Clustering of Adolescent Marijuana Use in Low-Income, Urban Neighborhoods

Background:  Limited data exist on the clustering of adolescent marijuana use, particularly in low-income, urban neighborhoods.  Clustering is an important aspect of the epidemiology of marijuana use that may shed light on aspects of the shared neighborhood environment driving higher or lower prevalence in certain neighborhoods, and thus inform community-wide prevention programs.

Method:  Seven hundred and ninety-nine primarily low income and African American youth were interviewed annually beginning in first grade as part of a longitudinal field study in Baltimore City.   In eighth grade, three hundred and twenty-one youth still resided in Baltimore City and had data available on marijuana use.  The alternating logistic regression (ALR) approach was used to estimate pairwise odds ratios between youth responses from the same neighborhood defined by census tract.  Using data provided by the Baltimore City police and the Baltimore City Data Collaborative, we modeled the extent to which clustering depended on neighborhood crime, violence, poverty and drug activity.  Models adjusted for gender, race, intervention and receipt of free or reduced-price lunch to ensure clustering was not an artifact of the composition of neighborhoods with respect to individual-level socio-demographic characteristics.

Results:  Being offered marijuana (PWOR=1.27; 95%CI=1.05, 1.55) and lifetime use of marijuana (PWOR=1.31; 95%CI=1.01, 1.70) significantly clustered within neighborhoods.  Clustering of opportunities to use marijuana was greater in neighborhoods with a larger percentage of adult arrests (PWOR=1.60; 95%CI=1.24, 2.06), juvenile drug arrests (PWOR=1.46; 95%CI=1.01, 2.10), juvenile violent crimes (PWOR=1.49; 95%CI=1.11, 2.00) and families in poverty (PWOR=1.38; 95%CI=1.01, 1.89).  Clustering of lifetime marijuana use was greater in neighborhoods with a larger percentage of adults arrested (PWOR=1.54; 95%CI=1.13, 2.11) and juvenile drug arrests (PWOR=2.13; 95%CI=1.29, 3.52).  

Conclusion:  This study provides evidence that opportunities to use marijuana and marijuana use in adolescence cluster in urban neighborhoods and, in particular, neighborhoods with higher rates of crime, violence, drug activity and poverty.  These findings have important implications by identifying shared aspects of the neighborhood environment that could be targeted in community-wide prevention efforts to reduce marijuana opportunities and use in adolescence.