Thornley John H M, France James
Centre for Nutrition Modelling, Department of Animal & Poultry Science, University of Guelph, Guelph, ON N1G 2W1, Canada.
Prev Vet Med. 2009 Jun 1;89(3-4):139-54. doi: 10.1016/j.prevetmed.2009.02.019. Epub 2009 Mar 27.
A simple generic model of foot and mouth disease (FMD) is presented. The dynamics of FMD at the index site (the farm where the initial infection occurs) is represented. Spread of disease from the index site is simulated with a three-term radial function, the terms corresponding to natural processes, short- and long-range movements. Parameterization includes the processes of threshold for clinical disease reporting, time delay before action is initiated, removal (slaughter) rate at the index farm; movement restrictions, culling of farms out to a given radius (from the index site), rate of culling, and vaccination (timing; rate of build-up of immunity). Between-farm basic reproductive ratio, R0bf, the number of primary (direct) infected farms arising from the initially infected (index) farm during the lifetime of the infection at the index farm, is calculated numerically. A "predictor-corrector" method is developed to estimate an upper limit to the number of secondary infections which could arise during this time. The model is easily programmed and is quick to execute. Simulation results are discussed in relation to the processes assumed in the model. They suggest that, given current UK farm practice, early disease detection, combined with immediate rapid slaughtering at the index site and restriction of short-range and long-range movements are relatively effective strategies of disease control. However, farm practice changes continually, and any control strategies need re-assessment in the light of current farm practice when and where an FMD outbreak occurs. A transparent and "simple" model can facilitate this process. The uncorrected model, without tuning, is compared with the cattle data from the 2001 UK FMD epidemic (Section 6; Fig. 8).
本文提出了一种简单的口蹄疫通用模型。该模型描述了口蹄疫在首发地点(即首次发生感染的农场)的动态变化。口蹄疫从首发地点的传播通过一个三项径向函数进行模拟,这三项分别对应自然传播过程、短程和远程移动。参数设定包括临床疾病报告的阈值、采取行动前的时间延迟、首发农场的扑杀率;移动限制、对距首发地点一定半径范围内农场的扑杀、扑杀率以及疫苗接种(时间安排;免疫建立速率)。计算了农场间基本再生数R0bf,即首发农场感染期间从该初始感染(首发)农场产生的原发性(直接)感染农场数量。开发了一种“预测 - 校正”方法来估计在此期间可能出现的二次感染数量上限。该模型易于编程且执行速度快。针对模型中假设的过程对模拟结果进行了讨论。结果表明,鉴于英国当前的农场实践,早期疾病检测,结合首发地点的立即快速屠宰以及对短程和远程移动的限制,是相对有效的疾病控制策略。然而,农场实践不断变化,在口蹄疫疫情发生时和发生地,任何控制策略都需要根据当前农场实践进行重新评估。一个透明且“简单”的模型有助于这一过程。将未经调整的原始模型与2001年英国口蹄疫疫情的牛数据进行了比较(第6节;图8)。