Abstract
Many infectious diseases are not maintained in a state of equilibrium but exhibit significant fluctuations in prevalence over time. For pathogens that consist of multiple antigenic types or strains, such as influenza, malaria or dengue, these fluctuations often take on the form of regular or irregular epidemic outbreaks in addition to oscillatory prevalence levels of the constituent strains. To explain the observed temporal dynamics and structuring in pathogen populations, epidemiological multi-strain models have commonly evoked strong immune interactions between strains as the predominant driver. Here, with specific reference to dengue, we show how spatially explicit, multi-strain systems can exhibit all of the described epidemiological dynamics even in the absence of immune competition. Instead, amplification of natural stochastic differences in disease transmission, can give rise to persistent oscillations comprising semi-regular epidemic outbreaks and sequential dominance of dengue's four serotypes. Not only can this mechanism explain observed differences in serotype and disease distributions between neighbouring geographical areas, it also has important implications for inferring the nature and epidemiological consequences of immune mediated competition in multi-strain pathogen systems. The population dynamics of multi-strain pathogens are often characterized by persistent and irregular fluctuations in disease incidence and strain prevalence levels over time. Previous theoretical approaches have often evoked strong immunological interactions between individual strains, such as cross-immunity, in order to explain these complex epidemiologies; however, spatial segregation between hosts and stochastic heterogeneities in transmission success are rarely considered in these studies. Here, with specific reference to dengue, we show that the stochasticities underlying disease transmission within a spatially explicit, agent-based model can give rise to multi-annual epidemic outbreaks and fluctuating pathogen population structures - even in the absence of immune competition. In contrast to previous modeling studies, which have resulted in ambiguous predictions about the exact nature and strength of interactions between dengue's four serotypes, our results present a parsimonious, demographic mechanism, that highlights the importance of spatial ecology for understanding and interpreting the epidemiological dynamics of dengue and other multi-strain pathogen systems.

This publication has 73 references indexed in Scilit: