Spatial and temporal analysis of the socioeconomic factors associated with breast cancer in Illinois using geographically weighted generalized linear regression

Keywords: Geographically weighted generalized linear regression, Breast Cancer, Illinois
Session Type: Virtual Poster Abstract
Day: Friday
Session Start / End Time: 2/25/2022 05:20 PM (Eastern Time (US & Canada)) - 2/25/2022 06:40 PM (Eastern Time (US & Canada))
Room: Virtual 21

Abstract

A lot of studies have attempted to examine the association between the risk of breast cancer and socioeconomic factors using traditional statistical approaches without considering the spatial variations of socioeconomic variables and their varying effects on the risk of breast cancer. As an alternative, geographically weighted regression (GWR) can address such weakness. However, not all factors are necessarily global factors or local factors. Under this context, this study employs geographically weighted general linear regression (GWGLR) model to analyze the relationship between socioeconomic characteristics and breast cancer risk for the state of Illinois from 1999 to 2013. The study first uses ordinary least square to identify the most influential variables. Then GWR and GWGLR models are constructed based on the same set of variables. The GWGLR models improved both the OLS and GWR models. The spatial and temporal analyses suggest that economic status has a global effect on breast cancer risk. Both low and high economic status could be associated with an elevated risk of breast cancer. Occupation, however, is associated with the risk of breast cancer more as a local factor. The GWGLR models also helped reveal the regional variations of those relationships.