P01-217 - Etiology of Adolescent Substance Use: Neighborhood Effects on Initiation

Abstract

Objectives

We investigated the timing of cigarette, alcohol, and marijuana use initiation in adolescents, focusing on three levels of predictors: (1) individual-, (2) family- and (3) neighborhood characteristics.

Method

A community-representative sample of 440 mother-child dyads was assessed 4 times over 6 years. Children were on average 10 (+ 2) years old at baseline (T1), with equal SES, gender and ethnic representation.

We examined the timing of first alcohol, cigarette, and marijuana use as reported by children on the Youth Risk Behavior Survey (YRBS) at T3 in relation to predictors collected at baseline (T1).

Individual-level predictors at T1 included child gender, race, and child reports of sensation seeking. Family-level predictors included household size, maternal education, family income, and child-reported exposure to home violence. Finally, maternal and child reports regarding gangs and drugs in the neighborhood, and child reports of neighborhood violence were factor-analyzed into a single scale, where higher scores reflected unsafer neighborhoods.

To examine the timing of first cigarette, alcohol, and marijuana use as a function of the above predictors, three separate Cox regression (i.e., survival) models were estimated.

Results

Controlling for all other predictors, neighborhood safety remained significant predictor of substance use initiation among adolescents across all three examined substances. Children living in unsafe neighborhoods started, on average, smoking, drinking, and using marijuana at an earlier age, as indicated by positive and significant regression coefficients.

Conclusions

These results point to the importance of broader contextual influences - such as neighborhoods - on underage substance use initiation.