The direction and rate of spread of chronic wasting disease
Source: PubMed Central Open Access, NCBI / U.S. National Library of Medicine
Introduction Chronic wasting disease (CWD) is a fatal transmissible prion disease of cervids that continues to expand across North America. Although the spatiotemporal distribution of CWD in the central United States has been extensively documented, the direction and rate of disease spread remain poorly understood, limiting the implementation of proactive surveillance and mitigation strategies. Methods We developed a data-driven spatiotemporal modeling framework to quantify the direction and rate of CWD spread across Kansas using surveillance data collected between 2005 and 2023. Kansas was partitioned into 20 kmspatial grid cells, and disease dynamics within each cell were modeled using a system of differential equations representing susceptible, infected, and environmental compartments. Spatial migration processes were incorporated through a stochastic mixing matrix informed by empirically observed patterns of first infection among neighboring cells. Model parameters were optimized through a grid search of plausible values derived from published literature, and predictive performance was evaluated using receiver operating characteristic (ROC) analysis. To characterize spread dynamics, we applied a weighted centroid approach at the zonal scale and a migration-based vector flow analysis at the local scale. Results The model successfully reproduced the observed spatiotemporal progression of CWD across Kansas with high predictive accuracy (mean AUC = 0.88). Zonal analyses revea
Abstract
Introduction Chronic wasting disease (CWD) is a fatal transmissible prion disease of cervids that continues to expand across North America. Although the spatiotemporal distribution of CWD in the central United States has been extensively documented, the direction and rate of disease spread remain poorly understood, limiting the implementation of proactive surveillance and mitigation strategies. Methods We developed a data-driven spatiotemporal modeling framework to quantify the direction and rate of CWD spread across Kansas using surveillance data collected between 2005 and 2023. Kansas was partitioned into 20 kmspatial grid cells, and disease dynamics within each cell were modeled using a system of differential equations representing susceptible, infected, and environmental compartments. Spatial migration processes were incorporated through a stochastic mixing matrix informed by empirically observed patterns of first infection among neighboring cells. Model parameters were optimized through a grid search of plausible values derived from published literature, and predictive performance was evaluated using receiver operating characteristic (ROC) analysis. To characterize spread dynamics, we applied a weighted centroid approach at the zonal scale and a migration-based vector flow analysis at the local scale. Results The model successfully reproduced the observed spatiotemporal progression of CWD across Kansas with high predictive accuracy (mean AUC = 0.88). Zonal analyses revealed a predominant northwest-to-southeast progression of infection, with substantial regional heterogeneity in both direction and velocity of spread. Local vector-flow analyses identified multiple transmission corridors and persistent hotspots that may function as regional sources of infection dissemination. Rates of spread quantified at both zonal and local scales corroborated the directional trends observed in the flow analyses and highlighted areas characterized by elevated transmission intensity and migratory spread. Importantly, the model demonstrated strong forecasting capability by anticipating emerging areas of infection prior to their confirmation through surveillance. Discussion By integrating mechanistic disease dynamics with empirically informed migration processes, our framework provides a quantitative characterization of both the direction and rate of CWD spread at two spatial scales. These findings offer actionable insights for wildlife disease management by supporting targeted surveillance, boundary monitoring, and region-specific intervention strategies.