Background: A success in preventing adolescent tobacco use is the most expeditious and effective way to reduce population-level tobacco use. Implementation of a series of tobacco control policies after the ratification of the Framework Convention on T...
Background: A success in preventing adolescent tobacco use is the most expeditious and effective way to reduce population-level tobacco use. Implementation of a series of tobacco control policies after the ratification of the Framework Convention on Tobacco Control led to a substantial decline in the cigarette smoking prevalence among South Korean adolescents. However, the recent deceleration in the reduction of smoking and evolving behavioral patterns regarding non-cigarette tobacco/nicotine product (NCTNP) among adolescents call for a new perspective that can drive societal transformation.
Generations are considered engines of social change, as each new generation with new experience shape new culture. Epidemiologic views define generation, or cohort effects, as differences in the risks regarding a health outcome by birth year. These effects are often coincided with shifts in population-level exposure. Therefore, epidemiologic studies of generations or cohorts require identifying cohort effects from a health outcome, and describing changes in the population-level exposure to risk factors. Such exposure changes represent generational characteristics.
Examination of cohort effects and related characteristics also serve as an evidence to implement a legislative measure of banning sales or supplies of tobacco products to those born in or after a certain year, known as the tobacco-free generation (TFG). The ultimate objective of the TFG proposal is to create normative differences between generations. Birth cohort differences occurring due to the impact of tobacco control policies, as well as other critical events would be an evidence in favor of measures considering generational traits.
In this study, I aimed to address the generational characteristics of adolescent smoking. Specifically, this study first seeks to identify birth cohort effects in adolescent smoking, and examine possible changes or events that may have shaped generational cohorts regarding smoking. Second, drawing on the social identity theory, which is in line with the generational identity theory, I aimed to determine the association between adulthood experiences and adolescent cigarettes smoking, along with the joint association of adulthood experiences, and significant adults (parents and teachers) smoking with adolescent cigarette smoking. The third focus of this study is examining the impacts of NCTNPs, which may serve as a foundation for the emergence of a new generation regarding tobacco use. Lastly, this study attempted to project the landscape of adolescent smoking with different scenarios regarding NCTNPs.
Methods: The four independent studies of this dissertation all used a nationally, representative cross-sectional data of school-attending Korean adolescents aged 12 to 18 years (mean age: 15 years).
In the first study, age-period-cohort (APC) analysis was applied to determine the influence of cohort effects on the trends of adolescent smoking prevalence during 2006-2017. Then, the long-term impacts of two denormalization policies: banning of smoking in public places including schools and adolescent facilities, and the largest-ever increase in cigarette prices from 2500 WON per pack to 4500 WON per pack, were analyzed by segmented regression models. In addition, birth year intervals were delineated by the risks of smoking obtained from the APC model, and potential generation shaping events and changes were described accordingly.
For the second study, the association of adulthood experience and adolescent cigarette smoking was examined by performing logistic regression analyses. The adulthood experiences examined in this study were height above the 95th percentile for age and sex, sexual development before 10-years old or younger, living arrangements away from parents, and having a job experience during the past 12-months. Tests for additive interaction between adulthood experiences, and parents or teacher smoking, with adolescent cigarette smoking, were also conducted. Moreover, additional analyses examining cohort differences in the exposure to adulthood experience and significant adult smoking were performed.
Third, the prevalence, and the impacts of electronic nicotine delivery systems (ENDS) and heated tobacco products (HTPs) on other tobacco/nicotine product use behaviors were assessed. First, participants were categorized into eight mutually exclusive tobacco or nicotine use states. A multistate Markov model was applied to identify transitions between the states to assess gateway effects to cigarettes and other NCTNPs. Moreover, the prevalence of ENDS and HTPs were examined solely or jointly with other products, and the association between NCTNPs and cigarette quit attempts were examined to identify polytobacco use among adolescents.
Lastly, a model based on the system dynamics methodology specialized in scenario testing, and handling complex systems was developed. The population of adolescents were categorized into three mutually exclusive groups of never smokers, current smokers and former smokers. The sectors included in the model were tobacco users, tobacco control policies, tobacco industry, tobacco products and social influence. The estimates derived from the exploratory model showed adolescent smoking prevalence from 2006 to 2018. Tests for validity, including tests for structure and behavior, were performed on the explanatory model. Then scenario testing assessing the impacts of NCTNP use, particularly HTPs, on adolescent cigarette smoking prevalence were conducted through 2019-2040.
Results: Between 2006 and 2017, the prevalence of cigarette smoking among Korean adolescents decreased from 16% to 9% in boys, and from 9% to 3% in girls. According to the APC analysis, the risk of smoking decreased with every consecutive year for adolescents born after 1997. After a complete ban on smoking in public places, there were significantly negative trends in the prevalence of cigarette smoking for both boys (β=-1.1; 95% CI: -1.9, -0.2) and girls (β=-0.4; 95% CI: -0.6, -0.1). Immediate decrease among girls was found after cigarette prices increased (β=-0.8; 95% CI: -1.5, -0.2). Three generational cohorts were delineated by risks of cigarette smoking. These generational cohorts depicted different characteristics.
From the second study, a positive correlation between tall stature (OR=1.18; 95% CI=1.07, 1.30), precocious sexual development (OR=1.47; 95% CI=1.32, 1.63), independent living from parents (OR=1.23; 95% CI=1.07, 1.41) and work experience (OR=4.38, 95% CI=4.17, 4.61), and adolescent smoking were found. The risk of adolescent cigarette smoking associated with a combination of job experience and parental smoking was larger than the sum of risk associated with each individual factor, indicating an interaction on an additive scale (RERI=0.47; 95% CI=0.11, 0.83). Moreover, the risk of adolescent cigarette smoking associated with a combination of precocious sexual development (RERI=0.51; 95% CI=0.12, 0.88), independent living (RERI=0.55; 95% CI=0.09, 1.00) and job experience (RERI=1.84; 95% CI=1.42, 2.25), and teacher smoking was more than the addition of risk associated with each individual factor, respectively. Additional analyses suggested exposure to parental smoking decreased from 36.7% in 2006 to 21.2% in 2018, and teacher smoking decreased from 26.1% in 2007 to 20.0% in 2018.
In all, 2.9% of adolescents were ever HTP users and 7.9% were ever ENDS users. Among HTP ever users, 75.3% were current cigarette users and 47.2% were current ENDS users. Similarly, among ENDS ever users, 55.4% were current cigarette users. Compared with never use, use of ENDS and HTP was associated with an increased risk of initiation of cigarette use (TIR=8.1; 95% CI=5.2, 12.4 and TIR=20.3; 95% CI=7.6, 53.7, respectively). The gateway effect was greatest in subjects who had used both ENDS and HTPs (TIR=64.0, 95% CI=41.4, 98.9). Transitions between ENDS and HTPs were frequent. Unlike ever use of ENDS, which was associated with a higher likelihood of making cigarette quit attempts (aOR=1.35, 95% CI: 1.15, 1.58), no difference in cigarette quit attempts were found by ever use of HTPs (aOR=1.06, 95% CI: 0.90, 1.26).
Adolescent cigarette smoking prevalence simulated by the explanatory model including the users, policy, industry, products, and social influence performed well in replicating the overall dynamics of adolescent cigarette smoking prevalence from 2006 to 2018, and showed the highest R2 coefficient of 0.95 among all other models. The status quo model showed gradual decrease in smoking prevalence from 8.2% in 2019 to 6.8% in 2040. On the other hand, the ten-fold increase in HTP use (HTP ever use: 28%) scenario showed that adolescent cigarette smoking increase by 3.2%p within three years. Under such scenario, smoking prevalence of 8.6% at 2040 was higher than that of 2019, even with tobacco control policies in effect.
Conclusion: This study identified birth cohort effects in adolescent cigarette smoking, elucidated generation-shaping events and changes, and suggested the possibility of the emergence of a new generation due to the introduction of NCTNPs such as HTPs. The strengthening of overall tobacco control policies, along with policies specifically targeting adolescents, and societal level movements away from smoking contributed to reduced risks of cigarette smoking among recent cohorts. However, rapid adoption of novel and emerging tobacco/nicotine products and the high likelihood of transitioning to other product use among adolescents lay foundation for the emergence of a new tobacco-related generation.
In creating generational identities, the role of collective experiences among those in critical stages are emphasized. The critical stage for formulating patterns of tobacco use are during early adolescence years. Thus, future tobacco control strategies must be focused on differentiating one generation from another, by implementing universal tobacco control policies reaching those at early adolescence. One innovative example is the TFG proposal. Implementation of the TFG, would be favorable in creating a new generational norm by widening the gap between the existing and new generations.
As mentioned, the results of the study pointed to a possibility of an emergence of a new generation regarding tobacco use, as a result of NCTNP penetration into the tobacco market. Both non-legislative and legislative discussions on attaining a generation free from tobacco should cover all tobacco/nicotine products, not just cigarettes.