S1E54 / A Smarter Way to Quarantine / Roxanne Khamsi & Jeffrey Townsend

It is really important right now to pay attention to this kind of quarantine issue. If you're going to be traveling, if you're going to be in a situation where you might be exposed to others or where you can protect others by quarantining yourself, now is the time to do it and to do it carefully because this is when it can be especially effective.

Hi, you're listening to Epidemic, the podcast about the science, public health, and social impacts of the coronavirus pandemic. I'm your host, Dr. Celine Gounder. Roxane Campsie is a science reporter based in Montreal. Like all of us, Roxane missed out on a lot of big life events this year because of the pandemic. The demands of quarantine have become a major issue for her.

Here in Canada, I can't go visit my sister in Colorado and come back without sitting at home for 14 days. And I'm not allowed to go to the grocery store. And I'm not allowed to, you know, like walk a dog if I had a dog. I really can't leave the house.

The word quarantine goes all the way back to the Middle Ages. Back then, the disease people were worried about was the plague. So the city of Venice started requiring ships to anchor in the harbour for 40 days before they could come ashore to prove they didn't have the plague. The word quarantine comes from the Italian words for 40 days. Makes a 14-day quarantine sound easy, right? But quarantine rules in Canada are serious. People are ordered to quarantine in their homes for 14 days after they return from abroad. There are big fines if someone doesn't comply. Fines as big as $1 million.

It's $1 million Canadian, so it's a little less.

All the same, that's still a lot of money. And if those fines weren't high enough, someone who violates their quarantine and infects other people could get jail time too.

It's definitely It's something that people here take really seriously. And I think that my friends in the States don't understand that. Like, a lot of them are surprised when I am like, guys, it's actually like a serious quarantine here for 14 days.

Even if you don't get a fine or jail time, quarantines are an important part of reducing the spread of coronavirus. But quarantining has been one of the most difficult and confusing parts of the pandemic. There are different rules for every country and even every state here in the United States, and it takes a toll on people.

A lot of people have jobs they have to go to where they can't quarantine, and 14 days just seems so cumbersome that they just throw their hands up and say it's not possible. It's cumbersome in practical ways and then also kind of emotional ways to stay at home for 14 days.

It's a lot. In the end, Roxanne decided that seeing her sister's new baby wasn't worth the hassle.

And so I didn't go see her and her new baby this fall. And, and I maybe would have if it was only 8 days.

But what if you could get all the public health benefits of a quarantine in 8 days or even less? Back in October, Roxanne wrote an article about some new research that caught my attention.

A biostatistician that I know reached out to me and said that his group was working on this new way of thinking about quarantine, and they'd done some calculations to suggest that quarantine didn't need to be as long.

That research is still under peer review at this recording, but it caught the CDC's attention too. This month, the agency cited the preprint in its updated quarantine recommendations. The CDC now recommends a 10-day quarantine without a test, or as little as 7 days if and only if someone tests negative on day 5 or later. In this episode, we're going to look at how quarantine for coronavirus could be cut by as much as 50% when combined with testing at just the right moment. We'll speak with the researcher whose work was cited by the CDC, and we'll find out what an oil rig has to do with all of this. Today on Epidemic, a smarter way to quarantine. The biostatistician Roxanne wrote about this fall is Jeffrey Townsend.

I'm the Elihu Professor of Biostatistics at Yale University, specifically in the Yale School of Public Health.

Before the pandemic, Jeffrey used his skills to do things like help oncologists estimate the growth rate of cancer in the body.

And I've always been someone who really likes that connection between the math that I'm doing and the real world.

So when the pandemic arrived, Jeffrey shifted his attention to COVID. He set out to see if an effective quarantine could be done in less than 14 days.

Quarantine in principle should be able to be shortened But how much was a question that nobody seemed to know the answer to, and it seemed something that was approachable from a mathematical perspective, so we decided to address that question.

Jeffrey thought testing could be the key. His model came up with two major findings.

One of the things that became very clear as we did the modeling was that testing on exit from quarantine was far better approach than testing on entry to quarantine. And then another thing that we found was just that you could significantly shorten the quarantine by doing that kind of exit testing.

Why might you miss somebody upon entry into quarantine? Why might they test negative?

Yeah, the easiest way to think about this is just to think a little bit about the biology of the disease. When the virus actually infects you, it probably doesn't infect a very large number of cells at the beginning. It probably infects a very small number of cells. So the point is, during that very early phase, often called the latent phase, when the virus is hardly producing very much virus, you're not very infectious. But it's also very hard in that early latent phase for any test to come out as positive and indicate that you actually have the infection.

So someone could be infected and have the virus multiplying inside them, but not have enough virus in their system when they're tested to get a positive result.

That means that we have to be extremely careful about trying to catch this disease early. And that's also why we sort of need to combine the test with a quarantine.

Jeffrey's model predicted that the tipping point for the virus was somewhere around day 7 of infection.

Past 5 days, you're gaining some benefit to 6 and 7. As you get to more days than that, 8, 9, 10, 11, 12, 13, 14, you are gaining benefit, but it's getting to be smaller and smaller benefit with each extra day.

Basically, the vast majority of people who are infected with the virus will test positive by about a week after exposure. So after waiting this period of time and then testing to verify if there's an infection would give you actionable results without someone waiting the full 14 days.

Industry immediately took interest because they have a major economic incentive to try to understand this issue.

Specifically, oil companies. BHP is an Australian oil company that operates offshore oil rigs around the world.

They really want to avoid any kind of sickness and, and God forbid, an outbreak on those rigs because they would probably, if they had a full-scale outbreak, have to shut down the entire rig, which would cost enormous sums of money.

The stakes were high for BHP. An offshore oil rig is not something that turns on and off like a light switch. The company needed to keep their workers safe and maintain their operations.

And we started collaborating with them and giving them some of our results early on. At the same time, they shared some data that they got, and that was a great opportunity for them to learn from us and for us to learn from them.

So what happened when Jeffrey's model got tested on an oil rig? We'll find out after the break. An offshore oil rig sounds like an environment that's pretty far removed from most people's everyday life, but Jeffrey says it was an ideal environment to test their model.

It's really valuable to us that it was an oil rig because we know there were no outbreaks on the oil rig because it's a closed population.

BHP tweaked Jeffrey's model.

Which is not exactly surprising. It's not like we did a cost-effectiveness analysis, which is ultimately what they need to figure out is how, you know, what are the costs of doing these things in terms of their personnel? There's a lot of operational questions that are not within the academic question that we asked. They decided to sort of accept that and look at that and say, okay, well, if most of the marginal returns are obtained by having at least 4 days, what we'll do is we'll do a 5-day quarantine We'll test at 96 hours, and then if they come out positive, then we will send them offshore.

BHP tested people at the end of quarantine, but they also tested their employees at the beginning. If somebody was already showing signs of the virus, the company needed to isolate them and find a replacement fast.

But that also provided a really nice test dataset for us to examine our results because then we had individuals who were tested on entry and we could observe if any of those individuals who tested negative at entry were testing positive at 96 hours.

Remember, these were people who tested negative at the beginning of their quarantine and then got tested again when they were getting ready to leave for the rig.

There was quite a few people who were caught by that second exit test, and because they were found to be infected during that exit test, they were not sent offshore. And that is estimated by our analysis anyway to have prevented as many as 9 outbreaks that might have occurred Jeffrey says knowing his research helped people was exciting. I mean, it was so awesome because for academics, most of the time we don't really get to see almost in real time our work applied to actually prevent disease, for instance. It usually takes years for a policy recommendation to be adopted by some regulatory authority, etc. But I think the thing that really brought it home the most was when they told us that they were telling other members of the industry how great their safety record was. So, you know, when industry starts bragging to other parts of industry about something, you know, it's good because that's mostly meant as, you know, that there's not really a lot of reason for them to do that except for out of pride.

So what are the implications here for the general public in terms of how they should be thinking about testing?

I think for the general public, the point would be if you think you might be encountering a risk situation by going somewhere, by visiting relatives, whatever you're doing, then it behooves you to have a quarantine period before you go to visit them. And if you can have that quarantine period to be about a week long, ideally even longer if you can, but a week long will get you most of the benefit. And then that test at the latest point you can where you'll still get the information about the test before you have to go and expose those individuals that you care about. Nothing is guaranteed in this world, but that means you're really limiting your risk considerably if you're doing that.

Consistency in messaging has been a big problem during the pandemic. Roxanne Kamsi says the different rules in different places make people feel like quarantine and other public health recommendations are arbitrary.

And I think what is going to happen is that public trust in these rules is going to erode because there's so much back and forth and so much contradiction.

Roxanne thinks research like Jeffrey's could help establish a common set of rules for how quarantine recommendations should be made.

I think it's important for public health officials to now talk to one another and to say, hey, let's be on the same page about this. Why should there be a 48-hour quarantine in one place and a 4-day quarantine in another and a 14 one in another. I think that what needs to be done is like almost like an evidence-based approach to quarantine rather than what seems to be happening, which is like what businesses think might be most logical or, you know, what the public is most easily convinced to do, which is not necessarily always the best thing to do.

But there are also limits to how this information could be scaled to the broader population.

They're kind of operating in like an ideal scenario where testing isn't a rate-limiting factor. But as you're very well aware, testing is certainly in the real world not something that's abundantly available as much as it should be, and the results don't come back as fast as they should. So I think that's a really, really good criticism and question about how to apply these findings.

Testing capacity has been a stubborn problem since the beginning of the pandemic. How can we implement these suggestions when turnaround times are slow and tests are still hard to come by?

The answer to both of those questions unfortunately depends on how well we do everything else in this pandemic. The ability to have fast turnaround and the ability to have enough testing capacity to do this on a large scale depends on our ability to get the currently out-of-control epidemic into control.

And there's a lot everyone can do to make that happen. They're the same basic public health recommendations we've been talking about for months: wear a mask, keep at least 6 feet apart from others, avoid indoor gatherings.

If we want to have the capacity to try to maintain that economic activity and the health and safety of those individuals who are at high risk, then everyone else needs to keep this epidemic at a low level. So that it enables us to apply those tests in the right place and to have the test turnaround that is necessary.

This podcast was released on December 11th. This week, the United Kingdom started distributing the Pfizer vaccine. Canada became the second Western country to approve Pfizer's vaccine, and the FDA found the same vaccine to be safe and effective. So as the vaccines start to roll out, What will the role of quarantines be in the coming months?

It's fantastic news that these vaccines appear to have very high efficacy. However, we need to be very cautious about that as a public health message because the very high efficacy of the vaccines in clinical trials does not at this point translate to any benefit whatsoever to individuals who are out there now.

It could take many months before enough people are vaccinated and the pandemic is under control. Jeffrey says this holiday season, as cases and deaths continue to spike, being thoughtful about quarantining could be more important than ever.

In fact, right now, we really need to do everything we can to exercise this kind of restraint on social interactions, social gatherings, where we're seeing most of the transmission happening. It is really important right now to pay attention to this kind of quarantine issue if you're going to be traveling If you're going to be in a situation where you might be exposed to others or where you can protect others by quarantining yourself, now is the time to do it and to do it carefully because this is when it can be especially effective and when it is important to do it.

Epidemic is brought to you by Just Human Productions. Were funded in part by listeners like you, were powered and distributed by Simplecast. Today's episode was produced by Zack Dyer and me. Our music is by The Blue Dot Sessions. Our interns are Tabitha Gordillo, Annabel Chen, and Brian Chen. If you enjoy the show, please tell a friend about it today. And if you haven't already done so, leave us a review on Apple Podcasts. It helps more people find out about the show. And Just Human Productions is now on Instagram. Check us out at Just Human Productions to learn more about the characters and big ideas we cover on Epidemic and our sister podcast American Diagnosis. You can learn more about this podcast, how to engage with us on social media, and how to support the podcast at epidemic.fm. That's epidemic.fm. Just Human Productions is a 501(c)(3) nonprofit organization, so your donations to support our podcast are tax deductible. Go to epidemic.fm to make a donation. We release Epidemic every Friday, but producing a podcast costs money. We've got to pay our staff. So please make a donation to help us keep this going. And check out our sister podcast, American Diagnosis. You can find it wherever you listen to podcasts or at americandiagnosis.fm. On American Diagnosis, we cover some of the biggest public health challenges affecting the nation today. In Season 1, we covered youth and mental health. In Season 2, the opioid overdose crisis. And in Season 3, gun violence in America. I'm Dr. Celine Gounder. Thanks for listening to Epidemic.