Abstract
E-cigarette use occurs with tobacco product use in youth.
Using the 2014 National Youth Tobacco Survey (NYTS), we examined past 30-day frequency of cigarette, cigar, smokeless, and e-cigarette use in the context of past 30-day and ever tobacco product use in US middle and high school students (N = 22 007). Frequency of product-specific use was examined by exclusive versus concurrent use with another product in the past 30 days (poly-use).
In 2014, the majority (83%) of US middle and high school students had not used tobacco or e-cigarettes in the past 30 days. In the 9.3% of youth reporting any past 30-day e-cigarette use, 63% also reported using a tobacco product; among the 3.3% past 30-day exclusive e-cigarette users, about two-thirds (2.1%) had ever used combustible or non-combustible tobacco products and one-third (1.2%) had not. Few never tobacco users had used e-cigarettes on 10 or more days in the past month (absolute percent < 0.1%). Among past 30-day cigarette and smokeless users, the two highest frequency categories were 1–2 days and daily use; among past 30-day e-cigarette and cigar users, prevalence decreased with increasing frequency of use. The majority of past 30-day cigarette, cigar, smokeless, and e-cigarette users reported poly-use.
Prevalence estimates for a single product mask the complex patterns of frequency, temporality, and poly-use in youth. Two-thirds of past 30-day exclusive e-cigarette users have ever used tobacco. Poly-use is the dominant pattern of tobacco and e-cigarette use among US middle and high school students.
Our study highlights the complexity of tobacco use patterns in US middle and high school students. Future studies addressing the full public health impact of movement into or out of combustible tobacco use will require longitudinal data with appropriate measures of tobacco and e-cigarette product-specific use (eg, frequency and intensity), as well as adequate sample size and a sufficient number of waves to determine how use of individual products, like e-cigarettes, impact progression into or out of more stable patterns of tobacco and e-cigarette use.
Introduction
According to the US Surgeon General’s 50th Anniversary Report, death from tobacco use is “overwhelmingly caused by cigarettes and other combusted tobacco products.”1 In the United States, relative to cigarette/combustible tobacco use, use of electronic cigarettes (e-cigarettes) among youth has increased since national data collection began in 2011.2 E-cigarette use prevalence estimates vary depending on frequency of use (eg, ever, monthly, or daily)3–5 and poly-use of e-cigarettes, combustible (eg, cigarettes, all types of cigars) and smokeless tobacco products is increasingly common in youth.3
Currently, fewer than 25% of youth ever cigarette users progress to daily cigarette use, indicating that a large proportion of tobacco product use that occurs in adolescence is experimental and does not lead to long-term regular use in adulthood.6–8 The value of the traditional surveillance definitions of “ever use” and “current use” (any use in the past 30 days) has been questioned because the frequency of cigarette use has diminished over time, and past 30-day use may therefore reflect a greater proportion of trial use compared to more frequent, intense use.6–10 Identifying youth tobacco users who are at high risk of progressing to persistent use is a priority for the US Food and Drug Administration (FDA).10–12
This study employed data from the 2014 National Youth Tobacco Survey (NYTS) to examine e-cigarette ever and past 30-day patterns of use, including exclusive use and concurrent use with tobacco products (poly-use). We compared these data with patterns of cigarette, cigar and smokeless tobacco use for which frequency data were available.
Methods
NYTS is a national survey of US middle and high school students (2014: N = 22 007; response rate: 73.3%).4 Ever and past 30-day use of various tobacco products and e-cigarettes was assessed; these included cigarettes, cigars (cigars, cigarillos or little cigars), smokeless tobacco, e-cigarettes, hookah, pipe, bidis, snus, and dissolvables. Ever tobacco use was defined as ever using, even just one time, any of these products, except e-cigarettes. Ever e-cigarette use was defined separately as ever having tried an e-cigarette such as Blu, 21st Century Smoke or NJOY. Past 30-day use measures defined product-specific prevalence and classified youth in one of four use categories: (1) Any past 30-day combustible use (including cigarettes, cigar, hookah, pipe, and bidis); (2) past 30-day non-combustible use (including smokeless tobacco, snus, and dissolvables); (3) past 30-day e-cigarette use only; and (4) no past 30-day tobacco or e-cigarette use. Data on frequency of use in the past 30 days were available for four products: cigarettes, cigars, smokeless tobacco, and e-cigarettes. Response categories were ordered from “0 days” to “all 30 days.” For each of these four products, we categorized past 30-day users as exclusive (ie, that product only) versus poly-users (ie, use of that product and at least one other product in the past 30 days). For past 30-day use items (ie, hookah, pipe, bidis, snus, and dissolvables), missing responses were categorized as nonusers to be conservative in estimating prevalence. For the product-specific past 30-day frequency questions (ie, for cigarettes, cigars, smokeless, and e-cigarettes), missing responses were excluded.
The prevalence of ever and past 30-day tobacco and e-cigarette use were estimated. Additional analyses examined the prevalence of ever tobacco use among past 30-day e-cigarette users. Bivariate analyses examined frequency of e-cigarette use by the four use categories and the frequency of cigarette, cigar, smokeless, and e-cigarette use by exclusive versus poly-use. Post-hoc analyses examined tobacco product purchasing among past 30-day tobacco and e-cigarette users by exclusive versus poly-use. Statistical analyses were weighted by primary sampling unit and sampling stratum using Stata 14.1 (www.stata.com). The confidence intervals for proportions using Stata’s svy commands are constructed using a logit transform so that their endpoints always lie between 0 and 1.13
Results
NYTS participants ranged in age from 9 to 19 or older, with mean age 14.6 (95% CI: 14.42, 14.81). Approximately half were male (50.2%). The majority of participants were white (65.2%), with 20.4% black, 14.2% other race, and less than 1% reporting more than one race. Hispanic participants comprised 21.7% of the sample.
Ever Use in the Full Sample
In 2014, the prevalence of ever use of any tobacco product, excluding e-cigarettes, was 31.5% in US middle and high school students. When examining the products that had been tried by youth as of 2014, 30.2% (95% CI: 28.4%–32.0%) of youth reported ever use of a combustible tobacco product, 9.5% (95% CI: 8.2%–10.9%) reported ever use of a non-combustible tobacco product and 19.4% (95% CI: 17.6%–21.3%) reported ever use of an e-cigarette in 2014.
Past 30-Day Tobacco and E-Cigarette Use in the Full Sample
Figure 1 presents frequency of days used e-cigarettes in the past 30 days with non-truncated full population percentages (0%–100%). Figure 2 focuses on the subsample of past 30-day e-cigarette users and magnifies the y-axis (0%–5%) to more clearly illustrate frequencies of e-cigarette use categories, ranging from 4.4% use on 1–2 out of 30 days to 0.9% use on all 30 days. Both figures present e-cigarette use in the context of four past 30-day use categories (any combustible tobacco use, non-combustible tobacco use, e-cigarette only use, and no past 30-day use). In the full sample, 90.7% of youth did not report any past 30-day e-cigarette use (Figure 1, “0 days”). This includes 83.0% of youth who did not report using any tobacco product or e-cigarette in the past 30 days; of these, 79% were never users of any product and 21% had ever used a tobacco product or e-cigarette, but not in the past 30 days. The remaining youth who had not used e-cigarettes in the past 30 days reported past 30-day combustible tobacco use (6.6%) or non-combustible use (1.1%).
Distribution of tobacco and e-cigarette co-use among US middle and high school students (weighted); National Youth Tobacco Survey, 2014. Tobacco use categories are mutually exclusive. Ever use, no past 30-day use was defined as reporting ever use of any of the following products, but no use of any product in the past 30-days: cigarettes (including roll-your-own cigarettes), cigars, pipe, kreteks, hookah, flavored little cigars, flavored cigars (including clove cigars), bidis, smokeless tobacco, snus, dissolvable tobacco products, and e-cigarettes. Any combustible tobacco use was defined as endorsing past 30-day use of any of the following products: cigarettes (including roll-your-own cigarettes), cigars, pipe, kreteks, hookah, flavored little cigars, flavored cigars (including clove cigars), and bidis; past 30-day e-cigarette users were included in this category if they reported use of any of these combustible products. Non-combustible tobacco use was defined as past 30-day use of smokeless tobacco, snus, or dissolvable tobacco products, but no past 30-day combustible use; past 30-day e-cigarette users were included in this category if they reported use of any of these non-combustible products. Exclusive e-cigarette users reported only past 30-day use of e-cigarettes. Respondents who did not endorse use of any tobacco or e-cigarette product were coded to the “no past 30-day use” group.
Distribution of tobacco and e-cigarette co-use among US middle and high school students who reported past 30-day e-cigarette use (weighted); National Youth Tobacco Survey, 2014. Tobacco use categories are mutually exclusive. Any combustible tobacco use was defined as endorsing past 30-day use of any of the following products: cigarettes (including roll-your-own cigarettes), cigars, pipe, kreteks, hookah, flavored little cigars, flavored cigars (including clove cigars), and bidis; past 30-day e-cigarette users were included in this category if they reported use of any of these combustible products. Non-combustible tobacco use was defined as past 30-day use of smokeless tobacco, snus, or dissolvable tobacco products, but no past 30-day combustible use; past 30-day e-cigarette users were included in this category if they reported use of any of these non-combustible products. Exclusive e-cigarette users reported only past 30-day use of e-cigarettes. Respondents who did not endorse use of any tobacco or e-cigarette product were coded to the “no past 30-day use” group.
In the full sample, 9.3% (95% CI: 8.0%–10.9%) of youth reported any past 30-day e-cigarette use. Of these, 87% (absolute percent 8.1%) had ever used a tobacco product and 13% (absolute percent 1.2%) had never used a tobacco product. Frequency of e-cigarette use varied in past 30-day users (Figures 1 and 2, Supplementary Table 1): 4.4% used on 1–2 days, 2.6% used on 3–9 days, and 2.4% used on 10 or more days. Past 30-day e-cigarette users who also reported the number of days used had a median of 3–5 days of use in the past month. The majority of never tobacco users who reported past 30-day e-cigarette use had used an e-cigarette 1–2 days in the past month (77%; absolute percent 0.9%); few never tobacco users used e-cigarettes on 10 or more days in the past month (absolute percent < 0.1%).
Past 30-Day Combustible and Non-Combustible Tobacco Use in Past 30-Day E-Cigarette Users
In the 9.3% of youth reporting any past 30-day e-cigarette use (Figure 2), approximately 63% also used a tobacco product in the past 30 days (5.5% any combustible and 0.4% noncombustible). The proportion using a tobacco product in the past 30 days increased to 79% when looking only at youth who had used e-cigarettes on 10 or more days in the past month (1.8% any combustible; 0.1% noncombustible; and 0.4% e-cigarette only).
The proportion of any past 30-day combustible use was lowest in the least frequent e-cigarette users (45.5% in the 1–2 days category) and highest in the daily e-cigarette users (77.8% in the all 30 days category). Conversely, the relative proportion of past 30-day e-cigarette only use was highest in the least frequent users (50.0% in the 1–2 days category) and lowest in the daily users (11.1% in the all 30 days category). Past 30-day non-combustible use varied with the frequency of past 30-day e-cigarette use, with the largest proportion in those who had used e-cigarettes on 10–19 of the past 30 days (10.0%).
Past 30-Day Exclusive Versus Poly-Use in Past 30-Day E-Cigarette Users
In collapsing the past 30-day any combustible and non-combustible categories to examine e-cigarette only use (exclusive use) versus use with tobacco products (poly-use), we note that 36% of past 30-day e-cigarette users were exclusive users. Among exclusive users, 65% (absolute percent 2.1%) reported ever use of a combustible or non-combustible tobacco product; this increased to 83% in those using e-cigarettes only on 10 or more days in the past month. Conversely, the proportion of youth reporting exclusive use decreased significantly as frequency of e-cigarette use increased (Figure 3). Poly-use ranged from 50.0% in the least frequent users (1–2 days in the past month) to 88.9% in the most frequent users (all 30 days).
Frequency of cigarette, cigar, and smokeless tobacco and e-cigarette use among past 30-day US middle and high school student users (weighted); National Youth Tobacco Survey, 2014.
Past 30-Day Exclusive Versus Poly-Use in Past 30-Day Cigarette, Cigar, and Smokeless Users
To provide further detail and context for findings related to poly-use in e-cigarette users, Figure 3 provides graphs of the frequency of past 30-day use of cigarettes, cigars, and smokeless tobacco by exclusive use of that product versus poly-use. In the full sample of US middle and high school students, 6.3% reported past 30-day cigarette use, 5.6% reported past 30-day cigar use, and 3.8% reported past 30-day smokeless tobacco use. For all products, a significant proportion of youth reported using a product 1–2 days in the past month; this accounted for 38% of past 30-day cigarette users, 50% of past 30-day cigar users, and 29% of past 30-day smokeless users. For cigarettes and smokeless tobacco, however, the next highest category of use was “all 30 days,” with 22% (absolute percent 1.3%) of past 30-day cigarette users reporting daily cigarette use and 29% (absolute percent 1.1%) of smokeless tobacco users reporting daily use. For cigars and e-cigarettes, the next highest category of use was “3–5 days,” at less than half the prevalence of what was reported for “1–2 days”; frequency of use decreased again at “6–9 days” and remained at that level through “all 30 days.” Past 30-day users of each product who also reported the number of days used had a median of 3–5 days (cigarettes), 1–2 days (cigars), and 10–19 days (smokeless) of use in the past month.
In comparing exclusive product use versus poly-use, there was a consistent pattern of poly-use across all products and frequency categories. The overall prevalence of poly-use was 81% (absolute percent 5.1%) in past 30-day cigarette users, 77% (absolute percent 4.3%) in past 30-day cigar users, 74% (absolute percent 2.8%) in past 30-day smokeless tobacco users, and 64% (absolute percent 6.0%) in past 30-day e-cigarette users. The prevalence of poly-use was highest in daily cigarette users (85.7%) and lowest in youth using e-cigarettes on 1–2 days in the past month (50.0%). When looking at each product by frequency category, the odds of exclusive use were significantly lower in those using cigarettes or smokeless on 6–9 days, 10–19 days, and all 30 days compared to those using on 1–2 days in the past month (p’s < .05; Supplementary Table 2). For cigars, the odds of exclusive use were significantly lower in those using on 6–9 days compared to 1–2 days in the past month (OR = 0.46, p = .021). The odds of exclusive use were significantly lower for all frequency categories of e-cigarette use beyond 1–2 days in the past month (p < .001).
Among youth who reported any past 30-day tobacco or e-cigarette use, the overall prevalence of purchasing tobacco products in the past 30 days was 22.3%. The proportion of youth who bought themselves tobacco products was 32% for past 30-day users of cigarettes, 38% for cigars, 38% for smokeless tobacco, and 22% for e-cigarettes. Fifteen percent or fewer of those who bought products for themselves were exclusive users of a particular product.
Discussion
A majority of US middle and high school students (83%) reported no tobacco or e-cigarette use in the past 30 days; most were never users of e-cigarettes or any other tobacco product. Of the 9.3% of youth who reported any past 30-day e-cigarette use, most had also used tobacco products in the past month (63%) or had ever used tobacco products (87%). For cigarettes, smokeless tobacco, and cigars, exclusive product use was also low. Most of those who report using e-cigarettes were ever users of other tobacco products, primarily combustible tobacco products. It was rare that tobacco naïve youth reported using e-cigarettes and if they did, rarer still to finding them using them frequently (ie, fewer than 0.1% used on 10 or more days per month).
The proportion of past 30-day poly-use increased significantly with increasing frequency of e-cigarette use among youth e-cigarette users. Cigarettes and smokeless tobacco use were bi-modally distributed, with peaks at the lowest (1–2 days) and highest levels of use (all 30 days). In contrast, cigar and e-cigarette use showed a dramatic decrease in frequency from 1–2 days to use on all 30 days. Different use patterns may reflect the prevalence of products available in youth social contexts in contrast to use when alone, or contextual factors like point of sale placement, marketing, and digital media influences.14–19
For e-cigarettes, curiosity about a novel product, harm perceptions, flavors, and ease of access (such as sharing with friends) may partly explain high prevalence of ever use and higher rates of infrequent experimentation compared with cigarettes, but low daily use (0.9%). Additionally, approximately two-thirds of the exclusive past 30-day e-cigarette users had historically used a combusted or non-combusted tobacco product, while the remaining one-third reported having never used a tobacco product. Not purchasing one’s own device, using with friends, use without nicotine, or use of products with inefficient delivery of nicotine (eg, due to use of less expensive, easily accessible, and less effective devices or inexperience with the devices) may partially account for very low use rates on 20 or more days per month.20,21 Cigarettes, the most toxic product, are the most dominant product for young daily tobacco users, perhaps due to their availability, appeal (eg, factors including menthol flavoring and marketing). Moreover, the delivery mode (the combusting of tobacco, that cigarettes share along with other combusted products like cigars) likely has the highest degree of chemosensory satisfaction and highest addiction liability of any nicotine containing products.
The different patterns between cigarettes and cigars are puzzling as they appear to be somewhat interchangeable (combustible) products. The lower cost of single or small packs (eg, cigarillos) and appealing flavors could make them attractive for experimentation, but given the high prevalence of poly-use, it is possible that the higher frequency of experimental use (1–5 days/month) but the low frequency of daily cigar use occurs with more frequent use of another product, like cigarettes (menthol or regular flavor), that become the primary choice of combustible product. Moreover, cigars are frequently sold individually or in small packs whereas cigarettes are sold in packs of 20; a young person with access to a full pack of cigarettes may be likely to smoke more frequently, hastening a transition to daily use of cigarettes. As youth progress from experimental and single product to daily and poly-use, longitudinal data and more fine-grained studies of the reasons and the patterns of use (eg, ecological momentary assessment and intensive web surveillance) are needed. Such fine grained analysis will enable better understanding these emerging patterns and their progression to daily cigarette use over time.22,23 Such mixed methods will allow for study of the sequence of trying each product and include detailed measures of product-specific reasons for use during transitions and trajectories of product use to further understand the differences in cigar, cigarette, smokeless and e-cigarette patterns of use to more precisely inform policy and practice.
Study limitations include the cross-sectional nature of the NYTS and inability to evaluate the sequence of tobacco product initiation other than historically by comparing the (retrospective) questions about ever use in one’s lifetime to past 30-day use. Additionally, it is not possible to determine the reasons for poly-use versus exclusive use, nor the reasons underlying any sequence in which youth transitioned from exclusive to poly-use or vice versa. Likewise, any process of progression to daily use cannot be discerned but can only be inferred from the frequency of use categories. Thus implications of temporal direction or of causality should be avoided.24 Lacking frequency measures for other tobacco products, like hookah, limited our ability to fully characterize these behavioral patterns.
Our study highlights the complexity of tobacco use patterns and dominance of poly-use in US middle and high school students. Future studies addressing the full public health impact of movement into or out of combustible tobacco use will require longitudinal data with appropriate measures of tobacco and e-cigarette product-specific use (eg, frequency and intensity), as well as adequate sample size and a sufficient number of waves to determine how use of individual products, like little cigars, e-cigarettes, and other innovations in emerging alternative nicotine delivery systems, impact progression into or out of more stable patterns of tobacco use.10,11 Currently, most surveillance studies that have reported on national data in the United States do not typically or completely: (1) account for the substantial differences in prevalence by frequency of use; (2) report on all possible patterns of concurrent poly-use; (3) characterize patterns in the context of historical (ever) use of both tobacco products and e-cigarette products compared to one another and to never users of any products; or (4) present metrics for total combustible product use, total tobacco use (excluding e-cigarettes) and total tobacco and e-cigarette use.2,8,25–27
Existing cohort studies that have prospectively examined sequential trajectories must also be viewed with caution due to: limited measures of low use frequency, such as ever use or use on only 1–2 days in the past 30 days; small sample sizes of respondents within specific product use categories at follow-up that are likely to be unstable; lack of considering the full context (eg, selectively omitting critical comparison subgroups of single-product and ever poly-users from analysis); and using locally recruited small samples that cannot be generalized to the US population.28–31 When interpreting findings for purposes of tobacco control strategy and policymaking,10,28–33 care must be taken to avoid biases, confounding, over or underestimates of trajectories and interpretations of results32 that go beyond the limitations of the data.11,12,24,26
Supplementary Material
Supplementary data are available at Nicotine & Tobacco Research online.
Funding
JLP was supported by National Institute of Drug Abuse and the Food and Drug Administration (K01DA037950). All other authors were financially supported by Truth Initiative. ACV and ALJ had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. The funders had no role in the design and conduct of the study; analysis and interpretation of the data; or preparation, review, or approval of the manuscript. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH or the Food and Drug Administration.
Declaration of Interests
None declared.
Acknowledgments
We would like to thank Ken Warner for his feedback on an earlier draft of the manuscript.
References
Comments