The dual use of traditional cigarettes and e-cigarettes and behavioral and psychological associated factors among Chinese adolescents.

The dual use of traditional cigarettes and e-cigarettes and behavioral and psychological associated factors among Chinese adolescents

Abstract

Background

The dual use of traditional cigarettes and e-cigarettes is common among adolescents. Dual use is likely more harmful than smoking traditional cigarettes or using e-cigarettes alone. However, few studies have explored the behavioral and psychological factors associated with this pattern of tobacco use among Chinese adolescents.

Methods

We surveyed high school students in all 31 provincial-level administrative divisions (PLADs) in mainland China. Tobacco use was categorized as current dual use, exclusive traditional cigarette use, exclusive e-cigarette use, and non-use. We collected 11 behavioral and psychological variables potentially associated with current tobacco use, including physical activity, screen time, sleep duration, diet, alcohol use, fighting, social self-control, self-esteem, self-efficacy, stress, and depression. Multinomial logistic regression adjusted for PLAD fixed effects was employed to analyze the associations.

Results

The weighted sample included 15,000 students with an equal distribution of sex and a median age of 17 years (IQR: 16–18). About 6.9% (95% confidence interval: 6.5%–7.3%) reported current dual use. Several behavioral and psychological factors, including excessive screen time (OR = 0.70, 95% CI = 0.61–0.80; appropriate screen time as the reference group), unhealthy dietary habits (OR = 1.34, 95% CI = 1.15–1.56; healthy diet as the reference group), alcohol use (occasional: OR = 5.53, 95% CI = 4.27–7.17; habitual: OR = 12.71, 95% CI = 9.24–17.47; never use as the reference group), fighting (occasionally: OR = 1.69, 95% CI = 1.41–2.03; habitual: OR = 1.83, 95% CI = 1.18–2.82; never fight as the reference group), strong social self-control (OR = 0.69, 95% CI = 0.57–0.82; moderate as the reference group), strong self-esteem (OR = 1.62, 95% CI = 1.36–1.94; moderate as the reference group), self-efficacy (strong: OR = 1.61, 95% CI = 1.38–1.88; weak: OR = 0.78, 95% CI = 0.62–0.97; moderate as the reference group), and strong stress (OR = 0.78, 95% CI = 0.64–0.95; moderate as the reference group), were found to be associated with the current dual use. Some of the above factors were also associated with the current exclusive use of traditional cigarettes or e-cigarettes.

Conclusions

This study reports on the phenomenon of current dual use and its behavioral and psychological associated factors among high school students in China. Families and schools should pay attention to adolescents’ health risk behaviors and psychological health, as these factors were found to be associated with tobacco use.

Clinical trial number

Not applicable.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s13690-026-02032-9.

Keywords: Cigarettes, E-cigarettes, Electronic nicotine delivery systems, Current Dual use, Adolescents

Text box 1. Contributions to the literature

This study offers nationally representative evidence on the dual use of cigarettes and e-cigarettes among adolescents in China.

It extends existing knowledge by investigating the roles of behavioral and psychological factors in adolescents’ dual use of cigarettes and e-cigarettes.

The findings provide empirical support for designing public health policies and interventions to prevent adolescent tobacco use in China and other countries with comparable socioeconomic and regulatory backgrounds.

Introduction

Tobacco use during adolescence is a significant global public health concern with profound health implications. Most adult smokers begin smoking during childhood or adolescence. Compared with adults, adolescents can develop nicotine dependence more rapidly and with fewer cigarettes [1]. In recent years, e-cigarettes have been very popular among adolescents. Evidence from international surveys suggests that e-cigarette use among adolescents is widespread across multiple world regions. Data from the 2022 Health Behaviour in School-aged Children (HBSC) survey showed that, on average, 19.4% of 15-year-olds across participating European countries and Canada reported using e-cigarettes in the past 30 days, with prevalence exceeding 30% in some countries [2]. The Global Youth Tobacco Survey (GYTS) has further demonstrated that past-30-day e-cigarette use among youths reached a global prevalence of 9.2% [3]. In mainland China, 9.5% of high school students were e-cigarette users [4], and the prevalence of e-cigarette use among adolescents is higher than that in other age groups [5]. However, the use of e-cigarettes has not led adolescents to completely give up traditional cigarettes. Studies have recently shown that the use of multiple tobacco products, particularly the dual use of traditional cigarettes and e-cigarettes, is common among adolescents. A survey conducted in 2020 revealed that 38.9% of American youth who were current e-cigarette users also used other tobacco products and that the dual use of e-cigarettes and combustible tobacco was the most common pattern [6]. Similarly, a study in China reported that 48.3% of college students who were tobacco users were dual users [7].

E-cigarette users are exposed to a different mixture of toxic chemicals than those found in traditional cigarettes, including compounds that are generated during the heating and aerosolization processes but are not present in the original e-liquid. As a result, dual users are exposed to a broader and more complex range of toxic substances than those who exclusively use either e-cigarettes or traditional cigarettes [8]. Therefore, due to the significant public health impact of dual use, it is important to better understand the factors associated with combined tobacco and nicotine use patterns to inform the most appropriate and effective public health responses.

Most previous studies have focused on single tobacco products, with limited research on the dual use of traditional cigarettes and e-cigarettes among adolescents or how the associations between dual use and various factors differ from the use of single products [6, 9, 10]. Some research on dual use has focused on social‒environmental factors, such as peer influence, family relationships, and advertising exposure [11–13], with relatively less attention given to individual behavioral and psychological factors. However, behavioral and psychological factors play critical roles in health across the lifespan [14]. Health risk behaviors (HRBs), such as unhealthy eating habits, insufficient physical activity, and tobacco use, are prevalent during adolescence and seldom occur in isolation [15, 16]. Therefore, when analyzing the associated factors for dual use, it is essential to recognize the associations of dual use with other health risk behaviors to develop more effective health promotion and prevention strategies [17].

In addition, the dual use of e-cigarettes and traditional cigarettes may be associated with psychological factors. Previous studies have demonstrated that psychological symptoms, such as stress and depression, predict adolescent smoking [18]. Low self-esteem and self-control are associated with a greater risk of smoking [19, 20]. Self-efficacy, defined as individuals’ beliefs about their ability to exercise control over their actions and associated outcomes, is also a risk factor for adolescent smoking [21].

Moreover, China is the world’s largest consumer of tobacco products and contributes substantially to the global burden of smoking-related diseases [22]. However, previous research on dual use has been conducted mainly in Europe and the United States, with limited evidence from China. China ratified the World Health Organization (WHO) Framework Convention on Tobacco Control (FCTC) in 2005, and the treaty came into force in the country on January 9, 2006. Although China has made notable progress in tobacco control since then, substantial gaps remain in fully meeting the FCTC requirements [23]. For instance, tobacco taxation remains relatively low, and both package warnings and Smoke and Aerosol-Free Environment (SAFE) environment policies in China are considerably weaker compared to those in Hong Kong and many other regions [24]. Additionally, China has the second-largest adolescent population in the world, accounting for 12.9% of the global adolescent population [25]. Therefore, researching the factors associated with current dual use among Chinese adolescents is both meaningful and urgent.

To fill these gaps, this study aims to examine the prevalence of different tobacco use patterns and the associations between behavioral and psychological factors and current dual use of traditional cigarettes and e-cigarettes among adolescents in China. This study seeks to enhance understanding of adolescent tobacco and nicotine use patterns and to inform the development of more targeted prevention strategies in China and other sociocultural contexts.

Methods

Survey design

We conducted an anonymous online cross-sectional survey using a two-step nationwide design, administered by KuRunData, a platform that has been extensively utilized in previous research for data collection [26, 27]. The survey was conducted between July and September 2021. Before the formal survey, a pilot study involving 104 high school students was conducted to refine the questionnaire and estimate the sample size. We used the sample size calculator developed by Wang et al. [28] to estimate the minimum required sample size. Based on the results of an online pilot survey, we set the proportion of current e-cigarette users among high school students (p) to 8.7%. We set the type I error rate (α) at 0.05 and the absolute error (d) at 0.5%. With these parameters, the minimum sample size was estimated to be 12,206. To simplify the sample size allocation process, we increased the overall sample size goal to 15,000.

In the first stage, the overall sample size goal was allocated to each of 31 provincial-level administrative divisions (PLADs) based on the proportion of high school students in each PLAD relative to the national total [29, 30]. In the second stage, we stratified the sampling within each PLAD by sex (male or female) [30], residence (urban or rural) [31], and school type (general high schools or vocational high schools) to obtain representative samples [29, 30]. We further allocated sample size goals to each stratum based on the proportion of high school students in each stratum relative to the total number in the corresponding PLAD. To minimize sampling errors, we oversampled strata with sample size goals below 30 to ensure that at least 30 students were surveyed per stratum. This resulted in the actual number of surveyed students exceeding the predetermined goal of 15,000, reaching 16,365 participants.

Considering that high school students may not have internet access during academic semesters, the survey was conducted during the summer break from July to September 2021. A total of 21,085 high school students accessed the e-questionnaire, and 17,539 completed it. Of the 17,539 responses, 613 were determined to be ineligible, and 561 were kept as backup data to avoid inadequate sample sizes caused by cleaning ineligible data. The 561 backup responses resulted from a technical delay in closing the online survey after the regional target sample sizes had been reached. Specifically, responses were classified as backup if they were submitted after the predefined target sample size for a given stratum had already been achieved. However, these 561 responses were not included in the final analysis. Ultimately, 16,365 respondents were included in our analysis.

To ensure that the composition of our sample aligned with official statistics for each PLAD in terms of sex, residence, and educational attainment, we assigned sample weights to each respondent [29–31]. These weights were calculated by dividing the sample size goal by the actual number of respondents within each stratum. This weighting approach enables our data to be nationally representative. Comprehensive details regarding the survey implementation and quality control procedures are provided in the Supplementary Methods.

Measurement

Outcome variable: tobacco use status

Tobacco use status was the outcome variable in our analysis. This variable was classified into four categories: current exclusive use of traditional cigarettes, exclusive use of e-cigarettes, dual use, and non-use of either e-cigarettes or traditional cigarettes. Based on previous research, the use of traditional cigarettes and e-cigarettes was determined by the self-reported frequency of use in the past 30 days. Any response indicating use on one or more days was considered current use [5, 32]. Respondents who reported current use of traditional cigarettes but did not use e-cigarettes were classified as current exclusive traditional cigarette users, whereas those who used only e-cigarettes and did not use traditional cigarettes were classified as current exclusive e-cigarette users. Individuals who had used both traditional cigarettes and e-cigarettes in the past 30 days were classified as current dual users, whereas those who had not used either traditional cigarettes or e-cigarettes were classified as current non-users.

Behavioral and psychological factors potentially associated with current tobacco use status

Based on previous studies [15, 16, 18–21], we collected data on 11 factors potentially associated with the current dual use of traditional cigarettes and e-cigarettes, including six behavioral and five psychological factors. Behavioral factors (physical activity, diet, sleep duration, screen time, alcohol use, and fighting) were included because they are modifiable lifestyle behaviors that can be targeted in preventive interventions. Psychological factors (social self-control, self-esteem, self-efficacy, stress, and depression) were examined as they may serve as indicators to identify adolescents at risk of engaging in risky behaviors.

Physical activity was assessed using items adapted from the International Physical Activity Questionnaire (IPAQ) [33]. Diet, alcohol use, and fighting were adapted from the Global School-based Student Health Survey (GSHS) core questionnaire modules [34]. Social self-control was measured using the Chinese version of the Social Self-Control Scale (Cronbach’s α = 0.76 in this study) [35]. Self-esteem was assessed using the Chinese version of the Rosenberg Self-Esteem Scale (Cronbach’s α = 0.90) [36]. Self-efficacy was measured using the Chinese version of the General Self-Efficacy Scale (Cronbach’s α = 0.88) [37]. Stress was measured using the Chinese version of the 10-item Perceived Stress Scale (PSS-10) (Cronbach’s α = 0.71) [38], and depression was assessed using the Chinese version of the Patient Health Questionnaire (PHQ-9) (Cronbach’s α = 0.87) [39], Other items were self-developed for this study. Given that psychological scale scores may not exhibit a strictly linear relationship with the outcome, these variables were treated as categorical variables in the regression analyses. The specific details can be found in Table S1.

Covariates

The following control variables were included in the analysis: age (years), sex (male, female), current parental tobacco use, defined as current use of traditional cigarettes, e-cigarettes, or both (neither parent, father only, mother only, and both parents), and body mass index (BMI) and its quadratic term (both centered to reduce potential collinearity, with both variables centered by subtracting the mean value), to account for potential non-linear associations between BMI and tobacco use [40–42].

Data analysis

We used descriptive statistics to summarize the characteristics of the survey participants. We present the numbers and proportions of the sample size for each PLAD. We described sample characteristics including age, sex, residence, school type, and grade. Continuous variables were presented as medians and interquartile ranges (IQRs). Categorical variables were presented as frequencies and percentages. Age was presented as both a categorical variable and a continuous variable. Additionally, we present the prevalence of current traditional cigarette use, e-cigarette use, and dual use among male and female high school students by PLAD using bivariate choropleth maps.

Because our outcome variable was classified into four categories, we used a multinomial logistic regression model to analyze the behavioral and psychological factors associated with current dual use status: exclusive use of traditional cigarettes, exclusive use of e-cigarettes, dual use, and non-use of either e-cigarettes or traditional cigarettes [4]. Compared with other regression models, the multinomial logistic regression model can provide odds ratios for each category compared with the reference category to offer insights into the likelihood of different outcomes. In this multinomial logistic model, current tobacco use status was designated as the outcome variable, with non-use of either e-cigarettes or traditional cigarettes as the reference category, and behavioral and psychological factors were included as exposure variables. Covariates, including age, sex, and centered BMI and its quadratic term, were also included in the model. Considering that some unmeasurable confounders were not included in the model but were consistent within each PLAD, we incorporated binary indicators for PLADs as fixed effects in the model. This methodological approach has been widely used in previous research [27, 43, 44]. The effect sizes of the associations between behavioral and psychological factors were presented as adjusted ORs with 95% confidence intervals (CIs) in a forest plot. Previous studies have reported sex and age group differences in tobacco use [45–48]. Therefore, in addition to performing multinomial logistic regression on the entire sample, we conducted subgroup analyses by sex and age group.

To account for the multistage sampling process in the survey, we performed complex survey data analyses for all statistical inferences of standard errors, 95% CIs, and statistical significance, with standard errors were clustered at the level of primary sampling units [49]. All statistical tests were two-sided, with statistical significance denoted at P < 0.05. All data analyses were conducted with Stata Statistical Software version 18 (StataCorp., Texas., USA) [50]. Bivariate choropleth maps and forest plots were generated using R 4.3.3 (R Core Team, 2024) [51].

Results

Sample characteristics

In the unweighted sample, the largest proportions of students were from Henan (8.3%), Guangdong (6.8%), and Shandong (6.1%) (Table S2). The weighted sample of 15,000 students included approximately equal proportions of males (49.9%) and females (50.1%). The median age of the students was 17 years (IQR: 16–18). The majority of the students resided in urban areas (59.5%), attended general high schools (66.5%), and were in the 12th grade (48.8%). Few differences were observed in sample characteristics between the unweighted (Table S3) and weighted (Table 1) samples.

Table 1Sample characteristics by e-cigarette and traditional cigarette current use status, weighted

Characteristica Full sample (nweighted = 15,000; 100%), n (%) Current dual users (nweighted = 1,029; 6.9%), n (%) Current exclusive traditional cigarette users (nweighted = 2,655; 17.7%), n (%) Current exclusive e-cigarette users (nweighted = 398; 2.7%), n (%) Current non-users (nweighted = 10,918; 72.8%), n (%)

Male, n (%) 7,483 (100%) 874 (11.7%) 2,351 (31.4%) 287 (3.8%) 3,971 (53.1%)

Age, median (Q1, Q3) 17 (16, 18) 18 (17, 18) 17 (17, 18) 17 (16, 18) 17 (16, 18)

Age (years), n (%)

 ≤ 16 2,004 (13.4%) 60 (5.8%) 224 (8.4%) 48 (12.1%) 1,672 (15.3%)

 16–17 3,709 (24.7%) 207 (20.1%) 629 (23.7%) 109 (27.4%) 2,765 (25.3%)

 17–18 5,374 (35.8%) 365 (35.4%) 1,163 (43.8%) 136 (34.1%) 3,711 (34.0%)

 > 18 3,914 (26.1%) 398 (38.6%) 639 (24.1%) 105 (26.5%) 2,771 (25.4%)

Urban residence, n (%) 8,924 (59.5%) 697 (67.7%) 1,866 (70.3%) 201 (50.5%) 6,160 (56.4%)

School type, n (%)

 General high schools 9,977 (66.5%) 355 (34.5%) 1,743 (65.6%) 118 (29.6%) 7,762 (71.1%)

 Vocational high schools 5,022 (33.5%) 674 (65.5%) 912 (34.4%) 280 (70.4%) 3,156 (28.9%)

Grade, n (%)

 10th 3,069 (20.5%) 130 (12.6%) 382 (14.4%) 81 (20.5%) 2,476 (22.7%)

 11th 4,605 (30.7%) 273 (26.6%) 936 (35.3%) 135 (34.0%) 3,260 (29.9%)

 12th 7,326 (48.8%) 626 (60.8%) 1,336 (50.3%) 181 (45.5%) 5,182 (47.5%)

The presented results were derived from a weighted sample. aPercentages are presented by column for all categorical variables, except for “male,” for which row percentages are reported to show the distribution of tobacco use status within males

n (%) frequency with percentage, IQR interquartile range

Current tobacco use status among Chinese adolescents

In this survey, 6.9% (95% CI: 6.5%–7.3%) of the students in the weighted sample reported current dual use of traditional cigarettes and e-cigarettes. The prevalence of current dual use among male students was 11.7% (95% CI: 11.0%–12.4%), while the prevalence among female students was 2.1% (95% CI: 1.8%–2.4%). Geographically, the highest prevalence of dual use was found in eastern areas such as Shanghai (11.7%, 95% CI: 4.4%–27.8%), Beijing (11.1%, 95% CI: 4.9%–23.4%), and Hainan (9.9%, 95% CI: 4.8%–19.3%). The lowest prevalence was observed in western areas such as Chongqing (4.1%, 95% CI: 1.8%–9.0%), Guangxi (4.2%, 95% CI: 1.5%–11.2%), and Shaanxi (4.9%, 95% CI: 2.4%–9.7%) (Figure S1a, Table S6). Nearly a quarter (24.6%, 95% CI: 23.9%–25.3%) of high school students reported the current use of traditional cigarettes. Among traditional cigarette current users, 27.9% (1,029/3,684) were dual users (Figure S1b, Table S4). Approximately one-tenth (9.5%, 95% CI: 9.1%–10.0%) of high school students used e-cigarettes. Among e-cigarette users, 72.1% (1,029/1,427) were dual users (Figure S1c, Table S5).

Behavioral and psychological factors associated with current tobacco use status

Several findings from the weighted sample highlight the behavioral and psychological characteristics of high school students. With regard to behavioral factors, we found that, on average, Chinese high school students participate in only 13 min of moderate-to-vigorous physical activity per day. Additionally, 55.7% of high school students spend more than two hours on screens daily, and 67.4% do not get sufficient sleep. Unhealthy diets are also prevalent, with 49.6% of students reporting poor dietary habits. Alcohol use is common, with 69.5% of students consuming alcohol; furthermore, 11.0% of students engage in physical fights. In terms of psychological factors, 11.6% of the students experienced symptoms of moderately severe or severe depression (Table S7).

A weighted multinomial logistic regression model revealed that most behavioral factors were associated with current dual-use status. Among the behavioral factors, excessive screen time (OR = 0.70, 95% CI: 0.61–0.80; appropriate screen time as the reference group) was inversely associated with dual use, whereas unhealthy diet (OR = 1.34, 95% CI: 1.15–1.56; healthy diet as the reference group), alcohol use (occasional: OR = 5.53, 95% CI: 4.27–7.17; habitual: OR = 12.71, 95% CI: 9.24–17.47; never use as the reference group), and fighting (occasional: OR = 1.69, 95% CI: 1.41–2.03; habitual: OR = 1.83, 95% CI: 1.18–2.82; never fight as the reference group) were positively associated with dual use. Among the psychological factors, when moderate levels were used as the reference, strong self-esteem (OR = 1.62, 95% CI: 1.36–1.94; moderate as the reference group), and strong self-efficacy (OR = 1.61, 95% CI: 1.38–1.88; moderate as the reference group) were positively associated with dual use, whereas strong social self-control (OR = 0.69, 95% CI: 0.57–0.82; moderate as the reference group), strong stress (OR = 0.78, 95% CI: 0.64–0.95; moderate as the reference group) and weak self-efficacy (OR = 0.78, 95% CI: 0.62–0.97; moderate as the reference group) was associated with a lower likelihood of dual use (Fig. 1, Table S8).

Many of the significant factors mentioned above were also associated with the current exclusive use of traditional cigarettes and e-cigarettes, with some exceptions. Habitual fighting was not associated with the exclusive use of traditional cigarettes despite being linked to dual use. In contrast, although insufficient sleep was not associated with dual use, it was associated with the exclusive use of traditional cigarettes. Weak stress (compared with moderate stress: OR = 1.17, 95% CI: 1.03–1.34) and weak self-efficacy were (compared with moderate self-efficacy: OR = 0.82, 95% CI: 0.70–0.97) associated with the exclusive use of traditional cigarettes but not with dual use. Additionally, insufficient sleep duration, habitual fighting, strong social self-control, strong self-esteem, weak self-efficacy, and stress were not associated with exclusive e-cigarette use. Weak social self-control (OR = 1.61, 95% CI: 1.25–2.08; moderate as the reference group) was associated with exclusive e-cigarette use but was not linked to dual use (Fig. 1, Table S8).

No apparent heterogeneity was observed when the analyses were stratified by sex and age group. For specific details, please refer to Table S9 and Table S10.

Discussion

In this study of a nationally representative, weighted sample of 15,000 Chinese high school students, we examined behavioral and psychological factors associated with current tobacco use. We found that over one-quarter of students used traditional cigarettes, about one-tenth used e-cigarettes, and 6.9% were dual users. Several behavioral and psychological factors—including screen time, diet, alcohol use, fighting, and self-regulation measures—were associated with dual use, with similar patterns observed for exclusive use. These findings underscore the importance of addressing adolescents’ health behaviors and psychological well-being in efforts to reduce tobacco use.

The proportion of dual users among the high school students in this study (6.9%) is comparable to the percentage reported among adolescents in Canada (6.3%) [53]. A previous study revealed that the prevalence of dual use in 67 countries ranged from 0.1% to 13.6%, and the prevalence of dual use exceeding 5% in only 15 countries. This suggests a high prevalence of dual use among Chinese adolescents [54]. Considering that 27.9% of traditional cigarette users and 72.1% of e-cigarette users are dual users, interventions that target either e-cigarette use or traditional tobacco use exclusively may not be highly effective because a significant number of adolescents are dual users. Therefore, more comprehensive intervention strategies that target the dual use of e-cigarettes and traditional cigarettes are needed.

We found that most behavioral factors were associated with dual use. Unhealthy diets, alcohol consumption, and fighting were all associated with a higher risk of dual use, reflecting the clustering of unhealthy behaviors. In particular, alcohol use was associated with greater risk of dual use among adolescents, which is consistent with the findings of previous studies [55]. These findings suggest that adolescents' health risk behaviors are closely related and that these behaviors should be regarded as associated indicators of traditional cigarette and e-cigarette use to enable early detection and timely intervention.

However, adolescents with more than two hours of daily screen time showed a lower likelihood of dual use, differing from prior studies [56, 57]. This finding may be explained by several context-specific factors. First, the survey was conducted during the COVID-19 pandemic, when China implemented the “Classes Suspended but Learning Continues” emergency plan, requiring all teaching and learning activities to move online [58]. As a result, much of the extended screen time was likely devoted to educational purposes — a type of screen use that existing evidence indicates is associated with lower odds of smoking, in contrast to social or entertainment-related screen activities, which tend to increase smoking risk [59, 60]. Second, China has implemented strict regulations on tobacco-related online content that prohibit direct and indirect forms of advertising including converting advertisements by influencers, films and other forms, aiming to limit adolescents’ exposure to tobacco-related content [24, 61, 62]. These considerations support the plausibility of our finding that longer screen time was associated with lower dual use risk and underscore the need for future research to differentiate types of screen activities when examining their relationship with adolescent tobacco use.

Given that the behavioral factors associated with dual use—such as alcohol use, fighting, and cigarette use—are strongly interconnected, preventing the dual use of e-cigarettes and traditional cigarettes among adolescents requires comprehensive interventions that target multiple health risk behaviors simultaneously, rather than focusing on a single behavior [63]. Accordingly, several strategies can be implemented. First, parents can be supported through structured family-based interventions delivered or reinforced by schools and public health programs, which provide guidance on healthy lifestyle practices and parenting skills. By improving parents’ own health literacy and behavioral modeling, such programs may help create a supportive home environment that reduces the likelihood of cigarette and e-cigarette use among adolescents [64–66]. Second, schools and families should strengthen collaboration, particularly for adolescents who exhibit behaviors such as tobacco use, alcohol consumption, or fighting, as these behaviors are closely linked to dual use. Close monitoring and timely intervention are therefore essential [67]. Evidence from existing programs supports this integrated approach. For example, the Adolescent Transition Program in middle schools in the United States combines a universal classroom-based intervention, the Family Check-Up, and family management treatments to target multiple problem behaviors simultaneously [68]. Similarly, the Family-School Partnership (FSP) program in the United States, which involves weekly home-school learning sessions, communication activities, and periodic parenting workshops, has demonstrated that coordinated efforts between families and schools can effectively reduce the risk of tobacco smoking [69].

Among the psychological factors, firstly, the observed association between social self-control and dual use supports previous evidence, highlighting the role of self-regulation in adolescents’ susceptibility to tobacco use [70]. Adolescents with poor social self-control are more likely to associate with deviant peers who are more likely to use substances, making them more susceptible to smoking under peer influence [71, 72]. Secondly, adolescents with higher self-esteem and self-efficacy were more likely to engage in dual use, contrary to previous findings [73]. This may be explained by the fact that individuals with low self-esteem may lack confidence in their abilities and avoid risky behaviors such as smoking, whereas those with high self-esteem may have an elevated sense of their capabilities, increasing their likelihood of engaging in risky behaviors [74]. Similarly, individuals with strong self-efficacy tend to have greater confidence in their ability and may believe that they can quit anytime or manage the negative consequences [75–77]. Finally, we found that higher stress were associated with a lower risk of dual use. In the Chinese context, adolescents’ stress primarily stems from intense academic demands, which are reinforced by stronger parental and teacher expectations compared with those faced by their peers in many Western countries [78, 79]. Those tend to increase students’ compliance with school rules and reduce the amount of free time and social opportunities available for engaging in risky behaviors such as smoking. As a result, academically driven stress may inadvertently limit adolescents’ likelihood of dual use.

Based on the observed associations between psychological factors and dual use, prevention initiatives may consider integrating psychological skill development into tobacco control programs [80]. For example, the Life Skills Training Program in New York State integrates cognitive-behavioral skill development, including strengthening self-esteem, resisting advertising influence, managing anxiety, and improving communication, which provides a valuable model for future interventions [81]. In addition, the European “Unplugged” program, evaluated in the EU-Dap cluster randomized controlled trial, incorporates social and emotional skill training, normative education, and critical thinking components to prevent tobacco and other substance use among adolescents, and has demonstrated effectiveness in reducing smoking initiation in randomized controlled trials, further supporting the value of psychological skill–based approaches in tobacco prevention [82].

Based on our findings and previously validated interventions, we recommend that China’s education authorities and other relevant government agencies adopt comprehensive strategies that simultaneously address behavioral and psychological factors associated with the use of traditional cigarettes and e-cigarettes to develop more effective interventions.

Our study has several strengths. First, it included a weighted sample of 15,000 participants from across all 31 PLADs in mainland China, making it both large and nationally representative. Second, the study investigated behavioral and psychological factors associated with current dual use of e-cigarettes and traditional cigarettes and compared them with those among exclusive users. This study, therefore, deepens the understanding of dual use as a specific pattern of tobacco use and provides some empirical evidence regarding associated factors of dual use. Finally, since each province has its own unique policies and cultural contexts for tobacco that are difficult to measure, we included binary indicators of PLADs as fixed effects in the model to effectively control for unmeasurable confounding factors [44].

Our study also has certain limitations. Firstly, it relied on data from an online survey of high school students in China, which limits the extent to which our findings can be generalized to other groups, such as adolescents without internet access or those in different countries. Secondly, among the 21,085 individuals who accessed the questionnaire, only 16,365 were included in the data analysis, with 4,720 excluded. This exclusion may have introduced potential bias into our study findings. Thirdly, because the data were self-reported, there may be biases due to the participants’ inaccurate recall, as well as possible underreporting or overreporting. Finally, the cross-sectional design of our study means that our findings should be viewed as associations rather than causal relationships. It is possible that some of the observed associations reflect reverse causation — adolescents who engage in dual use may subsequently report higher levels of self-esteem or self-efficacy due to perceived social acceptance, autonomy, or a sense of maturity gained from engaging in adult-like behaviors [75–77]. Future studies are encouraged to investigate the causal links between health risk behaviors, psychological factors, and dual-use in prospective cohort studies.

In conclusion, among this large, nationally representative sample, we found that more than one-quarter of students used traditional cigarettes, about one-tenth used e-cigarettes, and 6.9% were current dual users. Dual use was associated with multiple behavioral and psychological factors, including screen time, diet, alcohol use, fighting, social self-control, self-esteem, self-efficacy, and stress, many of which were also related to exclusive use of either product. These findings underscore the importance of family- and school-based efforts to monitor and support adolescents’ health behaviors and psychological well-being to address the growing challenge of dual tobacco and nicotine use.

Supplementary Information

Acknowledgements

Not applicable.

Abbreviations

PLADsProvincial-level administrative divisions

OROdds ratio

CIConfidence intervals

IQRInterquartile range

HRBHealth risk behavior

WHOWorld Health Organization

FCTCFramework Convention on Tobacco Control

BMIBody mass index

Authors’ contributions

KM conceptualized and designed the study. RM, ZM and LF conducted the analyses, interpreted the data, and wrote the article. KM, RM and LF critically revised the article. QL and JY critically revised the article. All authors approved the final version.

Funding

JY received grants from the State Key Laboratory Special Fund 2060204 and the Chinese Academy of Medical Sciences Innovation Fund for Medical Sciences, China (CIFMS2022-I2M-2–001, CIFMS2022-I2M-1–011, CIFMS2021-I2M-1–057, CIFMS2021-I2M-1–049, CIFMS2021-I2M-1–044, CIFMS2021-I2M-1–016, and CIFMS2021-I2M-1–001). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Data availability

The data and analytic codes used in this study will be available six months after the time of publication, with no end date, from the corresponding authors upon reasonable request.

Declarations

Ethics approval and consent to participate

The study was reviewed and approved by the Ethics Committee of the Institute of Basic Medical Sciences, CAMS & PUMC (No. 2021071). Our study adhered to the principles delineated in the Declaration of Helsinki. In accordance with the committee’s requirements, electronic informed consent was obtained from all participants, and the requirement for guardians’ informed consent was waived.

Consent for publication

Not applicable.

Competing interests

We declare no competing interests.

Footnotes

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Kaiyuan Min, Runping Ma and Lewen Fu contributed equally as co-first authors.

Contributor Information

Qing Li, Email: liqing@pumc.edu.cn.

Juntao Yang, Email: yangjt@pumc.edu.cn.

References

1.Smoking still a core challenge for child and adolescent health reveals WHO report. https://www.who.int/europe/news/item/05-06-2020-smoking-still-a-core-challenge-for-child-and-adolescent-health-reveals-who-report.

2.Charrier L, Van Dorsselaer S, Canale N, Baska T, Kilibarda B, Comoretto RI, et al. A focus on adolescent substance use in Europe, Central Asia and Canada. Health behaviour in school-aged children international report from the 2021/2022 survey. Volume 3. World Health Organization. Regional Office for Europe; 2024.

3.Sun J, Xi B, Ma C, Zhao M, Bovet P. Prevalence of e-cigarette use and its associated factors among youths aged 12 to 16 years in 68 countries and territories: global youth tobacco survey, 2012–2019. Am J Public Health. 2022;112(4):650–61.

4.Min K, Wang M, Wang C, Geldsetzer P, Bärnighausen T, Wang C, et al. Prevalence and associated factors of e-cigarette use and susceptibility among adolescents in China. Am J Respir Crit Care Med. 2024;209(12):1505–7.

5.Zhao Z, Zhang M, Wu J, Xu X, Yin P, Huang Z, et al. E-cigarette use among adults in China: findings from repeated cross-sectional surveys in 2015–16 and 2018–19. Lancet Public Health. 2020;5(12):e639–49.

6.Tashakkori NA, Park-Lee E, Roh EJ, Christensen CH. Multiple tobacco product use among youth e-cigarette users: national youth tobacco survey, 2020. J Adolesc Health. 2023;73(4):769–75.

7.Song H, Yang X, Yang W, Dai Y, Duan K, Jiang X, et al. Cigarettes smoking and e-cigarettes using among university students: a cross-section survey in Guangzhou, China, 2021. BMC Public Health. 2023;23(1):438.

8.Glantz SA, Nguyen N. Silva ALOd: population-based disease odds for e-cigarettes and dual use versus cigarettes. NEJM Evidence. 2024;3(3):EVIDoa2300229.

9.Jamal A, King BA, Neff LJ, Whitmill J, Babb SD, Graffunder CM. Current cigarette smoking among adults - United States, 2005–2015. MMWR Morb Mortal Wkly Rep. 2016;65(44):1205–11.

10.Dai H, Leventhal AM. Prevalence of e-cigarette use among adults in the United States, 2014–2018. JAMA. 2019;322(18):1824–7.

11.Singh T, Agaku IT, Arrazola RA, Marynak KL, Neff LJ, Rolle IT, et al. Exposure to advertisements and electronic cigarette use among US middle and high school students. Pediatrics. 2016;137(5):e20154155.

12.Betts JM, Dowd AN, Motschman CA, Tiffany ST. An evaluation of the contextual factors associated with the use of either tobacco or electronic cigarettes in individuals who dual use. Nicotine Tob Res. 2023;26(5):561–70.

13.Min K, Meng Z, Ma R, Li Q, Yang J. Cigarette and e-cigarette use among adolescents based on a nationwide survey — China, 2021. China CDC Wkly. 2026;8(15):448–53.

14.The National Institute on Aging: Strategic Directions for Research, 2020–2025. https://www.nia.nih.gov/about/aging-strategic-directions-research/goal-behavioral-psychological-factors.

15.Li X, Dessie Y, Mwanyika-Sando M, Assefa N, Millogo O, Manu A, et al. Co-occurrence of and factors associated with health risk behaviors among adolescents: a multi-center study in sub-Saharan Africa, China, and India. EClinicalMedicine. 2024;70:102525.

16.Busch V, Van Stel HF, Schrijvers AJ, de Leeuw JR. Clustering of health-related behaviors, health outcomes and demographics in Dutch adolescents: a cross-sectional study. BMC Public Health. 2013;13:1118.

17.Prochaska JJ, Spring B, Nigg CR. Multiple health behavior change research: an introduction and overview. Prev Med. 2008;46(3):181–8.

18.Do EK, Tulsiani S, Koris K, Minter T, Hair EC. Depression, anxiety, stress, and current e-cigarette use: results from the Truth Longitudinal Cohort of youth and young adults (2022–2023). J Affect Disord. 2024;365:628–33.

19.Daly M, Delaney L, Baumeister RF. Self-control, future orientation, smoking, and the impact of Dutch tobacco control measures. Addict Behav Rep. 2015;1:89–96.

20.Carvajal SC, Wiatrek DE, Evans RI, Knee CR, Nash SG. Psychosocial determinants of the onset and escalation of smoking: cross-sectional and prospective findings in multiethnic middle school samples. J Adolesc Health. 2000;27(4):255–65.

21.Tate C, Kumar R, Murray JM, Sanchez-Franco S, Montgomery SC, Montes F, et al. Socio-environmental and psychosocial predictors of smoking susceptibility among adolescents with contrasting socio-cultural characteristics: a comparative analysis. BMC Public Health. 2021;21(1):2240.

22.Zheng Y, Ji Y, Dong H, Chang C. The prevalence of smoking, second-hand smoke exposure, and knowledge of the health hazards of smoking among internal migrants in 12 provinces in China: a cross-sectional analysis. BMC Public Health. 2018;18(1):655.

23.Yang G, Wang Y, Wu Y, Yang J, Wan X. The road to effective tobacco control in China. Lancet. 2015;385(9972):1019–28.

24.Chan KH, Xiao D, Zhou M, Peto R, Chen Z. Tobacco control in China. Lancet Public Health. 2023;8(12):e1006–15.

25.Disease GIfCOL: Global strategy for prevention, diagnosis and management of COPD. In.: Global Initiative for Chronic Obstructive Lung Disease Fontana, WI, USA; 2022.

26.Homepage for KuRunData. https://www.kurundata.com/en/index.html.

27.Yu H, Du R, Wang M, Yu F, Yang J, Jiao L, et al. Attitudes toward the global allocation of Chinese COVID-19 vaccines: cross-sectional online survey of adults living in China. JMIR Public Health Surveill. 2022;8(6):e33484.

28.Wang X, Ji X. Sample size estimation in clinical research: from randomized controlled trials to observational studies. Chest. 2020;158(1s):S12-s20.

29.Number of Students in Secondary Vocational Schools. http://www.moe.gov.cn/jyb_sjzl/moe_560/2020/gedi/202109/t20210902_557961.html.

30.Number of Students in Regular Senior Secondary Schools(Total). http://www.moe.gov.cn/jyb_sjzl/moe_560/2020/gedi/202109/t20210903_558595.html.

31.National Bureau of Statistics of China. 2020 China statistical yearbook. Beijing, China: China Statistics Press; 2021.

32.Yoong SL, Hall A, Leonard A, McCrabb S, Wiggers J, Tursan d’Espaignet E, et al. Prevalence of electronic nicotine delivery systems and electronic non-nicotine delivery systems in children and adolescents: a systematic review and meta-analysis. Lancet Public Health. 2021;6(9):e661–73.

33.Craig CL, Marshall AL, Sjöström M, Bauman AE, Booth ML, Ainsworth BE, et al. International physical activity questionnaire: 12-country reliability and validity. Med Sci Sports Exerc. 2003;35(8):1381–95.

34.Bischops AC, Radev ST, Köthe U, Chen S, Geldsetzer P, Sarker M, et al. Data resource profile: the Global School-based Student Health Survey—behavioural risk and protective factors among adolescents. Int J Epidemiol. 2022;52(2):e102–9.

35.Sussman S, McCuller WJ, Dent CW. The associations of social self-control, personality disorders, and demographics with drug use among high-risk youth. Addict Behav. 2003;28(6):1159–66.

36.Chen F, Bi C, Han M. The reliability and validity of the Chinese version of the revised-positive version of rosenberg self-esteem scale. Adv Psychol. 2015;5:531–5.

37.Zhang J, Schwarzer R. Measuring optimistic self-beliefs: A Chinese adaptation of the General Self-Efficacy Scale. Psychologia: An International Journal of Psychology in the Orient. 1995;38:174–181.

38.Lu W, Bian Q, Wang W, Wu X, Wang Z, Zhao M. Chinese version of the Perceived Stress Scale-10: a psychometric study in Chinese university students. PLoS ONE. 2017;12(12):e0189543.

39.Wang W, Bian Q, Zhao Y, Li X, Wang W, Du J, et al. Reliability and validity of the Chinese version of the Patient Health Questionnaire (PHQ-9) in the general population. Gen Hosp Psychiatry. 2014;36(5):539–44.

40.Audrain-McGovern J, Benowitz NL. Cigarette smoking, nicotine, and body weight. Clin Pharmacol Ther. 2011;90(1):164–8.

41.Harrell FE. Regression modeling strategies. New York: Springer; 2019.

42.Farhat T, Iannotti RJ, Simons-Morton BG. Overweight, obesity, youth, and health-risk behaviors. Am J Prev Med. 2010;38(3):258–67.

43.Yu F, Geldsetzer P, Meierkord A, Yang J, Chen Q, Jiao L, et al. Knowledge about COVID-19 among adults in China: cross-sectional online survey. J Med Internet Res. 2021;23(4):e26940.

44.Bärnighausen T, Oldenburg C, Tugwell P, Bommer C, Ebert C, Barreto M, et al. Quasi-experimental study designs series-paper 7: assessing the assumptions. J Clin Epidemiol. 2017;89:53–66.

45.Ozbay N, Shevorykin A, Smith P, Sheffer CE. The association between gender roles and smoking initiation among women and adolescent girls. J Gend Stud. 2020;29(6):664–84.

46.Bleifertaitė R, Leonavičiūtė - Klimantavičienė M, Petraitienė S. Understanding gender differences in e-cigarette use among adolescents: survey findings. J Public Health. 2024;34:1149.

47.Sharapova S, Reyes-Guzman C, Singh T, Phillips E, Marynak KL, Agaku I. Age of tobacco use initiation and association with current use and nicotine dependence among US middle and high school students, 2014–2016. Tob Control. 2020;29(1):49–54.

48.Reitsma MB, Flor LS, Mullany EC, Gupta V, Hay SI, Gakidou E. Spatial, temporal, and demographic patterns in prevalence of smoking tobacco use and initiation among young people in 204 countries and territories, 1990–2019. Lancet Public Health. 2021;6(7):e472–81.

49.svy — The survey prefix command. https://www.stata.com/manuals/svysvy.pdf.

50.StataCorp. (2023). Stata Statistical Software: Release 18. College Station, TX: StataCorp LLC.

51.A Language and Environment for Statistical Computing_. R Foundation for Statistical Computing, Vienna, Austria. https://www.R-project.org/.

52.Shallcross L, Burke D, Abbott O, Donaldson A, Hallatt G, Hayward A, et al. Factors associated with SARS-CoV-2 infection and outbreaks in long-term care facilities in England: a national cross-sectional survey. Lancet Healthy Longev. 2021;2(3):e129–42.

53.Shan L, Manzione LC, Azagba S. Psychological well-being and dual-use of cigarettes and e-cigarettes among high school students in Canada. J Affect Disord. 2020;265:357–63.

54.Sreeramareddy CT, Acharya K, Manoharan A. Electronic cigarettes use and “dual use” among the youth in 75 countries: estimates from Global Youth Tobacco Surveys (2014-2019). Sci Rep. 2014;12(1):20967.

55.Cohn A, Brandon T, Armeli S, Ehlke S, Bowers M. Real-time patterns of smoking and alcohol use: an observational study protocol of risky-drinking smokers. BMJ Open. 2015;5(1):e007046.

56.Irvine DS, McGarity-Shipley E, Lee E-Y, Janssen I, Leatherdale ST. Longitudinal associations between e-cigarette use, cigarette smoking, physical activity, and recreational screen time in Canadian adolescents. Nicotine Tob Res. 2022;24(7):978–85.

57.Zahran S, Janssen I. Compositional associations of time spent in sleep, screen time, and physical activity with polysubstance use in adolescents. Addict Behav. 2023;144:107755.

58.Gu X, Li L. China’s experience of online education during the COVID-19 pandemic: policies, lessons and challenges. In: Dennen V, Dickson-Deane C, Ge X, Ifenthaler D, Murthy S, Richardson JC, editors. Global perspectives on educational innovations for emergency situations. edn. Cham: Springer International Publishing; 2022. p. 285–93.

59.Liu Y, Sun X, Zhang E, Li H, Ge X, Hu F, et al. Association between types of screen time and weight status during the COVID-19 pandemic: a longitudinal study in children and adolescents. Nutrients. 2023;15(9):2055.

60.Dos Santos PC, da Costa BGG, Lopes MVV, Malheiros LEA, Arundell L, da Silva KS. Cross-Sectional Associations of Screen Time Activities With Alcohol and Tobacco Consumption Among Brazilian Adolescents. Int J Public Health. 2023;68:1605816.

61.Lyu M, Lu W, Zou L, Xiong J, Yang J. The impact of new regulations on prevention and control of e-cigarettes on adolescents in middle schools - a city in China, 2022–2023. China CDC Wkly. 2024;6(14):289–93.

62. Standing Committee of the National People's Congress. (2021). Advertising Law of the People's Republic of China (2021 Amendment).

63.Hale DR, Fitzgerald-Yau N, Viner RM. A systematic review of effective interventions for reducing multiple health risk behaviors in adolescence. Am J Public Health. 2014;104(5):e19-41.

64.Mo X, Wang Z, Shao J. Parent-child attachment and good behavior habits among Chinese children: chain mediation effect of parental involvement and psychological Suzhi. PLoS ONE. 2021;16(1):e0241586.

65.Li MY, Luk TT, Cheung DYT, Guo ZQ, Siu KK, Guo J, et al. Effectiveness of family-based behavioral intervention for smoking cessation in low-income households: a systematic review and meta-analysis. EClinicalMedicine. 2025;87:103420.

66.Jackson C, Dickinson D. Enabling parents who smoke to prevent their children from initiating smoking: results from a 3-year intervention evaluation. Arch Pediatr Adolesc Med. 2006;160(1):56–62.

67.Villarreal DL, Nelson JA, Villarreal DL, Nelson JA. Parental monitoring and adolescent risk behaviors: the moderating role of adolescent internalizing symptoms and gender. J Child Fam Stud. 2018;27(11):3627.

68.Connell AM, Dishion TJ, Yasui M, Kavanagh K. An adaptive approach to family intervention: linking engagement in family-centered intervention to reductions in adolescent problem behavior. J Consult Clin Psychol. 2007;75(4):568–79.

69.Furr-Holden CD, Ialongo NS, Anthony JC, Petras H, Kellam SG. Developmentally inspired drug prevention: middle school outcomes in a school-based randomized prevention trial. Drug Alcohol Depend. 2004;73(2):149–58.

70.Pokhrel P, Sussman S, Stacy A. Relative effects of social self-control, sensation seeking, and impulsivity on future cigarette use in a sample of high-risk adolescents. Subst Use Misuse. 2014;49(4):343–51.

71.Daly M, Egan M, Quigley J, Delaney L, Baumeister RF. Childhood self-control predicts smoking throughout life: evidence from 21,000 cohort study participants. Health Psychol. 2016;35(11):1254–63.

72.Sutherland BD, Sutherland MT, Trucco EM. Electronic cigarette use intentions mediate the association between low self-control and future use by internalizing symptoms. Subst Use Misuse. 2022;57(12):1797.

73.Jones RD, Asare M, Lanning B. A retrospective cross-sectional study on the prevalence of e-cigarette use among college students. J Community Health. 2021;46(1):195–202.

74.Salmivalli C. Feeling good about oneself, being bad to others? Remarks on self-esteem, hostility, and aggressive behavior. Aggress Violent Behav. 2001;6(4):375–93.

75.Glendinning A, Inglis D. Smoking behaviour in youth: the problem of low self-esteem? J Adolesc. 1999;22(5):673.

76.Sussman S, Pokhrel P, Ashmore RD, Brown BB. Adolescent peer group identification and characteristics: a review of the literature. Addict Behav. 2007;32(8):1602–27.

77.Zimmerman MA, Copeland LA, Shope JT, Dielman TE. A longitudinal study of self-esteem: implications for adolescent development. J Youth Adols. 1997;26(2):117–41.

78.Li W, Xie Y. The influence of family background on educational expectations: a comparative study. Chin Sociol Rev. 2020;52(3):269–94.

79.Jiang S, Ren Q, Jiang C, Wang L. Academic stress and depression of Chinese adolescents in junior high schools: moderated mediation model of school burnout and self-esteem. J Affect Disord. 2021;295:384–9.

80.Thomas RE, McLellan J, Perera R. School-based programmes for preventing smoking. Cochrane Database Syst Rev. 2013;2013(4):Cd001293.

81.Botvin GJ, Baker E, Renick NL, Filazzola AD, Botvin EM. A cognitive-behavioral approach to substance abuse prevention. Addict Behav. 1984;9(2):137–47.

82.Faggiano F, Galanti MR, Bohrn K, Burkhart G, Vigna-Taglianti F, Cuomo L, et al. The effectiveness of a school-based substance abuse prevention program: EU-Dap cluster randomised controlled trial. Prev Med. 2008;47(5):537–43.

Associated Data

Supplementary Materials

Data Availability Statement

The data and analytic codes used in this study will be available six months after the time of publication, with no end date, from the corresponding authors upon reasonable request.