You are here:
A novel approach for measuring residential socioeconomic factors associated with cardiovascular and metabolic health
Citation:
Mirowsky, J., R. Devlin, D. Diaz-Sanchez, W. Cascio, S. Grabich, C. Haynes, C. Blach, E. Hauser, S. Shah, W. Kraus, K. Olden, AND L. Neas. A novel approach for measuring residential socioeconomic factors associated with cardiovascular and metabolic health. Journal of Exposure Science and Environmental Epidemiology . Nature Publishing Group, London, Uk, 27:281-289, (2017).
Impact/Purpose:
The study highlights the importance of neighborhood characteristics on health. After controlling for individual-level demographic factors, significant differences in disease status were found based on a residents’ cluster designation. When combined with health data, this method can provide significant insight into the relation of neighborhood factors and health status. In subsequent work we will look at genetic/epigenetic differences in the populations based on their assigned clusters as well as the distribution of air pollutants between the clusters.
Description:
Individual-level characteristics, including socioeconomic status, have been associated with poor metabolic and cardiovascular health; however, residential area-level characteristics may also independently contribute to health status. In the current study, we used hierarchical clustering to aggregate 444 US Census block groups in Durham, Orange, and Wake Counties, NC, USA into six homogeneous clusters of similar characteristics based on 12 demographic factors. We assigned 2254 cardiac catheterization patients to these clusters based on residence at first catheterization. After controlling for individual age, sex, smoking status, and race, there were elevated odds of patients being obese (odds ratio (OR) = 1.92, 95% confidence intervals (CI) = 1.39, 2.67), and having diabetes (OR = 2.19, 95% CI = 1.57, 3.04), congestive heart failure (OR = 1.99, 95% CI = 1.39, 2.83), and hypertension (OR = 2.05, 95% CI = 1.38, 3.11) in a cluster that was urban, impoverished, and unemployed, compared to a cluster that was urban with a low percentage of people that were impoverished or unemployed. Our findings demonstrate the feasibility of applying hierarchical clustering to an assessment of area-level characteristics and that living in impoverished, urban residential clusters may have an adverse impact on health.
