Minnesota leaders call national COVID-19 model 'overly optimistic'

A University of Washington model has gotten a lot of attention.
Publish date:

Gov. Tim Walz and Minnesota Department of Public Health Commissioner Jan Malcolm were clear Monday that COVID-19 models forecasting what could happen during the pandemic are not going to be perfect.

One model that has received a ton of attention is from the Institute for Health Metrics and Evaluation (IHME) at the University of Washington’s School of Medicine. It's a model that Gov. Walz referred to as "overly optimistic," while Malcolm called it "particularly optimistic." 

As we explained previously, the Washington model only goes out four months, until Aug. 4, which it considers to be the first wave of the novel coronavirus pandemic. The model has been updated five times since it went public March 26, starting with a projected death toll in Minnesota around 2,000. It has since lowered that estimate all the way down to 632. 

It's encouraging to see the numbers drop, but as Malcolm explained, the model has some "particularly optimistic assumptions" that don't match what's actually happening in Minnesota, much less the entire country. 

"The Washington model has some particularly, we would say, optimistic assumptions. First of all on social distancing, that Washington model assumes a level of stay at home social distancing that we just could not as a practical matter expect to see in the United States or Minnesota," said Malcolm. "It also only goes out four months. Our model goes out a year. The data just aren't apples to apples in terms of the totals." 

U of M infectious disease expert Dr. Michael Osterholm criticized the Washington model during a radio appearance on KFAN last week, saying it assumed a "Wuhan-style" lockdown in all 50 U.S. states. Wuhan, China, is where the novel coronavirus is believed to have began, and it led to the Hubei province issuing extreme quarantine guidelines for nearly three months. 

Meanwhile, a model from the Minnesota Department of Health and University of Minnesota is evolving as more Minnesota-specific data is being added daily. 

The first iteration of the Minnesota model estimated 50,000 to 55,000 deaths from COVID-19 over 12 months, according to a report from the Pioneer Press. Those numbers were based on data that Walz says was "too sparse at the time to drive any definitive conclusions." 

It's unclear if the new Minnesota-specific data will decrease that estimate, but either way it's a much more pessimistic forecast than the University of Washington model predicts, even if its a 12-month estimate compared to the four-month prediction from the Washington model. 

"I can't afford to be too short and I can't afford to be too late," said Gov. Walz. "I am still not comfortable that we have all of the ICU beds, all of the ventilators, and all of the personal protective equipment that will come. 

"We want to get the most accurate modeling, but those that want to lean towards overly optimistic modeling could be making the same mistakes in the criticism about leaning towards overly pessimistic models. 

"Yes, an overly pessimistic model will end up with more hospital beds than we might need, but that is a much better scenario than ending up with too few."

Malcolm agreed: "We'd rather plan for a higher level of need. We sure don't want to be underestimating what that need is."

It's expected that the Minnesota model will produce updated simulations in the very near future, at which point it is possible that the governor will share those findings with the public. 

Next Up