Doctoral Dissertations

Date of Award

8-2013

Degree Type

Dissertation

Degree Name

Doctor of Philosophy

Major

Geography

Major Professor

Liem Tran

Committee Members

Yinkgui Li, Nicholas Nagle, Agricola Odoi

Abstract

This research examines risk factors for sporadic cryptosporidiosis and Escherichia coli (E. coli) O157 infection in East Tennessee, using case-control and retrospective ecological approaches. Multiple models and approaches are used to identify risk factors for the two diseases, and to examine the effect of scale on risk for disease in the individual and in the population. Risk factors examined are animal density, land use, geology, surface water impairment, poverty rate and availability of private water supply. The research objectives are, first, to identify risk factors for E. coli O157 and cryptosporidiosis in East Tennessee by relating disease data to environmental data through statistical regression models and second, to examine the effect of scale by comparing risk factors for disease in the individual (case-control approach) and the population (ecological approach).

At the individual level Generalized Linear Models (GLMs), Generalized Additive Models (GAMs), and Spatial Logistic Regression Models are compared. At the population level, Spatial Lag, GLMs and GAMs are developed using Gamma, Tweedie and Poisson distributions.

Beef cow population density and proximity to karst geology are positively associated with both diseases at the individual scale. Land use variables representing developed land and pasture land are positively associated with both diseases at both scales. Poverty rate is positively associated with both diseases at the regional scale, and availability of private water supply is negatively associated with both diseases at both scales.

The results presented here show that the significance of environmental variables as risk factors for cryptosporidiosis and E. coli O157 depend on scale, and that an examination of risk factors for these diseases in the individual and the population can reveal the scale at which variables are important.

These results can be used to identify important environmental risk factors for the diseases and to identify the communities where background risk is highest. Limited public health resources can then be targeted to the risk factors and communities most at risk. These results can also be used as the framework upon which to develop a comprehensive epidemiological study that focuses on risk factors important at the individual level.

Download
Files over 3MB may be slow to open. For best results, right-click and select "save as..."

Share

COinS