Comparative Spillover Dynamics of Avian Influenza in Endemic Countries

禽流感A/H5N1在20世纪90年代末首次被发现。迄今为止，它已导致58个国家的家禽或野鸟暴发，导致至少571例人类病例，335人死亡，经济损失超过100亿美元，主要是由于扑杀家禽造成的。尽管采取了激进的、往往是严厉的控制措施。H5N1在旧世界的许多地区持续存在，并继续威胁全球健康和发展。在这项拟议的工作中，我们将结合使用三种不同的建模方法，收集家禽和野生鸟类的现场监测数据，以及在/VH5N1地方性国家进行接触调查，以调查导致A/H5N1地方性流行的因素。我们将整合动力、空间和网络模型，以发展对H5N1从家禽到人的溢出的预测能力。我们组建了一个由领先的疾病生态学家组成的多学科团队，利用在三个H5N1地方性疾病(孟加拉国、中国和埃及)和一个控制性国家(喀麦隆，那里H5N1不是地方性疾病)方面的多年经验。我们将收集有关野生水禽迁徙、家禽养殖场规模、市场动态和人类与这些感染源的接触网络的数据。我们将在每个国家的10个地点收集样本并检查动物种群中病毒发病率的数据。我们将测试3个假设，以了解为什么H5N1能够在这些国家持续存在：(1)家禽养殖场的模式、规模和分布推动H5N1地方性传播；(2)野鸟、家禽和人之间接触网络的模式、频率和强度影响溢出和延长传播的风险；以及(3)整合跨尺度的动态、网络和空间模型提供了一种更好地预测甲型流感传播风险的策略。这项工作将通过在H5N1流行病学模型中使用对现实接触网络的精细测量，极大地促进我们对H5N1长期动态的理解，使我们能够解释人们对H5N1在流行国家保持低流行率的能力知之甚少。此外，改进的预测模型将有助于在这些关键、高风险国家制定更有效的控制措施。

Avian influenza A/H5N1 was first identified in the late 1990s. To date, it has caused outbreaks in poultry or wild birds in 58 countries resulting In at least 571 human cases, and 335 human deaths, and more than $10 billion in economic losses, largely due to culling of poultry. Despite radical, often draconian, control efforts. H5N1 persists across many parts of the Old World and continues to threaten global health and development. In this proposed work we will use a combination of three different modeling approaches, collection of field surveillance data on domestic poultry and wild birds and a contact survey in /VH5N1 endemic countries to investigate what factors lead to the endemic persistence of A/H5N1. We will integrate dynamical, spatial and network models to develop a predictive capacity for the spillover of H5N1 from poultry to people. We have assembled a multidisciplinary team of leading disease ecologists, leveraging many years of experience in three H5N1 endemic (Bangladesh, China, and Egypt) and one control country (Cameroon, where H5N1 is not endemic). We will collect data on wild waterfowl migration, poultry farm size, market dynamics and human contact networks with these sources of infection. We will collect samples and examine data on viral incidence within animal populations at 10 sites in each country. We will test 3 hypotheses to understand why H5N1 is able to persist endemically in these countries: (1) The pattern, size and distribution of poultry farms drives H5N1 endemism; (2) The pattern, frequency and intensity of contact networks among wild birds, domestic birds and people influences the risk of spillover and prolongs transmission; and (3) Integrating dynamical, network and spatial models across scales provides a strategy to better predict influenza-A transmission risk. This work will significantly advance our understanding of the long-term dynamics of H5N1 by using fine-scale measurements of realistic contact networks in epidemiological models of H5N1, allowing us to explain the poorly understood capacity of H5N1 to persist at low prevalence in endemic countries. Furthermore, the improved predictive models will aid in developing more effective control measures in these critical, high risk countries.