The United States is facing a grim dilemma: either effectively shut down society for months to prevent transmission of the coronavirus or see health care systems overwhelmed by people needing treatment for severe infections.
That's the conclusion of a influential new analysis by a well-respected group at Imperial College London that does computer simulations of outbreaks.
Its findings put policymakers in a bind as they think about how to move forward in the weeks ahead when there's no vaccine or proven treatment.
"The take-home message of that Imperial College model is that in order to keep from overwhelming the health care system with severe cases, we are going to have to have very tight controls in place on transmission through social distancing. And those controls are going to be so tight that they will be economically and socially very damaging," says Marc Lipsitch, an expert on infectious disease modeling at Harvard University.
"There's no really good option that doesn't involve at least one of those two problems," he added. "And in fact, both could happen if we aren't very careful about how we manage things."
Predictions from models like this one appear to have had a major impact on top officials' decision to recommend that all Americans avoid groups of 10 or more people, stay away from bars and restaurants, and not travel unnecessarily.
"We had new information coming out from a model, and what had the biggest impact in the model is social distancing: small groups, not going in public in large groups," said Dr. Deborah Birx, who serves as the coronavirus response coordinator for the White House Coronavirus Task Force, as the White House announced the recommendations on Monday.
On Tuesday, however, President Trump said his task force had looked at a variety of models. There was no particular model, he said, that prompted him to move forward on the new social distancing recommendations.
"This is where we were going, this is what we had in mind," Trump said. "This is the next logical step."
The trouble is, according to the Imperial College model, it's not clear what steps should come after this one.
The researchers predicted that if the United States did absolutely nothing to mitigate the spread of coronavirus, about 2.2 million people would die.
The researchers then looked at the effect of different combinations of interventions — including home isolation of infected people, home quarantine for all household members of infected people, social distancing for all ages, and social distancing by those over the age of 70.
One option they considered was a relatively modest restriction on social activity. It combined home isolation of cases, home quarantine for household members, and social distancing for only those over 70. This approach cut deaths in half and reduced peak health care demand by two-thirds, they found.
However, it still resulted in many deaths and overwhelmed intensive care units. The simulation showed that the surge capacity of hospitals would be exceeded by at least eightfold. Even if all patients were treated, the U.S. would see more than 1 million deaths.
In the model, keeping the numbers of deaths low and keeping health care systems functioning required social distancing for the entire population, over a long period of time.
"It is likely such measures — most notably, large-scale social distancing — will need to be in place for many months, perhaps until a vaccine becomes available," researcher Neil Ferguson said in a written statement provided by the college. "The effects on countries and the world will be profound."
The researchers suggest that by closely monitoring trends in infections, it might be possible to lift the public health measures temporarily, but then reinstate them if numbers of cases began to rise.
"It's a good summary of where we are and the choices we have," says Cécile Viboud, who does infectious disease modeling at the National Institutes of Health. "Like every modeling study, it has a lot of uncertainty."
In the weeks ahead, the experience of China and South Korea could help show how slowly relaxing social distancing requirements might affect the spread of the disease and whether it leads to rebounds that might stress the health care system.
"We could learn a lot from that. I'm very hopeful about the South Korean situation and experience," Viboud says, noting that South Korea is testing a lot and isolating people with the virus and their contacts. "There's limited social distancing."
In the meantime, social distancing in the U.S. is buying time for hospitals to increase their capacity and for labs to ramp up diagnostic testing. It's also giving policymakers an opportunity to try to think of new approaches to the crisis that would get them out of the current bind.
DAVID GREENE, HOST:
Let's talk more about this - the White House coronavirus task force is going to sit down today for a meeting with infectious disease modelers. These are experts who use computer simulations to make predictions. And a new analysis from one research group is particularly sobering. NPR's Nell Greenfieldboyce says this will help us understand why officials have taken such dramatic steps to alter life in America.
NELL GREENFIELDBOYCE, BYLINE: To some, the dramatic measures being taken around the country may seem like overkill - telling people to work from home or to avoid bars and restaurants - but take a look at a new report put out by folks at Imperial College London, one of the world's top research teams for simulating infectious disease outbreaks. They calculate that if the U.S. did nothing to try to mitigate this epidemic, over 2 million people could die. Obviously, officials are doing something. And this study says if they want to keep U.S. hospitals from being completely overwhelmed, there's really only one option.
MARC LIPSITCH: We are going to have to have very tight controls in place on transmission through social distancing, and those controls are going to be so tight that they will be economically and socially very damaging.
GREENFIELDBOYCE: Marc Lipsitch is an expert on disease modeling at Harvard University. He says this new study takes everything researchers know now and provides a comprehensive assessment of the current dilemma.
LIPSITCH: There are two horrible options.
GREENFIELDBOYCE: Either everyone hunkers down to slow the spread of the virus, potentially for months, disrupting people's lives and the economy, or the health care system faces far more cases than it could possibly handle.
LIPSITCH: The only hope I see for resolving that is to find some ways of fighting this epidemic that we haven't got yet.
GREENFIELDBOYCE: Like a new treatment or a vaccine or a whole new approach to handling people who get sick. Another infectious disease modeller is Cecile Viboud at the National Institutes of Health. She says this new simulation is good, but she cautions that there's a lot of uncertainty in models. For example, no one understands how much the virus is spread by children. Still, she says widespread social distancing is buying some time.
CECILE VIBOUD: And that's time that we can use to prepare for the next phase, which includes more testing, and also to prepare hospitals.
GREENFIELDBOYCE: She says more testing could allow for more targeted public health efforts and that much can be learned from China and South Korea, which are experimenting with slowly lifting restrictions on social activity.
Nell Greenfieldboyce, NPR News.
(SOUNDBITE OF PENSEES' "LUNAMOTH") Transcript provided by NPR, Copyright NPR.