Impacto da exposição domiciliar ao tabaco e da composição familiar sobre a precocidade de uso de tabaco e álcool em adolescentes brasileiros

Abstract

Early use of tobacco and alcohol is associated with increased morbimortality and maintenance of their use during adulthood. This thesis investigates the role of household and family factors on adolescents' early onset of these substances. The data analyzed are from the Cardiovascular Risk Factors Study (ERICA), which investigated 74,589 adolescents from public and private schools, aged between 12 and 17 years old, living in 124 municipalities with more than 100 thousand inhabitants in the country. The first manuscript aimed to estimate the effect of the presence of smokers in the household on the earlier onset of smoking habit among adolescents. The objective of the second manuscript was to estimate the effect of family composition on the earlier onset of smoking and alcohol use among adolescents. In both analyses, unadjusted and adjusted Cox Regression Models were used to control for confounding variables. The Hazard Ratio estimated in manuscript 1 showed a 51% increase (95% CI: 31%-65%) in the chance that a nonsmoking adolescent exposed to tobacco at home would start smoking earlier than one who did not live with smokers. The results of the second manuscript depicted an important impact of family composition on the early onset of tobacco and alcohol use among adolescents. According to the estimated Hazard Ratios, there was a 68% increased chance (95% CI: 32%-60%) that a nonsmoking adolescent who did not live with both parents would initiate tobacco use earlier than an adolescent living with both parents. As for those who resided with only the mother or only the father, the increase was 46% (95% CI:32%-60%) and 54% (95%CI: 28%-84%), respectively. Regarding the onset of consumption of alcoholic beverages, there was a 35% increase (95% CI:21%-52%) in the chance of an adolescent who does not drink and does not reside with both parents to initiate alcohol use earlier than those living with the mother and the father. As for adolescents living only with the mother or only the father, this increase was 22% (95%CI: 17%-28%) and 36% (95%CI: 24%-50%), respectively. The third manuscript investigated the concurrent use of tobacco and alcohol through co-occurrence and clustering analyses. It also described the temporal relationship between the onset of tobacco and alcohol use among adolescents consuming both substances. For the clustering analysis, we calculated ratios between observed (O) and expected (E) values of tobacco and alcohol use. In the total sample, the O/E ratio was 3.51 (95% CI: 3.39-3.65), indicating the clustering of the use of these substances. The prevalence odds ratio was 14.7 (95% CI: 12.4- 17.4). Most of the users (45.2%) of both substances started with the use of alcohol. Our results reinforce the proposal that preventive actions should not be focused on each drug separately and should include the adolescents and their families.