|
|
|
| Abstract Title:
|
| The Effect of Contact Network Structure on Disease Dynamics
|
| Graduate Student Presenter:
|
Greg Ames
|
| Name of the Author(s) and Affiliation(s):
|
Greg Ames; Dylan George; Christian Hampson; Andrew Kanarek; Cayla McBee; Jeff Achter; and Colleen Webb; Colorado State University
|
|
Recent global outbreaks of infectious diseases such as SARS and Avian Influenza have highlighted the need for models that can allow us to accurately predict the dynamics of a spreading disease. Traditionally, such modeling efforts have relied on “well-mixed models”, such as the Kermack and McKendrik SIR model, which assume that all individuals are equally likely to come in contact with each other. Although these models have helped to increase our understanding of disease behavior, there has been increasing recognition among disease modelers that these models ignore elements of contact structure that can have potentially large effects on disease dynamics. In response, many researchers have begun using graph theory to develop models that explicitly incorporate heterogeneous contact structure among individuals in a population. Many of these studies have focused on the distribution of connections among individuals as an explanatory variable for the subsequent disease dynamics, with empirical studies finding that real disease contact networks tend be exponentially distributed. Our immediate goal is to investigate whether degree distribution is sufficient to characterize a contact network by examining disease processes on sets of graphs that have the same exponential degree distribution but differ in their other structural properties. Our ultimate goal is to use attributes of particular graphs to predict how disease dynamics differ for heterogeneously connected models. Further, we aim to “correct” traditional homogeneously mixed models, using these attributes, so as to incorporate important aspects of contact heterogeneity. This will make traditional disease models more informative when considering differing contact structures.
|
|
|
