Article Text
Abstract
Background Concerns about the effects of vaping have prompted calls to restrict e-cigarette flavours. Vaping proponents have criticised these proposals, which they argue may discourage smokers from taking up vaping or trigger relapse to smoking. We explored the role flavours play in vaping uptake and cessation among New Zealand cigarette smokers and vaping-susceptible never smokers (VSNS), and examined current vapers’ preferred flavours.
Methods We conducted an online survey of 1005 New Zealanders aged 18–70 years that included 324 current vapers (vaped in the last 30 days) and 302 ‘past’ vapers (reported past vaping, but not within the last month). We asked respondents their reasons for vaping and explored current vapers’ preferred e-cigarette flavours; we analysed the data using descriptive statistics and logistic regression.
Results Irrespective of smoking status, flavour was one of the main reasons respondents gave for vaping (smokers 83%; former smokers 77%; VSNS 80%). Flavour was less important to former vapers; 47% of smokers, 57% of former smokers and 64% of VSNS cited flavour as a reason for originally taking up vaping. Fruit flavours were most popular among all three groups; smokers also favoured tobacco flavour, while former smokers also favoured mint or menthol, and never smokers also favoured confectionery/sweets/lolly flavours.
Conclusions Flavours play a major role in vaping initiation for current smokers, former smokers and vaping-susceptible non-smokers, and remain important to those who continue vaping. Our findings highlight the need for regulation that allows some flavour diversity without the extravagant marketing currently used to promote vaping and e-liquids.
- electronic nicotine delivery devices
- public policy
- advertising and promotion
Statistics from Altmetric.com
Introduction
Concerns over deaths linked to use of electronic nicotine delivery systems (ENDS) have focused attention on vaping and the contents of e-liquids, and stimulated political interest. For example, in January 2020, President Trump confirmed plans to remove all e-liquid flavours, apart from tobacco and menthol, from the US market,1 a move that elicited both plaudits and condemnation. In New Zealand (NZ), ever use of ENDS is now 21.2% and the government is developing legislation to regulate vaping, which the Associate Minister of Health has advised will allow only three flavours: tobacco, menthol and mint.2 3
Two key reasons underpin these regulatory initiatives. First, fears that some flavour agents are harmful and may be associated with the recent spate of illnesses and, in some cases, deaths affecting people who have vaped.4 Analyses of e-liquids have found some contain cytotoxic organic compounds, known to cause mutations5–7; other studies have reported associations between exposure to e-liquid compounds and oxidative stress, inflammatory responses8 and respiratory diseases.9
Second, alluring flavour names, such as Honey Bear, Tropical Bomb and Good Morning Sunshine, may attract young people to vaping in the same way that evocative brand names and packaging once lured them to smoking.10–12 Uptake of vaping among young people will increase the potential risks they face relative to remaining smoke-free and vape-free, and public health researchers and policy-makers agree that non-smokers should not start vaping.13
ENDS proponents have responded to these concerns in several ways. Many agree that vaping is not risk-free, yet argue the known risks of smoking are so severe that vaping is likely to be considerably less harmful, even if the risks cannot yet be quantified.14 Known as ‘harm reduction’, this perspective differentiates between nicotine sources and promotes transition from those that are very harmful to those thought to pose fewer risks.15
Moves to limit flavours are seen as likely to disrupt this transition and have met with strong resistance from vaping associations and advocates. These groups argue that flavours are pivotal to vaping uptake, which they see as playing a crucial role in reducing smoking prevalence, and claim limiting vaping flavours will make switching more difficult and may foster relapse to smoking.16 17 They also speculate that reducing flavour options could lead people to mix their own flavours, putting them at risk of the adverse outcomes recently observed in the USA.17
These competing arguments raise important questions about the role flavours play in vaping uptake and transitions from smoking to vaping. We explored these questions and examined preferred flavours among current and former vapers.
Methods
Sample
In June 2019, we conducted an online survey of 1005 New Zealanders aged 18–70 years, including 324 current vapers (respondents who reported vaping in the last 30 days) and 302 past vapers (respondents who had vaped in the past, but who reported not vaping in the past month). The sample was designed to provide roughly equal numbers of vapers and non-vapers, balanced by age and sex, and with sufficient Māori and Pacific respondents to produce reliable estimates of ethnicity effects. Dynata, whose NZ panel currently stands at 285 000 active members (ie, have participated in a survey in the last 12 months), provided the sample.
We defined smokers as either daily or occasional cigarette smokers (ie, do not smoke daily but smoke at least once a month), and non-smokers as either former cigarette smokers or never smokers of cigarettes who were susceptible to vaping (ie, did not answer ‘definitely not’ to both of two questions about their likelihood of vaping). The gender split for smokers and current non-smokers was similar (approximately 55% female and 45% male) as was the ethnicity of the two groups (75% to 80% NZ European/Other and 43% Māori/Pacific, based on multiple ethnicity). Smokers tended to be younger and have less formal education than non-smokers. Eighty per cent of smokers, 36% of former smokers and 56% of vaping-susceptible never smokers (VSNS) had ever tried an e-cigarette; 65% of smokers, 38% of former smokers and 12% of VSNS reported they were still vaping at least once a month (the number of VSNS was only 11 but we report their results for completeness). Online supplementary file 1 outlines the sample characteristics.
Supplemental material
Procedure and survey instrument
We fielded the survey using the Qualtrics platform and set quotas for smoking status, ethnicity and gender to help ensure sample representativeness; this process produced a cleaned achieved sample of 1005 cases. Respondents answered items about their smoking and vaping attitudes and behaviours; they reviewed a list of vaping uptake and cessation motivations and identified those that applied to them, and indicated their flavour preferences after reviewing flavour genres developed following discussions with vapers and vape store staff (see online supplementary file 2). We also collected details of respondents’ demographic characteristics, including their age, gender, ethnicity, highest formal educational qualification, number of cigarettes smoked each day and time to first cigarette (the latter to calculate Heaviness of Smoking Index (HSI) scores).
Supplemental material
Using SPSS (V.25), we calculated descriptive statistics for each participant characteristic and developed logistic regression models to examine associations between participants’ characteristics and their e-cigarette flavour preferences.
Results
From a list of 12 possible reasons, smokers who reported current vaping cited flavour appeal as their main reason for vaping (nominated by 83% of the 270 smoking vapers); 77% of the 43 former smokers and 82% of the 11 VSNS who reported current vaping also cited liking flavours as one of the main reasons why they vaped. However, flavour was less important to past vapers, with 47% of 148 smokers and 57% of 70 former smokers and 64% of 84 VSNS who had formerly vaped citing flavour as one of the reasons they had begun vaping.
Among smokers who currently vaped, we found no significant relationship between liking e-cigarette flavours and age, gender, education, ethnicity or HSI. Māori and Pacific former vapers who smoked were more likely than NZ European/Other to report flavours as a reason for vaping, but the difference was not statistically significant (OR=1.64, CI 0.86 to 3.14, significance=0.137).
Among former smokers who currently vaped, males were significantly more likely than females to give flavour appeal as a reason for vaping (OR=4.50, CI 1.04 to 19.45, significance=0.044). For past vapers who were also former smokers, those aged under 35 were more likely than older respondents to cite liking e-cigarette flavours as a reason for vaping (OR=4.80, CI 1.71 to 13.47, significance=0.003). A similar, though weaker, association was revealed for the former vapers who were never smokers (OR=2.88, CI 0.99 to 8.36, significance=0.053).
Examination of specific flavour preferences showed that both smokers and non-smokers preferred fruit flavours. Current smokers also preferred tobacco and mint/menthol flavours; former smokers also preferred mint/menthol flavours and, to a lesser extent, tobacco flavour. The small number of VSNS preferred confectionery/sweets/lolly flavours to either tobacco or mint/menthol flavours (see table 1).
Preference for fruit flavoured e-liquid was most pronounced among those under 35 and among Māori and Pacific peoples. The adjusted ORs for those who smoked and vaped were 2.01 for those under 35 compared with those over 35 (CI 1.37 to 2.93, significance<0.001) and 1.39 for Māori and Pacific peoples compared with European/other (CI 0.95 to 2.03, significance=0.088). For former smokers who reported vaping, these associations had similar ORs, though these were not statistically significant.
Discussion
Current smokers, former smokers and VSNS all reported that e-cigarette flavours played a major role in vaping initiation, and in continued vaping among those who were current ENDS users. All three groups preferred fruit flavours, but current and former smokers were more likely than VSNS to prefer tobacco or mint/menthol flavours (the only flavours the NZ Government is considering permitting). VSNS preferred most other flavours over either tobacco or mint/menthol (though the subgroup numbers were very small).
Associations between demographic attributes and e-cigarette flavour preferences were mixed. However, our findings provide some tentative evidence that e-cigarette flavours, particularly fruit flavours, are more important to younger people than to older people. These findings are consistent with other studies highlighting the importance of flavours to young people,11 12 18 19 and with results regarding adults’ flavour preferences, which appear more varied.20
Unlike other studies, we also probed flavour preference by ethnicity, an important issue in NZ, where Māori and Pacific peoples bear a disproportionate burden of harm caused by smoking. Our findings suggest Māori and Pacific peoples may have stronger preferences than European/Other for fruit flavours, though replication studies are needed to clarify this association and identify the specific flavours with greatest appeal.
Evidence that confectionery and cake/dessert flavours may be more important to VSNS than to smokers or former smokers suggests restricting these flavours could reduce vaping’s appeal to never-smokers, while having relatively little impact on smokers, who generally prefer other flavours. However, while other non-tobacco flavours also influence non-smokers’ decision to begin vaping, they may help to promote smokers’ transition from smoking to vaping.14 Banning non-tobacco flavours, such as fruit, confectionery and cake/dessert flavours,21 would remove some potentially toxic compounds from the market place, but could make vaping less attractive to smokers. Given these flavours may appeal to groups at greatest risk of ill-health from smoking, further work is urgently required to test arguments that flavours are pivotal to vaping uptake and that limiting flavour availability would reduce switching and potentially promote relapse to smoking.
Future work could also address limitations of our study, including inherent limitations of cross-sectional studies, and the small sample of non-smoking vapers, which limited the study power and our ability to test for significant differences involving this group. Recruiting larger samples of vapers and stratifying by ethnicity would enable more detailed analysis of population groups most affected by smoking while cohort studies could examine relationships between flavour preferences and cessation-related outcomes.
Nonetheless, even if future work reveals that smokers require diverse flavours to promote transition to vaping, it is not clear they require thousands of flavours from which to choose or that regulators should allow the proliferation of e-cigarette flavours to continue. As the effects of intriguingly named and seductively marketed e-cigarette flavours on younger non-smokers are now becoming established, regulation to protect non-smokers is urgently needed. Future work could address this need by examining how flavour marketing could be reduced; for example, by allowing only short descriptive flavour names (or code numbers) and requiring e-liquid to use standardised (plain) packaging. Such moves could more effectively balance smokers’ desire for a pleasurable experience against the risk of vaping uptake by non-smokers.
What this paper adds
Limiting confectionery e-liquid flavours may discourage vaping uptake among non-smokers yet have relatively little effect on smokers’ willingness to switch to vaping.
Determining the e-liquid flavour range necessary to encourage vaping uptake among smokers requires urgent attention.
Limiting e-liquid flavour proliferation and seductive flavour marketing may help balance smokers’ needs against the risk of uptake among young non-smokers.
Acknowledgments
JH undertook work on this manuscript while she was a Fellow at the Institute of Advanced Study, Durham University, UK.
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
Contributors PG led the data collection and analysis, and the manuscript development. JH conceptualised and designed the project, obtained research funding, and contributed to iterations of the manuscript. Both authors have seen and approved the final manuscript version and are guarantors of the manuscript.
Funding This research was funded by the Health Research Council of New Zealand (grant 16/149).
Competing interests There are no competing interests to declare.
Patient consent for publication Not required.
Ethics approval The University of Otago Human Ethics Committee reviewed and approved the study, and we undertook formal consultation with the Ngāi Tahu Research Consultative Committee, which represents the interests of indigenous Māori peoples.
Provenance and peer review Not commissioned; externally peer reviewed.