In recent weeks a number of alarming stories have circulated suggesting that "You're Likely to Get the Coronavirus," as an article from the Atlantic put it. That story reported that Harvard epidemiologist Marc Lipsitch predicted that "some 40 to 70 percent of people around the world will be infected" within the coming year, though "many will have mild disease, or may be asymptomatic." What's more, a comment in The Lancet calculated that "approximately 60% of the population would become infected."
Having written about global warming since the 1990s, I'm profoundly accustomed to scientists warning us of dire things to come, while politicians and the media blithely ignore them. So I certainly didn't want to do that myself. But this time felt different. COVID-19 is a horrific public health threat. It's entirely possible that hundreds of thousands of people will die, conceivably millions. But at the same time, China's infection rate has been plummeting, and nearby nations like Taiwan and Singapore never saw an infection spike to begin with. South Korea's infection rate was declining as well. Here in the U.S., the Trump administration's bungling has been disastrous, but local public health responses have been vigorous, despite the fact that we're flying blind from lack of widespread testing.
So I wanted to get a sober assessment from a knowledgeable expert — not to dismiss worst-case fears, but to help put them into perspective, and shed light on how we might best avoid both the very real dangers of the virus itself, and additional dangers from over- or under-responding in various different ways. So, I reached out to Dr. Timothy Brewer, at UCLA's Fielding School of Public Health, who currently serves as chair of the board of directors for the Consortium of Universities for Global Health. The interview was conducted by phone on March 12, and has been edited for clarity and length.
A recent comment in The Lancet began by stating that governments won't be able to minimize both deaths from COVID-19 and the economic impacts. Keeping mortality low will be the highest priority for individuals; so governments must look at measures to address or ameliorate the inevitable economic downturn. What's your view of this from a public health perspective?
It's not that the public health response and economic response to an outbreak are antagonistic. In fact, they're synergistic. The things we do to contain the outbreak will also minimize the economic impact. If the government responds in a way that generates panic and anxiety, for example, not only will that probably facilitate the spread of whatever the pathogen is, it will certainly facilitate the economic consequences as well. However, if the government and public health agencies are able to respond in ways that help communities to calmly and rationally deal with the outbreak, and provide the necessary knowledge, that will not only minimize morbidity and mortality, it will minimize the economic effects as well. The best response from the public health perspective will also be the best response for minimizing economic disruption.
And that would have both acute and chronic aspects?
Yes, both acute and chronic aspects. Acutely, what we're really trying to do is to learn about the pathogen, understand where it came from, how serious it is, what populations are affected and how it might spread. From there, what we want to know is how can we limit that spread, limit the impact of the morbidity and mortality of it.
Chronically, what we're going to do, in addition to our acute response, to use our knowledge and technology to develop new interventions such as vaccines. We won't have a vaccine right away, but we know how to make influenza vaccines.
Are there other specific government actions, such as emergency paid leave legislation, that would be more effective from a public health perspective?
Definitely. Emergency paid leave legislation would be very valuable to have in place. We want people who are sick to be able to stay home. But if you stay home and that loss of income means you lose your house, you lose your job or you're not capable of supporting your family or buying food, chances are you're not going to do it, especially if you're not very sick. One of the things we know about COVID-19 is that people who are mildly symptomatic can spread the disease. So we need people to be able to stay home — and not just the people who can afford to stay home, but anybody who is sick and potentially capable of spreading the virus.
That Lancet comment calculated that "approximately 60% of the population would become infected," based on reproduction values of around 2.5 in the early stages of the epidemic in China, which it called "a very worst-case scenario." An article in the Atlantic cited Harvard epidemiology professor Marc Lipsitch predicting that "some 40 to 70 percent of people around the world" would become infected within the year, including those who would be asymptomatic. Flattening the curve so that health care facilities aren't overwhelmed is a major consideration, which makes numbers like this seem especially grim. What is your assessment of what constitutes a worst-case scenario?
A worst-case scenario is the continued and expanding COVID-19 spread around the world. I have no reason to believe that the numbers others are throwing around are at all accurate. So, let's look at some data. The first data we want to look at is in Hubei Province in China. This is where the outbreak started, and as best we can tell, there have been about 68,000 cases in a population of around 60 million people. So that's well under 1%. You could argue — as Mark and others might — that they underreported. So, let's multiply that by 10. Nobody assumes that they underreported tenfold, maybe it's twofold or threefold, but let's say it's 10. That still puts you around 1% of the population where the outbreak started. That means that 98 to 99% of the population did not get infected.
Let's take another pandemic: The last pandemic we had was the H1N1 avian influenza pandemic of 2009-2010. The Center for Disease Control and Prevention estimated that 61 million Americans got infected with that virus between April 2009 and April 2010. So, 61 million Americans is a lot of people, but that works out to about 20% of the population. It would be reasonable to say that recognizing that COVID-19 and H1N1 are different viruses, based on past pandemics it seems unlikely that 40% to 70% of the U.S. population will become infected with COVID-19, particularly as areas expand the implementation of public health measures to limit the spread of the virus. So I don't think there are any data right now to suggest that that level of infection will occur.
So what is the most likely scenario?
I think the most likely scenario is that we will continue to see outbreaks in different populations around the world and we will see some places where it will be contained and eliminated. Right now, in China, they're having under 30 [new] cases a day. So obviously they have been able to bring it under control. Taiwan has more interaction with mainland China, probably, than any other place in the world. Taiwan has had less than 120 cases since the outbreak started. Singapore has had about 100 cases, and Singapore is also very tightly connected to China. So there are examples around the world where governments and public health agencies have instituted effective responses. The case numbers are going down in [South] Korea now as well. But there are other places, like Iran and Italy and the United States, where the case numbers are still rising, where we have not yet seen the outbreaks brought under control, and that is concerning.
Next I was going to ask for the best-case scenario. I assume that would be every country in the world coming around to the level of success of Singapore and Taiwan, and that South Korea is headed toward.
Exactly. The best-case scenario is that the public health measures that are being instituted in different parts of the world, which we know work for respiratory viruses, are effectively put in place everywhere, and that the outbreak is contained.
What are the major things we need to do now?
So the first thing we need to do to get more testing out there, to go looking for where the virus is. There's probably community transmission going on in Washington state, in Northern California around San Francisco and the Sacramento area, and around New Rochelle, outside New York City. We don't really know what's happening anywhere else, and we don't necessarily know the extent of transmission in those locations. So the first thing we need to do is to roll out more testing, to get a better handle on where the virus actually is right now, and how it's spreading.
The second thing we need to do is to be a little more aggressive around our contact tracing, and around helping and supporting people to go into self-isolation if necessary. We need to be able to identify the people who are infected, and the people who are at risk for getting infected, and help to support them so they can remain isolated until the risk of infection and disease has passed. I think those are the two major things that we can be doing better than what we're doing right now.
You touched on what's happened in some other countries, What can we learn from that?
The good news is that the United States and the public in general should take heart from the fact that public health measures when applied effectively do work. That's a big take-home message.
Taiwan is very instructive here. What Taiwan did was they instituted, very early on, a very aggressive surveillance and contact tracing program. They rapidly tried to identify anybody coming into the island who might be at risk for infection, and then had them self-isolate or got them in a place where they would isolate, while supporting them and keeping in touch with them and then tracing contact of individuals who were infected. These are two basic tenets of public health control that all public health agencies in the United States will understand and know how to do. So I find that heartening and reassuring.
What we need to do as a country is to support our public health agencies and make sure they have the resources to be able to do these steps that they know very well how to do.
So where does the U.S. fall right now, compared with other countries?
I think we can say right now that the U.S. is probably somewhere in between Italy and say South Korea or Singapore, Taiwan or China, in that we clearly have a few pockets where things are getting worse. But not to the widespread extent that they currently are in Italy or Iran, for example. We probably have other places where things are relatively under control. So, in California, there have been a number of cases, I forget how many, I think about 40 to 45 cases where they were felt to be person-to-person transmission, and community-acquired. And yet we haven't seen the big outbreak that occurred in northern Italy. So that may suggest that in fact the California public health agencies, and the county agencies, are doing a reasonable job of containing this, at least so far.
So we don't know.
We don't know. Public health really is done at the local level. It's done by counties and cities and states, so we probably have jurisdictions where they're well-prepared and things are in place, and we probably have some places where they're less well prepared.
What help can modeling give us?
There have been several modeling papers that have already been published. There was one published in the Lancet on Feb. 29 looking at modeling and forecasting of international spread coming from Wuhan, China. There was another paper published on March 6 in Science looking at the impact of travel restrictions on COVID-19 spread. Also, there's another modeling paper in the Annals of Internal Medicine.
What models do is give you some sense of the likelihood of spread. Earlier in our discussion you mentioned the reproduction number, which is: One infected person leads to how many new cases? And right now the number being thrown around is about 2.5. So one infected COVID person results in two or three new infections, and that's been fairly consistent across models. So that's one thing models can do, is give you some sense of that.
Another thing that models can do is give you something called the epidemic doubling time. Which is basically: How quickly is the virus spreading? The Lancet paper that I mentioned predicted a doubling time of about six days, six and a half days. So, models can give you a sense of how likely things are to spread.
Then, in terms of the intervention side, you can mathematically describe something like a travel quarantine. So, what is the likelihood of a travel quarantine changing the spread of the virus? That was done in the Science paper. What they conclude is that it probably reduced or delayed the spread in China by about three to five days, but that it probably reduced the spread internationally by 80%, or several weeks.
So you can start to compare different things. You could compare increased contact tracing with quarantine, for example. Or you could look at additive interventions: contact tracing and a quarantine. Those are the kinds of things models can do, they can give you a sense. But you do have to remember that models are guesses. They don't predict the future.
In terms of possible interventions, then, what can we do?
There's a lot of things you can do. And you mentioned some of them: For example, we could pass legislation tomorrow that provided paid sick leave for anyone with a respiratory viral infection. That would be a terrific and wonderful thing to do. It would help us to control the outbreak, because it would relieve the financial anxiety and pressure of being out of work if you got sick. That's one of the things that we want to have happen.
A second thing we could do is make sure our public health departments are adequately financed and staffed, so that they can do the contact tracing which we know has been very successful in places like Taiwan and Singapore. So that's a second thing we can do.
A third thing we can do is try to make sure that there are adequate test kits available throughout the counties and cities, not bottlenecked in a few key locations, so that doctors and nurses and health care providers can test the people who need to be tested.
Finally we can try to get the message out to everybody that, look, we're all in this together, we will get through this if we help each other out. Don't overwhelm your doctor or your nurse practitioner if you're feeling well and just want to be tested. Please reserve their time and the test for the people who actually need them. And probably don't go and clear the supermarkets of toilet paper. You probably don't need to be sitting on 200 rolls of toilet paper right now. If we just kind of bring the anxiety down a little bit, that would be helpful.
I think the last thing I would say we can do, very concrete, is we can take care of each other. If you're aware of a friend or colleague or a neighbor who has had to self-isolate because either they have COVID-19 or they have been in contact, call them up. You cannot get COVID-19 over the phone. It just doesn't happen. So call up and make sure they're OK, see if they need anything, just check in on them. If you know someone who is elderly and home alone, call them up, make sure they're OK. Just check in on them. That's how we're going to get through this, is by using knowledge that we learn both about this virus and what we know about other viruses and public health in general, and compassion.
I'd like to know more about areas of uncertainty, beyond what you've already mentioned, most notably our lack of testing results. What are other major unknowns?
There's lots of room for research. We're now getting more and more data to show that people who are relatively asymptomatic or mildly symptomatic can spread this virus. So that's a very important piece of research. What we're still not sure about is, for how many days? There are estimates right now from early studies of anywhere from five to nine days. But that would be very useful information to know.
There's data out of Shenzhen, China, to show that children who are household contacts of patients with COVID-19 are as likely to get infected as adults, but they're much less likely to get disease. So how important are children and young adults as transmitters of infection? We know from influenza, for example, that children can be very important in transmitting influenza even if they don't get sick. Are children are important for COVID-19? We don't know that yet.
We also know, based on the same study, that only about 15% of household contacts do get infected. So that's still a number where we could get more data. How long do you need to be in contact with someone who is sick? How close does the contact need to be? I think we should get some additional information there.
There are data to show that you can isolate COVID-19 from surfaces, but whether that is really an important part of transmission or not is still unclear. There's one study out of Singapore, in a respiratory isolation hospital room with three patients, where they were able to isolate the virus from lots of different surfaces, but they were also able to show that routine cleaning eliminated the virus. So I think, again, that's an area where we probably want more information. But the preliminary data show that the kinds of routine cleaning and disinfection that we do should work for this virus, just like they do for other respiratory viruses.
Here in California and Los Angeles, authorities have announced measures about the size of public meetings, as has happened in many other places. How do these compare with what other countries have done and with what seems to be necessary to flatten the curve?
Social distancing is an effective intervention, and the question is when do you trigger it, and at what level? There's no single answer for that. In the H1N1 outbreak in 2009, Mexico did that. They basically shut down a city of 23 million people. Unfortunately, the influenza was so contagious that that did not prevent it from taking off and going around the world. But that doesn't mean what Mexico City did was wrong. They took the right steps. It's better to introduce them before the virus is widespread, because at that point it's probably too late.
The Atlantic article stated that the emerging consensus among epidemiologists is that we are likely to have a new seasonal disease, so the cold and flu season will become the cold and flu and COVID-19 season. What is your sense of the likelihood of that, and why?
I think we have no way to know. On the side that this will happen is that this virus does seem to be much more easily transmitted from person to person than the previous SARS coronavirus. So, in that sense, it seems better adapted for survival and transmission in humans than the first SARS coronavirus, or the MERS coronavirus for that matter. So it does suggest this could become an endemic virus, but I don't think we actually know that yet.
Should that happen, what will be the likelihood of developing annual vaccines, like for the flu?
It depends on the virus and the vaccine. For example, you don't need an annual vaccine for measles. You need two doses of measles vaccine and you're set for life. The reason why you need an annual vaccine for flu is because the flu virus changes so much as it replicates. Now coronaviruses are RNA viruses, like flu viruses, so they will change. But I think I would defer to a virologist and a vaccinologist as to whether or not we're going to need an annual vaccine. It may be possible to develop a vaccine that does not require being given every year. But I would defer to an expert about that. It's going to depend really on the target of the vaccine, how much that changes as the virus replicates.
Finally, what's the most important question I didn't ask? And what's the answer?
The most important question is, "How do we all get through this?" And we get through this by focusing on what we know, and what we need to learn, and taking care of each other — so not by discriminating, not by panicking, and not by trying to isolate ourselves with our 500 rolls of toilet paper. If you're my age, you have lived through two pandemic influenza outbreaks, 1968 and 2009, you have lived through HIV, you have lived through SARS and MERS and the Ebola outbreak in West Africa. And in every one of these cases, using what we know we have either been able to make huge advances in minimizing the impact of these viruses, or actually containing them completely. And that's going to help us with this virus too.