Effects of Infection-Induced Migration Delays on the Epidemiology of Avian Influenza in Wild Mallard Populations

Wild waterfowl populations form a natural reservoir of Avian Influenza (AI) virus, and fears exist that these birds may contribute to an AI pandemic by spreading the virus along their migratory flyways. Observational studies suggest that individuals infected with AI virus may delay departure from migratory staging sites. Here, we explore the epidemiological dynamics of avian influenza virus in a migrating mallard (Anas platyrhynchos) population with a specific view to understanding the role of infection-induced migration delays on the spread of virus strains of differing transmissibility. We develop a host-pathogen model that combines the transmission dynamics of influenza with the migration, reproduction and mortality of the host bird species. Our modeling predicts that delayed migration of individuals influences both the timing and size of outbreaks of AI virus. We find that (1) delayed migration leads to a lower total number of cases of infection each year than in the absence of migration delay, (2) when the transmission rate of a strain is high, the outbreak starts at the staging sites at which birds arrive in the early part of the fall migration, (3) when the transmission rate is low, infection predominantly occurs later in the season, which is further delayed when there is a migration delay. As such, the rise of more virulent AI strains in waterfowl could lead to a higher prevalence of infection later in the year, which could change the exposure risk for farmed poultry. A sensitivity analysis shows the importance of generation time and loss of immunity for the effect of migration delays. Thus, we demonstrate, in contrast to many current transmission risk models solely using empirical information on bird movements to assess the potential for transmission, that a consideration of infection-induced delays is critical to understanding the dynamics of AI infection along the entire flyway.

野生水禽种群是禽流感（Avian Influenza, AI）病毒的天然储存宿主，学界普遍担忧这些鸟类会沿其迁徙路线散播病毒，进而引发禽流感大流行。观测研究显示，感染禽流感病毒的个体可能会推迟从迁徙停留地出发的时间。本研究聚焦于绿头鸭（Anas platyrhynchos）迁徙种群，探究禽流感病毒的流行病学动态，旨在明确感染诱导的迁徙延迟对不同传播能力毒株传播的影响机制。我们构建了宿主-病原体模型，将流感病毒的传播动力学与宿主鸟类的迁徙、繁殖及死亡过程相结合。模型预测显示，个体迁徙延迟会同时影响禽流感病毒暴发的时间与规模。研究发现：（1）相较于无迁徙延迟的场景，迁徙延迟会降低每年的总感染病例数；（2）当毒株传播率较高时，病毒暴发始于鸟类秋季迁徙早期抵达的停留地；（3）当传播率较低时，感染多发生在季末，而迁徙延迟会进一步延后感染的发生时间。据此推测，水禽种群中高致病性禽流感毒株的扩散，可能会导致年末感染率升高，进而改变养殖家禽的暴露风险。敏感性分析结果表明，世代时间与免疫力丧失对迁徙延迟的影响效应至关重要。因此，与当前诸多仅依靠鸟类移动的实证信息评估传播潜力的传播风险模型不同，本研究证实，纳入感染诱导的延迟因素，对解析整条迁徙路线上的禽流感感染动态具有关键意义。