Quantifying Associations between Environmental and Social Stressors

Abstract

Introduction: Association rule mining (ARM) has been widely used to identify associations between various entities in many fields. Though some studies have utilized it to analyze the relationship between chemicals and human effects, fewer have used this technique to identify and quantify associations among environmental and social stressors. Methods: We created socio-demographic variables based on U. S. Census tract-level income, race/ethnicity population percentage, education level, and age information from 2010-2014, 5-year summary files in the American Community Survey database, and generated chemical variables by utilizing the 2011 National-Scale Air Toxics Assessment (NATA) tract-level air pollutant exposure concentration data. We then applied ARM to quantify and visualize associations between these variables. Boostrapping, random sampling with replacement, was used to estimate the 95% confidence interval for certain statistical measures of the association rules. Results: Tracts with an average population age of 40 to 50 years old, a low percentage of racial/ethnic minorities (0-10%), and moderate income levels (20 ~ 40% of the residents have income lower than predefined poverty line) were more likely to have lower estimated chemical exposure concentrations (1st quartile). Tracts with a high percentage of racial/ethnic minorities (70 ~100% of the tract) and populations (70 ~ 100% of the tract) with low income tended to have higher estimated chemical exposure concentrations (4th quartile), especially for diesel PM, 1, 3-butadiene, and toluene. Interestingly, tracts with low-education population percentages (10 ~ 30% of the residents have an associate degree or above) tended to be associated with low estimated chemical exposure concentrations (1st quartile). Conclusions: Unsupervised data mining methods can be used to evaluate potential associations between environmental inequalities and social disparities, providing support to public health decision making.