Geostatistical software for the analysis of individual-level epidemiologic data
R44 Phase II (End date: 06/30/2011)
This SBIR project is developing the first geostatistical software to offer tools that are specifically designed for the geostatistical analysis of both aggregated and individual-level health data, providing: (1) the description of spatial patterns of cancer incidence and mortality rates and identification of nested scales of variation; (2) the spatial interpolation of individual-level data to create isopleth maps of relative risk accounting for area-based data (e.g. contextual factors), model for population density and locational uncertainty due to geocoding errors; (3) the identification of clusters and hotspots of significantly high or low incidence or mortality rates; (4) the detection of significant disparities between rates for different ethnic groups and genders; and (5) the visualization of changes in disparity through time. This project will accomplish 6 aims:
- Conduct a requirements analysis to identify methods and functionality to incorporate into the software.
- Develop innovative non-parametric geostatistical techniques for identifying nested scales of variation, exploring the combined impact of covariates on incidence rates, and the creation of incidence risk maps that incorporate both individual-level and area-based data.
- Develop new methodologies for incorporation of locational uncertainty into the geostatistical analysis, disaggregation of population data to allow the isopleth mapping of mortality rates, assessment and propagation of the uncertainty attached to mapped rates through hypothesis testing, such as the detection of significant interactions between covariates or the delineation of areas of significant gender disparities.
- Build and test a complete set of functionalities based on the research results and simulation studies, and incorporate them into Biomedware’s space-time visualization and analysis technology.
- Apply the software and methods to demonstrate the approach and its unique benefits for the investigation of geographic, ethnic and gender variations in cancer stage at diagnosis and survival data, and the exploration of relationships between health outcomes and potential factors, such as environmental and occupational exposures, socio-economic conditions, and proximity to screening facilities.
- Create instructional materials, including a short course, to foster the adoption of this approach in health science.
Feasibility of this project was demonstrated in Phase I. This Phase II project will accomplish aims three through six. These technologic, scientific and commercial innovations will revolutionize our ability to interpret variation in cancer incidence at multiple spatial scales and across time, to understand the causes underlying observed racial, gender and geographic disparities in incidence, mortality and survival rates, and to quantify the longterm benefits of current strategies for reducing the disproportionate incidence of cancer morbidity and mortality among minorities and the medically underserved in the United States. The substantial benefit of this research is its utility in accessing, linking, mapping and analyzing diverse individual-level and population-based data including cancer vital events and possibly area wide distributions of social class measures. The methods developed in this project will help generating hypotheses for in depth individual studies of risk factors that are causal, or impact survival or morbidity, and establishing the rationale for targeted cancer control interventions, including consideration of health services needs, and resource allocation for screening and diagnostic testing.