U.S. COVID-19 Cases on the Rise: Q&A with RAND Experts


Mar 18, 2020

A woman visits her mother who has tested positive for coronavirus at a Seattle-area nursing home, the epicenter of one of the biggest outbreaks in the United States, in Kirkland, Washington, March 11, 2020, photo by Jason Redmond/Reuters

A woman visits her mother who has tested positive for coronavirus at a Seattle-area nursing home in Kirkland, Washington, March 11, 2020

Photo by Jason Redmond/Reuters

COVID-19, the disease caused by the new coronavirus, has now been detected in all 50 states. More than 6,000 cases and 100 deaths have been reported so far in the United States. With local orders to close schools, stores and restaurants going into effect in cities around the United States, we asked several RAND researchers to answer some questions about the crisis:

  • Jennifer Bouey, the Tang Chair in China Policy Studies at RAND, is an epidemiologist whose research focuses on global health strategies and the social determinants of health.
  • Courtney Gidengil is a senior physician policy researcher who also practices infectious diseases at Boston Children's Hospital.
  • Laura Faherty is a physician policy researcher and assistant professor of pediatrics at the Boston University School of Medicine.

They spoke on a conference call with RAND media relations director Jeffrey Hiday on March 16. What follows is an edited transcript of the conversation, with some updates made the following day to reflect changing information.

Are we seeing the right approaches to the outbreak here in the U.S.? Is it happening in good time or a little too late?

Courtney Gidengil: Over the last 7 to 10 days, there's been a real shift in tone and an urgency around this outbreak. We're seeing everyone move into mitigation mode. Early on there were travel bans and screening of travelers, all trying to prevent it from taking hold within the United States. Now we're trying to slow the spread of the virus.

What that means is a phrase that probably everyone has heard now countless times, which is to flatten the curve. If you picture a bell curve, the idea is that the peak of the curve stays as close as possible to the actual capacity of the health care system. We may not be able to prevent the number of overall cases, but we can spread them out over time so they don't overwhelm the health care system like we're seeing in Italy. We're also trying to avoid health care workers getting infected and sick because ultimately that means we can't take good care of our patients.

Goals of community mitigation for pandemic influenza, illustration by CDC

Illustration by CDC

There aren't that many serious cases yet. How are you seeing hospitals prepare?

Gidengil: Hospitals are likely to be thinking through a lot: How to cross-train staff to work in different roles. How to get more ventilators and where to put them if there isn't enough room in ICUs, and how to get more personal protective equipment. How can hospitals work together in communities? Could children's hospitals potentially treat young adults, which was thought through in the H1N1 pandemic? How to schedule health care workers, have clear backups, and encourage health care workers to think through their childcare.

We have been looking toward what's being learned in Italy, China and elsewhere. But because we're operating within the particulars of our own health care system here in the United States, people are looking with great interest at what is going on in Seattle, New York City, Santa Clara and other spots where they are some days ahead of a lot of communities in the country.

Are there any lessons to be taken from what we're seeing in Asia or Europe?

Jennifer Bouey: Yes. After an initial fumble in understanding transmission, on Jan. 20, China initiated its public health emergency and in just two days they locked down Wuhan. Soon after we saw a steep increase of case numbers in Wuhan and its health care system was totally overwhelmed for about two weeks. But within three to four weeks of complete shutdown, China pretty much was able to keep the number of cases stable. And nowadays we see that there are few new cases coming out of China. That shows that social distancing does work if you go to a certain extreme.

We see that there are few new cases coming out of China. That shows that social distancing does work if you go to a certain extreme.

Share on Twitter

Jennifer Bouey

Now China is focusing on dealing with the large number of expats coming back, especially college students. We have about 300,000 Chinese students in the U.S. and now many of them are heading back to China and their families. But 80% of China's cases are coming from within family clusters; and the government doesn't trust self-quarantining at home. So China is struggling with how to isolate these potential new cases coming from elsewhere.

What are some of the social distancing measures that are being used in the U.S. and how are they working?

Laura Faherty: “Social distancing” is a bit of a misnomer. We should call it “physical distancing”—while trying to maintain social connectedness. The idea is to keep yourself as far away from other people possible and to avoid touching surfaces that other people have touched: doorknobs, elevator buttons, things like that.

The simplest way to think of social distancing is, if you are able to stay home, stay home. Your outdoor activities or exercise should be solo or with your immediate family members who you're already exposed to. Getting basics like essential medications, restocking food supply, etc. are obviously necessary, but we should limit trips out to the very minimum possible. All sorts of recreational gatherings at bars and restaurants, movie theaters and the like should be put on hold for the time being.

We're seeing different guidance on this. Some places are shutting down all restaurants, some are going to take-out only, some are saying it's fine so long as you keep everyone distant from one another.

Faherty: Yes, these guidelines are rapidly shifting and are dependent on local context and the amount of known community transmission. It also is dependent on an individual's health and age. A person with underlying medical conditions or over age 60 needs to take more stringent precautions.

The more aggressive we are at staying home and avoiding nonessential trips outside the house, the better.

Share on Twitter

Laura Faherty

In the United States, we're not used to these regulations that say don't go out to a restaurant, don't send your children to school. So I think that there's been an effort to adjust the aggressiveness of measures to the level of risk in specific communities at the time. I think that's adding to some of the confusion.

But the overall message is: This will hopefully be a time-limited situation and the more aggressive we are at staying home and avoiding nonessential trips outside the house, the better.

How long will we need to practice social distancing?

Faherty: That's a tough question. At least several weeks. Potentially much longer than that. We should be mentally preparing ourselves for this to be a marathon, not a sprint.

Bouey: Looking at the experience of China, many cities have been on lock-down for over five or six weeks now. They're gradually trying to bring the manufacturing sector back because of the mounting hit to their economy. But on the other hand, the health epidemic has been well-contained so far.

Is it a good idea to close schools preventively?

Faherty: There are some compelling reasons to close schools—especially in areas with suspected or confirmed community transmission. That said, the decision is incredibly complex with significant downsides.

For instance many, many students in disadvantaged communities rely on their school for meals, sometimes multiple meals a day. They rely on the school as a safe place to be if they're currently homeless. A lot of students with special needs receive services like physical therapy and occupational therapy.

And of course there's the question of who takes care of the children if their parents aren't able to stay home. Will it be a grandparent who's potentially at the highest risk of severe illness? Will a parent have to risk losing his or her job?

Some school districts are doing a wonderful job of setting up resource centers where students can come for meals or to be safe and taken care of. We are also seeing some innovative strategies to maintain continuity of learning to the extent possible in these extraordinary circumstances.

If the period from contracting the virus to the end of contagion is, say, three weeks, could a country that isolates itself for three weeks stamp out the virus?

Bouey: You'd have to isolate the patients who are infected from those who are healthy and quarantine the healthy people. So even if there's a public lockdown with implementation of quarantine for three weeks, we still have to think, would there be outbreaks in family clusters?

That's what we see in Wuhan. Even though they have a city blocked off, new cases are still coming from within families when the infected were not isolated.

So now in China they have complete isolation. Everyone who has even a very slight chance of getting the disease is isolated in a different setting. But even with that, once they bring workers back to businesses, you'll still have cases with mild symptoms that can spread the disease. This will be a long battle.

Is it feasible to utilize hotels and motels as quarantine quarters for infected people and that this could potentially solve two problems at once?

Bouey: China is doing that right now. Anyone entering China from another country has to be quarantined for 14 days, even if they have had no known contact and no symptoms. They usually have to pay for the hotel themselves.

With students coming back from college, some families would rather have the kids stay at a home where they can quarantine while the family moves to a hotel during the 14-day quarantine. The healthy family members associated with the traveler stay in a hotel but can still go out and get food and supplies.

Initially we heard that if you don't show symptoms, chances are you are not spreading the disease. What do we know about that at this point?

Gidengil: This is a big question with important implications, so I'm hoping we'll get more evidence about this. But there are two scenarios in which a person might not have symptoms but be contagious.

The first is that person might be recently infected and about to develop an infection; we call such people pre-symptomatic. And I think that there is emerging agreement that such people do have the potential to spread the infection because of this. Some studies have measured the level of virus on day one of symptoms and find quite high levels of virus in the nose and throat of patients. And the fact that it's in their nose and throat—not deep in the lungs—also means it's easier to spread potentially.

There is some evidence from young children and babies, particularly one study of a baby who had remarkably high levels of virus in their nose and throat, but no symptoms. And that has an implication for the role children might play in spreading the virus, although there is still a lot of uncertainty as to what role children may or may not play in this outbreak.

There are a lot of headlines in the news about asymptomatic people. Some of this has been in non–peer-reviewed articles and more information expected to come out. But in speaking to infectious disease doctors, most say that for the majority of this to spread by totally asymptomatic people would make it quite unlike other diseases that we've ever seen. And the guidance we had been getting from CDC is that asymptomatic people are not likely to be driving this pandemic. That being said, there is still a lot to be learned and these assumptions could change over time; it's possible that people with quite mild symptoms may be driving most of the spread of the infection.

But it's important to know for sure whether asymptomatic people can be contagious. Containing the virus with methods such as screening for symptoms would be exceedingly difficult or almost impossible if we can't tell people to quarantine themselves or to get tested based on symptoms.

Is it true that for most people who will be sick with COVID-19, the symptoms would show up by the fifth day after exposure?

Gidengil: The range is 2 to 14 days. Many (about half) do show up by day 5, and the vast majority by 10 days. Very few show up between days 10 and 14 and a tiny tail after that. We're still sticking to 14 days as the length of quarantine, because that's reasonably the latest we think you can develop symptoms.

Does the virus eventually need to make its way through a community to confer immunity and spend out the virus?

Gidengil: There have been concerns that the UK may be adopting this approach. How safe it is to allow that to happen depends on the burden on the health care system.

It's also important to think about immunity and how much of a hold we want to allow the virus to have. Certainly, this infection could become more endemic; rather than outbreaks, it may become something that just always exists in particular communities. There's also a question of how long-lasting immunity is.

Does the virus spread faster in cold climates? And if so what does that mean for the curve flattening in the summer in the U.S.?

Bouey: There are some hypotheses like that based on SARS, which appeared around the same time in November and December, and when the World Health Organization identified the virus and epidemic, it was already in April. When China implemented quarantines, it was May–June and the new cases soon disappeared.

Based on that example, lots of people think, OK, this is another coronavirus, and in addition to the quarantine measures implemented, maybe the warm weather also helped. That's the hypothesis.

But so far we have seen cases of community transmission in Singapore, in Guangzhou, and many warm-weather places. So it's a hypothesis that no one has confirmed yet.

If over the last few months people who'd had flu shots got something severe and flu-like, any chance they had coronavirus and, if so, what would the implications for that be?

Gidengil: It's hard to know. There has been a fair number of severe respiratory viruses this season. It has been a very, very difficult flu season. But I don't think anyone can say with certainty whether those people had COVID-19 rather than the flu because we haven't had the ability to test those people for coronavirus.

Hopefully we'll soon have antibody testing as well and that'll give us a much better handle on whether someone was exposed to COVID-19 or had it. I doubt that we'll be able to test everybody, but we'll be able to do some population studies that will be very helpful.

Have scientists figured out whether a person who has recovered from COVID-19 has immunity, or can they get it again?

Gidengil: I don't think we have a handle on that yet. With SARS, we saw at least some immunity, but it's also clear that SARS behaves differently. With the new coronavirus, there have been news reports of individual people testing negative and then they redeveloped symptoms and tested positive. It's not clear why this is happening—whether their throat wasn't swabbed well enough or some other reason the test didn't go well, or whether they are getting the infection again.

As I was saying, we need antibody or titer testing to understand some of these cases. That would help us know not only who was infected, even if they didn't know it, and speak to some of this asymptomatic question. It would also help us understand what level of antibody you need to have immunity.

It would be unusual for people to not have any immunity whatsoever to this virus. That would be unlike most viruses we've observed. Then again, as I said, this virus is new and may behave differently.

Where do we stand with the amount of testing being done in the country?

Gidengil: We still don't have enough tests to go around for everyone who is presenting sick to emergency rooms. There have been more tests available in recent days, and my sense is that hospitalized patients can now be more relatively easily tested.

In addition to testing in the state labs and at the CDC we also have some commercial labs that over the last week have started to take tests. I don't know how much backlog they may eventually acquire or what volume of testing they're getting. I think we can expect the testing will increase, although I think there's still uncertainty as to whether it can truly meet capacity or not.

We still don't have enough tests to go around for everyone who is presenting sick to emergency rooms.

Share on Twitter

Courtney Gidengil

The other big issue is who is going to actually do the testing. Primary care providers don't have enough personal protective equipment to test someone. If you're testing someone and they're coughing and sneezing in your face, you're supposed to wear particular types of personal protective equipment and they just don't stock that in the primary care offices. Or they don't have the swabs to do the test.

We are seeing some pretty innovative approaches: drive throughs and tents outside of hospitals. There are many aspects to ramping up the testing besides just the test kits.

And what about the supply chain? Do we have enough reagents, for example?

Bouey: The three or four weeks of shutdown after Chinese New Year in China caused lots of anxiety globally about supply chain. China produces about half of the global production of chemicals, metals, electronics and textiles.

API, the active pharmaceutical ingredients, has become a big concern in the U.S. and other countries. China provides about 80% of the API for both generic and other medicines. What I've heard is that most of the generic drug makers in India are using China's API, but because of the Chinese New Year and seasonal storage, they usually have three months of stock. If China's manufacturing sector doesn't come back to normal production until April, that will be become more of a problem.

China is also a big producer of face masks and all of that personal protection gear. Hopefully, production there is much enhanced because they put a priority on those medical supplies and soon they will be willing to start exporting again.

Can you talk about surveillance testing, and why that would be helpful if we had sufficient tests available?

Gidengil: About three or four weeks ago, the CDC had announced that they were going to do surveillance in five major cities including New York City, Los Angeles, San Francisco, Chicago and Seattle. I thought that was proactive. They were going to use the existing flu surveillance network—outpatient practices that send flu swabs from a certain proportion of all patients who have compatible symptoms.

The hope would have been to detect some early signal ahead of seeing people present very sick, because there was a delay between when you get infected to when you're so sick that you maybe need to be in the hospital.

Unfortunately, to my knowledge, that has not come to fruition. The presumed reason is CDC's test kits were not working as expected and there was an inadequate supply. But once we get enough, that will be incredibly important to do. It could let communities know what to expect, or if the number of cases detected indicates there's a surge coming. Surveillance is a cornerstone of public health and we can't know what's happening without taking some action.

What are the main differences in the underlying assumptions used in the models that suggest contradictory approaches to handling the crisis, such as in the U.S. and UK?

Bouey: One of the things we talk about when we assess interventions is political feasibility and viability. We have to think about the cultural and social environment. Those are all part of the reason that different policies can come out in different countries.

However, I still feel that any policy implementation should be based on evidence and science in terms of whether we want to have the immunity passive or active. Passive immunity means that you just let people get the disease and then produce a herd immunity. Whereas active immunity means that we have a vaccine and we protect large populations that way. Before the vaccine is available, the only public health tool is social distancing.

So we want to make sure that we understand the disease transmission and the capacity of our health care system, as well as the population's willingness to get tested and do the social distancing. All those things should be understood and factored into these policies.