The use of cigarettes, hookahs, electronic cigarettes, and other tobacco indicators among Brazilian schoolchildren: data from National School Health Survey 2019

: Objective : To describe the prevalence of smoking indicators among Brazilian students according to sociodemographic characteristics in 2019, and compare the prevalence between 2015 and 2019. Methods : Data from the National Survey of School Health 2015 and 2019 were used. Indicators related to tobacco use were analyzed. Indicators were compared between the 2015 and 2019 editions. Prevalence and respective 95% Confidence Interval (95%IC) were calculated for the total population and according to sex, age group and type of school. Results : 22.6% (95%CI 21.7–23.4) of the students had tried any cigarette and it was higher between 16 and 17 years of age (32.6%; 95% CI 31.4–33, 8) and in males (35.0%; 95%CI 33.6–36.4). The experimentation of hookah, electronic cigarette and other tobacco products are also high, with 26.9% (95%CI 26.0–27.8), 16.8% (95%CI 16.2–17.4) and 9.3% (95%CI 8.8–9.8), respectively, being higher among boys aged 16 to 17 years. It is noteworthy that there were no changes in the indicators of cigarette experimentation, smoking for the first time before the age of 13, smoking in the 30 days prior to the survey, and at least one of the smoking parents. Conclusion : Although smoked tobacco indicators are stable between 2015 and 2019, the high prevalence of experimentation with products such as hookah and electronic cigarettes is highlighted, drawing attention to the need for new regulatory measures.


INTRODUCTION
Tobacco is an important avoidable risk factor for several types of cancer, and cardiovascular and respiratory diseases 1,2 . Health risks arise from both direct consumption of tobacco and exposure to secondhand smoke [1][2][3][4] .
Tobacco use among adolescents is a global problem, as it is highly addictive and has health consequences 5 . A study found that all-cause mortality in adulthood was higher by 151, 83, and 56%, respectively, among individuals who started smoking at ages 5-9, 10-14, and 15-19, compared with those who have never smoked 6 .

Palavras
Adolescence is a phase of initiation of new social behaviors, which can be determinant for health during adulthood, such as the development of NCDs. The high prevalence of smoking and the introduction of new tobacco products among adolescents become a worrying scenario, so it is necessary to monitor the consumption of tobacco products in this population, in order to support national and global agendas and commitments. In this sense, this study aimed to describe the prevalence of smoking indicators among Brazilian students according to sociodemographic characteristics in 2019 and compare the prevalence between 2015 and 2019.

METHODS
Data from the National School Health Survey (Pesquisa Nacional de Saúde do Escolar -PeNSE) for the years 2015 and 2019 were used. PeNSE is a periodic survey, carried out since 2009 and every three years by the Brazilian Institute of Geography and Statistics (Instituto Brasileiro de Geografia e Estatística -IBGE) in partnership with the Ministry of Health. Its objective is to know and measure the risk and protective factors for the health of adolescents 16 .
The research is carried out by sampling, using as a reference for selection the register of public and private schools of the National Institute of Educational Studies and Research Anísio Teixeira (Instituto Nacional de Estudos e Pesquisas Educacionais Anísio Teixeira -INEP). The sampling plan is a conglomerate in two stages, in which the schools correspond to the first stage and the classes of students enrolled to the second. In the selected classes, all students were invited to answer the survey questionnaire 16 .
In 2015, two different sampling plans were used, which included, respectively, schoolchildren who attended the 9 th grade of elementary school (sample 1) and schoolchildren aged 13 to 17 who attended the 6 th to 9 th grade of elementary school (former 5 th to 8 th grades, in Brazil) and from 1 st to 3 rd grades of high school, in the reference year of the research (sample 2). In sample 2, we investigated 10,926 Brazilian students enrolled and attending 371 schools and 653 classes, in the five main geographic regions of the country, and the general total for Brazil in public and private schools 17 . In 2019, IBGE used a single sample of students aged 13 to 17, from public and private schools, for the following geographic levels: Brazil, Major Regions, Federation Units, Capital Cities, and the Federal District. Samples were collected in 4,242 schools, 6,612 classes, and among 159,245 students 16 . Considering enrolled students and non-respondents, the sample loss was approximately 2.4% in 2015 and 15.4% in 2019. More details on the sample can be found in other publications 16,17 .

REV BRAS EPIDEMIOL 2022; 25: E220014
Students, using smartphones, answered the structured and self-administered questionnaire, which included information on socioeconomic status, family context, experimentation and use of cigarettes, alcohol, and other drugs, violence, safety, accidents, and other living conditions of these adolescents who attend the school 16 .
The following indicators referring to tobacco use in 2019 were used: 1. Cigarette use ever-assessed by the "yes" answer to the question: "Have you ever smoked cigarettes, even if a puff or two?". 2. Cigarette use for the first time at age 13 or under -according to the question: "How old were you when you first smoked cigarettes?".

3.
Smokers in the 30 days prior to the survey -we considered those who answered "one day" or "more days" to the question: "In the last 30 days, on how many of them did you smoke cigarettes?". The prevalence and respective 95% confidence intervals (95%CI) of tobacco use indicators were estimated for the total population of adolescents and according to gender (female and male), age groups (13-17, 13-15, and 16-17 years old) and administrative dependency (public and private school). In addition, indicators referring to smokers in the last 30 days and experimentation with hookah and electronic cigarettes were presented according to Federative Units (FU). Finally, the local where the cigarette was obtained was analyzed. Differences between groups were considered significant when there was no overlap in the 95%CI.

RESULTS
The PeNSE 2015 sample consisted of 10,926 schoolchildren aged 13 to 17 years, 50.3% male and 49.7% female. In 2019, 159,245 schoolchildren aged 13 to 17 years were evaluated, of which 49.3% were male.
When analyzing the indicators according to the type of school, it appears that the prevalence of all indicators were higher in public schools, except for cigarettes experimentation between 16 and 17 years old and electronic cigarettes experimentation in all age groups, which were higher in private schools. Experimentation with other tobacco products in the 13 to 17 and 16 to 17 age groups had no significant difference (Supplementary Table 1).
The percentage of students who smoked in the 30 days prior to the survey in Brazil was 6.8% (95%CI 6.3-7.3). The state of Acre leads, with 10.9% (95%CI 9.0-12.8), followed by Mato Grosso do Sul and Roraima, with 9.7% (95%CI 8.5-10.9) and 9.2% (95%CI 7.8-10.6), respectively. Sergipe and Bahia are the states with the lowest percentages (3.2%; 95%CI 2.4-4.1 and 3.7%; 95%CI 2.6-4.7, respectively) ( Figure 1). 2) and Piauí (8.7%; 95%CI 6.9-10.5) are the states with the lowest percentage ( Figure 3).   When analyzing tobacco indicators in 2015 and 2019, it was observed that there were no changes in cigarette experimentation, smoking for the first time before age 13, smoking in the 30 days prior to the survey, and having at least one of the parents who smoked (Figure 4).
Supplementary Figure 1 reveals the way of purchasing cigarettes among those who smoked 30 days before the interview, with the most common purchase being in stores, bars and the like, corresponding to 37.5% (95%CI 35.2-39.9) in students aged 13 to 17 years and 43% (95%CI 39.7-46.4) among those aged 16 to 17 years. The second most frequent way was to "ask someone" (19.9%; 95%CI 18.1-21.7, in students aged 13 to 17 years).

DISCUSSION
The results of the present study indicate that cigarette experimentation occurred in onefifth of the adolescents. Cigarette use in the last 30 days was 6.8%, and there was a higher prevalence in one-tenth of older adolescents (16-17 years). It can be seen that other tobacco products experimentation is high: a quarter have already tried hookah, 16.8% electronic cigarettes, and a tenth of the students have tried other tobacco products, for which experimentation among older adolescents was higher. A quarter of the adolescents reported that at least one of their parents smoked, a third of their friends smoked, and being a passive     smoker at home occurred in a third of the students. The local where tobacco was obtained were shops, bars, and the like, with 37.5%. Comparing with 2015, there were no changes in the prevalence of: cigarette experimentation, smoking for the first time before the age of 13, smoking in the 30 days prior to the survey, and having at least one of the parents who smoked. Data from the current study also showed that tobacco use increases with age, both for cigarettes and for other products.
The increase in the prevalence of tobacco use with increasing age found in the present study can be explained by: the search for new experiences and exposure to risk, part of the transition process from childhood to adulthood 18 ; peer pressure 19,20 , which is also very worrying in this study, given that around 30% reported that their friends smoke, which can be a stimulus for use; greater access to money among older adolescents, which can increase the ability to buy cigarettes 19,20 ; by the pressure of the industry and the different formats of tobacco advertising, which are attractive and portray smoking as a sign of maturity, of reaching adulthood, of freedom, and transgression 21 . These results are consistent with international 22,23 and national 12,24 studies. The GTYS showed an increase in smoking according to age for both cigarettes and other tobacco products, being about twice as high at age 15 when compared to age 13 25 . In a study with data from the Global School-Based Student Health Survey (GSHS) 26 , it was observed that the prevalence of use of any tobacco product increased by 61% at the age of 14 to 15 years compared to that of 12-13 years.
The present investigation found that female students aged between 13 and 15 have experimented with cigarettes more than male students. Girls tend to be more mature than boys at this stage of life, which can temporarily increase the habit, although, between the ages of 16 and 17 and in adulthood, men smoke more 27 . Monitoring should be continued to identify new trends, as has been the case with alcohol use, indicating a progressive increase among young women, which tends to converge in this consumption in adult life 11 .
It is noteworthy that a quarter of the students reported that their parents were smokers. This indicator should be viewed with concern, as studies demonstrate the relationship of cigarette use by adolescents with smoking among their parents or other close people 12 , which can be explained by the theory of social learning 28 and by the naturalization of the habit.
The results of the present study serve as a warning because, although the prevalence of smoked cigarettes is not so high, other tobacco products, such as hookah, electronic cigarettes and others are quite frequent. The rise of new tobacco products has reflected a global trend 29 for such products to be attractive to young people 30 . Similar results have been described in several countries, such as Europe, the Mediterranean, and Eastern European regions 31,32 . Products such as hookah and electronic cigarettes can be the gateway to establishing the habit and dependence of smoking 33,34 . This evidence is important to guide policies to combat smoking. The hookah contains a high amount of nicotine, and its use involves important health risks. Hookah smoke contains the same substances as tobacco (nicotine, carbon monoxide, others) 29 . A 1-2-hour hookah session can equate to smoking between 100 and 150 cigarettes 30 .
A study with data from PeNSE 2015 12 already drew attention to the increase in the use of other tobacco products among adolescents, which was 27% between 2012 and 2015.

REV BRAS EPIDEMIOL 2022; 25: E220014
The current survey included new questions, which makes it difficult to compare it with that of 2015; however, the prevalence of hookah experimentation was higher than that of smoking cigarettes in 2019. The high experimentation of electronic cigarettes is also noteworthy. Thus, new tobacco products must be a major concern today, and the introduction of new regulatory measures for these products as well as the use of clear messages about their harm must be a priority for managers, particularly at the federal level. Some countries have adopted measures to regulate hookah use, banning flavors added to tobacco in alternative products or devices 35 . In Brazil, the 2014 presidential decree included a ban on the use of hookah in closed environments, which may have contributed to this increase not being even more expressive 36 .
The National Health Surveillance Agency (Agência Nacional de Vigilância Sanitária -Anvisa), in 2009, prohibited the sale of electronic cigarettes, however these products have been widely used and sold in the country, not being affected by adequate inspection 37 . Brazilian legislation also prohibits the sale, offer or availability of any tobacco product to persons under 18 years of age 36 .
As for secondhand smoke, the question in the 2015 edition was different: "In the last seven days, in how many of them people smoked in your presence?" -that is, the act of smoking could have taken place at home, at work, at school or in other environments. In 2015, half of adolescents were exposed to secondhand smoke 12 . In 2019, the question was specific about passive smoking at home and showed a prevalence of 27.4%. Secondhand smoke causes as many harms as active smoking and must be addressed as a target of regulatory measures. Globally, research in 142 countries showed that the prevalence of secondhand smoke at home was 33.1% (95%CI 32.1-34.1), slightly higher than in Brazil 38 . A study with adults shows that the frequency of secondhand smoke at home has decreased over the years 39 , which may be a result of the regulatory measures adopted 36 . Furthermore, it is observed that secondhand smoke at home has higher prevalence among women, low-income populations, and younger people 39 . We emphasize that more vulnerable populations tend to suffer more from the problem of secondhand smoke.
Among the limitations of the present study, it should be noted that the data obtained were based on the students' reports, which may have led to information bias. Although most Brazilian adolescents are in school (97%), it is known that those who are outside the school environment present greater health risks, as well as more risk behaviors, which would tend to underestimate the prevalence found. This is a descriptive study, without covariate adjustment. Furthermore, the changes to several questions in There is stability in the indicators of tobacco smoked between 2015 and 2019, however the prevalence of experimentation with cigarettes and other tobacco products such as hookah and electronic cigarettes are high. These results highlight the need for new regulatory measures for tobacco in Brazil.