Evaluation of individual and ensemble probabilistic forecasts of COVID-19 mortality in the United States.

Short-term probabilistic forecasts of the trajectory of the COVID-19 pandemic in the United States have served as a visible and important communication channel between the scientific modeling community and both the general public and decision-makers. Forecasting models provide specific, quantitative, and evaluable predictions that inform short-term decisions such as healthcare staffing needs, school closures, and allocation of medical supplies. Starting in April 2020, the US COVID-19 Forecast Hub (https://covid19forecasthub.org/) collected, disseminated, and synthesized tens of millions of specific predictions from more than 90 different academic, industry, and independent research groups. A multimodel ensemble forecast that combined predictions from dozens of groups every week provided the most consistently accurate probabilistic forecasts of incident deaths due to COVID-19 at the state and national level from April 2020 through October 2021. The performance of 27 individual models that submitted complete forecasts of COVID-19 deaths consistently throughout this year showed high variability in forecast skill across time, geospatial units, and forecast horizons. Two-thirds of the models evaluated showed better accuracy than a naïve baseline model. Forecast accuracy degraded as models made predictions further into the future, with probabilistic error at a 20-wk horizon three to five times larger than when predicting at a 1-wk horizon. This project underscores the role that collaboration and active coordination between governmental public-health agencies, academic modeling teams, and industry partners can play in developing modern modeling capabilities to support local, state, and federal response to outbreaks.

The challenges of data usage for the United States' COVID-19 response.

During the coronavirus pandemic, policy makers need to interpret available public health data to make decisions affecting public health. However, the United States' coronavirus response faced data gaps, inadequate and inconsistent definitions of data across different governmental jurisdictions, ambiguous timing in reporting, problems in accessing data, and changing interpretations from scientific institutions. These present numerous problems for the decision makers relying on this information. This paper documents some of the data pitfalls in coronavirus public health data reporting, as identified by the authors in the course of supporting data management for New England's coronavirus response. We provide recommendations for individuals to collect data more effectively during emergency situations such as a COVID-19 surge, as well as recommendations for institutions to provide more meaningful data for various users to access. Through this, we hope to motivate action to avoid data pitfalls during public health responses in the future.

Can Comorbidity Data Explain Cross-State and Cross-National Difference in COVID-19 Death Rates?

Many efforts to predict the impact of COVID-19 on hospitalization, intensive care unit (ICU) utilization, and mortality rely on age and comorbidities. These predictions are foundational to learning, policymaking, and planning for the pandemic, and therefore understanding the relationship between age, comorbidities, and health outcomes is critical to assessing and managing public health risks. From a US government database of 1.4 million patient records collected in May 2020, we extracted the relationships between age and number of comorbidities at the individual level to predict the likelihood of hospitalization, admission to intensive care, and death. We then applied the relationships to each US state and a selection of different countries in order to see whether they predicted observed outcome rates. We found that age and comorbidity data within these geographical regions do not explain much of the international or within-country variation in hospitalization, ICU admission, or death. Identifying alternative explanations for the limited predictive power of comorbidities and age at the population level should be considered for future research.

A systems approach for resources management during the COVID-19 pandemic: Multi-agency perspectives from New England.

The emergence of COVID-19 in the United States has overwhelmed local hospitals, produced shortages in critical protective supplies for medical staff, and created backlogs in burials and cremations. Because systemic disruptions occur most acutely at a local scale, facilitating resource coordination across a broad region can assist local responses to COVID-19 surges. This article describes a structured systems approach for coordinating COVID-19 resource distribution across the six New England states of the United States. The framework combines modeling tools to anticipate resource shortages in medical supplies, personnel needs, and fatality management for individual states. The approach allows decision makers to understand the magnitude of local outbreaks and equitably allocate resources within a region based on the present and future needs. This model contributed to determining material distribution in New England as the 2020 COVID-19 surges unfolded in the spring and fall seasons. Using a systems analysis, the model demonstrates the translation of anticipated COVID-19 cases into resource demands to enable regional coordination of scarce resources.

AOPERA: A proposed methodology and inventory of effective tools to link chemicals to adverse outcome pathways.

New approaches, like the Adverse Outcome Pathway (AOP) framework, have been developed to describe how chemicals cause toxicity by linking in vitro assays to adverse health outcomes. However, approaches, tools and resources for development of AOPs have not been well described. Here we review information resources for AOP development and define a streamlined process for linking a chemical to an existing AOP. We propose a four step process to facilitate AOP development: link the uncharacterized chemical directly to Molecular Initiating Events, Key Events, or Adverse Outcomes; identify analogs with toxicological information for the uncharacterized chemical; link the characterized chemical (initial chemical if characterized, a characterized analog if initial chemical is not) to Molecular Initiating Events, Key Events, or Adverse Outcomes; and identify AOPs that contain the Molecular Initiating Events, Key Events, or Adverse Outcomes that were found in Steps 1 and 3. The process and library of informational resources proposed and tested here served as the foundation for an informational online tool (AOPERA) that helps practitioners identify their current-state knowledge gaps, navigate the four-step process, and connect to relevant resources. AOPERA can be found at https://igbb.github.io/AOPERA_HTML. Additionally, we anticipate that by simplifying and standardizing the process of linking a chemical to a known AOP, we will lower the barrier to entry for this objective and increase its accessibility to new practitioners. In turn, this may increase the demand for new or improved AOPs to which practitioners can link chemicals, thereby contributing to the expansion of the library of known AOPs.

An Analytical Perspective on Pandemic Recovery.

After implementing restrictions to curb the spread of coronavirus, governments in the United States and around the world are trying to identify the path to social and economic recovery. The White House and the Centers for Disease Control and Prevention have published guidelines to assist US states, counties, and territories in planning these efforts. As the impact of the coronavirus pandemic has not been uniform, these central guidelines need to be translated into practice in ways that recognize variation among jurisdictions. We present a core methodology to assist governments in this task, presenting a case for appropriate actions at each stage of recovery based on scientific data and analysis. Specifically, 3 types of data are needed: data on the spread of disease should be analyzed alongside data on the overall health of the population and data on infrastructure-for example, the capacity of health systems. Local circumstances will produce different needs and present different setbacks, and governments may need to reinstate as well as relax restrictions. Transparent, defensible analysis can assist in making these decisions and communicating them to the public. In the absence of a widely administered vaccine, analysis remains one of our most important tools in addressing the coronavirus pandemic.