RAPID: COVID-19 Scenario Modeling Hub to harness multiple models for long-term projections and decision support

正在进行的新冠肺炎疫情伴随着政策制定者做出许多艰难的管理决策。该项目支持新冠肺炎情景建模中心，该中心将来自不同科学背景的多个建模小组聚集在一起，帮助制定有关控制措施对SARS-CoV-2感染、住院和死亡的长期潜在影响的决策。通过考虑在对即将到来的大流行进程的不同假设下对这些结果的预测，研究人员将通过向政府官员和公众提供及时的信息，为美国的应对工作提供信息，从而帮助制定决策。在大流行持续的情况不确定的情况下，需要考虑多种可能的情景并涉及多个建模团队的投入尤其重要。这包括对病原体传播性和疾病严重性的影响，这些影响可能伴随着新的变异。病原体传播的环境也有很大的不同，往往是不可预测的，这取决于人类的行为和干预，如社交距离和疫苗接种。这项研究产生的疫情预测将有助于决策在快速变化的情况下如何管理新冠肺炎干预。决策分析、专家诱导和模型聚合的方法将被用于收集来自多个群体的模型预测，然后将这些结果合成一个统一的集成预测。这种综合将特别有用，因为需要及时作出管理决定，以减少对公共卫生的破坏性影响，同时考虑到不确定性和有限的资源。最新数据将直接提供给利益相关者，如美国疾病控制中心和白宫新冠肺炎数据团队。还将与其他有关各方(例如，世界卫生组织)分享进展情况。将通过网络界面向公众提供个别模型投影和整体投影的可视化，并通过开放获取出版物提供科学见解。这一框架的发展也将有利于未来在建模小组之间快速建立合作，以帮助在面对其他新兴和地方性病原体时为限制公共卫生和经济负担的决策提供信息。这一奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

The ongoing COVID-19 pandemic has been accompanied by many difficult management decisions for policymakers. This project supports the COVID-19 Scenario Modeling Hub, which brings together multiple modeling groups from different scientific backgrounds to help inform decisions about the long-term potential impact of control measures on SARS-CoV-2 infections, hospitalizations, and deaths. By considering projections of these outcomes under different assumptions about the upcoming course of the pandemic, researchers will help inform decisions by providing timely information to government officials and the public to inform response efforts in the United States. The need to consider multiple potential scenarios and involve input from multiple modeling teams is particularly important when the conditions under which the pandemic will continue are uncertain. This includes effects on pathogen transmissibility and disease severity that may accompany novel variants. The environment in which the pathogen spreads also varies greatly in often unpredictable ways, depending on human behavior and interventions, such as social distancing and vaccine administration. Epidemic projections generated from this research will help inform decisions about how to manage COVID-19 interventions under rapidly changing circumstances.Approaches from decision analysis, expert elicitation, and model aggregation will be used to collect model projections from multiple groups and then synthesize these results into a unified ensemble projection. This synthesis will be particularly useful as timely management decisions need to be made in order to reduce devastating effects on public health while also accounting for uncertainty and limited resources. Updates will be provided directly to stakeholders, such as the United States Centers for Disease Control and the White House COVID-19 Data Team. Progress will also be shared with other interested parties (e.g., the World Health Organization). Visualizations of the individual model projections and the ensemble projection will be made accessible to the public via a web interface and scientific insights will be made accessible through open access publishing. The development of this framework will also benefit future endeavors to quickly establish collaborations across modeling groups to help inform decisions to limit public health and economic burden in the face of other emerging and endemic pathogens.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.