IHBEM: Using socioeconomic, behavioral and environmental data to understand disease dynamics: exploring COVID-19 outcomes in Oklahoma

现代最关键的挑战之一是更好地了解大规模疾病爆发的地点、原因和方式。研究表明，随着时间的推移，全球大规模疾病暴发的频率正在增加，但对结果的差异仍然知之甚少。这项研究将探讨导致俄克拉荷马州各县和大都市区COVID-19结果变化的因素，特别是哪些地区的病例比根据其总体人口规模预期的多或少。研究人员将研究环境因素，如天气模式和空气质量，以及社会经济因素，如医生数量和愿意接种疫苗的个人比例的差异。调查人员还将对全州的个人进行调查，以试图更好地了解人们为什么会做出这样的医疗选择，以及行为如何导致结果的差异。理解所有这些因素需要一个具有不同专业知识的团队。传统上，大多数疾病动力学的数学和定量模型都是由数学家、生态学家和计算机科学家开发和研究的。然而，了解对卫生保健措施的态度差异及其起源更多的是社会科学家和历史学家的职权范围。通过建立一个跨越所有这些学科的合作者团队，研究团队将能够更全面地了解俄克拉荷马州的COVID-19结果。这反过来又将表明，在未来，哪些行动可能最有效地减轻COVID和其他大规模疾病事件的影响。这项工作的最终成果将包括一个新的数据存储库和一个面向公众的智能流行病学建模平台，该平台由Jupyter Notebooks提供支持。该项目还将提供外联和培训，包括向来自代表性不足群体的学生提供培训。暴发频率的增加似乎与全球化和其他人类活动有关。然而，大多数人类行为、社会和经济因素对疫情风险的影响很少被量化。相关的社会因素很难衡量，通常需要专家来生成和解释数据。然而，具有这方面专业知识的社会科学家很少接受疾病动力学数学建模方面的培训。为了应对这些挑战，研究人员将专注于开发可用于探索俄克拉荷马州COVID-19结果的数据源和数学模型。该项目将是社会科学家和传染病建模专家之间的真正合作。俄克拉何马州还没有得到充分的研究，并且在空间上是异质的，因此俄克拉何马州的疾病动态模型很可能可以推广到美国的许多其他地区。调查人员将制定规范现有行为和社会经济因素数据的程序，并开发新的数据来源。该小组将开发过去疫情的统计模型，以及反映经实证证明驱动COVID-19动态的因素的数学模型。后一项工作将证明如何修改基线类sir模型以反映人类行为因素。研究人员还将对比基于现有社会经济因素数据的模型与纳入有关一级和二级预防的行为和态度变化的新调查数据的模型的表现。生成的代码和数据集将在智能流行病学建模框架中免费提供和搜索，这将使其他研究人员能够轻松地对其进行迭代。该项目由数学和物理科学理事会（MPS）的数学科学部（DMS）和社会、行为和经济科学理事会（SBE）的社会和经济科学部（SES）共同资助。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

One of the most critical modern challenges is to better understand the where, why and how oflarge disease outbreak occurrence. Research shows that the frequency of large disease outbreaks is increasing over time globally, and yet differences in outcomes remain poorly understood. This research will explore the factors that drove variation in COVID-19 outcomes across the counties and metropolitan areas of Oklahoma, particularly which areas had more or fewer cases than would be expected based on their overall population size. The investigators will look at both environmental factors, such as weather patterns and air quality, and socioeconomic factors such as numbers of doctors and differences in the proportion of individuals that were willing to be vaccinated. The investigators will also conduct surveys of individual across the state to try and better understand why people made the healthcare choices that they did and how behavior drove differences in outcomes. Understanding all of these factors requires a team with diverse expertise. Traditionally, most mathematical and quantitative models for disease dynamics have been developed and studied by mathematicians, ecologists, and computer scientists. However, understanding differences in attitudes towards health care measures and how they originate is more the purview of social scientists and historians. By building a team of collaborators spanning all of these disciplines, the research team will be able to build a more complete picture of COVID-19 outcomes in Oklahoma. This will in turn suggest what actions may be most effective to try and best mitigate the effects of both COVID and other large-scale disease events in the future. The final product of this work will include a new data repository and a public-facing intelligent epidemiological modeling platform powered by Jupyter Notebooks. The project will also provide outreach and training, including to students from underrepresented groups.Increases in outbreak frequency seem to be related to globalization and other human activities. Yet the effects of most human behavioral, social and economic factors on outbreak risk are rarely quantified. Relevant social factors can be hard to measure, often needing specialists to generate and interpret data. However social scientists with expertise to do so are rarely trained in mathematical modelling of disease dynamics. To address these challenges, the investigators will focus on developing data sources and mathematical models that can be used to explore COVID-19 outcomes in Oklahoma. The project will be a true collaboration between social scientists and experts in modelling infectious diseases. Oklahoma is understudied, and is spatially heterogeneous such that models of disease dynamics in Oklahoma are likely to be generalizable to many other regions of the US. The Investigators will generate protocols for standardizing existing data on behavioral and socioeconomic factors as well as develop new data sources. The team will develop statistical models of past outbreaks, and mathematical models reflecting factors shown to have driven COVID-19 dynamics empirically. The latter work will demonstrate how baseline SIR-like models can be modified to reflect human behavioral factors. The Investigators will also contrast the performance of models based on existing data on socioeconomic factors with models incorporating new survey data on variation in behaviors and attitudes related to primary and secondary prevention. The code and datasets to be generated will be made freely available and searchable in an intelligent epidemiological modeling framework, which will enable other researchers to easily iterate on them.This project is jointly funded by the Division of Mathematical Sciences (DMS) in the Directorate of Mathematical and Physical Sciences (MPS) and the Division of Social and Economic Sciences (SES) in the Directorate of Social, Behavioral and Economic Sciences (SBE).This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.