Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)


Geography and Anthropology

First Advisor

Nina Lam


Cancer is one of the most serious health problems in the U.S. Louisiana mortality rates for all sites of cancer have exceeded national rates since 1950. Even though concern about the issues of environmental cancer is growing, conclusive results have not been reported. Previous studies differed in data source, data quality, time of observation, and statistical procedure, making comparison of results problematic. The spatial patterns of the most common cancer sites in Louisiana (from 1953 to 1993) and their relationships with environmental factors were investigated, using GIS software and statistical software. To provide a clear understanding of how the spatial distributions of cancer mortality rates in Louisiana compared with those in the nation, this research was conducted using data at different geographic levels. Cancer mortality rates for most parishes in South Louisiana were higher than those of the entire U.S. for all sites combined as well as lung and stomach cancer, individually. Spatial clusterings of cancer mortality rates were tested using factor analysis, spatial autocorrelation and correlograms, and scan statistic. Cancers of all sites combined, lung, and stomach exhibited the strongest positive spatial autocorrelation. In searching for the presence of geographical clusters for lung cancer deaths at parish and census tract levels from 1988 to 1993, results revealed that there was a statistically significant and geographically distinct cluster of lung cancer deaths in southeastern Louisiana from 1988 to 1993. Stepwise regression analyses were applied to data to examine the relationships between environmental factors and cancer mortality rates. Results showed a positive relationship between cancer (all sites combined) and urban population and a negative relationship between cancer and persons employed in health and education services. Breast cancer mortality rates were significantly positively related to urban residence and education level. Colorectal cancer mortality rates were associated with the type of drinking water. Lung cancer mortality rates were more closely related to occupational variables and agricultural chemicals than any other types of cancer mortality rates. Prostate and stomach cancer mortality rates were positively associated with nonwhite populations.