Abstract

Introduction

In the tenth revision of the International Classification of Diseases (ICD-10) ¹1. Centro Colaborador da OMS para a Classificação de Doenças em Português. Classificação Estatística Internacional de Doenças e Problemas Relacionados à Saúde. Décima revisão. http://www.datasus.gov.br/cid10/V2008/cid10.htm (acessado em 21/Set/2016).
http://www.datasus.gov.br/cid10/V2008/ci... , smoking is classified in the group of mental and behavioral disorders. This grouping includes many disorders that differ from each other in severity and symptomatology, but which have in common a relationship with the use of psychoactive substances with or without a medical prescription. Smoking is the leading global preventable cause of morbidity and mortality and one of the main risk factors for chronic noncommunicable diseases (NCDs) ²2. Pinto MT, Pichon-Riviere A, Bardach A. Estimativa da carga do tabagismo no Brasil: mortalidade, morbidade e custos. Cad Saúde Pública 2015; 31:1283-97..

Smoking has decreased in Brazil according to data from VIGITEL ³3. Ministério da Saúde. Vigitel Brasil 2016: vigilância de fatores de risco e proteção para doenças crônicas por inquérito telefônico: estimativas sobre frequência e distribuição sociodemográfica de fatores de risco e proteção para doenças crônicas nas capitais dos 26 estados brasileiros e no Distrito Federal em 2016. http://portalarquivos.saude.gov.br/images/pdf/2017/junho/07/vigitel_2016_jun17.pdf (acessado em 22/Set/2017).
http://portalarquivos.saude.gov.br/image... , the Ministry of Health’s system aimed at monitoring the principal determinants of NCDs in the adult population of the country’s 26 state capitals and the Federal District. In 2006, 15.7% of Brazilian adults were smokers. By 2016, smoking prevalence had dropped to 10.2%, higher in men (12.7%) than in women (8%). Even with this reduction in prevalence in recent years, smoking is still considered a serious public health problem, especially in terms of the costs to the health system ²2. Pinto MT, Pichon-Riviere A, Bardach A. Estimativa da carga do tabagismo no Brasil: mortalidade, morbidade e custos. Cad Saúde Pública 2015; 31:1283-97.. The fight against smoking has gained momentum with campaigns and educational activities by the Federal government, through the Ministry of Health, in collaboration with the Brazilian National Cancer Institute (INCA), World Health Organization (WHO), and Pan American Health Organization (PAHO) ⁴4. Silva ST, Martins MC, Faria FR, Cotta RM. Combate ao tabagismo no Brasil: a importância estratégica das ações governamentais. Ciênc Saúde Coletiva 2014; 19:539-52..

Schools are known to be a prime locus for such campaigns and educational activities, since schoolteachers and their attitudes can influence students’ choices to start or quit smoking ⁵5. Chen P, Huang W, Chao K. Factors associated with Taiwanese junior high school personnel advising students to quit smoking. J School Health 2011; 81:91-9.^,66. Erick PN, Smith DR. Prevalence of tobacco smoking among school teachers in Botswana. Tob Induc Dis 2013; 11:24.^,77. World Health Organization. Promoting health through schools. Geneva: World Health Organization; 1997. (WHO Technical Report Series, 870).. Since 2007, Brazil has implemented the National School Health Program (PSE), an inter-sector policy of the Ministries of Health and Education in response to recommendations by PAHO for the development of Regional Initiatives for Health Promoting Schools ⁸8. Figueiredo TA, Machado VL, Abreu MM. A saúde na escola: um breve resgate histórico. Ciênc Saúde Coletiva 2010; 15:397-402.. The strategy involves the entire school community, with an emphasis on students’ health promotion, providing for the creation and maintenance of healthy physical spaces, including cigarette- and drug-free environments, among other measures ⁸8. Figueiredo TA, Machado VL, Abreu MM. A saúde na escola: um breve resgate histórico. Ciênc Saúde Coletiva 2010; 15:397-402..

Work characteristics influence individuals’ vulnerability to harmful behaviors and illness ⁹9. Clougherty JE, Souza K, Cullen MR. Work and its role in shaping the social gradient in health. Ann N Y Acad Sci 2010; 1186:102-24.. The negative effects of work on schoolteachers’ health can be identified by examining smoking, since it is related to the work’s characteristics ⁹9. Clougherty JE, Souza K, Cullen MR. Work and its role in shaping the social gradient in health. Ann N Y Acad Sci 2010; 1186:102-24.^,1010. Heikkillä K, Nyberg ST, Fransson EI, Alfredsson L, De Bacquer D, Bjorner JB, et al. Job strain and tobacco smoking: an individual-participant data meta-analysis of 166 130 in 15 European studies. PLoS One 2012; 7:e35463. and is considered indicative of the individual’s overall health status ¹¹11. Pavão AL, Werneck GL, Campos MR. Autoavaliação do estado de saúde e a associação com fatores sociodemográficos, hábitos de vida e morbidade na população: um inquérito nacional. Cad Saúde Pública 2013; 29:723-34..

There is now a growing trend to examine individual habits associated with increased odds of illness, rather than focusing on classical occupational risks or diseases, as in the traditional occupational health approach ¹²12. International Labour Office. Emerging risks and new patterns of prevention in a changing world of work. Geneva: International Labour Office; 2010.. This suggests taking a more cautious look at how work favors or impedes adherence to healthy habits, which in turn explains the growing prevalence of noncommunicable diseases ¹³13. Schimidt MI, Duncan BB, Azevedo e Silva G, Menezes AM, Monteiro CA, Barreto SM, et al. Chronic non-communicable diseases in Brazil: burden and current challenges. Lancet 2011; 377:1949-61..

There is irrefutable evidence that individuals’ relationship with their surroundings is heavily associated with the production of tensions, known jointly as stress, the effects of which include adverse psychological situations. In the case of Brazilian schoolteachers, this includes anxiety produced by overlapping demands in carrying out their activities and the worries resulting from the teaching-learning process in the current context of heavy pressure and multiple, often mutually conflicting tasks. This unfavorable scenario is aggravated by low pay, time pressures, and violence and lack of discipline in the classroom, thus increasing the odds of substance use ¹⁴14. Schrijvers CT, van de Mheen HD, Stronks K, Mackenbach JP. Socioeconomic inequalities in health in the working population: the contribution of working conditions. Int J Epidemiol 1998; 27:1011-8..

Smoking - both previous and current - is a constant variable in studies on schoolteachers’ health conditions and lifestyles ¹⁵15. Santos MN, Marques AC. Condições de saúde, estilo de vida e características de trabalho de professores de uma cidade do Sul do Brasil. Ciênc Saúde Coletiva 2013; 18:837-46.^,1616. Vedovato TG, Monteiro MI. Perfil sociodemográfico e condições de saúde e trabalho dos professores de nove escolas estaduais paulistas. Rev Esc Enferm USP 2008; 42:290-7.^,1717. Yue P, Liu F, Li L. Neck/shoulder pain and low back pain among school teachers in China, prevalence and risk factors. BMC Public Health 2012; 12:789., especially in research on voice disorders ¹⁸18. Jardim R, Barreto SM, Assunção AA. Condições de trabalho, qualidade de vida e disfonia entre docentes. Cad Saúde Pública 2007; 23:2439-61.^,1919. Valente AM, Botelho C, Silva AM. Distúrbio de voz e fatores associados em professores da rede pública. Rev Bras Saúde Ocup 2015; 40:183-95.. However, it tends to be a secondary event in studies on outcomes, rather than highlighting it as the target event.

The current study thus aimed to investigate smoking prevalence and associated factors in Brazilian schoolteachers in a representative national sample.

Methods

This cross-sectional study focused on a sample of Brazilian schoolteachers currently working in basic education, which in Brazil includes preschool (children under 5 years), primary (6 to 14 years), and secondary (15 to 17 years) ²⁰20. Secretaria de Educação Básica, Ministério da Educação. Diretrizes curriculares nacionais gerais da Educação Básica. http://portal.mec.gov.br/index.php?option=com_docman&view=download&alias=13677-diretrizes-educacao-basica-2013-pdf&Itemid=30192 (acessado em 21/Set/2016).
http://portal.mec.gov.br/index.php?optio... .

This study on smoking in Brazilian schoolteachers analyzes data from a nationwide survey on Brazilian schoolteachers’ health, called Educatel Brazil, conducted through an agreement between Federal University of Minas Gerais (UFMG) and the Ministry of Education. The survey aimed to investigate factors associated with absenteeism and to present a map of the health status and work conditions of Brazilian schoolteachers in a representative national sample ²¹21. Maia EG, Claro RM, Assunção AA. Múltiplas exposições ao risco de faltar ao trabalho nas escolas da Educação Básica no Brasil. Cad Saúde Pública 2019; 35 Suppl 1:e00166517..

At the time of the data collection, from October 2015 to March 2016, the total population for the Educatel Study was 2,229,269 teachers ²²22. Diretoria de Estatísticas Educacionais, Instituto Nacional de Estudos e Pesquisas Educacionais Anísio Teixeira. Censo Escolar da Educação Básica 2013: resumo técnico. Brasília: Instituto Nacional de Estudos e Pesquisas Educacionais Anísio Teixeira; 2014.. Selection of the schoolteachers for the study sample started with stratified sampling based on the following criteria: (a) major geographic regions of the country; (b) school’s census area, namely urban versus rural; (c) schoolteachers’ age brackets; (d) gender; (e) school’s administrative jurisdiction; (f) type of teaching contract; and (g) grade level in which the teacher was working. After stratification, simple random sampling was performed in each of the resulting strata. Sample calculation assumed the following parameters: 38% prevalence of absenteeism among schoolteachers ¹⁵15. Santos MN, Marques AC. Condições de saúde, estilo de vida e características de trabalho de professores de uma cidade do Sul do Brasil. Ciênc Saúde Coletiva 2013; 18:837-46., 95% confidence interval, 2% predicted error, and 20% losses. A sample size of 6,500 schoolteachers was reached. More information on the procedures used in the sample design are provided in a complementary methodological study ²³23. Vieira MT, Claro RM, Assunção AA. Desenho da amostra e participação no Estudo Educatel. Cad Saúde Pública 2019; 35 Suppl 1:e00167217..

A team of 30 previously trained interviewers conducted the computer-assisted telephone interviews. The team also included two supervisors and a coordinator, besides researchers from the Center for Studies on Health and Work at UFMG who monitored the entire process. Teachers were excluded if they worked in schools with no possibility for telephone contact or if they had not been contacted after 15 attempts. After identification of the potentially eligible individuals for the sample (7,642 schoolteachers), the telephone calls began, totaling 119,378 calls. Details on the survey design are described in a complementary study ²⁴24. Assunção AA, Medeiros AM, Claro RM, Vieira MT, Maia EG, Andrade JM. Hipóteses, delineamento e instrumentos do Estudo Educatel Brasil, 2015/2016. Cad Saúde Pública 2019; 35 Suppl 1:e00108618..

The study’s dependent variable, “current smoker”, was based on the answers to the following questions: “Are you now or have you ever been a smoker, that is, have you smoked at least 100 cigarettes (five packs) in your life? (no/yes)” and “Do you smoke cigarettes now? (no/yes)”. Schoolteachers that answered yes to both questions were classified as “current smokers” and those who answered no to one of the two were classified as “non-smokers”.

The independent variables used in the analysis included the schoolteachers’ individual and work characteristics and were organized in four blocks. The individual variables comprised two blocks:

(1) Sociodemographic information: gender (male/female), age (≤ 34 years/35-44 years/45-54 years/≥ 55 years), teacher’s educational level (primary or middle school - completed or in progress/university - completed or in progress), marital status (single or without partner/married or with partner), children (no/yes), self-reported race (white/black or brown/Asian-descendant or indigenous/other), monthly income (≤ 3 times the minimum wage/> 3 times the minimum wage), and major geographic region (Northeast/North/Central/South/Southeast).

(2) Health status: physical activity (sufficient/insufficient), missed work in the last 12 months (no/yes), work problems due to voice (no/sometimes/frequently), frequent sleep loss due to worries (no/yes), use of anxiolytics or antidepressants in the four weeks prior to the interview (no/yes), and self-rated health (good or very good/fair/bad or very bad).

The variable “physical activity” was defined as the combination of type, weekly frequency (1-2 days a week/3-4 days a week/5-6 days a week/every day, including Saturday and Sunday), and duration (less than 10 minutes/10-19 minutes/20-29 minutes/30-39 minutes/40-49 minutes/50-59 minutes/≥ 60 minutes) of physical activity reported by the interviewee. “Sufficient physical activity” was defined as practicing moderate physical exercise for at least 30 minutes a day on at least five days of the week ³3. Ministério da Saúde. Vigitel Brasil 2016: vigilância de fatores de risco e proteção para doenças crônicas por inquérito telefônico: estimativas sobre frequência e distribuição sociodemográfica de fatores de risco e proteção para doenças crônicas nas capitais dos 26 estados brasileiros e no Distrito Federal em 2016. http://portalarquivos.saude.gov.br/images/pdf/2017/junho/07/vigitel_2016_jun17.pdf (acessado em 22/Set/2017).
http://portalarquivos.saude.gov.br/image... .

“Voice problems at work” was based on the question: “In the last four weeks, have you had problems with your voice at work or in your professional development?” The options were “never or almost never” and “rarely” were grouped as “no”, and the options “sometimes” and “frequently” were maintained as such.

The variable “frequent loss of sleep due to worries” was based on the question: “In recent weeks, how often have you lost sleep due to worries?” The options “not at all” and “no more than usual” were grouped as “no”, and the options “a little more than usual” and “much more than usual” were grouped as “yes”.

“Self-rated health” was examined with the question: “In general, would you say that your health is: (very good/good/fair/bad/very bad)”. The answers were grouped as “good or very good”, “fair”, and “bad or very bad”.

The work-related variables formed the other two blocks:

(1) Job information: time working in basic education (less than 10 years/10-20 years/more than 20 years), currently also working in another school (no/yes), type of teaching contract (public admissions/covered by labor legislation/tenured, stable and private school system/private school system/temporary contract), other paid activity besides teaching (no/yes), workweek exceeding 40 hours (no/yes).

(2) Occupational stressors: commuting time to and from school (≤ 30 minutes/31-60 minutes/61-120 minutes/≥ 121 minutes), loud noise in classroom (no/yes), lack of discipline in classroom (no/yes), suffered verbal violence by students in the 12 months prior to the interview (no/yes), suffered physical violence by students in the 12 months prior to the interview (no/yes), heavy work demands (no/yes), sufficient time for work tasks (yes/no).

The variables dealing with loud noise and lack of classroom discipline were examined with the question “How often is the noise at work so strong the you have to raise your voice to talk to others? (frequently/sometimes/rarely/never or almost never)” and “How often is your workplace agitated due to lack of student discipline? (frequently/sometimes/rarely/never or almost never)”, respectively. Both were transformed into dichotomous variables, and the categories were grouped in the options “no” (rarely/never or almost never), and “yes” (frequently/sometimes).

Verbal or physical violence by students was measured by the questions: “In the last 12 months, have you ever suffered verbal violence from students? (never/once/twice or more)” and “In the last 12 months, have you ever suffered physical violence from students? (never/once/twice or more)”. Here as well, the variables were transformed into dichotomous, and the options were grouped as “no” (never) and “yes” (once/twice or more).

The variable “heavy work requirements” was examined with the question: “Does your work demand too much of you? (frequently/sometimes/rarely/never or almost never)”. Concerning sufficient time to complete the tasks at work, the following question was used: “Do you have enough time to complete all the tasks at your work? (frequently/sometimes/rarely/never or almost never)”. For both variables, the answers “sometimes/rarely/never or almost” were grouped as “no”, and “frequently” was transformed into “yes”.

Data were analyzed with Stata, version 10.0 (https://www.stata.com), in five stages: (1) a descriptive analysis considering sociodemographic variables, health status, employment variables, and occupational stressors; (2) estimate of smoking prevalence according to the categories of explanatory variables; (3) univariate logistic regression; (4) multivariate logistic regression by blocks (sociodemographic, health status, and work characteristics) including the variables associated with smoking (p ≤ 0.20) in the univariate analysis; and (5) multivariate logistic regression including the variables associated with the outcome at p ≤ 0.05 in the multivariate analyses by blocks.

Selection of the variables to construct the final model for the multivariate analysis used the backward stepwise method. Thus, the explanatory variables selected in the previous stages were included in the logistic regression analysis and were removed one by one until the final model consisted only of variables with p ≤ 0.05.

The Educatel Brazil study was approved by the Ethics Research Committee of the School of Medicine of UFMG (CAAE: 48129115.0.0000.5149).

Results

A total of 6,510 Brazilian schoolteachers were interviewed (85.2% response rate). Women constituted 63.2% of sample; 64% were 44 years old or younger (mean 40.3 ± 10.6 years); 92.2% had completed university or were in progress; 60.4% were married or lived with a partner; 66.7% had children; 51% were white and 43% were black or brown; 62.4% had a monthly income at the school of up to three times the minimum wage; and two-thirds worked in the Central, Southeast, and South of the country.

Current smoking prevalence was 4.4%. Among male teachers, prevalence was 5.9% and was higher in those over 55 years of age (10.7%). Prevalence was 3.5% in female teachers as a whole and 5.5% in those from 45 to 54 years of age. Former smokers accounted for 11.3% of the participants (those who answered “Yes” to the question “Are you now or have you ever been a smoker, that is, have you smoked at least 100 cigarettes (five packs) in your life?” and “No” to the question “Do you smoke cigarettes now?).

Figure 1 shows the regional differences in teachers’ smoking: prevalence rates of 16.9%, 12.7%, and 8.1% in São Paulo, Rio Grande do Sul, and Minas Gerais, respectively, and 0.4% in Roraima and Alagoas.

Among the sociodemographic characteristics, white men over 45 years of age that lived alone and had children showed higher smoking prevalence rates (Table 1). The univariate analysis further indicated a positive statistical association between smoking and university education (complete or under way), monthly income greater than three times the minimum wage, and living in regions of Brazil other than the Northeast (Table 1). As for health status, smoking was positively associated with insufficient physical activity, work absenteeism in the previous 12 months, and use of anxiolytics or antidepressants, and negatively associated with occasional voice problems at work (Table 1).

As for the characteristics of teaching work, a workweek exceeding 40 hours was positively associated with smoking, while longer commuting time to and from school and lack of classroom discipline were negatively associated with the outcome (Table 2). Among schoolteachers that took more than two hours to commute to and from school, 79.3% commuted by automobile, motorcycle, or public transportation, while the others walked, bicycled, or use some other form of transportation.

The final multivariate model included sociodemographic variables, health status, and an indicator for occupational stress. Female gender was inversely associated with smoking. Age was maintained in the final model, indicating that higher age is positively associated with smoking. Being married or living with a partner, occasional voice problems at work, and longer commuting time to and from school remained inversely associated with the outcome. Living in the South and Southeast of Brazil, insufficient physical activity, and use of anxiolytics or antidepressants were positively associated with smoking (Table 3).

Discussion

This is the first nationwide study in Brazil, representative of schools’ census areas (rural versus urban), states, grade levels, and teachers’ gender and age to examine a behavior that represents overall individual health and serves as an indicator of social inequalities.

Two groups of results stood out: the negative association between smoking, female gender, being married/living with a partner, occasional voice problems, and longer commuting time and the positive association with age, living in the South and Southeast regions of the country, insufficient physical activity, and use of anxiolytics or antidepressants.

Prevalence of smoking (4.4%) was similar to that among municipal schoolteachers in Maceió, Alagoas State (5.6%) ²⁵25. Alves LP, Araújo LTR, Xavier Neto JA. Prevalência de queixas vocais e estudo de fatores associados em uma amostra de professores de ensino fundamental em Maceió, Alagoas, Brasil. Rev Bras Saúde Ocup 2010; 35:168-75.. However, when compared to other studies in Brazil, the prevalence rates in the current study were lower than among municipal preschool and primary schoolteachers in three other Brazilian cities: Ceballos & Santos ²⁶26. Ceballos AG, Santos GB. Fatores associados à dor musculoesquelética em professores: aspectos sociodemográficos, saúde geral e bem-estar no trabalho. Rev Bras Epidemiol 2015; 18:702-15. found 9.9% of smokers among the 525 schoolteachers in a city in Pernambuco State; Fuess & Lorenz ²⁷27. Fuess VL, Lorenz MC. Disfonia em professores do ensino municipal: prevalência e fatores de risco. Rev Bras Otorrinolaringol 2003; 69:807-12. found 8.9% prevalence among 451 schoolteachers in a city in São Paulo State; and Santos & Marques ¹⁵15. Santos MN, Marques AC. Condições de saúde, estilo de vida e características de trabalho de professores de uma cidade do Sul do Brasil. Ciênc Saúde Coletiva 2013; 18:837-46. observed 14.1% smoking prevalence in 414 schoolteachers in a city in the State of Rio Grande do Sul. Furthermore, a study of 258 primary and secondary schoolteachers in nine state schools in two cities in the interior of São Paulo State found that 9.3% were smokers ¹⁶16. Vedovato TG, Monteiro MI. Perfil sociodemográfico e condições de saúde e trabalho dos professores de nove escolas estaduais paulistas. Rev Esc Enferm USP 2008; 42:290-7..

These differences may be due to regional specificities. The current study showed significant differences in smoking prevalence between regions of the country, with higher proportions in the South and Southeast when compared to the Northeast. The result confirms that “Brazil is a country with major economic, social, and cultural diversities that can impact the patterns of tobacco consumption” ²⁸28. Barros AJ, Cascaes AM, Wehrmeister FC, Martínez-Mesa J, Menezes AM. Tabagismo no Brasil: desigualdades regionais e prevalência segundo características ocupacionais. Ciênc Saúde Coletiva 2011; 16:3707-16. (p. 3714).

The regional differences evidenced here are in line with the results of the VIGITEL study. According to data for 2016, smoking prevalence in adults varied from 5.1% in Salvador, Bahia State, to 14% in Curitiba, Paraná State ³3. Ministério da Saúde. Vigitel Brasil 2016: vigilância de fatores de risco e proteção para doenças crônicas por inquérito telefônico: estimativas sobre frequência e distribuição sociodemográfica de fatores de risco e proteção para doenças crônicas nas capitais dos 26 estados brasileiros e no Distrito Federal em 2016. http://portalarquivos.saude.gov.br/images/pdf/2017/junho/07/vigitel_2016_jun17.pdf (acessado em 22/Set/2017).
http://portalarquivos.saude.gov.br/image... .

The low prevalence of smoking among teachers is consistent with the literature. A study in the United States compared 40 occupational groups, and schoolteachers were the profession that smoked the least ²⁹29. Bang KM, Kim JH. Prevalence of cigarette smoking by occupation and industry in the United States. Am J Ind Med 2001; 40:233-9.. The number of former smokers among American schoolteachers was also similar to our survey, with a nearly twofold proportion of former smokers compared to current smokers. Among Brazilian schoolteachers, there were 2.5 times more former smokers than current smokers.

Teachers’ own educational level is also associated with smoking. In Brazil ³3. Ministério da Saúde. Vigitel Brasil 2016: vigilância de fatores de risco e proteção para doenças crônicas por inquérito telefônico: estimativas sobre frequência e distribuição sociodemográfica de fatores de risco e proteção para doenças crônicas nas capitais dos 26 estados brasileiros e no Distrito Federal em 2016. http://portalarquivos.saude.gov.br/images/pdf/2017/junho/07/vigitel_2016_jun17.pdf (acessado em 22/Set/2017).
http://portalarquivos.saude.gov.br/image... ^,3030. Silva GA, Valente JG, Almeida LM, Moura EC, Malta DC. Tabagismo e escolaridade no Brasil, 2006. Rev Saúde Pública 2009; 43:48-56. and elsewhere in the world ³¹31. Centers for Disease Control and Prevention. Current cigarette smoking prevalence among working adults - United States, 2004-2010. MMWR Morb Mortal Wkly Rep 2011; 60:1305-9., prevalence of smoking is higher among individuals with less schooling. This is another factor that favors the low prevalence found here in Brazilian schoolteachers, with a sample that included 92.2% of teachers with university education (complete or under way).

Smoking was less common among women teachers. Studies have shown that women smoke less than men, regardless of the type of sample (overall population ³3. Ministério da Saúde. Vigitel Brasil 2016: vigilância de fatores de risco e proteção para doenças crônicas por inquérito telefônico: estimativas sobre frequência e distribuição sociodemográfica de fatores de risco e proteção para doenças crônicas nas capitais dos 26 estados brasileiros e no Distrito Federal em 2016. http://portalarquivos.saude.gov.br/images/pdf/2017/junho/07/vigitel_2016_jun17.pdf (acessado em 22/Set/2017).
http://portalarquivos.saude.gov.br/image... or teachers ³²32. Al-Naggar RA, Jawad AA, Bobryshev YV. Prevalence of cigarette smoking and associated factors among secondary school teachers in Malaysia. Asian Pac J Cancer Prev 2012; 13:5539-43.^,3333. Rahman MM, Karim MJ, Ahmad SA, Suhaili MR, Ahmad SF. Prevalence and determinants of smoking behaviour among the secondary school teachers in Bangladesh. Int J Public Health Res 2011; Special Issue:25-32.^,3434. Savadi P, Wantamutte AS, Narasannavar A. Pattern of tobacco use among primary school teachers in Belgaum city, India - a cross sectional study. Global Journal of Medicine and Public Health 2013; 2(4).^,3535. Vendrametto MC, Silva MC, Gomes MF, Mella-Júnior SE, Mella EA. Prevalência de tabagismo em docentes de uma instituição de ensino superior. Arq Ciências Saúde UNIPAR 2007; 11:143-8.). The low prevalence of women smokers may be attributed to the social norm that establishes different male and female roles ⁶6. Erick PN, Smith DR. Prevalence of tobacco smoking among school teachers in Botswana. Tob Induc Dis 2013; 11:24.^,3333. Rahman MM, Karim MJ, Ahmad SA, Suhaili MR, Ahmad SF. Prevalence and determinants of smoking behaviour among the secondary school teachers in Bangladesh. Int J Public Health Res 2011; Special Issue:25-32. according to what society expects and accepts from men and women.

Since early ages, social influences modulate the paths by which men, differently from women, cope with adversities and build their experience in the world, with impacts on patterns of illness: disease rates are higher in women, while early mortality tends to affect men more frequently ³⁶36. Afifi M. Gender differences in mental health. Singapore Med J 2007; 48:385-91.. Male mortality is historically due to smoking, alcohol abuse, occupational exposure to carcinogenic agents, traffic accidents, and violent deaths involving weapons ³⁷37. Gove WR. Gender differences in mental and physical illness: the effects of fixed roles and nurturant roles. Soc Sci Med 1984; 19:77-91..

Marital status remained negatively associated with smoking, in keeping with other studies ⁶6. Erick PN, Smith DR. Prevalence of tobacco smoking among school teachers in Botswana. Tob Induc Dis 2013; 11:24.^,3232. Al-Naggar RA, Jawad AA, Bobryshev YV. Prevalence of cigarette smoking and associated factors among secondary school teachers in Malaysia. Asian Pac J Cancer Prev 2012; 13:5539-43.^,3535. Vendrametto MC, Silva MC, Gomes MF, Mella-Júnior SE, Mella EA. Prevalência de tabagismo em docentes de uma instituição de ensino superior. Arq Ciências Saúde UNIPAR 2007; 11:143-8.. Married schoolteachers or those living with a partner showed lower smoking prevalence than those living alone. Companionship is a situation that fosters greater support and security to continue on “life’s journey”, thereby constituting a barrier to recourse to negative coping strategies like substance use. The family exerts a well-documented influence on the success of smoking cessation strategies ³⁸38. Cardoso DB, Coelho AP, Rodrigues M, Petroianu A. Fatores relacionados ao tabagismo e ao seu abandono. Rev Med (São Paulo) 2010; 89:76-82..

Schoolteachers that reported occasional voice problems at work showed lower prevalence of smoking. There is a known association between smoking and voice problems, such as voice fatigue, hoarseness, and clearing one’s throat ³⁹39. Vieira AC, Behlau M. Análise de voz e comunicação oral de professores de curso pré-vestibular. Rev Soc Bras Fonoaudiol 2009; 14:346-51.^,4040. Ferreira LP, Latorre MR, Giannini SP, Ghirardi AC, Karmann DF, Silva EE, et al. Influence of abusive vocal habits, hydration, mastication, and sleep in the occurrence of vocal symptoms in teachers. J Voice 2010; 24:86-92.. The presence of voice symptoms or alterations may lead the teacher to avoid adopting harmful behaviors or even to abandon them ⁴¹41. Servilha EA, Bueno SS. Estilo de vida e agravos à saúde e voz em professores. Disturb Comun 2011; 23:153-63.. In this study, smoking prevalence was also lower in schoolteachers that reported frequent voice problems at work, but the difference was not statistically significant.

Longer commuting time to and from school showed a negative association with smoking, indicating that teachers who took longer to commute tended to smoke less. This result was unexpected. The duration of commuting adds time to the workday, considering the time dedicated to work, which can be considered a stressor, which would increase the odds of substance use ¹⁴14. Schrijvers CT, van de Mheen HD, Stronks K, Mackenbach JP. Socioeconomic inequalities in health in the working population: the contribution of working conditions. Int J Epidemiol 1998; 27:1011-8.. The fact that longer commuting time was associated with lower prevalence of smoking could be explained by the impossibility of smoking during automobile or motorcycle commuting or on public transportation, the means used by 79.3% of schoolteachers that reported longer commuting time to and from school.

There was a higher prevalence of smoking, with an upward gradient, among teachers over 45 years of age. In short, the older the individual, the lower the smoking rate. The results corroborate those of a study on the overall population: according to data from the VIGITEL study in 2016, smoking prevalence was higher in adults from 45 to 64 years of age ³3. Ministério da Saúde. Vigitel Brasil 2016: vigilância de fatores de risco e proteção para doenças crônicas por inquérito telefônico: estimativas sobre frequência e distribuição sociodemográfica de fatores de risco e proteção para doenças crônicas nas capitais dos 26 estados brasileiros e no Distrito Federal em 2016. http://portalarquivos.saude.gov.br/images/pdf/2017/junho/07/vigitel_2016_jun17.pdf (acessado em 22/Set/2017).
http://portalarquivos.saude.gov.br/image... .

The lower proportion of smokers among teachers with sufficient physical activity can be interpreted as an expected gain among those that have already manifested better adherence to healthy habits. In other words, it is consistent with adherence to physical activity to be less prone to unhealthy habits such as tobacco consumption ³⁸38. Cardoso DB, Coelho AP, Rodrigues M, Petroianu A. Fatores relacionados ao tabagismo e ao seu abandono. Rev Med (São Paulo) 2010; 89:76-82..

The group that reported use of anxiolytics or antidepressants in the four weeks prior to the survey showed a high prevalence of smoking when compared to individuals that had not used such medications during the same period. Prevalence of smoking is known to be higher in patients with mental disorders ⁴²42. Stramari LM, Kurtz M, Silva LC. Prevalence of and variables related to smoking among medical students at a university in the city of Passo Fundo, Brazil. J Bras Pneumol 2009; 35:442-8.. Smoking may also act here as a negative strategy for coping with emotional tensions by individuals with anxiety and depressive disorders ⁴³43. Moylan S, Jacka FN, Pasco JA, Berk M. Cigarette smoking, nicotine dependence and anxiety disorders: a systematic review of population-based, epidemiological studies. BMC Med 2012; 10:123.^,4444. Rondina RC, Gorayeb R, Botelho C. Psychological characteristics associated with tobacco smoking behavior. J Bras Pneumol 2007; 33:592-601..

Previous studies have pointed to associations between psychosocial job demands and intensity of tobacco consumption, attempts at smoking cessation, and relapses after attempts at quitting ¹⁰10. Heikkillä K, Nyberg ST, Fransson EI, Alfredsson L, De Bacquer D, Bjorner JB, et al. Job strain and tobacco smoking: an individual-participant data meta-analysis of 166 130 in 15 European studies. PLoS One 2012; 7:e35463.^,4545. Yasin SM, Retneswari M, Moy FM, Darus A, Koh D. Job stressors and smoking cessation among Malaysian male employees. Occup Med 2012; 62:174-81.. Unexpectedly, even when included among the independent variables, work conditions and demands at schools did not reach statistical significance in relation to smoking in this study (which does not mean that the teachers considered their work conditions adequate). Most schoolteachers in the sample had workweeks exceeding 40 hours and identified the following occupational stressors: loud noise and lack of discipline in the classroom and heavy work demands.

One can logically assume that the reduction in smoking prevalence in Brazilian schoolteachers is partly due to the country’s tobacco control initiatives, which have been acknowledged as efficacious in the general population ⁴4. Silva ST, Martins MC, Faria FR, Cotta RM. Combate ao tabagismo no Brasil: a importância estratégica das ações governamentais. Ciênc Saúde Coletiva 2014; 19:539-52.. The specific case of schoolteachers suggests the hypothesis that health promotion activities in schools ⁷7. World Health Organization. Promoting health through schools. Geneva: World Health Organization; 1997. (WHO Technical Report Series, 870).^,88. Figueiredo TA, Machado VL, Abreu MM. A saúde na escola: um breve resgate histórico. Ciênc Saúde Coletiva 2010; 15:397-402. have contributed to the drop in the number of smokers, even in the presence of occupational stressors that might otherwise maintain or increase the smoking rate among teachers.

Health promotion activities in Brazilian schools include a ban on tobacco use in schools by teachers and school staff, in addition to visitors and the students themselves. A systematic literature review that included 26 studies in different tobacco-free environments associated the ban on smoking with the reduction in prevalence of smoking or with the decrease in cigarette consumption by workers ⁴⁶46. Fichtenberg CM, Glantz SA. Effect of smoke-free workplaces on smoking behaviour: systematic review. BMJ 2002; 325:188.. In Japan, the results of a study with 1,534 teachers in tobacco-free schools suggest that restrictions on smoking produce two-way effects, by protecting students from becoming passive or active smokers while encouraging schoolteachers to cut down or quit smoking ⁴⁷47. Kiyohara K, Kawamura T, Itani Y, Matsumoto Y, Takahashi Y. Changes in teachers' smoking behaviour following enforcement of a total smoke-free school policy. Public Health 2012; 126:678-81..

Study limitations and strengths

The current study has some limitations to consider when analyzing the results: the inclusion of only healthy workers (since teachers on leave were excluded from the data collection); the cross-sectional design, which produce an image of health status characteristic of a specific moment in time and reduces the possibility of drawing causal inferences between the target factors; recall bias; and lack of direct contact with the interviewer. There was also a lack of complementary information on smoking: age at initiation and time transpired (in years), number of cigarettes consumed, and history of any previous attempts to quit.

The study’s strengths also deserve mentioning. The survey achieved a nationwide scope and included a sample that represented all Brazilian schoolteachers in basic education, regardless of the schools’ location or size. The questionnaire used in the data collection was based on an extensive literature review that allowed the researchers to produce an explanatory manual on the questions and conduct validity tests to measure the answers’ internal consistency ²⁴24. Assunção AA, Medeiros AM, Claro RM, Vieira MT, Maia EG, Andrade JM. Hipóteses, delineamento e instrumentos do Estudo Educatel Brasil, 2015/2016. Cad Saúde Pública 2019; 35 Suppl 1:e00108618..

Since face-to-face interviews allow for clarifying questions and answers, they are generally considered advantageous in comparison to remote interviews. However, in this study the relative disadvantage of telephone interviews was outweighed by the gain in scope, agility, and cost reduction, allowing to reach a sample representing the more than two million teachers working in basic education in Brazil ²³23. Vieira MT, Claro RM, Assunção AA. Desenho da amostra e participação no Estudo Educatel. Cad Saúde Pública 2019; 35 Suppl 1:e00167217..

Conclusion

Prevalence of smoking in Brazilian schoolteachers was relatively low in comparison both to the general population and to local samples of teachers. Factors such as gender, age, marital status, geographic region, physical activity, voice problems, use of medications, and commuting time to and from school were associated statistically with smoking in the study group. Unexpectedly, other occupational stressors did not remain associated with smoking, even though the majority of the schoolteachers reported loud noise and lack of discipline in the classroom, in addition to heavy work demands. The results highlight the need to continue and expand initiatives with the potential to generate two-way effects, by protecting students from becoming smokers and encouraging schoolteachers to cut down or quit smoking. Future studies may consider investigating the intensity of cigarette consumption, attempts to quit, and the association between smoking and other factors such as the use of alcohol and illicit drugs, overall school and classroom conditions, and shortage of teaching materials.

Acknowledgments

The authors wish to thank the Department of School Systems Linkage of the Brazilian Ministry of Education, the Brazilian National Institute for Educational Studies and Research “Anísio Teixeira” (INEP), and the Brazilian National Research Council (CNPq).

References

Publication Dates

History