Modeling spatiotemporal patterns of chronic wasting disease using surveillance data
No Thumbnail Available
Authors
Meeting name
Sponsors
Date
Journal Title
Format
Thesis
Subject
Abstract
Chronic wasting disease (CWD) is a neurodegenerative disease of cervids that has spread in North America uncontrollably since the late 1960s. This disease has no cure or effective prevention strategies, and its spread has wildlife disease management concerns as well as concerns for rural economies as many in the Midwestern states depend on big game hunting revenues. Surveillance of this disease by state wildlife health departments is done through nonprobabilistic sampling and it is often used for understanding the spatiotemporal distribution of the disease and its background prevalence in different areas. Making interpretations based on such surveillance data is problematic. In this study, I fit progressively complex Generalized Additive Models with a surveillance dataset collected in Kansas between 2005 - 2023 that was balanced spatially and spatiotemporally while comparing different ratios of negatively diagnosed data records to the positively diagnosed data sets. The best fitting model, with spatiotemporal smoothing functions and a tensor product interaction effect, produced a spatiotemporal probability map with an Area Under the Curve of 0.95, and a deviance explained of 65 percent. The prediction map reveals the potential spatiotemporal distribution of CWD in the state of Kansas and the spatiotemporal progression over the years. The northwestern, central, and southcentral portions of the state have progressively higher probability of cervids that could test positive for CWD. The potential spread of the disease to the eastern side of the state is also minimally evident.
Table of Contents
DOI
PubMed ID
Degree
M.S.