A new forecast from Oregon Health & Science University predicts that the state could see its COVID-19 hospitalizations spike to more than 3,000 by the end of February, more than double the peak level during the delta surge in September. That forecast assumes no additional interventions, but even the 1,200 hospitalizations experienced in September pushed hospitals beyond their capacity. State officials hope the additional interventions and vaccine efforts can keep the hospitalization number at approximately the same level of the delta peak.
What actually happened?
Total hospitalized patients peaked just below 1,200, not 3,200, not 2,000, not 1,600 as they kept revising their forecast up, down, then up again.
OHSU’s Peter Graven forecast on Feb 3rd, that the state would not drop below 400 patients until the end of March. Hospitalization numbers fell under 400 on March 4th, and to 108 by March 31st. Thus, even a short-term forecast was wrong.
The only thing we learned from disease models the past two years is that they are useless, no better than eyeballing chart trends with the Eye-Ball-o-Matic Stat-o-Meter, and generally worse than that. These models were used to set public policies that caused actual harm.
There are reasons I no longer believe a single word coming from epidemiologists.
