PARENT SESSION
Oral Session # 22: Pathogens, Toxins, and Disease I: Modeling ; Mammalian.
Presiding: J Foufopoulos
Tuesday, August 5. 8:00 AM to 11:30 AM, SITCC Meeting Room 201.

A priori prediction of disease invasion dynamics in a novel environment.

Russell, Colin^*,1, Smith, David ², Waller, Lance³, Childs, James ⁴, Real, Leslie⁵, ¹ University of Cambridge, Cambridge, England, United Kingdom² University of Maryland, Baltimore, Maryland³ Rollins School of Public Health, Atlanta, Georgia⁴ Centers for Disease Control and Prevention, Atlanta, Georgia⁵ Emory University, Atlanta, Georgia

ABSTRACT- Predictive theories of disease emergence require the development of a priori models of spatio-temporal host-pathogen dynamics. A major challenge to the development of such models has been the incorporation of small scale heterogeneity. Initially we blended small-scale date-rich assessments of spatial heterogeneity into global models of disease spread, in a mechanistic model for the spread of raccoon rabies across the state of Connecticut. Our model incorporated spatial heterogeneity in local transmission rates depending on presence or absence of rivers and global long-distance transmission independent of local habitat conditions. We applied the Connecticut model in an a priori attempt to model the spread of raccoon rabies across New York State. Confronting the model predictions with the observed data from New York townships provided us with an unusual and powerful opportunity to test the predictive capabilities of our model as well as examine the importance of initial conditions. Using the model with multiple points of disease introduction into the state we were able to correctly predict 80% of the New York data (R² = 0.80, y = 0.95x, p < 0.0001). We were also able to use deviations from model predictions to elucidate the anisotropic effects of rivers and the interaction of wave direction and other geographic features.

Key words: predictive modeling, raccoon rabies, spatio-temporal dynamics