PURPOSE: Introduction: A substantial proportion of adolescents try smoking, despite the restrictions on consumption.
METHODS: Objectives: To identify factors associated with the initiation of tobacco use by adolescent students and their early use (11-14 years old). Methodology: Design: cross sectional study with a representative sample. Reference population: adolescent students (elementary and high-school) from public and private schools in Salvador, Brazil. An anonymous questionnaire was applied to the pupils in the period of april 07th to June 13th, 2008. The selection of schools, classes and students was obtained by a multiple-stage sampling. Of the 6,500 questionnaires, 5,347 were considered valid for the study. Selected variables: Dependent: lifetime tobacco use. Independent: socio-demographic and educational variables, alcohol consumption and tobacco advertising. Statistics analysis: descriptive measures, 95%CI (precision of estimates)and logistic regression (RL). The level of statistical significance was p<0.17 to RL and 83%CIwas considered.
RESULTS: Results: 16.1% (95%CI=15.1-17.2)tried tobacco. The average age of the experimental use of the tobacco (n=757) was 13.9(±1.8) years old, 61.2% used before fifteen years old and 11.8% of young people tried tobacco at 11 years old. In the adjusted model (n=423), the father's smoking (OR=1.53;83%CI=1.07-2.19), advertising that promotes tobacco use (OR=2,43;83%CI=1.46-4.04), male gender (OR=0.54;83%CI=0.41-0.73%), white skin/Asian (OR=0.73;83%CI=0.54-1%) and economy class B1-A1 (OR=0.63;83%CI=0.46-0.86%)were associated with early consumption. Curiosity (64%;95%CI=56.9-63.8)and peers influence (17.6%;95%CI=15-20.4)were the main determinants for the experimental use reported by students. The consumption of alcohol (OR=6.04;83%CI=4.62-7.88), smoking friends (OR=3.18;83%CI=2.54-3.99) and boyfriend or girlfriend (OR=3.42;83%CI=1.78-6.56), tobacco advertising (OR=1.99;83%CI=1.39-2.86), father`(OR=2.16;IC83%=1.66-2.87)and mother` smoking(OR=2.2;83%CI=1.53-2.93), brother smoking (OR=1.63;83%CI=1.17-2.28), age 15-19 years (OR=3.41;83%CI=2.38-4.88), parents living separately (OR=1.43;83%CI=1.16-1,77) and aggressive-coercive intervention by parents (OR=1.60;83%CI=1.26-2.04) were related to trial of tobacco. Attend private school(OR=0.69;83%CI=0.54-0.88) and high school (OR=0.66;83%CI=0.49-0.9)reduced the chances of experimental tobacco use.
CONCLUSION: Conclusions: Alcohol consumption was strongly associated with tobacco experimentation, as well as the influence of peers and relatives. Tobacco consumption by the father and the advertising of these products were associated for the early use.
CLINICAL IMPLICATIONS: Knowing the determinants of tobacco use among school age adolescents may prevent this use.
DISCLOSURE: Adelmo Machado Neto, University grant monies No; Grant monies (from sources other than industry) No; Grant monies (from industry related sources) No; Fiduciary position (of any organization, association, society, etc, other than ACCP This project was suported by:Foundation for Research Support of the State of Bahia, Brazil.National Council of Science and Technology (CNPq)Department of Science and Technology; Other No; Product/procedure/technique that is considered research and is NOT yet approved for any purpose. There are no conflicts of interest in this study, considering it was a population-based study with questinário anonymous self-applicable.