Estimating an area-level socioeconomic status index and its association with colonoscopy screening adherence
Reproducibility of Results
Permanent link to this recordhttp://hdl.handle.net/20.500.12613/4888
MetadataShow full item record
Abstract© 2017 Wheeler et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Socioeconomic status (SES) is often considered a risk factor for health outcomes. SES is typically measured using individual variables of educational attainment, income, housing, and employment variables or a composite of these variables. Approaches to building the composite variable include using equal weights for each variable or estimating the weights with principal components analysis or factor analysis. However, these methods do not consider the relationship between the outcome and the SES variables when constructing the index. In this project, we used weighted quantile sum (WQS) regression to estimate an area-level SES index and its effect in a model of colonoscopy screening adherence in the Minnesota-Wisconsin Metropolitan Statistical Area. We considered several specifications of the SES index including using different spatial scales (e.g., census block group-level, tract-level) for the SES variables. We found a significant positive association (odds ratio = 1.17, 95% CI: 1.15-1.19) between the SES index and colonoscopy adherence in the best fitting model. The model with the best goodness-of-fit included a multi-scale SES index with 10 variables at the block group-level and one at the tract-level, with home ownership, race, and income among the most important variables. Contrary to previous index construction, our results were not consistent with an assumption of equal importance of variables in the SES index when explaining colonoscopy screening adherence. Our approach is applicable in any study where an SES index is considered as a variable in a regression model and the weights for the SES variables are not known in advance.
Citation to related workPublic Library of Science (PLoS)
Has partPLoS ONE
ADA complianceFor Americans with Disabilities Act (ADA) accommodation, including help with reading this content, please contact firstname.lastname@example.org