Test of Biosocial Models of Adolescent Cigarette and Alcohol Involvement

• NCJ Number: 217418
• Journal: Journal of Early Adolescence Volume: 27 Issue: 1 Dated: February 2007 Pages: 4-39
• Author(s): Vangie A. Foshee; Susan T. Ennett; Karl E. Bauman; Douglas A. Granger; Thad Benefield; Chirayath Suchindran; Andrea M. Hussong; Katherine J. Karriker-Jaffe; Robert H. DuRant
• Date Published: February 2007
• Length: 36 pages
• Grant Number(s): CA16086
• Sponsoring Agency: National Cancer Institute
• Publication Format: Article
• Publication Type: Report (Study/Research)
• Language: English
• Country: United States of America

Annotation

This study tested biosocial models of interactions between biological factors and social context factors for the prediction of adolescent cigarette and alcohol use.

Abstract

Findings indicated that the models that included the interactions of the biological and contextual factors explained significantly more of the variance in adolescent cigarette and alcohol use than the models that did not include an analysis of interaction effects. The relationships between the biological and substance use variables grew stronger as the context became more harmful. For example, while findings on the direct link between alcohol use and testosterone were weak, the interaction analysis showed that testosterone was significantly related to alcohol involvement for girls living in harmful family contexts and harmful peer contexts. Indeed, the most significant effect involved interactions between biological factors and the peer context. The nature of the interactions indicated that the biological propensity for cigarette and alcohol use coupled with peer models of these behaviors might be particularly toxic. The findings suggest that researchers should not ignore the possibility that biological influences vary by context. Moreover, researchers who focus on social influences should not ignore biological variables. Participants were 409 students at a middle school in central North Carolina who completed questionnaires and provided saliva samples for hormone assays. The self-administered questionnaires focused on pubertal development, family context variables, peer context variables, perceived school context variables, perceived characteristics of their neighborhood, and cigarette and alcohol use. Salvia samples were tested for testosterone and estradiol. Data analysis involved the use of bivariate correlations and two-step hierarchical linear regression models. Future research should focus on identifying the mechanisms that are responsible for exacerbating the negative impact of biological factors on substance use in harmful contexts. Tables, references