Since COVID-19 spread globally in early 2020 and was declared a pandemic by the World Health Organization (WHO) in March, many countries are managing the local epidemics effectively through intervention measures that limit transmission. The challenges of immigration of new infections into regions and asymptomatic infections remain. Standard deterministic compartmental models are inappropriate for sub- or peri-critical epidemics (reproductive number close to or less than one), so individual-based models are often used by simulating transmission from an infected person to others. However, to be realistic, these models require a large number of parameters, each with its own set of uncertainties and lack of analytic tractability. Here, we apply stochastic age-structured Leslie theory with a long history in ecological research to provide some new insights to epidemic dynamics fuelled by external imports. We model the dynamics of an epidemic when R₀ is below one, representing COVID-19 transmission following the successful application of intervention measures, and the transmission dynamics expected when infections migrate into a region. The model framework allows more rapid prediction of the shape and size of an epidemic to improve scaling of the response. During an epidemic when the numbers of infected individuals are rapidly changing, this will help clarify the situation of the pandemic and guide faster and more effective intervention.

自 2020 年初 COVID-19 在全球蔓延并于 3 月被世界卫生组织 (WHO) 宣布为大流行以来，许多国家正在通过限制传播的干预措施有效地管理当地流行病。新感染者进入地区和无症状感染者的挑战依然存在。标准的确定性区室模型不适用于次临界或准临界流行病（生殖数量接近或小于 1），因此基于个体的模型通常用于模拟从感染者到他人的传播。然而，为了现实，这些模型需要大量参数，每个参数都有自己的一组不确定性和缺乏分析易处理性。这里，我们应用在生态研究中有着悠久历史的随机年龄结构 Leslie 理论，为外部输入推动的流行病动态提供一些新的见解。我们对流行病的动态进行建模，当R ₀低于 1，代表成功应用干预措施后的 COVID-19 传播，以及感染迁移到一个地区时预期的传播动态。该模型框架允许更快速地预测流行病的形状和大小，以改善应对措施的规模。在感染人数快速变化的流行期间，这将有助于澄清大流行的情况，并指导更快、更有效的干预。