Bayesian analysis for inference of an emerging epidemic: citrus canker in urban landscapes

Authors: NERI, F - University Of Cambridge; COOK, A - University Of Cambridge; Gottwald, Timothy; GILLIGAN, C - University Of Cambridge

Submitted to: PLoS Computational Biology
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 2/1/2014
Publication Date: 4/24/2014
Publication URL: http://doi:10.1371/journal.pcbi.1003587
Citation: Neri, F.M., Cook, A.R., Gottwald, T.R., Gilligan, C.A. 2014. Bayesian analysis for inference of an emerging epidemic: citrus canker in urban landscapes. PLoS Computational Biology. 10(4):e1003587.

Interpretive Summary: Outbreaks of infectious diseases require a rapid response from policy makers. The strength and efficacy of the responses depend upon available knowledge of how diseases spread through time and through the landscape. Yet, when a new pathogen is introduced into an alien environment, such information is often lacking or of no use, and many characteristics of disease spread must be estimated from the first observations. This poses a challenge to researchers concerning how quickly and reliably the characteristics of spread of an emerging disease be estimated, and how soon can the future progress of the epidemic be reliably predicted? We investigate these issues using a unique dataset for the invasion of a plant disease, Asiatic citrus canker in urban Miami. We use mathematical models to analyse rates and extent of spread of the disease. Doing so we were able to determine the complex epidemic behaviour of the disease. We can use the model to make reliable predictions from the early stages of the epidemic. Such a model is very important to regulators and policy makes to better understand and anticipate the kinds of epidemics that will occur from newly introduced pathogens.

Technical Abstract: Outbreaks of infectious diseases require a rapid response from policy makers. The strength and efficacy of the responses depend upon available knowledge of the spatial and temporal parameters governing pathogen spread, affecting, amongst others, the predicted severity of the epidemic. Yet, when a new pathogen is introduced into an alien environment, such information is often lacking or of no use, and epidemiological parameters must be estimated from the first observations of the epidemic. This poses a challenge to the epidemiologists: how quickly and reliably can the parameters of an emerging disease be estimated, and how soon can the future progress of the epidemic be reliably predicted? We investigate these issues using a unique, spatially- and temporally-resolved dataset for the invasion of a plant disease, Asiatic citrus canker in urban Miami. We use epidemiological models, Bayesian Markov-chain Monte Carlo methods, and dedicated goodness of fit tests, to analyse rates and extent of spread of the disease. A rich and complex epidemic behaviour is revealed. The spatial scale of spread is approximately constant over time, and can be estimated soon with extreme precision, although the evidence for long-range transmission (confounded by import of inoculum from outside the region of interest) is inconclusive. In contrast, the rate of infection is characterised by strong, environmentally-driven monthly fluctuations, strikingly similar among distinct geographical regions. Uninformed predictions from the early stages of the epidemic, assuming complete ignorance of the future environmental drivers, fail because of the unpredictable variability of the infection rate. Conversely, predictions improve dramatically if we assume prior knowledge of either the main environmental trend, or the main future environmental events. A clear contrast emerges between the high detail attained by modelling in the spatio-temporal description of the epidemic, and the bottleneck imposed on epidemic prediction by the limits of meteorological predictability: we argue that identifying such bottlenecks will be a fundamental step in modelling weather-driven epidemics.