Adolescent gender differences in the determinants of tobacco smoking: a cross sectional survey among high school students in São Paulo

Sanchez, Zila M.; Opaleye, Emerita S.; Martins, Silvia S.; Ahluwalia, Jasjit S.; Noto, Ana R.

Background: Diverse psychosocial factors have been associated with the use of cigarettes by adolescents. We investigated gender differences in tobacco smoking, and factors correlated with smoking among boys and girls. Methods: Data was collected on recent cigarette smoking (CS) and related factors, with a focus on religious beliefs, leisure activities, family structure, relationships and parental monitoring from 2,691 private school-attending youths from 28 schools in São Paulo, Brazil, selected via probability sampling. Estimates were derived via weighted hierarchical logistic regression models. Results: There was no difference in the prevalence of recent cigarette smoking between boys and girls (14.2%). Older age (a_OR_boys = 1.71[1.33-2.21]; a_OR_girls = 1.73[1.35-2.23]), second-hand smoke exposure at home (a_OR_boys = 1.52[1.00-2.29]; a_OR_girls = 1.74[1.13-2.68]) and not having a religion (a_OR_boys = 1.99[1.41-2.81]; a_OR_girls = 1.78[1.14-2.78]) were associated with CS in boys and girls. Adolescents who went out often at night were more likely to be tobacco smokers (a_OR_boys = 8.82[3.96-19.67]; a_OR_girls = 14.20[6.64-30.37]). For girls, data suggest that CS was also associated with a lack of parental attention and care (a_OR_girls = 4.37[1.19-16.04]) and no participation in youth religious activities (a_OR_girls = 2.76[1.49-5.12]). For boys, CS was additionally associated with the loss of one or both parents (a_OR_boys = 3.75[1.78-7.85]). Conclusions: Although older age, living with smokers at home and lack of religion were common contributing factors to cigarette smoking among all adolescents, girls were influenced to a greater degree by family relationships and religiosity than boys. The study results may be materially important to the development of prevention programs that influence determinants connected to gender and the implementation of single-core models of prevention; gender differences must be considered in order to reduce adolescent CS.

Geographic Areas


Also Published In

BMC Public Health

More About This Work

Academic Units
Published Here
January 7, 2014