A mathematical framework for improving targeting of surveillance in complex pathosystems
Alexander Mastin: University of Salford
<div>Emerging plant pathogens are an increasing source of economic loss and a risk to food security worldwide, and early detection surveillance is a key method of preventing their establishment. Our previous work has demonstrated the importance of accounting for the ecology and epidemiology of the pathosystem in question when planning surveillance, but inherent heterogeneities within natural pathosystems can make this challenging. One way to consider these heterogeneities is by dividing the system into distinct epidemiological groups, such as plant hosts and insect vectors, between which a pathogen spreads. Using this approach, we can create a mathematical model of pathogen transmission and, by linking it to a statistical model of a sampling process, derive a simple heuristic for estimating the incidence of infection at the time of first detection for any amount of sampling effort from the constituent groups. We demonstrate the application of this heuristic by considering the vector-borne causative agent of the citrus disease huanglongbing, and find that it performs well in comparison to computer simulation. Through this process, we have also found that targeting surveillance efforts towards either hosts or vectors, rather than both, can maximise the cost effectiveness of surveillance, with the optimal group to sample being influenced by the transmission rates between the groups and the relative sampling costs.</div>
View Presentation |
