Modelling the spread of infections in a first-world child care facility: coding, analysis and policy implications

加拿大赛诺菲巴斯德是赛诺菲的疫苗部门，是世界领先的医疗保健公司之一。 以及该国最大的疫苗公司，对公共卫生有着坚定的承诺。它们提供了重要的 加拿大用于预防和治疗传染病、常见病和癌症的疫苗。 加拿大赛诺菲巴斯德对世界上第一个国家的日托中心的数学模型感兴趣 作为加拿大)，以便将其用作感染预防行为和预防性治疗的情景测试。 他们对一种模型特别感兴趣，在这种模型中，人口中的个人是孩子、工作人员和父母，并且 其中他们的交互被明确地建模：人与人和人与环境，在典型的 暴发持续时间约为4-5个月。他们希望进一步使用模型的输出来 降低日托中心传染性所产生的社会经济影响分析。 为了响应公司的兴趣，我们建议开发一种基于小人口代理的模型 一种日托设施，目的是测试和告知预防病毒感染的政策(肠道、 流感等)这可以在儿童保育环境中实施。这项研究不仅在应用上有帮助 日托环境；它代表了感染基本数学模型的一种新方法 在小而半封闭的人群中传播。这种类型的数学模型与类似的模型有很大的不同 区间、博弈论或基于代理人的流行病学模型，因为它指的是非常小的人口，与 平均接触模式，并有可能在几分钟的时间尺度分辨率上进化，而不是 一次几小时或几天。

Sanofi Pasteur in Canada, the vaccines division of Sanofi, is one of the world's leading healthcare companies, and the country's largest vaccine company with a strong commitment to public health. They provide important vaccines to Canadians for the prevention and treatment of infectious diseases, common illnesses and cancers. Canada Sanofi Pasteur is interested in a mathematical model of a day care center in a 1st world country (such as Canada), in order to use it as scenario testing for infection prevention behaviour and prophylactic treatments. They are specifically interested in a model where individuals in the population are kids, staff and parents, and where their interactions are modeled explicitly: person-to-person and person-to-environment, during typical outbreak episodes roughly of 4-5 months duration. They want to further use the model's output for socio-economic impact analyses arising from lowering infectiousness in day care centers. In response to the company's interest, we propose the development of a small-population agent-based model of a day care facility for the purpose of testing and informing prevention policies against viral infections (enteric, influenza, etc.) that can be implemented in a child care setting. The research is not only useful in applications to day care environments; it represents a novel approach in basic mathematical modelling of infection transmission in small, semi-closed populations. This type of mathematical model differs greatly from similar compartmental, game theoretic or agent-based epidemiological models, since it refers to a very small population, with different than average contact patterns, and with the possibility of being evolved on a time scale resolution of minutes, versus hours or days at a time.