Citation

Lê F, Ahern J, Galea S. Drug Alcohol Depend. 2010; 112(1-2): 18-26.

Affiliation

Division of Epidemiology, University of California, Berkeley School of Public Health, 101 Haviland Hall, Berkeley, CA 94720, USA; Department of Epidemiology, University of Michigan School of Public Health, 109 Observatory Street, Ann Arbor, MI 48109, USA.

Copyright

(Copyright © 2010, Elsevier Publishing)

DOI

10.1016/j.drugalcdep.2010.05.005

PMID

20541875

PMCID

PMC2967648

Abstract

BACKGROUND: The neighborhood distribution of education (education inequality) may influence substance use among neighborhood residents. METHODS: Using data from the New York Social Environment Study (conducted in 2005; n=4000), we examined the associations of neighborhood education inequality (measured using Gini coefficients of education) with alcohol use prevalence and levels of alcohol consumption among alcohol users. Analyses were adjusted for neighborhood education level, income level and income inequality, as well as for individual demographic and socioeconomic characteristics and history of drinking prior to residence in the current neighborhood. Neighborhood social norms about drinking were examined as a possible mediator. RESULTS: In adjusted generalized estimating equation regression models, one-standard-deviation-higher education inequality was associated with 1.18 times higher odds of alcohol use (logistic regression odds ratio=1.18, 95% confidence interval 1.08-1.30) but 0.79 times lower average daily alcohol consumption among alcohol users (Poisson regression relative rate=0.79, 95% confidence interval 0.68-0.92). The results tended to differ in magnitude depending on respondents' individual educational levels. There was no evidence that these associations were mediated by social drinking norms, although norms did vary with education inequality. CONCLUSIONS: Our results provide further evidence of a relation between education inequality and drinking behavior while illustrating the importance of considering different drinking outcomes and heterogeneity between neighborhood subgroups. Future research could fruitfully consider other potential mechanisms, such as alcohol availability or the role of stress; research that considers multiple mechanisms and their combined effects may be most informative.

Language: en