Click here for the latest coronavirus news, which the BDN has made free for the public. You can support our critical reporting on the coronavirus by purchasing a digital subscription or donating directly to the newsroom.
A new model predicts that both Maine and the nation likely have a couple more weeks before COVID-19 deaths will reach their peak and start to decline.
The projections by the University of Texas at Austin set out the probability of a peak over three separate time frames. In the most likely scenario, there is a 75 percent chance that COVID-19 deaths will peak in two weeks in Maine.
There is a 43 percent probability that Maine will reach a mortality peak in seven days, according to the model. In the least likely scenario, there’s a 15 percent chance that the peak in deaths has already passed. The model reflects only confirmed COVID-19 deaths.
Regardless of when a peak in deaths occurs, public health officials have urged people to continue to stay home. Even after a peak, there is still a good possibility of getting sick.
“The bottom line here, folks, is we’re still in the middle of things,” Nirav Shah, the director of the Maine Center for Disease Control and Prevention, said on Monday. “I urge everyone to continue with those physical distancing measures.”
Though people may want to know when they can return to their everyday lives, some experts have instead turned to devising parameters for reopening the country.
An American Enterprise Institute report identified four goals that should be met before a state lifts social distancing measures: There should be a sustained reduction in cases for at least 14 days; hospitals must be able to safely treat all patients requiring hospitalization without resorting to crisis standards of care; the state must be able to test all people with COVID-19 symptoms; and the state must be able to actively monitor confirmed cases and their contacts.
If the University of Texas predictive model holds true, deaths in Maine would likely begin to fall during the beginning of May. The model is unique because it takes into account the impact of social distancing using aggregated and anonymized geolocation data from tens of millions of cell phones across the country.
In this way it differs from a widely-cited model by the University of Washington’s Institute for Health Metrics and Evaluation, which estimates that deaths from COVID-19 have already peaked in the United States.
However, the IHME model doesn’t factor in real-time daily social-distancing data. Instead it takes into account the dates of statewide policy changes, such as shelter-in-place orders, and bases its predictions for the United States on patterns observed in China and Europe.
The new model at the University of Texas quantifies changes to how often people gather together in restaurants, bars, schools, parks, pharmacies or grocery stores, and how much time they spend at work and at home. Researchers found that statewide policies were far from the only influence on people’s behavior.
For instance, large cities in Texas issued stay-at-home orders that stopped travel many days before a statewide policy took effect, a fact that is visible in the data and affects predictions about future deaths.
For detailed technical information, visit this report on the COVID-19 mortality forecasting model.
Watch: Maine CDC coronavirus press conference, April 20