Determinants Number of Cigarette Smoked with Iranian Adolescents: A Multilevel Zero Inflated Poisson Regression Model

Background: The present study was conducted to know about the association between amount of smoking and mental health plus demographic factors in the Iranian adolescents. Methods: Data of this research was drawn from the National Health Survey in Iran from April 1999-March 2000. All 1745 ado­lescents aged 15-20 yr that were residence in the eight northwest provinces of Iran, were included in this study. In order to evident zero-inflation and because of sampling design, multilevel zero-inflated Poisson (ZIP) regression applied for analy­sis of data. We analyzed data with programs written with s-plus. Results: Zero-inflated part of ZIP model shows that gender of female effects to increase the chance of "not to smoke ciga­rette at all" (P< 0.001) and older adolescents are less at risk of smoking than younger (P= 0.036). It also shows that, unem­ployed (P= 0.028) and Housewife (P= 0. 003), adolescents are more at risk of smoking compare with student adolescents. Un­der Poisson part of this model, depression (P= 0.012) and gender of female (P< 0.001) are indirectly associated with num­ber of cigarette smoked per day by adolescents. Conclusion: Among adolescent smokers, males smoke more number of cigarettes than females, younger teenager are more at risk of being a smoker than older ones and dropping out of high school is an important risk factor for smoking. We did not find any relation between mental disorders and being a smoker while we did for the relation between Number of Smoked Cigarette (NSC) and depression.