Geographically-weighted regression analysis of percentage of late-stage prostate cancer diagnosis in Florida

Abstract

This study assessed spatial context and the local impacts of putative factors on the proportion of prostate cancer diagnosed at late-stages in Florida during the period 2001–2007. A logistic regression was performed aspatially and by geographically-weighted regression (GWR) at the nodes of a 5 km spacing grid overlaid over Florida and using all the cancer cases within a radius of 125 km of each node. Variables associated significantly with high percentages of late-stage prostate cancer included having comorbidities, smoking, being Black and living in census tracts with farmhouses. Having private or public insurance, being married or diagnosed in a for-profit facility, as well as living in census tracts with high household income reduced significantly this likelihood. Geographically-weighted regression allowed the identification of areas where the local odds ratio is significantly different from the ratio estimated using aspatial regression (State-level). For example, the local odds ratios for the comorbidity covariates were significantly smaller than the State-level odds ratio in Tallahassee and Pensacola, while they were significantly larger in Palm Beach. This emphasizes the need for local strategies and cancer control interventions to reduce the percentage of prostate cancer diagnosed at late-stages and ultimately eliminate health disparities.

Impaired Water Hazard Zones: Mapping Intersecting Environmental Health Vulnerabilities and Polluter Disproportionality

“…This presentation of the GWLR results maintains comparability with previous GWR-based research on environmental inequality outcomes [70]. This new visualization strategy also contributes to a growing literature that uses FDR-correction techniques in the presentation of GWLR results [58,61,65].…”

“…Since the introduction of geographically weighted regression (GWR) techniques [54,55], studies have used GWR to model spatial non-stationarity in a wide variety of multivariate statistical relationships. These include applications of geographically weighted logistic regression (GWLR) to model local variation in the predictors of binary outcomes spanning multiple subfields in human and physical geography [56][57][58][59][60][61][62][63][64][65][66][67][68]. Within the environmental inequality outcomes literature, spatial non-stationarity has been identified in the aggregate-level population vulnerability predictors of industrial air-toxic releases in New Jersey, USA [69], estimated lifetime cancer risk from cumulative ambient air-toxic pollution in Florida, USA [70], and vegetation land cover in shrinking and growing US cities [71].…”

“…More recent research is attentive to these issues of abnormality, multiplicity, and spatial dependence [20,58,61,65]. Guided by this work, I used the 1.96 and FDR-corrected critical t-value thresholds in the process of classifying the GWLR results in the Bay-Delta.…”

Area Deprivation and Liver Cancer Prevalence in Shenzhen, China: A Spatial Approach Based on Social Indicators

“…Previous research has indicated that health outcomes always present spatial autocorrelation at fine geographical scale (Chalkias et al 2013;Goovaerts et al 2015). In this context, the traditional OLS (ordinary least square regression) should be limited in identifying the local influential factors, since the homoscedasticity assumption of OLS is not satisfied (LeSage and Pace 2009).…”

“…The routinely recorded statistical data make it possible to establish an ADI at a fine geographical resolution. In addition, a typical characteristic of health outcome at the fine geographical sale is the spatial autocorrelation in local concentrations (Chalkias et al 2013;Goovaerts et al 2015). Several studies have demonstrated the role of spatial autocorrelation in explaining the geography of cancer prevalence associated with environmental pollution (Eitan et al 2010;Muir et al 2004;Sundmacher and Busse 2011).…”