Association for Public Policy Analysis & Management

Introduction Although cigarette smoking prevalence has continued to decline, the use of non-cigarette (i.e. novel) tobacco products, most particularly e-cigarettes, has grown dramatically in the past decade. Between 2011 and 2018, current e-cigarette use by high school students grew from 1.5% to 20.8%. In addition to the health risks these products present on their own, experimentation with these products has been found to lead to use of conventional cigarettes.

E-cigarettes and other novel products are available in an array of characterizing flavors (e.g. mango, menthol) and bright packages designed to appeal to youth. While it is well established that youth tobacco users are more likely to consume flavored tobacco than adults are, less is known about the specific role of flavors in progression from experimentation to established use. The Food and Drug Administration has identified characterizing flavors as a key factor in the youth tobacco epidemic and local governments in the US, including San Francisco and Chicago, have recently passed partial or complete bans on the sale of flavored tobacco products. Identifying the role of flavors in tobacco use progression will help inform the extent to which policies regulating flavored tobacco will prevent addiction among youth.

Methods We analyzed a representative sample of US youth (age 12-17) from four waves of the Population Assessment of Tobacco and Health (PATH) Study (2013-2017). The population included baseline never-users of five tobacco products: cigarettes, e-cigarettes, cigars, smokeless tobacco, and hookah. For each tobacco product, survey-weighted logistic regression estimated the associations between flavored tobacco experimentation (vs. non-flavored experimentation) reported at first and second follow-up and (1) future established use of that product and (2) future established use of cigarettes. To address selection bias (e.g. flavors might appeal to lower-risk youth), models used propensity score adjustment to account for observed differences in demographic, social, behavioral, and environmental covariates between treated (flavored tobacco experimenters) and comparison (non-flavored tobacco experimenters) youth at baseline.

Results Preliminary results indicate a positive and statistically significant association between flavored experimentation with smokeless tobacco and future use of smokeless tobacco, but not for other products. Flavored experimentation with hookah was negatively and statistically significantly associated with future cigarette use. Flavors were not significantly associated with future cigarette use for other products.

Discussion Findings indicate an effect of flavors on tobacco use progression for smokeless tobacco and that the role of flavors in tobacco progression varies across tobacco products. These findings suggest that research on novel tobacco products is strengthened when all products are considered. Recent flavored tobacco bans might reduce progression from experimentation to established use of smokeless tobacco.