The Effects of Cumulative Neighborhood Risk on Substance Use Initiation Among Latino and African American Adolescents

Background and Purpose:Adolescent substance use continues to be a major health concern. Environmental characteristics are known to be critical factors in explaining substance use among adolescents. Yet, empirical evidence is limited particularly about the effects associated with timing of environment exposure in substance use initiation. To address this concern, this study: (1) examines the effects of cumulative exposure to neighborhood risk and protective factors during preadolescence (age 8 to 11years) on the likelihood of initiating alcohol, cigarette and marijuana use during adolescence (ages 12 to 18years); (2) investigates whether the effects of the aforementioned cumulative risk and protective factors are different across the three types of substances; and (3) examines whether the effects of cumulative neighborhood risk and protective factors vary by ethnicity.

Methods:This study uses administrative and retrospective survey data from a natural experiment in Denver whereby subsidized housing residents were quasi randomly assigned to their initial housing units and neighborhoods. The sample includes 736 Latino and African American youth between the ages of 12 and 18 years who resided for at least 2 years in public housing in Denver before reaching the age of 19. The statistical model employed is a Cox Proportional Hazard Cause-Specific model with clustered robust standard errors. The dependent variable is the initiation of drinking alcohol, smoking cigarettes and marijuana use during adolescence (None = 0, cigarettes = 1, alcohol = 2, and marijuana = 3) and after their initial random assignment to subsidized housing. The cumulative measures of neighborhood social disorder, neighborhood violent and property crime rates, and neighborhood social capital are used as primary independent variables along with individual and household covariates. In addition, the model is stratified by Latino and African American ethnicity.

Results:Adolescents who initiated one of the three types of substance use included 5.6% who initiated cigarette use and 5.4% who initiated alcohol use. Only 5.2% of the youth used marijuana for the first time during adolescence. The results of the Cox PH Cause-Specific models indicate that exposure to neighborhood social disorder during preadolescence is a significant risk factor, especially for the initiation of cigarette use among Latino adolescents. Specifically, greater exposure to neighborhood social disorder during preadolescence was associated with a 36% higher hazard of initiation of cigarette use during adolescence for all youth (HR = 1.36, SE = 0.17, p < .05) and 42% for Latino youth (HR = 1.42, SE = 0.23, p < .05).

Conclusions and Implications: The findings from this study suggest that neighborhood characteristics do matter for substance use initiation, particularly cigarette use, among adolescents. Additionally, the results suggest the need for further research on the relationship between exposure to neighborhood social disorder and adolescent substance use initiation in order to develop and implement community-based prevention and intervention programs to reduce substance use initiation and facilitate healthy adolescent development.