The results of this study indicate that there are other statistically significant socioeconomic and environmental variables associated with lung cancer mortality aside from smoking prevalence, and that their relationships vary across space. The county-level geographic scale and geographically-weighted regression used in this analysis allow for the identification of patterns in the distribution and spatial variability of lung cancer mortality and associated explanatory variables. This allows for a greater local contextualization of lung cancer mortality across the eastern United States.
The application of GWR in this study highlights the extent to which geographic information systems and spatial analysis can be used to understand lung cancer. By using GWR and considering the different degrees to which parameters such as income or race may be associated with lung cancer mortality at the local scale, we can better understand the disease and population it affects across space. These results can be visualized, facilitating greater communication of information. It becomes possible for governing bodies to understand observed illnesses and the characteristics of populations burdened by them. Policy and planning can, and should, be targeted to local populations in context-specific approaches in order to address lung cancer mortality efficiently and effectively. This study ultimately highlights the need for a geographic perspective in medical and health studies; it calls for finer scale, context-specific approaches to be carried forward in future research and practice.