Epidemiology and Health (Oct 2019)

Alcohol-related emergency department visits and income inequality in New York City, USA: an ecological study

  • Kathleen H. Reilly,
  • Katherine Bartley,
  • Denise Paone,
  • Ellenie Tuazon

DOI
https://doi.org/10.4178/epih.e2019041
Journal volume & issue
Vol. 41

Abstract

Read online

OBJECTIVES Previous research has found that greater income inequality is related to problematic alcohol use across a variety of geographical areas in the USA and New York City (NYC). Those studies used self-reported data to assess alcohol use. This study examined the relationship between within-neighborhood income inequality and alcohol-related emergency department (ED) visits. METHODS The study outcome was the alcohol-related ED visit rate per 10,000 persons between 2010 and 2014, using data obtained from the New York Statewide Planning and Research Cooperative System. The main predictor of interest was income inequality, measured using the Gini coefficient from the American Community Survey (2010-2014) at the public use microdata area (PUMA) level (n=55) in NYC. Variables associated with alcohol-related ED visits in bivariate analyses were considered for inclusion in a multivariable model. RESULTS There were 420,568 alcohol-related ED visits associated with a valid NYC address between 2010 and 2014. The overall annualized NYC alcohol-related ED visit rate was 100.7 visits per 10,000 persons. The median alcohol ED visit rate for NYC PUMAs was 88.0 visits per 10,000 persons (interquartile range [IQR], 64.5 to 133.5), and the median Gini coefficient was 0.48 (IQR, 0.45 to 0.51). In the multivariable model, a higher neighborhood Gini coefficient, a lower median age, and a lower percentage of male residents were independently associated with the alcohol-related ED visit rate. CONCLUSIONS This study found that higher neighborhood income inequality was associated with higher neighborhood alcohol-related ED visit rates. The precise mechanism of this relationship is not understood, and further investigation is warranted to determine temporality and to assess whether the results are generalizable to other locales.

Keywords