Mediation in a family-directed program for prevention of adolescent tobacco and alcohol use

Abstract

BACKGROUND: Family Matters is a universal intervention designed to prevent adolescent tobacco and alcohol use through involvement of family members and by targeting family risk factors for tobacco and alcohol use. Previously reported findings suggest that the program reduced the prevalence of both adolescent smoking and drinking in the 12 months after program completion. This paper reports analyses conducted to identify the mediators through which the program influenced adolescent smoking and drinking. METHODS: One thousand fourteen adolescents ages 12 to 14 years and their families, identified by random-digit dialing, were entered into a randomized trial. Adolescents and their parents provided data by telephone for measuring mediator and behavioral variables at baseline, 3 months, and 12 months after program completion. Repeated-measures logistic regression with generalized estimating equations was used to assess mediation processes. RESULTS: The program resulted in statistically significant changes in several substance-specific aspects of the family, such as rule setting about tobacco and alcohol use. However, the intermediate family effects did not account for the program effects on adolescent behavior. CONCLUSIONS: The variables hypothesized to explain program effects were not identified by direct empirical examination