Abstract Introduction. The prevalence of tobacco use among students is apparently higher than in the general population. Several studies carried out in school population have found an increasing tendency in tobacco consumption, proportionally higher in women, which allows to identify similar consumption patterns in men and women. One of these studies estimated that the consumption frequency increases between 0.7% and 7% in three years. Nonetheless, another study points out to the fact that the Latin American and Caribbean youth population structure is an element that will favor the increase of the rate of smokers in the next years. Some of the risk factors identified are the tobacco consumption by other family members, a low perception of consumption risk, inclination towards tobacco use, to not deem the possibility of becoming a regular smoker, social tolerance, high availability and access to cigarettes, low school performance, to have friends that smoke, the search of a social image, a means of weigh control, search for a glamorous behavior, assert the transition between childhood and adulthood, the constant tobacco publicity in the media, the low supervision and the low monitoring by parents. Objective. To determine the prevalence and influence of the family and school environmental factors on tobacco use by adolescent students in Highschool No. 5 of the University of Guadalajara, Jalisco, Mexico, which has both an outstanding academic excellence and extracurricular program. Method. A cross sectional, prolective, and comparative study was performed. The population survey included 3,056 students enrolled at the above mentioned highschool in the 2004 term (36% men and 64% women); of the total population surveyed, 60.5% attended the morning shift and 39.5% the evening shift, their average age was 16.1+1.1 years. Due to the fact the previous studies report different prevalences between men and women, the size of the sample was estimated separately. For men, for whom a tobacco use of 27% has been estimated (and expecting a maximum deviation of 5% from the population prevalence) the required sample number was 190 subjects. Regarding women, of which a 16% prevalence was estimated (and expecting a maximum deviation of 3% from the population prevalence) the required sample number was 327 subjects. The subjects were recruited by drawing lots and their selection was carried-out by an aleatory numbers table. Each selected subject was located in its corresponding classroom and shift, and was invited to participate in this survey. They were asked to give an informed consent and there was a 100% rate of acceptance. The National Addictions Survey questionnaire was used to obtain the data, which assessed the socio-demographic data, consumption once in a lifetime, age of first use, number of times of consumption in a lifetime, time of use since the first time, use in the past 12 months, consumption in the last month, and number of cigarettes smoked daily. The tobacco consumption by the mother, father, siblings, and friends, was also included. The current tobacco addiction was defined by the consumption in the last month. The use pattern was considered experimental when it referred to one to 99 instances of consumption, and regular when it referred to more than 100 instances of use. For the statistical analysis, the categorical varia-bles were compared using the square Ji simple tests with Yates adjustment or Fisher’s exact test according to requirement. The numerical variables were compared using the test for independent samples. A model of logistic regression was used to analyze the independent association of factors linked to the present consumption of tobacco use. Results. The prevalence of consumption once in a lifetime was 58.2%; in the last 12 months, 32.4%; and in the last month, 23.9% (IC of 95%: 20.4-27.6). The average age of first pathway of consumption was 13.9± years (limits 5-18). The subjects with present tobacco use were older (16.4± 1.0 years of age) than the non tobacco users (16.1±1.1 years of age, p=0.002) and they also had a lower school perfor-mance in the previous semester (81.1±6 versus 86.0±7 p<0.0001). We did not found any difference between the prevalence of tobacco use among men (45/194; 24.7%) and women (83/353; 23.5%; p=0.41). The assessment of the prevalence according to the consumption pattern showed that 48% of students had an experimental use, while 10% of the adolescents studied already had a regular tobacco consumption pattern. Using the univaried analysis, the presence of a regular tobacco consumption pattern had a statistical association with the performance in semesters at highschool, the tobacco use by the mother, father, siblings, and groups of friends, as well as the fact of studying in the evening shift, studying and also working, and being an irregular student. Likewise, in the case of present tobacco use, in the univaried analysis were associated the performance in semesters at highschool, studying in the evening shift, being an irregular student, studying and also working, tobacco use by the father (not the mother), siblings, and friends. After the logistic regression analysis, only studying in the evening shift, tobacco use by the father and friends, and being an irregular student kept their statistical association regarding present tobacco use. Discussion. The prevalence of tobacco consumption by adolescent students found in this survey was high in experimentation (48.2%) and regularity (10%) in view of the characteristics of the highschool where this survey was carried out. Nevertheless, this prevalence is lower than those reported on other recent surveys that were done in high school students from similar settings. One of them, the 2003 National Student’s Survey found a 68.4% prevalence of tobacco use. This figure is 20% higher than the prevalence found in our study. We believe that those differences can be explained because the school where we performed our study has been awarded as an academic excellence high school. Another unexpected result was the absence of any differences in the consumption pattern in men and women, which also agree with the results of another recent study. This similarity in the behavior of men and women should be explored qualitatively. One contribution of our study is the fact that the father’s use but not the mother’s or siblings, has implications on adolescents consumption. Previous studies had reported the family’s consumption as a risk factor, without identifying the above mentioned differences. Nonetheless, further studies regarding the influence of the father figure or image should be carried-out. Meanwhile, as a working hypothesis, we consider that the father’s consumption defined attitudes of tolerance that are cognitively significant due to the cultural image of authority that he represents in the family. Even if the adolescent seeks to separate himself from family bond, and is more susceptible to the influence of the social environment, most of the family behavior has already been impressed on them. In the population studied by us, having friends that use tobacco was the most important association factor for its consumption. These findings have also been reported previously; nevertheless, the dynamics of use and their relationship with consumption should be further studied, as well as the social conceptualization attributed to friendship. The fact that studying in the evening shift was present in the unvaried analysis as well as in the logistic regression may be due to older students, lower performance and consequently, higher use probability. In short, the prevalence of tobacco use in our population of highschool adolescents was rather high, and we found no difference in the consumption pattern of men and women. The independent factors associated to tobacco use were tobacco consumption by the father and friends, as well as being in the evening shift. Finally, we consider that this study could represent the basis for: 1) Studies regarding new issues using qualitative tools in order to understand the social conceptualization concerning consumption and thus design more efficient educational actions; and 2) Having a baseline that will allow the assessment of the modification of risk factors on the decrease of tobacco consumption.
resumen está disponible en el texto completo