Minnesota officials unveil modeling behind state's COVID-19 strategy

State health officials on Friday unveiled the thinking and data modeling behind their extended stay-at-home strategy to curb the COVID-19 pandemic, which has caused 64 deaths as of Saturday and 1,427 lab-confirmed infections in Minnesota.

With updated understanding of the coronavirus causing this pandemic, Minnesota researchers found two scenarios that would be achievable and cause the greatest reductions in deaths and pressure on hospitals. One modeled the impact of a plan that included an extended statewide order until May 8, while the other assessed restrictions only on people 60 and older — and others at greatest risk of severe cases — until July 10.

Both would halve the predicted number of deaths in Minnesota over the next year when compared to doing nothing at all to combat the virus, the modeling results showed. Gov. Tim Walz opted for the statewide extension — albeit until May 4 rather than the modeling date of May 8 — because it would push the peak of the outbreak to mid-July.

Both models assumed Minnesota would have 2,200 intensive care beds available, but the delay would give hospitals the chance to add more beds as well as ventilators, which could improve outcomes, said Stefan Gildemeister, state health economist.

"As that capacity goes up, mortality could shrink," he said.

Modeling by the University of Minnesota and the Minnesota Department of Health has generated immense interest in a public that is anxious about the COVID-19 pandemic but also about the economic impact of the shutdown that has caused more than 400,000 people to apply for state unemployment benefits.

Some have raised concerns about modeling in Minnesota that has predicted thousands of deaths over 365 days, when a widely cited model out of the University of Washington predicted 442 deaths in Minnesota and 60,000 in the U.S. — albeit only over 120 days.

Health officials stressed that the Minnesota modeling is designed to identify pandemic response strategies that work best — not to forecast death tolls. For transparency, though, they released the estimated death figures — which would be around 50,000 if the state had done nothing, and 22,000 (with a range of 9,000 to 36,000) under the model that most closely resembles the governor's order.

"They are not about specific-point-in-time estimates," state health Commissioner Jan Malcolm said of the models. "They are about directional changes."

The latest death numbers were lower than the predictions that fueled Walz's initial decision on a two-week, stay-at-home order. Gildemeister said the assumptions in the model have changed substantially due to the latest research on the coronavirus.

Length-of-stay in intensive care was originally set at 23 days, but was reduced in the latest modeling to 13 days based on U.S. experience in treating COVID-19. That alone means more estimated bed capacity for COVID-19 patients.

The researchers ran dozens of models under different scenarios, though Gildemeister said they focused on solutions that were likely in a democratic U.S. compared to stern measures in authoritarian countries such as China. The solutions that appeared most effective included the expansion of intensive care beds and restrictions on people at greatest risk.

Minnesota has 2,770 intensive care beds — including 525 that could be converted in 72 hours, according to the state COVID-19 dashboard. However, 844 ICU beds are currently in use, so the count of available beds is lower at the moment than the 2,200 included in the modeling.

Minnesota's COVID-19 case count on Friday included outbreaks involving at least one staff member or resident of 55 congregate care facilities. Five of the seven newly reported deaths on Friday involved such residents, almost all of whom are older and have other medical problems. The median age of deaths in Minnesota is 87.

COVID-19 is a viral respiratory illness caused by a coronavirus that emerged in December in Wuhan, China. Nobody has immunity to it, and no proven vaccines or treatments exist.

Minnesota so far is weathering the pandemic about as well as any state, according to an analysis by the U.S. Centers for Disease Control and Prevention. The cumulative incidence of COVID-19 cases in Minnesota was 20.1 per 100,000 people on April 7, according to the CDC report.

Minnesota also had the lowest increase in its case rate since March 31, which state health officials have credited to public compliance with the state stay-at-home order.

A shortage of testing means that Minnesotans have unconfirmed infections, but all states have such shortages so this CDC comparison is reliable, said Kris Ehresmann, state infectious disease director.

"Our testing, although not as widespread as we would like, is keeping pace with our neighbors," she said. "So the lower rate is not a function of not testing."

What happens after the extended stay-at-home order ends May 4 is unclear. The modeling that most closely resembled the governor's order included an extension for four weeks for the entire state, but then lesser social distancing requirements for everyone through June 6, and stay-at-home restrictions that could continue for high-risk individuals into August.

Malcolm said the modeling is only one tool guiding the governor, whose first stay-at-home order exempted 78% of jobs deemed essential. Walz will be looking this month to allow more people to return to work under conditions that limit face-to-face contact and possible virus transmission.

While business leaders trust the advice Walz is getting from the U and Mayo Clinic, the governor's commitment to getting more businesses reopened was a key to maintaining their support for the extended order, said Charlie Weaver, executive director of the Minnesota Business Partnership.

Companies are willing to comply with conditions such as checking worker temperatures at the door or maintaining social distancing in offices, he added.

"It wasn't a quid pro quo," Weaver said. "We might have supported it anyway. But it sure did help."

The coronavirus is circulating faster than expected, partly because as many as 1 in 5 infected people have no symptoms and don't know they are spreading it. The latest Minnesota modeling estimated that one infected person spread the virus to 3.9 others — compared to initial estimates of 2.2.

Gildemeister said the researchers will continue to challenge their assumptions about the coronavirus and the outbreak, which appears to be farther along in Minnesota than originally suggested. Even estimates about the effectiveness of social distancing need to be checked.

The modeling was based on an estimated 80% reduction in face-to-face contact and opportunities for virus transmission under the stay-at-home order. Gildemeister said surveys of Minnesotans are tallying their daily levels of contact with others to see if that 80% figure holds up.

"A model is as good as its inputs," Gildemeister said, "and as more evidence emerges, our inputs will become better. We will learn more of the disease and, more importantly, we will learn more of the disease in Minnesota."

Staff writer Mary Lynn Smith contributed to this report.