Socioeconomic Status & Health Disparities: Utilizing a Composite Index across Health Datasets

Abstract

Research indicates that the relationship between socioeconomic status (SES) and health outcomes are robust, though no ‘gold standard’ as to what best captures SES exists. Many studies use individual proxies to control for SES, but there are a number of limitations to doing so. Additionally, little research has been conducted as to how to develop a US-based composite SES index that can be utilized across various datasets. The aim of the study is to generate a weighted SES index, following the National Assessment of Educational Progress (NAEP)’s call for an understanding of SES groupings and the guidance of the Bureau of Justice’s SES index. After generating a composite SES index, we evaluate whether or not each unique index delivers a similar classification of SES across four national datasets and empirically compare the resulting grouping across datasets utilizing multiple regression analyses. Using weighted descriptive statistics and weighted logistic regression analyses, we reveal the key variations in the distribution of determinants of the socioeconomic index. While previous studies conflated race and ethnicity with SES to resulting health indicators, this study captures both, highlighting the importance of each covariate and a composite index. Our findings suggest similar outcomes with regards to the assigned SES classifications (i.e. low, middle, and high SES) across the four national datasets.