Please Note

The Washington Post is providing this important information about the coronavirus for free. For more free coverage of the coronavirus pandemic, sign up for our daily Coronavirus Updates newsletter where all stories are free to read.

D.C. Mayor Muriel E. Bowser announced Friday that the city projects that 1 in 7 residents of the city will contract the novel coronavirus, with the nation’s capital seeing a peak in hospitalized patients around late June and early July.

The city government is relying on a different model from the one used by the federal government when White House officials said this week that the outbreak would peak later this month.

In neighboring Virginia, Gov. Ralph Northam (D) said Friday that state models project a surge in May, but he declined to say how many Virginians he expects to become infected.

Fran Phillips, Maryland’s deputy secretary of public health, said Friday that state officials are studying models to show when Maryland is likely to hit its peak.

Here’s an explanation of why the models are different and what that means for residents in the Washington region.

Are there multiple models?

There have been a number of different models projecting the trajectory of the coronavirus outbreak constructed by private consultants, universities and others.

No model is perfect, and none should be read for confident predictions of the virus’s eventual toll. Projections will change frequently, but governments use models to aid planning and decision-making. Two in particular appear to be shaping government policy.

The University of Washington’s Institute for Health Metrics and Evaluation, or IHME, was cited by White House officials this week and is the source of many graphs and projections making their way around the Internet this week.

The COVID-19 Hospital Impact Model for Epidemics, or CHIME, developed by Penn Medicine, is the one used by the D.C. government.

Why does Bowser expect a later peak?

The key difference is how the models calculate the benefits of social distancing.

The D.C. government thinks the IHME model overestimates the impact of social distancing on reducing new infections.

According to city officials, the CHIME model takes into account social distancing measures from the closure of nonessential businesses and schools and the ban on mass gatherings — but it also calculates that many people will not comply.

The model says some 93,000 D.C. residents will contract the coronavirus during the pandemic, with the vast majority of them recovering.

“My experience with human beings is we don’t do everything perfectly, and our model accounts for people who are very strictly social distancing . . . and for all the variations we will experience with social distancing,” Bowser said.

Projections for hospital bed needs are based on a 31 percent reduction in contact, as shown in the chart below. The city has not detailed other assumptions in its CHIME model calculations.

What is the peak, anyway?

The summer peak projected by the District government refers to when the city will need the most hospital beds.

The IHME model says that the peak will come April 15 — but that the city will have enough hospital beds and ventilators at that time.

Using the CHIME model, the D.C. government has determined that it will need far more hospital beds for a summer peak.

Projections based on the CHIME model show the District needing an additional 2,705 intensive-care beds, while the IHME model shows the city would need only 47 more.

The city’s model also shows that D.C. hospitals would need more than 1,000 additional ventilators at the end of June, while the IHME model says the District would not need any more than it has today.

What does the difference in models mean?

In short, the D.C. government is bracing for the worst.

And using the more dire CHIME model is essentially a way to be safe, rather than sorry.

“We pray it’s wrong. We do,” Bowser said. “But we are looking at all of the modeling, and we don’t think one that says we don’t need any additional beds is a model that would be safe preparation for the residents of the District of Columbia.”

The city is still preparing for the possibility of a late-April peak by providing an additional 1,000 hospital beds by April 15, using a mix of existing and new facilities.

But Bowser is trying to avoid a situation in which the city is not prepared for an overloaded health-care system in the summer if social distancing and other measures are not effective. The projected strain on the D.C. health-care system using the IHME model is shown below.

How will this affect business and school closures?

The mayor said she has not decided whether to extend the public-health emergency and the associated restrictions beyond April 25.

But she said that on the basis of current modeling, D.C. Public Schools will not reopen on April 27 as initially planned.

That’s a sign that Washingtonians should brace to hunker at home beyond April.

Are there projections for a peak in deaths?

The IHME model projects daily fatalities peaking in the District and the rest of the country in mid-April.

The District, which reported 15 deaths as of Friday morning, did not release a similar projection for when deaths would peak.

But city officials estimated that 220 of the city’s approximately 700,000 residents would die in the best-case scenario, and that the city would see more than 1,000 fatalities in the worst-case scenario.

The IHME model presents the best-case scenario, with 218 fatalities, with a surge in deaths in April and a leveling off after May.