Benjamin Thompson
Welcome to Coronapod.
Noah Baker
In this show, we’re going to bring you Nature’s take on the latest COVID-19 developments.
Benjamin Thompson
And we’ll be speaking to experts around the world about research during the pandemic.
Amy Maxmen
We’re entering a new era now. We have new COVID strategies, there’s some new unknowns and we’ve got a vaccine.
Benjamin Thompson
Hi, listeners. I’m Benjamin Thompson, coming to you from the South London basement once more. Welcome to this week’s edition of Coronapod. Noah Baker is here once again and returning is Amy Maxmen. Hello to you both.
Amy Maxmen
Hi.
Noah Baker
Hi there, Ben. Hi, Amy.
Benjamin Thompson
Amy, we haven’t had you on for a few weeks. How are things in California?
Amy Maxmen
Things are going really well. The vaccine is rolling out here in the US. My older relatives are now getting it, and that means I get a lot of text messages with questions, and my Aunt Mimi who’s wonderful, she’s sort of like a socialite living in New York City. She’s in her late 70s so she’s gotten her second vaccine. She can’t wait to see her friends. And her first questions are things like, ‘Can I go out to eat with my friends who are also vaccinated?’ which is actually sort of tricky to answer.
Noah Baker
This is not the first time that Aunt Mimi has been on Coronapod. Quite some time ago you referenced Aunt Mimi from early on in the pandemic when she was asking you questions. I really love that she’s made another appearance.
Amy Maxmen
She used to like hate science. She would laugh about how she would throw out the science section of newspapers, but now like daily she is texting me that she’s like ‘Amy, Tony Fauci’s on PBS.’
Noah Baker
It’s good for you to know these things. Is this where you’re getting your tips, Amy?
Amy Maxmen
Laughs. This is how I’m keeping my ear to the ground.
Benjamin Thompson
I mean, in fairness to Aunt Mimi, she is asking some of the questions that researchers really do want to know the answers to, right?
Noah Baker
Yeah, absolutely. It’s one of the things last week when we were discussing potential futures for where the COVID-19 pandemic might be in year and a half’s time. It seems that many of these vaccines protect individuals from severe disease, but do they stop transmission of the virus? And the answer to that question is really, really vital to work out what the kind of future of the pandemic might look like because that really changes the game. And that’s not an easy question to answer. It needs a lot of data, it needs a lot of studies, and people are starting to get closer to answers there. Amy, you’ve been looking into a database this week that is maybe one step towards finding some of these answers.
Amy Maxmen
Yeah, so we’re going to talk later about some studies that are directly trying to study people and see what their viral levels are like after the vaccine and if they can transmit it to the people in their household. But there’s other ways to get at these questions as well, and something that epidemiologists do is they go for big numbers and sort of big data. So, yeah, I wrote about a database that launched this week. It’s exciting. I think they’ve collected so far individual-level data from 24 million people. And they’re anonymised, but this is different from, say, like the Johns Hopkins Coronavirus Tracker that looks at kind of aggregate numbers, total numbers of cases in different states or different countries. This is trying to look at individuals, so it’s listing up to 40 variables per individual, such as what date did they feel sick, what date did they go to the hospital, what date did they die, do they have any contacts, questions like that. And this is going to help epidemiologists crunch the numbers and make estimates of, say, like how long does this virus incubate for inside somebody, when are they most infectious, kind of the things you really need to know to roll out responses. And this is also going to help, moving forward, when you can flood in things like do they have a variant, were they vaccinated, might they have gotten COVID from somebody who was vaccinated.
Benjamin Thompson
So, this is maybe in somewhat marked difference to what we’ve talked about previously where data was siloed by maybe different government departments or different countries or different university institutions, for example. This is all of it together in one place.
Amy Maxmen
Exactly, that’s kind of the big sort of gamechanger here. So, usually, it’s siloed, maybe one health department reports it well or maybe it’s just news sources. So, to kind of back up, the way this all started a year ago is, January 2020, a lot of epidemiologists were following this strange new pneumonia in China. And they did what they do when there’s an emerging epidemic – they made their own kind of homemade spreadsheets. And in the beginning of an outbreak especially, a lot of the news sources are just articles in whatever language they’re written in, and so they’ll comb these different articles and they’ll plug things into their spreadsheet, like one man who’s 64, fishmonger, got sick, had coughing, died on this date. And then son of fishmonger, got sick the day he died, something like that. And that’s where they begin to be able to learn things. So, they were doing what they do, making their own spreadsheets. One of the spreadsheets was on Google Sheets so it was kind of open for people in the community, and then there was like this debate on Twitter in late January where some epidemiologists were saying, ‘This is definitely frequently spreading between humans.’ Others were like, ‘I don’t know if it’s definitely. We knew maybe it’s sometimes but now how often.’ But then because people started looking at this one database, they were able to really make that assessment not based on their sort of gut feeling for how quickly cases would have gone up between humans but based on the actual data in the database. So, this sort of grew and grew. Lots of people starting contributing to the database and today it comes out with some support from Google and from the Rockefeller Foundation that will make it be like in the cloud and have some automated bits and also accessible to anyone. You just need to log in, agree to terms and conditions, and I guess you can download all that data.
Noah Baker
With access to data researchers can do a lot of things, and with access to data that is transparent and reliable they can do even more things. I mean, I say ‘reliable’ – I’m sure there’s caveats with everything as there are always caveats. I think back to the beginning of the pandemic when a lot of the best data we had about case fatality rates and things were coming from individual cruise ships, which were sort of natural experiments for what happened, and it’s been difficult to do really, really large-scale epidemiological work, especially across the world when you think about different cultures and things because there are so many different data sources, and this maybe is the first place where some of the sort of signals in all of that noise might be able to be teased out a little bit to add to all of those earlier studies.
Amy Maxmen
Exactly, because anything under the Sun can happen when you’ve got millions of people. You will get things you never thought you’d see. So, we’ve heard a lot of reports about somebody being reinfected, getting COVID twice. I know at Berkeley I was just told there’s a student at Berkeley who’s reinfected and they have a variant, so maybe it means you lose immunity to the new variants. But it’s hard to say from one person because we’re all a little bit different. One is one, but if you have a huge number, now we can start saying, ‘Is this what we would expect or this much more than what we would expect?’
Benjamin Thompson
How hard is this to do, to pull together hundreds of millions of records? You say it’s been done sort of a little bit homemade to begin with. What’s been the stumbling block to getting to this point now?
Amy Maxmen
Yeah, it’s a great question because it’s one of those things that it’s hard to get anybody excited about except for if you’re an epidemiologist because it’s actually super hard. So, one stumbling block is the information. Everybody reports in different ways, especially when you’re talking about combing news articles and getting the information out of that. That’s one difficulty. Then some countries like, for example, Sam Scarpino, who is part of this project who I spoke with – and I should say the database is a website called global.health –he was saying back when they were combing a bunch of data from different Asian countries like South Korea and Singapore especially, it was like they had all the information right there. There’s ten cases at this shoe store at this mall in this town. So, it was super simple to get a lot of the information. A lot of it was standardised so it was pretty easy to plug into the website manually. And then with a grant they actually got from Google.org, they also were able to write some codes that made some data sources automatically feed into their database if they were reported in really standardised ways. So, certain countries have centralised systems were their governments report data on every case in a very standard way. Remember, we’re not just wanting numbers. We want to know, say, their age or their race and when they had symptoms. We want all that stuff. When it gets trickier is when data is not computed from a centralised source in a way that’s really standardised, and that’s a lot of countries. It’s definitely the US. Sam was saying when the outbreak in the US got super huge and they wanted to track all of that data, a huge part of the problem was just finding it and manually having to put it in. He said even in like the state of Massachusetts, they changed their data reporting systems at least four times, so then they had to kind of redo how they were kind of entering everything. So, that’s what makes it really tough. There’s a lot of manual labour that goes into this. I think mid-point last year they had 100 epidemiologists volunteering around the world to enter in data, so it’s just a lot of human power. And then of course, this was being done on a Google spreadsheet – that just doesn’t work after a while. It’s just too much data and too many people working on it. So, that’s when they applied for this fellowship so that they could get some assistance to put the database on the cloud and to start writing some codes.
Noah Baker
And I think it’s also worth mentioning that one of the key things that they need to do here when gathering all this data is to make sure it’s all properly anonymised. Data security is vital here and privacy is really, really important, and that’s no small task either.
Amy Maxmen
No, so actually, that’s something that’s really interesting because I had heard researchers complaining that they loved to do studies but hospitals and health departments wouldn’t allow them access to the data because of privacy concerns, which aren’t nothing. We do want privacy. But sometimes, it was almost as if people are just worried so they would rather err on the side of caution and say, ‘No, I don’t want to share it.’ Because they don’t really know how to properly share it and what’s legal and what’s not legal. It’s actually quite tricky, so if you don’t want to navigate it, just go ahead and say it’s too private, we can’t share it. So, what the group did is they spoke with a lot of legal experts and ethics experts and tried to go about this in the best way that’s compliant to a lot of privacy laws, and they’re hoping this way they can present people who have data with enough assurances that they’re going to keep the data anonymised. And I should say, not allowing anybody to see anything is definitely most the private thing you can do, but there’s always a balance here.
Noah Baker
And so, this database is now live at global.health, and it’s huge, as you mentioned, and it’s having more and more data added to it every day. I think it’s important to mention that that doesn’t mean that therefore scientists now have all the data they need and all the problems can be solved, like there’s still gaps. There’s still places that have very little data. If I’m looking at the map right now in Europe, Germany is listed as having more than 2 million cases in there whereas the UK is listed as having 299,000. Now, I know there’s been a lot more than 299,000 cases in the UK and certainly more than there has been in Germany, and yet those numbers as yet don’t really reflect that, and that’s, I’m assuming, because there’s still hurdles to get over to get all the data from countries into this database in an appropriate way.
Amy Maxmen
You got it, and I think what’s so fascinating about this map is you can see Syria, Venezuela, Libya, Yemen, there’s no data coming from those places, and this is individual case data –
we know that there are cases in these places – but this means the government either isn’t recording or isn’t reporting data on individual cases as in the age or something that makes this an individual compared to a lump sum. But here’s what’s fascinating. I’ve listed all those countries. Guess what else is in that group – Italy and France.
Noah Baker
Yeah, and Spain.
Amy Maxmen
Yeah, that is more a matter of maybe they don’t feel comfortable releasing the data. So, those are sort of the things that this group actually is also hoping to sort of overcome.
Noah Baker
So, potentially, these data are coming in. How could these data be used to answer questions like the question we were asking at the beginning – how transmissible is the virus after you’ve had a vaccine or even how do the variants change the transmission dynamics of the virus?
Amy Maxmen
Yeah, so the database is great when you want to find kind of like big numbers because anything can happen once. You can always get some sort of exception to the rule, so that’s why individual stories aren’t great. And studies that are just on a cohort are just on a cohort. So, this data could help depending on what information they plug in per individual. So, if we have, say, data on what variant a person had if they’re positive, we can start to ask questions like, ‘How many people who are now having their second case of COVID have the variant?’ That helps us think maybe this variant is overcoming antibodies a person naturally generated. If we see that somebody who’s positive was vaccinated, can we also see that they have a variant, meaning that the variant evaded a vaccine? Also, if we include this variable called ‘Ct levels’, which is sort of an indicator of the viral load in people, how much virus they have inside of them, we can start asking other questions about the vaccine and variants as well. So, if you see that there’s a high Ct level, which means you have a low level of the virus because the machine really had to try and find it, does that occur in people who were vaccinated, meaning that maybe they’re getting infected but they’re keeping it at a low level. And then if we see a very low Ct level, which means they have a high amount of the virus, are we seeing that in people who have some of these variants? Now, that would suggest that somehow the variants are allowing the virus to multiply a lot inside of them, and then you might think that there’s a bigger chance that they’re breathing this virus out more and possibly infecting more people. That gets you an essence of mechanism.
Noah Baker
And that particular measure – viral load – is really key to understanding how transmissible the virus is and how that might be impacted by things like variants or things like whether or not someone has been vaccinated, and before we have these large data to look at, there are other ways to try to get at that question, and the vaccine trials that were run for the various vaccines that are currently in circulation, several of them did a version of this during their trials or are continuing to do this. So, they tested viral load as a part of their trial, it wouldn’t necessarily have been everyone but it would have been a smaller cohort, and they used that to then make inferences about what the vaccine was doing to transmission as well as how effective it was at reducing disease, which is what this 90-95% number is always talking about is how effective it is at reducing disease. And those trials were promising and there is more of these data that are being gathered as vaccines roll out. Can you tell us a little bit more about the trials that are ongoing, Amy?
Amy Maxmen
Yeah, so there’s two studies that have launched pretty recently. One is in the UK and it’s looking at health workers who are vaccinated and it’s going to test them as well as the people in their households to see if they end up getting infected. And in Israel they’re doing something similar only with households where one person in the household has the vaccine. There’s also data that we have from some of the trials on vaccines. So, in the AstraZeneca trial they had swabbed participants every week, and they estimated that there was a 50% reduction in asymptomatic infections among the people who are vaccinated compared to the placebo group. Kind of the asterisk there is this was just a subset of the people in the trial so it’s not everyone, but it does suggest that the vaccine is preventing infection. And in Moderna’s vaccine trial, they tested all of their participants after the first shot and they saw a two thirds drop in the number of asymptomatic infections after people got the first shot compared to the placebo group. And the caveat there is, well, they just tested them that one time and then again at the end of the trial, so we might be missing things.
Noah Baker
And it’s also of course important to remember that changes to asymptomatic infections don’t necessarily track on to whether or not someone can transmit the infection further. Just because you don’t get sick doesn’t mean you can’t transmit it, although not coughing is known to reduce the risk of transferring it to somewhere else because you’re not actively shoving it out there in the environment in the same way. But really to get a clear answer on how things are happening, we need that big data. We really need to not just look at subsets. We need to look at large-scale trials which can say with some kind of statistical confidence that, yes, those who have the vaccine do not appear to be passing the infection on to other people at the same rates, and that’s a really key piece of information that scientists are sort of clambering for and will really change the course of how this pandemic progresses.
Benjamin Thompson
And they’ll make a big difference to your Aunt Mimi as well, I should think, once this answer is known, Amy.
Amy Maxmen
It’s true, and I mean, I’ll tell you what I told her in the meantime. I’m not a doctor of medical science. I’m a doctor of sea spider evolution. With that caveat, this is kind of based on reading what I’ve read and talking with friends and I posted something on Twitter and public health people responded. So, right, we don’t have all the data, so I told her just that. We don’t know if you can spread this. But the data looks pretty good and so I think my advice was, ‘Yes, you can go out to eat in New York City’s restaurants,’ which are at 35% capacity. If she was in a warm place I’d say eat outside but it’s cold there so, yeah. She can go out to eat with her friends, keep a distance, since they’re also vaccinated, and when any servers come close to the table, put that mask on just in case. And also, I told her to wear the mask otherwise, when she’s at grocery stores and on the streets. One, to protect other people but also because it’s really unnerving for people who are at risk and who haven’t been vaccinated. They don’t know you’re vaccinated so it’s kind of a nice thing to do to just keep wearing that mask.
Noah Baker
And I suppose we also should add that there is always the possibility that variants could change the game about the efficacy of vaccines, so it’s still always good to take precautions at all times to make sure that you’re doing what you can to protect yourself, even if you have been vaccinated.
Amy Maxmen
That’s absolutely correct. And we should know more maybe. I heard at a press conference – I just read an article on this – apparently a 90 year old woman named Esther Cohen raised her hand and asked Anthony Fauci if she could return to Mahjong games with her friends who are vaccinated and he said, ‘Esther, we are thinking about that and the CDC should be coming out with guidelines soon.’
Benjamin Thompson
So, I think what we’re all saying then is read your local guidelines and stick to them, everyone, please.
Noah Baker
Yeah, and listen to the Esthers and the Mimis of this world that are asking the really core questions that are getting to the heart of these issues.
Benjamin Thompson
Well, on that note then let’s leave it there for the time being for this edition of Coronapod. Amy and Noah, thank you so much for joining me.
Amy Maxmen
Thanks.
Noah Baker
Thanks, Amy. Thanks, Ben.