RAPID: Statistical inference of incidence of SARS-CoV-2 in the US using multiple data streams to identify levels of immunity and the impact of non-pharmaceutical interventions

这项研究的目标是整合多个独立的数据源，以估计随着时间的推移美国各地SARS-CoV-2的感染率。以人群为基础的SARS-CoV-2血清学检测对于了解累积发病率和人群免疫水平至关重要。美国疾病控制与预防中心与一些实验室合作，在全国范围内进行了血清学调查，这有助于追溯评估累计感染人数。然而，这些调查的数据可能很难解释，因为不同个体的抗体反应不同，不同的检测方法，以及感染后一段时间的差异。在血清阳性率中观察到的模式与其他数据来源，包括报告的新冠肺炎病例和死亡人数相一致，可以解释美国疾控中心血清阳性率随空间和时间的变化。此外，该项目将估计最近发生免疫事件(感染或疫苗接种)的人口比例，以了解2021-2022年美国奥米克龙变异驱动的浪潮之前的免疫格局。该项目将开发工具来联合分析血清学、病例和死亡数据，并为博士后学者的培训做出贡献。这项研究的主要目标是使用统计和机械模型整合多个独立的数据流，以估计美国各地血清调查中使用的分析中的血清复原率，并估计各州随着时间的推移的血清感染率和累积发病率。该模型将采取多目标方法，采用我们开发的快速推理技术，从病例、住院和死亡数据中提供有关SARS-CoV-2传播的信息。这样的方法已经被应用于州一级的新冠肺炎发病率数据，包括这个小组。该项目是与疾控中心合作资助的，以支持快速反应研究项目，以进一步推进联邦传染病建模能力。该奖项反映了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 goal of this study is to integrate multiple, independent data sources to estimate the rate of SARS-CoV-2 infections across the US over time. Population-based SARS-CoV-2 serological assays are critical for understanding cumulative incidence and population-level immunity. The US CDC, in partnership with a number of laboratories, has conducted nationwide serosurveys which can help retrospectively assess the cumulative number of total infections. However, data from these surveys may be difficult to interpret due to heterogeneity in antibody response across individuals, by assay, and over time since infection. Reconciling patterns observed in seroprevalence with other data sources including reported COVID-19 cases and deaths can explain variation in seroprevalence across space and time in the US CDC. In addition, the project will estimate the proportion of the population with recent immunizing events (infection or vaccination) to understand the immunity landscape prior to the Omicron-variant-driven wave in 2021-2022 in the US. The project will develop tools to jointly analyze serology, caseand death data, and contribute to the training of a post-doctoral scholar.The primary objective in this study is to integrate multiple independent data streams using statistical and mechanistic models to estimate the rate of seroreversion in assays used in serosurveys across the US, and estimate seroprevalence and cumulative incidence over time by state. The model will provide information about SARS-CoV-2 transmission from case, hospitalization and death data by taking a multi-objective approach and adapting fast inference techniques that we have developed. Methods such as these have been applied to state-leveldata on COVID-19 incidence, including by this group. This project was funded in collaboration with the CDC to support rapid-response research projects to further advance federal infectious disease modeling capabilities.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.