Background: Specific hypotheses regarding putative mechanisms by which stressful life events might be related to smoking initiation among adolescents have been tested cross-sectionally on a cohort of 1598 grade 6 students in Scarborough, Canada. The purpose of this study was to examine the utility of these cross-sectional models in accounting for current smoking as compared to never smoking when this cohort was in grades 8 and 11.
Methods: Current smoking was defined as reported regular or occasional smoking. Logistic regression and multiple regression models were tested separately for each gender and grade with variables entered in prespecified steps.
Results: Current smoking is more strongly related to psychosocial variables and environmental variables among older male adolescents as compared to younger ones. In older female adolescents, current smoking appears to be more strongly related to attitude variables and less strongly related to psychosocial variables than among younger female adolescents. There is some evidence that grade 11 males may use cigarettes as a coping strategy for depression.
Conclusions: Male and female students differ in how stress and a range of psychosocial factors are interrelated with regards to current smoking and these relationships appear to change over time.