Centre for Infectious Diseases, University of Edinburgh, Kings Buildings, Edinburgh, UK.
Epidemiol Infect. 2012 Jul;140(7):1215-26. doi: 10.1017/S0950268811001774. Epub 2011 Sep 19.
The kernel of modelling transmission dynamics of infectious diseases lies in constructing the force of infection (FOI). Traditionally, it was based on mass-action law. In this paper, we show, based on survey data of Escherichia coli O157 infection on Scottish cattle farms, that the actual form of FOI deviates greatly from mass-action law. Further, control effectiveness deviates qualitatively: the epidemic of mass-action FOI can be controlled with a control effort larger than the so-called herd immunity, while the epidemic inferred from the survey data of E. coli O157 infection was shown to be difficult to control. This indicates that, at least for E. coli O157 infection on cattle farms, it is risky to rely on models of transmission dynamics that were based on mass-action law to design the optimal intervention programme for infectious diseases. This suggests the importance of collecting epidemic data and model selection from data-driven models to infer the most likely model of transmission dynamics.
传染病传播动力学建模的核心在于构建感染力（FOI）。传统上，它是基于质量作用定律的。本文基于苏格兰牛场大肠杆菌 O157 感染的调查数据，表明 FOI 的实际形式与质量作用定律有很大的偏差。此外，控制效果也存在质的差异：具有大于所谓群体免疫的控制力度的情况下，可以控制质量作用 FOI 的流行，而从大肠杆菌 O157 感染的调查数据推断出的流行则很难控制。这表明，至少对于牛场的大肠杆菌 O157 感染，依赖于基于质量作用定律的传染病传播动力学模型来设计传染病的最佳干预方案是有风险的。这表明从数据驱动模型中收集流行数据和模型选择来推断最可能的传播动力学模型的重要性。
