Defining high-value information for COVID-19 decision-making

Joshua A Salomon

In this paper, we use a simple epidemiological model and a stylized example of how priorities for collecting new information might be evaluated with respect to their value for decisionmaking in the COVID-19 response. In an environment with considerable uncertainty, it is important to understand the implications that obtaining additional information could have for policy and planning. In the short-term, our results suggest that given baseline knowledge of the number of tests and number of hospitalizations, short-term planning on the scale of weeks would benefit most from understanding the impact of NPIs. Over a longer term, information about the fraction of infections that are not captured in confirmed case reports will be critical to forecasting the extended trajectory of the epidemic, including the timing and level of its peak.