Death projections issued during the coronavirus pandemic have thus far proven to be controversial and imperfect for modeling the spread amid the ever-changing response and research into COVID-19.
One of the oft-cited models, the Institute for Health Metrics and Evaluation (IHME) at the University of Washington’s School of Medicine, has issued a new series of projections for COVID-19 in Minnesota.
The model is estimating Minnesota will reach 2,399 deaths by Dec. 1, though the range is between 1,945 and 3,418. To reach 2,399, Minnesota will need to average approximately 6.5 deaths per day over the next three-plus months.
That rate is strikingly similar to the four-week projection published late last week by the Centers for Disease Control and Prevention (CDC), which uses 34 model groups, one of them being the IHME, to create ensemble guidance to forecast deaths.
The CDC's model has Minnesota reaching 1,796 total deaths by Sept. 1. Based on that number, Minnesota will need to average approximately 6 deaths per day for the estimate to be accurate.
The subject of modeling has been a thorny one in Minnesota, with the Walz Administration using in its decision-making process about COVID-19 related shutdowns a University of Minnesota model that has been criticized for its huge death projections over the course of the year.
In mid-May, the Minnesota Department of Health (MDH) released the second iteration of its COVID-19 model that estimated at least 22,000 deaths in Minnesota within the first 12 months of the pandemic. Now over five months into the health crisis, there have been 1,660 confirmed deaths in Minnesota and another 41 deaths assumed to have been caused by the virus.
At the current rate, the model's assumptions will be greatly overstated – barring a huge surge in deaths over the winter.
Meanwhile, the IMHE model has proven to have projections a little closer to what has actually happened, albeit the model is updated more regularly and has also been shown to underestimate death and case rates.
On March 26, just 20 days after the first case of COVID-19 was detected in Minnesota, the IHME model projected approximately 2,000 deaths by August 4. Ten days later the model reduced the Aug. 4 death toll estimate to around 630, with a range of 224 to 1,362 deaths.
Gov. Tim Walz, on April 6, referred to the IHME model as "overly optimistic," while MDH Commissioner Jan Malcolm said the model was "particularly optimistic" with its then-latest projection.
Minnesota had 1,620 COVID-19 deaths on Aug. 4, so the model did prove to undercount the significance of the virus' impact during that stretch of time.
The newest model update from the IHME, however, is more in line with what Malcolm would expect.
“While we would prefer to see no deaths from COVID-19 over the next few months, we know that is not a realistic expectation, considering the level of disease spread in the state currently and our expectation that disease activity could increase this fall," Malcolm said Monday.
"The IHME model’s prediction of 740 deaths over the next 100-plus days does not seem out of line with recent trends in deaths: it’s an average of approximately 7 deaths per day, a level we have certainly seen before. But we as Minnesotans do have it in our power to lower that number. We should see this as a call to action to do those things that can help prevent the spread of COVID-19 and thus reduce the number of deaths in our communities.”
Signs of Minnesota's death rate beginning to increase?
On July 15, Minnesota State Epidemiologist Dr. Ruth Lynfield said that due to increasing COVID-19 hospitalizations in Minnesota she feared "an increase in deaths in the coming weeks."
Have the deaths begun to increase as Dr. Lynfield feared?
From July 1 to July 15, Minnesota reported 77 deaths, or an average of just over 5 per day. From July 16 to July 31, Minnesota reported 82 deaths, or an average of just over 5 per day. Since the beginning of August, Minnesota has reported 60 deaths, or an average of 6 per day.
The six deaths per day matches the short-term projection from the CDC and the longer-term estimate from the IHME model.
But what's notable is that Lynfield made her remarks when there were 236 COVID-19 patients hospitalized, including 106 in the ICU. Nearly one month later there are 320 patients hospitalized and 159 in the ICU, which begs the question: Is the rise in hospitalizations going to be followed by a rise in deaths?
According to the IHME model, "there is generally a 17-21 day lag between infection and deaths." Considering the bulk of the ICU increase has happened in the past two weeks, Lynfield's fears may be on the verge of being realized in the coming days and weeks.
Then the question will become how effective the statewide mask mandate is. The mandate went into effect July 25. State officials said it would likely take three weeks to see how effective the mandate is, meaning next week will start producing results, possibly in the form of reduced infections and ultimately, hospitalizations and deaths.
Two weeks ago, Malcolm said Minnesota was in a "vulnerable" stage of the pandemic, adding that it would be critical for residents to follow the mask mandate and other preventative measures to avoid racing up the charts like other states have in recent weeks.
"We certainly don't want Minnesota to experience the same catastrophic results of that very, very fast growth that we have seen in Florida, Texas and other states," she said. "We as citizens have a great deal of influence over whether that happens in Minnesota or not. We have it in our power to prevent that really rapid and uncontrolled spread by doing the right thing for each other."