

• Violation rates are related to overall cigarette sales and to youth smoking.

• Juvenile sales prohibition laws appear to have some effect on youth smoking.

Abstract

Background

Scholars who examine the efficacy of juvenile tobacco sales restrictions, especially the 1992 “Synar Amendment” that led all of fifty U.S. states to enact prohibitions on tobacco sales to minors, are notably divided as to impact on youth smoking. Some researchers claim that such policies have failed and ought to be abandoned (Craig and Boris, 2007, Etter, 2006, Glantz, 2002), while others insist that enforcement has indeed led to reduced tobacco use (DiFranza, 2011b, SAMHSA, 2011). The present study is the first to combine data on Synar violation rates from all states and years available since the amendment's implementation, assessing the connection to national rates of cigarette sales and youth smoking behavior.

Methods

Using national data from the United States Substance Abuse and Mental Health Services Administration, the Tobacco Institute, and the Centers for Disease Control's Youth Risk Behavior Surveillance System across all U.S. states between 1996 and 2007, we employ hierarchical linear modeling to examine the connection between retailer Synar violations and youth smoking.

Results

Controlling for state-level demographic variables, results indicate that retailer violation rates are significantly associated with greater youth smoking prevalence, as well as higher overall cigarette sales.

Conclusion

While critiques of Synar policies are substantive and should be addressed, laws prohibiting the sale of tobacco to juveniles appear to have had some degree of success.

Introduction

The 1992 federal Synar Amendment eventually brought about state-level youth tobacco sales prohibitions in all states, but researchers remain at odds about questions of the validity of enforcement measures (undercover compliance tests) and the public health impact of reducing youth smoking. While some scholars insist that Synar policies are ineffective (Craig and Boris, 2007, Etter, 2006, Glantz, 2002), others maintain that restricting sales of tobacco to juveniles has led directly to reductions in youth smoking (DiFranza, 2011b, SAMHSA, 2011).

This study is the first to examine retailer violation rates of Synar compliance testing across all U.S. states in all the years for which data have been released (1996–2007), assessing the extent to which violation is related to overall tobacco sales and to rates of youth smoking.

Using combined data from (1) the Substance Abuse and Mental Health Services Administration (SAMHSA) on Synar compliance-check violation rates, (2) the now-disbanded Tobacco Institute's (TI) data on per capita cigarette sales, (3) the Center for Disease Control's (CDC) Youth Risk Behavior Surveillance System (YRBSS) on juvenile smoking rates, and (4) social-demographic control measures from the U.S. Census, we assess the efficacy of Synar laws by testing hypotheses about the relationships between compliance-check violation, overall cigarette sales, and the prevalence of youth smoking.

The last two decades have seen a great deal of change in the implementation of regulations on youth access to tobacco, as well as a degree of controversy on the issue of whether such policies are effective. The debate has occurred not only between the tobacco industry and health organizations, but also within the public health community (DiFranza, 2011b, Fichtenberg and Glantz, 2002).

In 1986, twelve states (Colorado, Georgia, Kentucky, Louisiana, Missouri, Montana, New Hampshire, new Mexico, South Dakota, Virginia, Wisconsin, and Wyoming) still lacked any restrictions on tobacco sales to minors (DiFranza & Dussault, 2005). By the early 2000s, all fifty states had implemented “youth access” laws (Glantz, 2002); however, states have varied widely in the severity of their regulations, including requirements about sealed packaging, sales clerk intervention, purchaser identification, vending machine bans, graduated penalties, and compliance check frequency (Gallet, Hoover, & Lee, 2009).

The transition began in the early 1990s with the Synar Amendment, a U.S. congressional provision attached to a health administration reorganization act and named for its sponsor, Oklahoma congressman Mike Synar. The bill was enacted in 1992 but did not go into effect until 1996 when the final protocols were established (DiFranza, 2011a, Gruber, 2001, SAMHSA, 2009, U.S. Title, 1996). In that year, the Food and Drug Administration (FDA) implemented a regulation that made sales of tobacco to minors a federal offense, and had begun developing a national enforcement program in coordination with the states when the Supreme Court ruled that tobacco was outside of the FDA's jurisdiction (Food and Drug Administration vs. Brown and Williamson Tobacco, 2000). The Substance Abuse and Mental Health Services Administration (SAMHSA) was then tasked with coordinating implementation.

The statue required states to enact “… a law providing that it is unlawful for any manufacturer, retailer, or distributor of tobacco products to sell or distribute any such product to any individual under age 18” (U.S. Title 45: 96.130b). Additionally, the Synar amendment provided for enforcement and accountability by stating that “the State shall, at a minimum, conduct annually a reasonable number of random, unannounced inspections of outlets to ensure compliance with the law” (U.S. Title 45: 96.130c-e). The language further specified that the compliance-check inspections must be conducted across a probability sample of outlets throughout the state, and with a sample of youth decoys representative of the state's under-18 population.

States are furthermore required to provide detailed annual reports on their methodological compliance and the results of the inspections, and to maintain violations rates less than 20 percent to avoid reductions in SAMHSA funding – up to 40 percent of Block Grant funds for Substance Abuse Prevention and Treatment (DiFranza and Dussault, 2005, SAMHSA, 2011).

SAMHSA touts the success of the Synar program by noting that average violation rates have dropped from 40 percent to 10 percent between 1997 and 2007, and that all 50 states, as well as the District of Columbia, have been in compliance with the minimum 20 percent violation rate since 2006 (SAMHSA, 2009, SAMHSA, 2011, State Synar Rate Tables, 2011). Furthermore, the agency claims that Youth Risk Behavior Survey (YRBS) results support Synar's effect on youth sales and youth tobacco use. In 1995, 38.7 percent of student smokers under age 18 purchased cigarettes from retail outlets, a percentage that dropped to 16.0 percent by 2007, and the percentage of students reporting current cigarette smoking declined from 36.4 percent in 1997 to 20.0 percent in 2007 (SAMHSA, 2011).

The agency's self-accolades did not prevent substantial criticism from some public health scholars, however, who saw the new policy as an echo of the tobacco industry's disingenuous attempts to co-opt the issue of youth smoking prevention (Ling, Landman, & Glantz, 2002). During the late 1980s and early 1990s, the Tobacco Institute, a Philip Morris sponsored trade organization, in conjunction with The Coalition for Responsible Tobacco Retailing (CRTR), had been promoting programs alleged to prevent youth smoking, particularly the popular “We Card” campaign. As one health research study described:

…the coalition's activities for more than a decade focused primarily on handing out materials, without any effort to determine whether they changed behavior, collecting endorsements, publicizing We Card, and praising the program's sponsors… the program's goal is to undermine enforcement of existing laws, prevent passage of effective state legislation, establish the tobacco industry as a ‘partner’ with state agencies, and burnish the public images of tobacco companies and retailers (Appolonio & Malone, 2010: 1193, 1196).

Some argued that tobacco industry support of juvenile sales prohibitions, including the “It's the Law,” “We Card,” and “Action Against Access” programs, simply reinforced the industry's marketing message that smoking is an adult activity, thereby making it more attractive to teenagers (Etter, 2006, Ling et al., 2002). While this “forbidden fruit” claim is eschewed by health proponents who believe that youth access restrictions can be effective when enforced (DiFranza, 2011b), evidence does support the contention that these programs were fronts for tobacco industry lobbying efforts. One internal Philip Morris document described strategy for combating “state and local laws and ordinances on ‘youth-related’ issues, such as advertising bans, vending machine bans and restrictions, punitive licensing schemes and other maters, couched as measures designed to ‘protect’ young people from tobacco” (Malmgren, 1992: 3). The document indicated that the program was a good way to coordinate tobacco industry lobbying efforts:

For monitoring purposes, we fund our allies in the convenience store group to regularly report on ordinance introductions and assist in campaigns… that reporting is complemented by other reporting mechanisms and channels such as member company sales representatives and other allied groups. Promotion of The Institute's [Tobacco Institute] ‘It's the Law’ program and other industry programs play a helpful role as well (Malmgren, 1992: 5)

Evidence also suggests that Philip Morris failed to impose its own prescribed sanctions (pricing discount suspensions) for stores that were found violating Synar compliance checks (Feighery, Schleicher, Ribisl, & Rogers, 2009).

One early test of the “It's the Law” campaign utilized Synar-style compliance checks with underage purchasers. Conducted seven months after the campaign's implementation, the study failed to find a difference in compliance rates between retailers who were participating in the program and those that were not (DiFranza & Brown, 1992). The authors noted that the tobacco industry's legislative activities surrounding the program included lobbying efforts to preempt existing tobacco control efforts and even to prohibit official Synar-style compliance operations. In contrast, an evaluation of one of the first compliance-check programs in Woodbridge, Illinois, showed substantial reductions in juvenile tobacco use (Jason, Ji, Anes, & Birkhead, 1991). By 1996, the new Synar amendment required enforcement efforts through compliance checks in all states, but not everyone agreed on whether enforcement efforts had any impact on either youth access to tobacco, or youth smoking.

The compliance check method, in which a juvenile is recruited to attempt a tobacco purchase from a retail outlet, is the federally sanctioned method for assessing Synar Amendment compliance/violation (DiFranza, Savageau, & Bouchard, 2001), and has also been utilized in Ontario, Canada (Dubray & Schwartz, 2010) and Hong Kong (Kan & Lau, 2010). Critics are quick to point out, however, that states develop many of their own procedural protocols for these checks, some of which restrict the behavior of the youthful recruits; preventing them from being smokers, appearing older than average, manipulating physical cues to appear older (e.g., make-up, jewelry, clothing), providing false information about themselves (e.g., telling stories or making conversation to insinuate being of legal age), or participating in compliance checks for stores where they are known to the clerk. Ironically, juvenile smokers report that knowing the vendor is an important factor in successfully completing illegal purchases (DiFranza & Coleman, 2001), and vendors have learned to be cautious of unknown youths while selling to known underage “regulars” (Landrine & Klonoff, 2003). One researcher thus complained that “the standard protocol prohibits the exact strategies that minors use to buy tobacco” (Etter, 2006: 48).

Additionally, states self-monitor the implementation of protocols for Synar compliance-check enforcement, and their federal funding for substance abuse treatment is at stake. Unsurprisingly, they tend to find themselves in compliance. Besides the strategies already described, other methods available to enforcement agents include using boys (who have lower purchase success rates) instead of girls, using younger rather than older teens (Clark et al., 2000, DiFranza and Coleman, 2001, DiFranza et al., 2001, Etter, 2006), and using white youths (who also have lower purchase success rates) instead of racial minority youths (Landrine et al., 2008).

Despite claims of flawed enforcement protocols, a good deal of research has been conducted on the impact of state Synar laws on youths’ ability to obtain tobacco, as well as the public health outcome of reducing youth smoking behavior. Some studies have noted a degree of success (Cummings et al., 2003, DiFranza et al., 2009, Ma et al., 2001, Siegel et al., 1999, Widome et al., 2007), but other research failed to find connections between retailer compliance and youth smoking (Bagott et al., 1998, Hublet et al., 2009, Rigotti et al., 1997, Staff et al., 2003), and some evidence even existed of racial and geographic disparities in violation rates (Asumda & Jordan, 2009).

Given such varied results and contentions about Synar policies and their alleged success/failure, some researchers attempted to implement extensive meta-analyses, but their conclusions were mixed, as well (Craig and Boris, 2007, Etter, 2006). Stead and Landcaster (2000) examined 27 studies that addressed retailer compliance, youths’ perceived access, and changes in smoking behavior. The integration of findings was qualitative, but greater weight was given to research that used controls. The authors concluded that (1) giving information to retailers – the strategies employed by “It's the Law” and “We Card” programs – was less effective than strategies that included enforcement, (2) youth-access programs generally did not appear to reduce teens’ perceptions about the ease of purchasing tobacco, and (3) the effect of youth-access programs could not be conclusively linked to reductions in youth tobacco use. The authors were cautious about interpreting the studies’ conclusions, and quick to point out the uncertainty in making assertions about the impact, if any, of the assessed programs. Richardson et al. (2008) were similarly ambiguous, reviewing 21 studies of Synar laws’ impact on sales to juveniles and teen smoking behavior. Only two of the studies assessed the latter question, and were inconclusive; however, active enforcement did tend to be modestly associated with reductions in sales to youths.

The two meta-analyses that have claimed definitive conclusions about Synar's efficacy reached opposing conclusions. Fichtenberg and Glantz (2002: 1088) conducted a meta-analysis of nine studies that assessed community-level connections between merchant compliance and teen smoking behavior. The authors concluded that greater compliance with Synar ordinances does not reduce teen smoking, and recommended that “given the limited resources available for tobacco control, as well as the expense of conducting youth access programs, tobacco control advocates should abandon this strategy.” DiFranza (2011b: 6), however, recently conducted a review of 19 studies that assess the impact of Synar prohibitions on both the prevalence of youth purchasing and tobacco use. The author confirmed the findings of previous meta-analyses that enforcement is essential for reductions in sales to juveniles, but went further, asserting that “every intervention that has successfully disrupted the sale of tobacco to minors has been associated with an observed reduction in tobacco use by youth. In this regard, the literature is unambiguous.”

Confounding these attempts to discern the true efficacy of tobacco sales prohibitions to juveniles is the fact that studies’ research questions vary in their definitions of different independent variables (e.g., prohibition campaigns without enforcement, with enforcement, type and intensity of enforcement) and dependent variables. However, if the goal of Synar laws is to reduce youth access, one consideration of research connecting enforcement to youth smoking is the extent to which – even if access from retailers really does decline – juveniles may be getting cigarettes from other sources.

Some public health scholars have asserted that reduced retail purchasing, due to effective Synar enforcement, resulted in increased acquisition of tobacco from social networks (Ahmad and Billimek, 2007, Huang et al., 2002). Anti-tobacco opponents of youth access restrictions, still convinced that these ordinances were too close to tobacco industry efforts and thus deflected resources from more effective tobacco control measures, utilized the displacement argument. Calling Synar programs “A Failed Intervention,” Glantz (2002: 301) claimed that “…while these programs do make it difficult for teens to purchase cigarettes, they do not affect teen smoking prevalence… as youth access laws make it harder to purchase cigarettes, teens simply get their cigarettes elsewhere.” With two colleagues (Ling et al., 2002: 4), he likewise concluded that “it is time for public health practitioners to recognize that the balance of empirical evidence shows that youth access is a failed strategy and abandon it.”

Is youth access to tobacco entirely displaced by alternative acquisition strategies? Some evidence does suggest that juveniles obtain cigarettes through social networks that reduce the effectiveness of retail restrictions (Craig and Boris, 2007, DiFranza and Coleman, 2001, Philip Morris USA, 2003). One survey of over 8000 seventh to tenth graders revealed that reduced retail access increased the likelihood of teens providing cigarettes to other teens (Pokorny, Jason, & Schoeny, 2006). Another survey of 250 college students found that nearly one-third (33.2 percent) had been asked to purchase cigarettes for minors, and nearly one-third of those (30.8 percent) had agreed. If the sample was representative of its intended population, the implication is that about one-in-ten young college men and women (more than 90 percent of the sample were between 18 and 24) have provided cigarettes to minors (Shive, Ma, & Shive, 2001).

Forster, Chen, Blaine, Perry, and Toomey (2003) examined responses from more than four thousand teens in the eighth, ninth, and tenth grades (ages 13–16 years), and found that almost 90 percent had obtained a cigarette from another teenager, while about 75 percent had provided cigarettes to another teen. Meanwhile, a survey of 1089 eighth, ninth, and tenth graders in Minnesota who identify themselves as smokers indicates that about two-thirds have provided cigarettes to other teens, with heavier smokers most likely to be providers (Wolfson, Forster, Claxton, & Murray, 1997).

Still, retailers appear to play a role in the social network chain. Teen smokers’ social networks may stem from smokers who have established regular methods for accessing retail cigarettes (e.g., establishing a relationship with a clerk, using a fake ID once and then utilizing the same source once known), thus counteracting the efforts of laws against juvenile sales (Jason, Pokorny, Muldowney, & Velez, 2005). In the UK, a different version of circumvention was observed in a qualitative study using focus groups of twelve to fifteen year olds (N = 85). While many did use social connections to obtain cigarettes, a majority purchased from local shops – they often knew which shops would sell to them – or solicited proxy sales by asking strangers (Robinson & Amos, 2010).

An internal Phillip Morris document, using 2001 data from the CDC's Youth Risk Behavior Survey, indicates that of high school students who smoked one or more cigarettes in the previous 30 days, 19 percent respond that they “usually” obtain their cigarettes directly from retailers. An additional 33 percent obtained cigarettes by “Giving someone Else Money to Buy,” and 27 percent by “Borrowing,” (Philip Morris USA, 2003: 4). These figures markedly contrasted with those from the Robert Wood Johnson Foundation's 1996 National Study of Tobacco Price Sensitivity, Behavior, and Attitudes among Teenagers and Young Adults (N = 17,287), in which slightly less than half of respondents aged 13–19 usually obtained cigarettes from noncommercial sources. However, the teens who usually used noncommercial sources were younger and smoked few cigarettes per day than those who made retail purchases themselves (Castrucci, Gerlach, Kaufman, & Orleans, 2002). Widome et al. (2007) observed data likewise suggesting that the displacement of retail sales to social networks is commensurate with a reduction in smoking overall, as teens in communities with greater prevalence of obtaining cigarettes from social sources than from retailers were less likely to become heavy smokers. Thus, while displacement is an important phenomenon that counteracts the efficacy of Synar laws, restricting youth access from retailers – to the extent that such laws actually achieve this result – may still significantly impact the prevalence of youth smoking.

Conclusions about the efficacy of Synar laws are difficult to draw from the available literature because, as discussed earlier in this paper, results are contradictory, operational definitions of measures vary, and questions surround the validity of enforcement (compliance-check) methodology. However, despite the issues surrounding Synar enforcement protocols, data released from SAMHSA on reported violation rates show considerable variation (see Table 1, Table 2, as well as Fig. 2), and to the extent that comparative violation rates may be related to actual cigarette sales, as well as to youth smoking, the data reflect a valid measure of retailers’ inclination to sell cigarettes to minors.

Previous studies have tended to compare rates of tobacco sales or youth smoking across samples of geographic areas for communities that implemented prohibitions versus those that did not (before Synar), or compared Synar violation/compliance with rates of youth smoking in particular communities or states. The present study is the first to combine data on Synar violation rates from all states and years available since the amendment's implementation, as well as the first to evaluate the impact of these data at a macro level to cigarette sales rates and youth smoking prevalence across the United States.

The central empirical questions regarding Synar laws and their public health impact involve whether such prohibitions lead to reduced cigarette access (fewer retail sales) to youth, and in turn to reduced youth smoking. Controlling for social demographic characteristics, we expect Synar violation rates to be positively associated with per capita cigarette sales. Although retail cigarette sales cannot be measured with regard to the proportion sold illegally to youth versus legally to adults, we suggest that greater willingness to engage in illegal sales to juveniles will lead to greater sales overall. Such an association would also support the contention that Synar compliance-check rates are not entirely farcical and contain some degree of validity in reflecting retailers’ inclination to sell tobacco to juveniles. Additionally, we expect that per capita cigarette sales will be positively associated with the prevalence of youth smoking. As the theoretical framework for Synar's public health impact calls for the law to hinder youth access and thus reduce youth smoking, reduced sales to juveniles should – ceteris paribus – be reflected in reduced sales overall. Finally, as depicted diagrammatically in Fig. 1, we expect that Synar violation will be positively associated with youth smoking, and that this relationship will be accounted for by per-capita sales.

We conceptualize sales as an intervening variable between the policy and its intended public health impact. Thus, greater failure to comply with Synar should lead to greater overall cigarette sales, and in turn to a greater prevalence of youth smoking.