Quantitative measurement of the urban-rural continuum for examination of U.S. rural health disparities is a relatively new research area, where only a handful of studies have investigated health disparities using quantitative rural measures and even fewer have attempted to integrate health variables within said measures. Most U.S. rural health disparity studies and more specifically, rural colorectal cancer disparity studies, utilize various categorical and demographics/economics-based rural coding systems, which were not created for health disparity research. Further, both categorical rural classification schemes and more recent attempts to build quantitative health-focused measures are spatially and temporally static, which reduces translatability for study of cancer disparities across spatial units and time frames. In other words, to the knowledge of the author, no previous research has produced health-focused quantitative rural metrics that can be both flexibly translated to match the relevant time frames of health datasets and upscaled/downscaled to the desired spatial unit of analysis. Finally, spatial principles are inconsistently applied in rural colorectal cancer disparity studies, reducing inferential ability from results. Colorectal cancer is considered one of the most burdensome cancers for U.S. rural areas, so improvement of both measurement of the urban-rural divide for study of health disparities and application of spatial methods may help solidify findings of previous work. In this manuscript, there were two overarching goals: 1) construct spatiotemporal health disparity-focused continuous measurements of rural disadvantage that could be upscaled and downscaled and 2) utilize statistical and spatial methods to assess relationships between rural disadvantage and U.S. colorectal cancer mortality and screening. In Chapter 1, a county-level rural disadvantage index with integrated health factors was constructed using principal component analysis to weight ten rural indicator variables in three rural dimensions, followed by application to overall county-level cancer mortality via quantile regression. To the knowledge of the author, the index produced in this chapter is the first county-level quantitative rural measure with integrated health variables. Based on choropleth mapping, the constructed index showed improved numeric range and gradient over a popular existing rural measure while still retaining expected urban/rural trends. Spatiotemporal analysis showed only gradual change in index values for most U.S. counties, indicating stability over time. Results of the quantile regression showed that higher rural disadvantage index values were associated with higher cancer mortality rates, reflecting previous rural cancer disparity work. However, this effect was only present in the upper deciles of the probability distribution of mortality rates, indicating more complexity than previously understood. The county rural disadvantage index computed in this chapter should be considered a first step in attempting to integrate health variables into county-level quantitative rural measures. In Chapter 2, I applied the county-level rural disadvantage index to both global and local spatial models to explore rural colorectal cancer mortality and screening disparities. For the global mortality and screening linear models, Moran eigenvector spatial filtering was utilized to remove spatial autocorrelation from the residuals, while for the local models, geographically weighted regressions were used to determine if spatial non-stationarity existed in the relationships between rural disadvantage and both mortality and screening rates. To the best knowledge of the author, this paper constitutes the first instance of Moran eigenvector spatial filtering being used for spatial analysis of colorectal cancer mortality and screening. The global spatially filtered models displayed increasing colorectal cancer mortality rates and decreasing colorectal screening rates, respectively, as rural disadvantage increased, which reflected findings from previous work. In comparison to base global linear models, however, the magnitudes of effect of the spatially filtered models were reduced, displaying the importance of modeling spatial nuance. The geographically weighted regressions suggested that spatial non-stationarity existed in relationships between rural disadvantage and both mortality and screening, indicating the utility of local modeling. This Chapter provided a spatial modeling framework on which future rural colorectal cancer disparity analyses can account for spatial autocorrelation and spatial non-stationarity. In Chapter 3, the same rural indicator variables used in Chapter 1 were extended to construction of a sub-neighborhood grid-based rural disadvantage index for the state of Texas. A negative binomial hurdle model was then fit to examine the association between gridded index values and high spatial resolution colorectal cancer incidence-based mortality rates, while empirical Bayesian kriging and a spatial union procedure were also utilized to identify high colorectal cancer mortality risk-at-diagnosis areas. The rural disadvantage index produced in this chapter is the first sub-neighborhood quantitative rural measure produced for the state of Texas and the third sub-neighborhood quantitative rural measure for the U.S. more generally. Moreover, this paper is the first instance of empirical Bayesian kriging being utilized for cancer outcome spatial point data. Choropleth mapping showed that the constructed index had improved numeric range and gradient over an existing high resolution rural measure while mostly retaining expected urban/rural structure. The negative binomial hurdle model found that among Texas grid cells with at least one death, incidence-based mortality rates increased significantly as rural disadvantage values increased. The empirical Bayesian kriging procedure successfully identified high colorectal cancer mortality risk-at-diagnosis areas for the state of Texas, while the spatial union procedure displayed where these areas overlap with high rural disadvantage areas. The resulting sub-neighborhood maps have potential to inform where funding, colorectal cancer screening, and/or clinical services may best be micro-targeted.