COVID-19 Forecasts: Hospitalizations
Interpretation of Forecasts of New Hospitalizations
- This week CDC received forecasts of daily, new reported COVID-19 hospitalizations over the next 4 weeks from ten modeling groups.
- Four national forecasts predict a likely increase in the number of new hospitalizations per day over the next four weeks, one forecast predicts a likely decrease, and three forecasts are uncertain about the trend or predict stable numbers. For November 16, the forecasts estimate 2,600 to 6,200 new COVID-19 hospitalizations per day.
- State-level forecasts also show a high degree of variability, which results from multiple factors. Hospitalization forecasts use different sources of data for COVID-19 cases or deaths, with different limitations, and make different assumptions about social distancing.
- The national forecasts show the predicted number of new COVID-19 hospitalizations per day for the next four weeks in the United States.
- The forecasts make different assumptions about hospitalization rates and levels of social distancing and other interventions and use different methods to estimate the number of new hospitalizations.
State-level forecasts show the predicted number of new COVID-19 hospitalizations per day for the next four weeks by state. Each state uses a different scale, due to differences in the number of new COVID-19 hospitalizations per day in each state.
Additional forecast data and information on forecast submission are available at the COVID-19 Forecasting Hubexternal icon.
These forecasts make different assumptions about social distancing measures and use different methods and data sets to estimate the number of new hospitalizations. Information about individual models is available here: https://github.com/cdcepi/COVID-19-Forecasts/blob/master/COVID-19_Forecast_Model_Descriptions.mdexternal icon.
Social distancing is incorporated into the forecasts in two different ways:
- These modeling groups make assumptions about how levels of social distancing will change in the future:
- Columbia Universityexternal icon (Model: Columbia)
- Covid-19 Simulator Consortiumexternal icon (Model: Covid19Sim)
- Institute of Health Metrics and Evaluationexternal icon (Model: IHME)
- Johns Hopkins University, Infectious Disease Dynamics Labexternal icon (Model: JHU-IDD)
- These modeling groups assume that existing social distancing measures in each jurisdiction will continue through the projected four-week time period:
- Georgia Institute of Technology, College of Computing,external icon (Model: GT-DeepCOVID)
- Google and Harvard School of Public Healthexternal icon (Model: Google-HSPH)
- Johns Hopkins University, Applied Physics Labexternal icon (Model: JHU-APL)
- Karlen Working Groupexternal icon (Model: Karlen)
- Los Alamos National Laboratoryexternal icon (Model: LANL)
- University of California, Los Angelesexternal icon (Model: UCLA)
The rate of new hospitalizations is estimated using one of four approaches:
- These modeling groups assume that a certain fraction of infected people will be hospitalized:
- Columbia Universityexternal icon
- Covid-19 Simulator Consortiumexternal icon
- Google and Harvard School of Public Healthexternal icon
- Johns Hopkins University, Applied Physics Labexternal icon
- Johns Hopkins University, Infectious Disease Dynamics Labexternal icon
- Los Alamos National Laboratoryexternal icon
- University of California, Los Angelesexternal icon
- The Georgia Institute of Technology, College of Computing,external icon uses COVID-19 hospitalization data reported by some jurisdictions to forecast future hospitalizations.
- The Institute of Health Metrics and Evaluationexternal icon estimates numbers of new hospitalizations based on numbers of forecasted deaths.
- The Karlen Working Groupexternal icon uses the rate of reported infections to estimate the number of new hospitalizations in a given jurisdiction, unless the rates of reported infections and hospitalizations differ. In that case, the rate of reported hospitalizations is used to forecast new hospitalizations.
1 The full range of the prediction intervals is not visible for all state plots. Please see the forecast data for the full range of state-specific prediction intervals.