Beware the gap: Bridging scales from within-host processes to between-host disease dynamics

Background/Question/Methods: A rapidly developing arena of inquiry in infectious disease ecology and evolution focuses on linking within- and between-host disease dynamics. Explicitly tracking dynamics at both scales is important because between-host parameters, such as transmission and virulence, are intimately tied to processes occurring at the within-host scale. This approach provides the opportunity for new insights that are not possible without explicitly nesting within- and between-host processes. While there has been tremendous progress in developing theory linking within- and between-host disease dynamics, the majority of empirical support is derived from laboratory studies rather than from natural systems and do not account for inherent sampling biases of field data. I developed a novel modeling framework that nests stage-structured N-mixture models and multi-state Jolly-Seber models to estimate demographic rates of within- and between-host disease dynamics, respectively, from field data. The field data requirements within and across seasons are: (i) host capture-mark-recapture data and (ii) repeated diagnostic samples of pathogen presence/absence taken from a single host individual. My modeling framework can be used to make inferences on host abundance, survival, recruitment, infection and recovery rates, and detection probability, as well as pathogen abundance, reproduction, and environmental re-infection rates.

Results/Conclusions: I show the utility of my modeling framework by applying the approach to a population of emerald glass frogs, Espadarana prosoblepon, that are afflicted by the pathogenic amphibian fungus Batrachochytrium dendrobatidis. I found that the within-host dynamics, such as reproduction rates of zoospores, influenced between-host disease dynamics, including transmission, recovery and mortality rates. This, in turn, affected the number of infectious individuals and the number of infectious spores in the environment, which altered the inoculum size, or initial pathogen load, of individuals picking up infections from the environment. I also found that inoculum size influenced the progression of disease within a host, which led to a feedback from between-host dynamics to within-host dynamics. This modeling framework provides the opportunity to study the evolution of pathogen virulence in association with transmission rates and explore the possibility of a conflict between natural selection at the individual and population level for directly transmitted diseases. This is especially critical in today’s world where infectious diseases cause mass mortality of species and populations worldwide.