Association between perceived changes in electronic cigarette retailer density and regulation support among adolescents in China Authors: Xibin Sun, Yang Wang, Linnea Laestadius Published: 2025 ABSTRACT AbstractBackgroundAdolescents’ support for the regulation of electronic cigarettes (e-cigarettes) has been documented to be influenced by various factors, but the commercial environment has rarely been studied. We examined the relationship between adolescents’ perceived changes in e-cigarette retailer density and their support for e-cigarette regulations in China and further identified potential underlying pathways.MethodsData were collected from a national online cross-sectional survey of Chinese adolescents during 2022–2023 (N=1988). Using structural equation modelling, we examined the association between perceived changes in e-cigarette retailer density and support for e-cigarette regulations, including flavour bans, online sales bans, taxation and comprehensive bans. Specifically, the analyses tested the mediating effects of e-cigarette-related positive outcome expectancy and perceived health risk, along with the moderating effects of e-cigarette use status.ResultsMore than half of the participants supported the flavour ban (58.3%), the online sales ban (67.7%) and taxation (71.7%), but only 43.5% supported the comprehensive ban on e-cigarettes. Perceived increases in retailer density were significantly negatively associated with adolescents’ regulation support (β=−0.097, 95% CI −0.142 to –0.054), mediated by positive outcome expectancy (β=−0.009, 95% CI −0.017 to –0.003) and perceived health risk (β=−0.045, 95% CI −0.063 to –0.027). However, subgroup analysis found this association to be significant only in never users, rather than those who had tried e-cigarettes.ConclusionsPositive outcome expectancy and perceived health risk simultaneously influenced the negative association between perceived increases in e-cigarette retailer density and regulation support among adolescents who never vaped. Measures are needed to address the perceptions of increasing e-cigarette retailer density among adolescents.Keywords: Awareness, Electronic nicotine delivery devices, Public policy WHAT IS ALREADY KNOWN ON THIS SUBJECT Public support, particularly among adolescents, is crucial for the development and enforcement of e-cigarette regulations. Prior studies have found youth support for e-cigarette policies to be influenced by numerous factors, including demographic characteristics, use status, information exposure and perceptions of e-cigarettes. Public support, particularly among adolescents, is crucial for the development and enforcement of e-cigarette regulations. Prior studies have found youth support for e-cigarette policies to be influenced by numerous factors, including demographic characteristics, use status, information exposure and perceptions of e-cigarettes. WHAT IMPORTANT GAPS IN KNOWLEDGE EXIST ON THIS TOPIC Little research has examined the relationship between adolescents’ support for e-cigarette policies and the e-cigarette retail environment, specifically retailer density. Little research has examined the relationship between adolescents’ support for e-cigarette policies and the e-cigarette retail environment, specifically retailer density. WHAT THIS STUDY ADDS This study reveals a negative association between perceived increases in e-cigarette retailer density and adolescents’ support for e-cigarette policies, mediated by positive outcome expectancy and perceived health risk; however, the association was not significant among adolescents who have ever tried e-cigarettes. To help create adolescent support for regulation, it is critical to address perceptions of e-cigarette retail density via stricter measures such as controlling access to retailers near schools, restricting advertising known to capture youth attention, displaying mandatory signage explicitly prohibiting minors from entering stores and graphically presenting health hazards. This study reveals a negative association between perceived increases in e-cigarette retailer density and adolescents’ support for e-cigarette policies, mediated by positive outcome expectancy and perceived health risk; however, the association was not significant among adolescents who have ever tried e-cigarettes. To help create adolescent support for regulation, it is critical to address perceptions of e-cigarette retail density via stricter measures such as controlling access to retailers near schools, restricting advertising known to capture youth attention, displaying mandatory signage explicitly prohibiting minors from entering stores and graphically presenting health hazards. Introduction The global prevalence of adolescent electronic cigarette (e-cigarette) use within the past month was 9.2% from surveys administered between 2012 and 2019,1 partially driven by aggressive youth-targeted marketing. In China, the proportion of secondary school students who had vaped peaked at 16.1% in 2021.2 Research has increasingly associated e-cigarette use with adverse effects on respiratory, cardiovascular and neurological health, in addition to nicotine addiction.3 To mitigate the potential health threat to adolescents, the Chinese government has implemented several regulations on the domestic e-cigarette market, including a flavour ban, prohibition of underage and online sales, and taxation at the same level as combustible cigarettes, with a strategic focus on reducing youth access and interest in these products. Public support is critical to developing and enforcing policies. It can build regulators’ confidence against potential stakeholder resistance and reduce community noncompliance after policy implementation. Discerning public priorities can also help inform communication strategies to create broader support for policy change.4 Fostering youth support is essential for widespread compliance with e-cigarette regulatory policies.5 As adolescents are targeted by e-cigarette marketing and susceptible to peer-driven norms, regulatory frameworks need to be largely aligned with their attitudes to maximise compliance. A previous study indicated that adolescent support for an outdoor smoking-free policy is crucial for its success. Perceiving such policies as threats to individual autonomy may trigger resentment and rebellious tobacco use, whereas understanding the policy rationale increases compliance.6 Prior studies have found youth support for e-cigarette policies to be influenced by demographic characteristics, use status, information exposure and perceptions of e-cigarettes.4 5 7 The local e-cigarette retail environment might affect support for regulations by shaping individual perceptions of and attitudes towards these products; however, its impact has been rarely explored. Only one study, conducted in California, USA, has shown a negative association between tobacco retail density and policy support for increased tobacco taxes among adults.8 To our best knowledge, no studies have explored how retail density shapes policy support among youth. Greater retailer density and exposure to tobacco at the point of sale are both associated with increased adolescent smoking.9 10 In recent years, the number of e-cigarette physical stores in China has dramatically risen and then shrunken in response to regulatory actions. The impact of changes in retailer density on youths’ attitudes towards e-cigarette regulation has yet to be explored. The objective of this study is to examine the relationship and potential pathways between perceived changes in retail density and adolescents’ support for e-cigarette-related policies. Methods We administered a national online survey of Chinese adolescents aged 15–19 years on the research platform Wenjuanxing from December 2022 to February 2023, through their Sample Service.11 Through quota sampling by gender and geographic area, we recruited 2144 adolescents from mainland China (online supplemental table S1). Our analytical sample was 1988 after excluding those who were unaware of e-cigarettes or had logical errors in responses. Our study measured adolescents’ support for four types of e-cigarette regulations: (1) a ban on all flavours except tobacco, (2) a ban on online sales, (3) taxation at the same rate as combustible cigarettes and (4) a comprehensive ban on all manufacturing, promotion and sales, with a 5-point Likert scale from ‘strongly disagree’ to ‘strongly agree’. With the same scale, we assessed perceived health risk of e-cigarettes (eg, lung damage) and positive outcome expectancy (eg, social enhancement). Perceived changes in e-cigarette retail density were evaluated by observed changes in the number of physical stores during the past year, with five possible responses coded from 1 to 5 corresponding to significant decrease, slight decrease, no change or never noticed, slight increase and significant increase. We adopted subjective measures of retailer density because prior literature suggests they influence adolescents’ susceptibility to tobacco use more strongly than objective ones.12 Logistic regressions were used to examine the associations between perceived changes in retailer density and support for each e-cigarette policy in adolescents. We also performed parallel mediation analysis of the role of perceived health risk and positive outcome expectancy, which treated the mediators and outcome as latent variables and further investigated whether e-cigarette use status moderated the indirect effects. Gender, age, education level, monthly allowance and advertising exposure were adjusted as covariates in the models, and all analyses were conducted in Stata V.16.0 and Mplus V.8.0. A p value less than 0.05 was considered statistically significant. We additionally conducted a sensitivity analysis by excluding ‘never noticed’ respondents. (Details regarding the sampling, measurements and statistical analyses are provided in online supplemental appendix.) Results Among adolescents surveyed, 19.37% had tried e-cigarettes. The mean sum score of perceived increases in e-cigarette retailer density was 2.67 (SD=0.97) (online supplemental table S2). Except for the comprehensive ban, e-cigarette policies were supported by more than half of participants, ranging from 58.3% to 71.7%. Adjusted for sociodemographic characteristics and advertising exposure, adolescents who perceived an increase in retail density and those with higher positive outcome expectancy were less likely to support all e-cigarette policies except taxation equal to combustible cigarettes. Risk perception was positively associated with support for each e-cigarette policy. Compared with never users, those who had tried e-cigarettes had a lower likelihood of supporting the e-cigarette policies, except for the comprehensive ban (online supplemental table S3). Perceived increases in retail density were significantly negatively associated with support for e-cigarette policies in adolescents (β=−0.097, 95% CI −0.142 to –0.054), partially mediated by positive outcome expectancy and perceived health risk (Root Mean Square Error of Approximation (RMSEA)=0.048, Comparative Fit Index (CFI)=0.926, Tucker-Lewis Index (TLI)=0.909; table 1, online supplemental figure S1). The indirect effects through positive outcome expectancy and perceived health risk accounted for 9.4% and 46% of the total effect, respectively. However, the mediation paths differed significantly by participants’ e-cigarette use status (RMSEA=0.045, CFI=0.917, TLI=0.904; table 1). The negative association between perceived increases in retail density and support for e-cigarette policies among adolescents who never tried e-cigarettes (n=1603) was significant and simultaneously mediated by positive outcome expectancy (β=−0.007, 95% CI −0.016 to –0.001) and perceived health risk (β=−0.058, 95% CI −0.081 to –0.036); while among adolescents who have ever tried e-cigarettes (n=385), this association was no longer significant. Online supplemental figure S1 visually depicts the structural equation model frameworks for both the entire population and sub-populations. After excluding ‘never noticed’ respondents (n=715), the main results remained consistent (online supplemental table S4). Table 1Parallel mediating effect of perceived health risk and positive outcome expectancy on the relationship between perceived changes in e-cigarette retail density and adolescents’ support for e-cigarette regulations in China, 2022–2023EffectEntire population (n=1988)Adolescents who have ever tried e-cigarettes (n=385)Adolescents who never tried e-cigarettes (n=1603)βBoot 95% CIβBoot 95% CIβBoot 95% CITotal −0.097 (−0.142 to −0.054) −0.057(−0.148 to 0.030) −0.110 (−0.161 to −0.057) Direct −0.043 (−0.085 to −0.003) −0.031(−0.120 to 0.054)−0.044(−0.092 to 0.003)Indirect −0.054 (−0.075 to −0.034) −0.026(−0.071 to 0.020) −0.065 (−0.090 to −0.041) Perceived health risk −0.045 (−0.063 to −0.027) −0.016(−0.050 to 0.020) −0.058 (−0.081 to −0.036) Positive outcome expectancy −0.009 (−0.017 to −0.003) −0.010(−0.037 to 0.014) −0.007 (−0.016 to −0.001) The model for overall was controlled for gender, age, education level, monthly allowance, advertising exposure frequency; the two sub-groups were the same with the exception of e-cigarette use status.Significant associations are highlighted in bold.Boot 95% CI, CI estimated by 5000 bootstrap resampling; β, unstandardised regression coefficient. Parallel mediating effect of perceived health risk and positive outcome expectancy on the relationship between perceived changes in e-cigarette retail density and adolescents’ support for e-cigarette regulations in China, 2022–2023 The model for overall was controlled for gender, age, education level, monthly allowance, advertising exposure frequency; the two sub-groups were the same with the exception of e-cigarette use status. Significant associations are highlighted in bold. Boot 95% CI, CI estimated by 5000 bootstrap resampling; β, unstandardised regression coefficient. Discussions Our findings from an online national survey suggest broad adolescent support for regulating e-cigarettes, with a flavour ban, online sales ban and increased taxation all supported by more than half of adolescents in China. Support for a comprehensive ban was notably lower. Overall, perceived increases in retailer density were associated with a lower likelihood of supporting e-cigarette regulation, partially mediated through perceived health risk and positive outcome expectancy. However, this mediation effect was only significant in those who had never tried e-cigarettes. The decrease in e-cigarette experimentation by Chinese secondary school students between 2021 and 20232 might be partly attributable to the recently implemented regulations and their high level of support among adolescents. Noteworthily, Hong Kong, as a special administrative region of China, has completely prohibited e-cigarettes, which includes a total ban on the importation, manufacture, sale, distribution and promotion of all such products and their accessories; however, mainland China has not yet implemented a comprehensive ban on these products. The weak support for a comprehensive ban may be attributed to youths’ mixed risk perceptions, which may be driven by an absence of academic consensus and the high volume of online and advertising misinformation about e-cigarette safety.13 In prior research, nearly half of adolescents believed e-cigarettes to be an effective tool for smoking cessation and less harmful than combustible cigarettes.14 This suggests that opposition to a comprehensive ban may partially stem from not wanting to deprive people who use cigarettes of access to harm reduction. However, current evidence regarding the effectiveness of e-cigarettes for cessation remains mixed.15 16 Future studies are warranted to qualitatively investigate adolescents’ attitudes toward comprehensive bans and potential drivers of opposition to such bans. Similar to previous findings from county-level statistics of retailers and adult voters in California,8 we demonstrate that adolescents who perceived an increase in e-cigarette retail density were less likely to support e-cigarette regulations. Our use of individual-level data also precludes possible ecological fallacies. Our results have further revealed the mediating roles of positive outcome expectancy and perceived health risk in the association between changes in retail density and policy support among adolescents. High retail density increases their exposure to marketing by stores, including external advertising, promotional pricing, tobacco power walls and product placement and visibility within stores, the scent of flavoured products and free samples.17 Adolescents may perceive e-cigarette stores as trendy rather than dangerous due to these environmental cues. Notably, a prior study reported that half of Chinese e-cigarette retailers near schools did not post the required ‘not for sale to minors’ signs in shops,18 attempting to conceal the fact that e-cigarettes are not age-appropriate products for adolescents. These jointly contribute to the normalisation of e-cigarette use, decreasing perceptions of harmfulness19 and increasing adolescents’ expectations of positive outcomes, such as stress relief and improved socialisation,20 and may therefore lower support for e-cigarette regulations. However, the mediating effect of harm perceptions and positive outcome expectancy existed only among never users. Previous work reports a profound effect of adolescent smoking behaviour on future attitudes towards tobacco policies in adulthood.21 This suggests that those who have tried e-cigarettes may have developed a solidified attitude towards e-cigarette regulation and are thus less likely to be influenced by other social and environmental factors, including retail density. Notably, the negative association between perceived health risk and regulation support was significant for adolescents who had ever tried e-cigarettes, suggesting that their attitude is possibly influenced by deliberative risk assessments rather than environmental cues. We must acknowledge the following limitations. Although the study attempted to delineate the relationship between perceived changes in retailer density and regulation support, the cross-sectional nature precluded us from drawing causal inference. To address this concern, future studies should consider longitudinal designs or instrumental variable approaches. The online survey only included those who were registered on the Wenjuanxing platform, potentially leading to selection bias. We used perceived changes in retailer density rather than objective measures since tobacco use risk is more influenced by subjective measures of tobacco retailer density.12 Future research could use GIS-based methods to measure objective density and explore the causes of discrepancies between objective and subjective density perceptions. To help create adolescent support for regulation, it is critical to address perceptions of e-cigarette retail density via stricter measures such as controlling access to retailers near schools, restricting advertising known to capture youth attention, displaying mandatory signage explicitly prohibiting minors from entering stores and graphically presenting health hazards. Further, strict enforcement enabled by adequate resources needs to be implemented in response to the rise of online sales and underground e-cigarette markets driven by the recent restrictions on the e-cigarette market. Supplementary material Footnotes Funding: This work was supported by China Medical Board, Cambridge, Massachusetts, USA [grant number: #21-430]. Provenance and peer review: Not commissioned; externally peer reviewed. Patient consent for publication: Consent obtained from parent(s)/guardian(s) Ethics approval: This study involves human participants. The project was approved by the Institutional Review Board at Peking University (No: IRB00001052-22133). Participants gave informed consent to participate in the study before taking part. References Associated Data Supplementary Materials