Chris Murray, IHME Director, Univ Washington, Seattle — modeling scenario for COVID-19 outcome

This email information was received today from Professor David Eaton, and we agreed that it be shared on these GEITP pages. Chris Murray [Director of the Institute for Health Metrics and Evaluation (IHME), University of Washington, Seattle] is among the most respected “health data modelers” in the world. On this URL is his computer-modeling projection of COVID-19 resource count, deaths per day, and accumulative deaths in the U.S.:

https://covid19.healthdata.org/projections?fbclid=IwAR3-mat0CYVlXC7sszkqFrqtLVuQKiefgp2zcWwB6ZNBCbhTADsQ2Rz4LCA

The peak is predicted to be reached on April 14th — with hospital beds overloaded ☹ — but the good news is that the country is projected to go back to “near normal” (coinciding with the onset of hot summer weather) between the beginning of June, on into July and August… 😊

COMMENT: Dan, I believe there are several caveats here. First, I believe that Professor Murray’s modeling scenario representis the predicted peak for AVERAGE nationwide deaths. Many individual metropolis areas will peak earlier and later, and this will be a huge problem. Second, no one knows with certainty that COVID-19 will die down in the summer. While it does appear that the virus is spreading more slowly in tropical areas, Singapore, and Australia (where it is now summer) are having significant outbreaks. Third, many experts are still worried about the virus becoming much more virulent during its second and third waves (i.e. the 1918 Spanish flu was FAR worse than the 1917 flu).

Our best hope is effective treatment (several are promising) until an effective vaccine can somehow be fast-tracked, and then herd immunity will hopefully develop — because it does appear that lots of people have the virus asymptomatically, as you know. Rapid testing for IgG to COVID-19 will hopefully be available in the next few weeks; this will give us a better idea about how much herd immunity we have developed. But I am not sure if rapid testing will be cost-effective as a mass screening tool. As you know, the new 5-minute Abbott test is for presence of COVID-19 RNA (i.e. you can be negative today, and positive tomorrow).

COMMENT: I do not disagree with any of this. However, Chris Murray does include upper and lower confidence-intervals on all his projections; thus, it is a bit misleading to say he is only predicting the average. And I also expect that Chris would agree with the large uncertainties — regarding ‘rebound’ of the virus in the fall, and whether UV light/higher summer temperatures will moderate COVID-19 as much as other viruses are affected.

That said, I believe the modeling of hospital beds, ICU beds, and respirators — by State — is a very useful tool for States to help manage this crisis. However, as with any model, a model is only as good as the input data (i.e. ‘garbage in, garbage out’). 😉

DEaton, PhD; professor emeritus, University of Washington, Seattle

This entry was posted in Center for Environmental Genetics. Bookmark the permalink.