School of Environment and Life Sciences, University of Salford, Manchester, M4 5WT, United Kingdom; email:
Computation and Systems Biology, Rothamsted Research, Harpenden, AL5 2JQ, United Kingdom.
Annu Rev Phytopathol. 2017 Aug 4;55:591-610. doi: 10.1146/annurev-phyto-080516-035334. Epub 2017 Jun 21.
The rise in emerging pathogens and strains has led to increased calls for more effective surveillance in plant health. We show how epidemiological insights about the dynamics of disease spread can improve the targeting of when and where to sample. We outline some relatively simple but powerful statistical approaches to inform surveillance and describe how they can be adapted to include epidemiological information. This enables us to address questions such as: Following the first report of an invading pathogen, what is the likely incidence of disease? If no cases of disease have been found, how certain can we be that the disease was not simply missed by chance? We illustrate the use of spatially explicit stochastic models to optimize targeting of surveillance and control resources. Finally, we discuss how modern detection and diagnostic technologies as well as information from passive surveillance networks (e.g., citizen science) can be integrated into surveillance strategies.
新兴病原体和菌株的增加导致人们呼吁加强植物健康的更有效监测。我们展示了关于疾病传播动态的流行病学见解如何提高采样时间和地点的针对性。我们概述了一些相对简单但功能强大的统计方法来提供监测信息，并描述了如何对其进行改编以纳入流行病学信息。这使我们能够解决以下问题：在首次报告入侵病原体后，疾病的可能发病率是多少？如果没有发现疾病病例，我们能确定疾病不是仅仅因为偶然而未被发现吗？我们通过使用空间显式随机模型来说明如何优化监测和控制资源的靶向定位。最后，我们讨论了如何将现代检测和诊断技术以及来自被动监测网络（例如公民科学）的信息整合到监测策略中。
