RAISE: IHBEM: Understanding and Predicting Behavioral Responses to Epidemic Risks and Control Policies: Implications for Epidemiological Models and Policy Design

2019冠状病毒病全球大流行表明，公共卫生措施的成功至关重要地取决于了解人类对感染风险和政策建议的行为反应。核心数学流行病学模型提供了关于大流行风险的有用见解，但通常不能解释人们对疾病风险的各种反应，以及这些反应影响持续传播的方式。行为影响疾病传播：自然影响人，人影响自然——将自然系统和人类系统联系在一起，形成决定疫情进程的反馈循环。多学科研究小组将在学科、工具包和数据之间架起桥梁，推进有关人类行为和传染病结果之间这些反馈的知识。通过纳入对风险和社会规范（例如，接受口罩/疫苗）的行为反应的异质性，pi将扩展数学流行病学模型，涉及来自总体公共卫生结果和个人不同反应的反馈。这些模型将以来自三个国家的数据为依据，这些国家具有代表健康风险、公共政策、信息来源和政府信任的独特特征：美国、挪威和瑞典。通过调查和行为实验室实验，在受控环境中收集的数据将补充真实世界的数据，以更深入地了解影响人们对流行病风险反应的个人因素和社会过程。该项目将提高流行病学模型预测疾病结果和公共卫生条例或准则的经济影响的能力，提高公共决策者在未来疫情爆发期间设计和评估流行病控制措施的能力。该项目将开发和估计行为反应函数，这些函数可包含在构成大多数传染病预测模型的常微分方程系统中，目的是在新型流行病期间更好地为政策设计提供信息。PIs将重点关注两个总体研究问题：Q1人们的行为反应如何通过对病原体传播的反馈影响传染病爆发的演变？在疫情爆发期间，人们的行为如何受到作为新型流行病特征的明显不确定性和频繁的政策变化的调节？为了解决这些问题，pi将：(i)开发一个新的耦合ode的流行病学-行为系统，（ii）使用观察、调查和实验数据和方法来估计人们对流行病风险和自上而下控制政策的反应，（iii）将我们的实证研究结果整合到我们的新epi模型中，以及（iv）使用他们的参数化epi模型进行回顾性和前瞻性政策模拟和比较。此外，pi将使用来自三个发达国家的观测数据，这三个国家对正在发生的COVID-19大流行采取了不同的政策方法：美国、挪威和瑞典。此外，PIS将设计调查和行为实验室实验，以更深入地了解在类似于新型流行病的情况下对风险的反应。这种多方法方法将提供机会来检验关于观测数据关联机制的假设，并检验调查和实验室研究的外部有效性。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。

The global COVID-19 pandemic has revealed how critically the success of public health measures depends on understanding human behavioral responses to both infection risks and policy recommendations. Core mathematical epidemiological models have provided useful insight about pandemic risks, but typically do not account for the wide variety of people’s responses to the risk from the disease and the ways these responses shape ongoing transmission. Behavior affects disease spread: nature affects people, and people affect nature—connecting the natural and human systems in feedback loops that determine the course of outbreaks. The multi-disciplinary research team will bridge disciplines, toolkits, and data to advance knowledge about these feedbacks between human behaviors and infectious disease outcomes. The PIs will extend mathematical epidemiological models by including heterogeneity in behavioral responses to risks and social norms (e.g., mask/vaccines acceptance) involving feedback from both aggregate public health outcomes and the diverse responses from individuals. The models will be informed by data from three countries with unique characteristics representing health risks, public policies, information sources, and government trust: the United States, Norway, and Sweden. Real-world data will be supplemented by data collected in controlled settings—through surveys and behavioral laboratory experiments—to gain a deeper understanding of the individual factors and social processes that shape people’s responses to epidemic risks. The project will improve the abilities of epidemiological models to predict both disease outcomes and the economic impacts of public health regulations or guidelines, enhancing the capacity of public policy-makers to design and evaluate epidemic control measures during future outbreaks.This project will develop and estimate behavioral reaction functions that can be included in systems of ordinary differential equations (ODEs) that comprise most epi-models, with the goal of better informing policy design during novel epidemics. The PIs will focus on two overarching research questions: Q1 How do people’s behavioral reactions influence the evolution of an infectious disease outbreak through feedbacks on pathogen spread? and Q2 How are people’s behaviors during an outbreak moderated by the pronounced uncertainties and frequent policy changes that are characteristic of novel epidemics? To address these questions, the PIs will: (i) develop a new epidemiological-behavioral system of coupled ODEs, (ii) use observational, survey, and experimental data and methods to estimate people’s reactions to epidemic risks and top-down control policies, (iii) integrate our empirical findings into our new epi-model, and (iv) use their parameterized epi-model to conduct retrospective and prospective policy simulations and comparisons. In addition, the PIs will use observational data from three developed countries that undertook distinct policy approaches to the on-going COVID-19 pandemic: the United States, Norway, and Sweden. Moreover, the PIS will design surveys and behavioral laboratory experiments to gain a deeper understanding of responses to risk in contexts that are similar to novel epidemics. This multi-method approach will provide opportunities to test hypotheses about the mechanisms that underlie associations in the observational data and examine the external validity of the survey and laboratory studies.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.