Friday, June 19, 2020

CDC Updated COVID-19 Forecasts: Cumulative Deaths & Hospitalizations




#15,332

The CDC has updated its weekly (4 week) forecast for COVID-19 deaths and hospitalizations, and while the models vary widely, they provide some inkling of what we should expect from the pandemic in the short term.
As British Statistician George E. P. Box (18 October 1919 – 28 March 2013) - Professor Emeritus of Statistics at the University of Wisconsin - famously said : “All models are wrong, but some models are useful.”
And so, while best taken with a large grain of salt, we have the latest model runs.  First up, forecast deaths through July 11th.

Forecasts of Total Deaths
Updated June 18, 2020

Interpretation of Cumulative Death Forecasts
  • This week CDC received 21 individual national forecasts.
  • This week’s national ensemble forecast suggests that there will likely be between 129,000 and 145,000 total reported COVID-19 deaths by July 11th.
  • The state-level ensemble forecasts suggest that the number of new deaths over the next four weeks in Alaska, Arizona, Arkansas, Florida, Hawaii, North Carolina, Oregon, South Carolina, and Utah will likely exceed the number reported over the last four weeks. For other states, the number of new deaths is expected to be similar or decrease slightly compared to the previous four weeks.

National Forecast

The figure shows cumulative reported COVID-19 deaths and forecasted deaths for the next four weeks in the United States.
Models make various assumptions about the levels of social distancing and other interventions, which may not reflect recent changes in behavior. See model descriptions below for details.
State Forecasts
State-level forecasts figures show observed and forecasted state-level cumulative COVID-19 deaths in the US. Each state forecast uses a different scale, due to differences in the numbers of COVID-19 deaths occurring in each state.
Forecasts fall into one of two categories
  • The Auquan, CAN, ERDC, Geneva, GT-DeepCOVID, Imperial, ISU, LANL, LSHTM, MIT, MOBS, Navigator, SWC, UA, UCLA, UMass-MB, and UT forecasts assume that existing control measures will remain in place during the prediction period.
  • The Columbia, Covid19Sim, GT_CHHS, JHU, NotreDame, PSI, and YYG forecasts make different assumptions about how levels of social distancing will change in the future.
Download state forecasts pdf icon[12 pages]
Download forecast data excel icon[1 sheet]
(Continue . . . )

Unlike the cumulative deaths forecasts which predict a range of 16,000 deaths between the high and low estimates, the hospitalization forecasts have as much as a 10,000 cases per day spread by the middle of July. 
  • Deaths, at least in the short term, are easier to predict since we know what the current hospital census of COVID-19 cases is, and have a pretty good idea of what percentage are unlikely to survive. 
  • Hospitalizations, on the other hand, can increase quickly over a matter of days based on the social distancing practices used by the public. 
  • It should be noted that there are fewer models used in the hospitalization ensemble, and this is only the third week it has been published. 
I've only posted some excerpts, so follow the link to read both forecasts in their entirety.

 
Updated June 17, 2020
Interpretation of Forecasts of New Hospitalizations
  • Numbers and trends in new COVID-19 hospitalizations per day vary among national forecasts. This week, national forecasts from two models suggest an increase in the number of daily hospitalizations over the next four weeks, while the other four models forecast declines. Near the end of the forecast period, there may be 1,000 to 11,000 new COVID-19 hospitalizations per day.
  • Forecasts estimate the number of new hospitalizations from a variety of data sets of COVID-19 cases or deaths. The use of different data sets, with different limitations—along with the use of different assumptions about social distancing—may result in the high variation among forecasts. 
National Forecasts 


  • The six 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. See models below for details. 
State Forecasts
Eight state-level models predicting the number of new hospitalizations were submitted this week. These forecasts show the predicted number of new COVID-19 hospitalizations per day for the next four weeks in each state. Each state forecast uses a different scale, due to differences in the number of new COVID-19 cases occurring per day in each state.
Download state forecast 
Download forecast data 
(Continue . . . . ) 
Posted by at