Divide and conquer? Infection dynamics in metapopulations

Persistence of infectious agents in populations is an important issue in epidemiology. It is often assumed that spatial fragmentation of a population of hosts, in a so-called metapopulation, increases the probability of persistence of the infectious agents, and that increased movement of hosts between patches, i.e. increased connectivity between the subpopulations, is beneficial for this. In the first part of this thesis, a stochastic model is used that describes both within- and between-patch dynamics and explicitly models movement of the hosts. It is shown that the relation between persistence of the infectious agents and the movement of hosts is less straightforward than previously thought. As a concrete example, the persistence of the plague bacterium Yersinia pestis is studied in a metapopulation of great gerbils in Kazakhstan. In the second part, the concept of the invasion indicator Rm , an analogy to Ro specifically adapted for metapopulations, is introduced to determine whether an infectious agent can successfully invade a metapopulation. This invasion indicator is then used to determine whether a mutant-infectious agent can invade a population and successfully replace the resident-infectious agent. This is a relevant problem for the evolution of infectious disease agents in spatially fragmented populations of hosts.