In 2018, the Food and Drug Administration issued an advanced notice of proposed rulemaking to gather information on the role that flavors play in tobacco products. Specifically, the notice sought comments about whether and to what degree flavors compel youth to initiate tobacco product use, as well as whether and how flavors may help adults who smoke combustible cigarettes reduce their cigarette use and switch to potentially less-harmful products.

In proposing a rule that targets the availability of (non-tobacco) flavored e-cigarettes, the FDA is suggesting that these flavors are responsible for youth uptake of e-cigarettes. However, in order to justify removing a flavor from the market, the FDA would have to show not only that a specific flavor is the cause of e-cigarette use, but that such use is not displacing use of combustible cigarettes.

In an attempt to elucidate the role that flavors play in the uptake, maintenance of use (defined as steady use over time) and escalation of use (defined as increased use over time) of e-cigarettes among adolescents, Janet Audrain-McGovern et al. evaluated several factors, including flavors, that may exacerbate the risk of continued and escalating e-cigarette use following initial exposure. For 18 months, the investigators followed ninth-grade students who had reported using an e-cigarette at baseline and asked about their frequency of e-cigarette use through surveys. The initial survey asked the students about a variety of factors, including their sex, race, close peer and parental use of e-cigarettes, the age at which they first used e-cigarettes and the characteristics of the e-cigarette at first use. In this latter category, the characteristics surveyed included the nicotine content of the e-cigarette and whether it was flavored or unflavored/tobacco-flavored.

In follow-up surveys, students were asked about current (post-30 day) use following their first exposure to e-cigarettes as well as frequency of use following first exposure. This allowed the investigators to examine continued and escalating use of e-cigarettes and to assess the factors that may have contributed to each condition.

In applying the model to baseline and continuing levels of e-cigarette use, the authors found that initiation of e-cigarette use is significantly associated with family and peer use, as well as prior combustible cigarette smoking. Furthermore, continued e-cigarette use is significantly associated with gender, sensation-seeking behaviors and the use of non-tobacco flavors. These findings should come as no surprise, as many of these factors are routinely associated not only with e-cigarette use, but with a tendency to engage in most risk behaviors.

When examining escalation of e-cigarette use, the authors looked at the change from the baseline of e-cigarette use over time. They found that age of first use, sensation-seeking behaviors, and use of flavored products were associated with increased e-cigarette use over time.

The Audrain-McGovern et al. study shows that many factors may contribute to initiation and maintenance of e-cigarette use, including use among family members and peers as well as prior use of combustible cigarettes. The most important information it provides in the context of the FDA notice is twofold: a) that the factors that lead adolescents to initiate e-cigarette use have nothing to do with e-cigarettes themselves, but are rather the same risk factors associated with initiation of dangerous behaviors more generally; and b) that the factors that contribute to sustained e-cigarette use in adolescents are the same factors that contribute to adults successfully switching from combustible products to e-cigarettes.

In terms of the first finding, initiation of any risk behavior is strongly tied to social determinants of health, including wealth, education, and community and social support. Indeed, studies routinely demonstrate that populations with less income or education have higher rates of problematic drug use, including higher smoking rates, than those with more income or education. This appears to be true of e-cigarettes as well. In fact, the Audrain-McGovern et al. study indicates that family and peer use, as well as sex and race, are more predictive of initial e-cigarette use than any features of e-cigarettes themselves, including flavor.

Furthermore, given the findings indicating that combustible cigarette smoking is strongly associated with current use of e-cigarettes, this study (and others) suggests that e-cigarettes may have the potential to act as a pathway away from smoking rather than a gateway to smoking. While non-tobacco flavors may be a factor in the maintenance of e-cigarette use in adolescents, this is also likely to be true for adults who use e-cigarettes as a smoking cessation device. Indeed, evidence suggests that non-tobacco flavors enable adults to dissociate tobacco flavor from nicotine delivery, thus increasing their likelihood of successfully switching from combustible cigarettes to e-cigarettes.

However, the Audrain-McGovern et al. study lacks crucial data and context. To start, of the 2,000 students who were originally eligible to participate in the study through parental consent, only 368 ninth-graders confirmed ever using e-cigarettes in the first year of the study. While not the original intention of the study, the authors do not provide raw data on the number of students who had ever tried e-cigarettes and either used e-cigarettes following their first try or escalated e-cigarette use. This information could have provided valuable context through which to assess the magnitude of experimental, regular or intensifying use of e-cigarettes at later times.

Furthermore, the investigators did not assess individual flavors but instead categorized all flavored products together. This categorization may have been appropriate in the context of this particular study, but based on this data alone, it is impossible to determine whether so-called ‘kid-appealing’ flavors are the driving force behind the escalation of e-cigarette use.

When using research to drive public policy, assembling a more complete picture will be crucial. Taken alone, this study demonstrates that initiation of e-cigarette use is not associated with features of e-cigarettes like flavors, and that flavors might well play a role in acting as a pathway out of combustible cigarette use. In promulgating the proposed rule, policymakers at the FDA would do well to use these findings to help them understand the factors that may contribute to initiation, maintenance and escalation of e-cigarette use, especially if these factors are unique to never-smokers or underage users.