This transcript was automatically generated using speech recognition technology and may differ from the original audio. In citing or otherwise referring to the contents of this podcast, please ensure that you are quoting the recorded audio rather than this transcript.

Gavin: Hello, welcome to another episode of The Lancet Voice. I'm Gavin Cleaver. 

Jessamy: And I'm Jessamy Bagonal. It's July 22nd, 2020, and coming up on today's episode, we've actually got a little bit of hope in the COVID 19 era. And we speak with Editor in Chief Richard Horton about the successful phase two trials of two COVID 19 vaccines and what that means for now and the future.

Gavin: And then Jess and me chats with two public transport experts about the risks of virus transmission in trains, planes, and automobiles. And then we chat modeling. You'll have seen, especially recently, that scientific modeling is a prominent tool for showing how decisions made now affect the future. But how does it work?

We asked Kathleen O'Reilly of the London School of Hygiene and Tropical Medicine. to explain how a modeler predicts the future. 

Jessamy: But first, you'll almost certainly have heard in the last couple of days about the successful early trials of what we're calling the Oxford vaccine and the China vaccine.

Gavin spoke with Lancet Editor in Chief Richard Horton about what this progress means and how far there is yet to go, and what's gone wrong with global cooperation during the COVID 19 crisis. 

Gavin: Richard, in a nutshell then, how promising are the results of these papers? 

Richard: Thanks Gavin. The results of these two randomised trials are extremely encouraging.

And I think it should give us optimism that we're making good progress on the way to a vaccine to Protect human populations from Cove in 19. They are only phase one phase two studies. So we do need to wait for the full randomized trials in phase three populations. But I think that we have good reason to be hopeful for the future.

Gavin: So these two papers consist of the commonly referred to as the Oxford vaccine and also the China vaccine. So what's the difference between the two vaccines? And are they currently at the same stage of development, roughly? 

Richard: They, the China vaccine is a recombinant adenovirus vectored vaccine. And what it's doing is expressing the, what's called the spike gene from the coronavirus and it's based upon a virus that was widely prevalent in The Oxford virus is a chimpanzee adenovirus viral vector, and that also expresses the spike protein as well.

They do have differences, but they are also related. They both have a very impressive impact on generating both antibody related and what's called cellular related immunity. And that, I think, is the great step forward that both of these studies are telling us that both arms of the immune response can be provoked, can be stimulated by the, these two vaccines.

And that's why I think we can be hopeful as we go into the phase three studies. 

Gavin: Now, of course, as you mentioned, these are phase one and phase two combined results. So obviously that's something to be cautious about, but could you summarize, What we still need to remain, keep in mind about these vaccines, if you will, what we should remain cautious about with these results.

Richard: What these results are telling us are that these two candidate vaccines are safe. There were no major adverse reactions from either vaccine. The kind of adverse reactions that one saw were symptoms like muscle fatigue. General malaise, feeling feverish, headache, and of course, because it's an intramuscular injection, pain and sensitivity at the injection site.

And these kinds of adverse reactions can be relatively easily prevented by taking, for example, paracetamol before you actually receive the vaccine shot. So that's the first thing. The vaccine is safe from these studies. It's also, and this is a word that is used by vaccinologists. And I'll explain what it means in a second.

It's immunogenic. And by immunogenic, I mean it's stimulating the immune response. As I was saying, these two arms of the immune response, the humoral response and the cellular response Antibodies and T cells. These two vaccines are stimulating both arms of that immune response, and that's very important.

It's not telling you that these two vaccines are efficacious or effective in preventing COVID. They're simply telling you that the first base, so to speak, or maybe even the second base, has been reached. That we've got safety. They're well tolerated, and they're doing the right things immunologically that we need them to do if the vaccine is going to work.

Right, 

Gavin: but we have no proof that these vaccinations prevent COVID 19. 

Richard: No, we're not at that stage yet. Right now, if you take, for example, the Oxford trial, that's being tested in phase three studies. In not just in the United Kingdom, but also in Brazil and in South Africa, where, of course, there's a lot of transmission of the virus at the moment, and they're also testing that vaccine in some of the high risk populations that we know about, people who are older, those who have chronic disease people who are health workers on the front lines of the pandemic.

So we will get extremely valuable and useful information from those phase three studies, but they're not going to be available for some months. 

Gavin: So looking at the broader picture then, obviously vaccine development is a huge priority for pretty much, every country in the world at the moment.

It's something that the world is in desperate need of. So we have this massive acceleration of vaccine research. Of course, the kind of thing that would normally take years is actually being done in weeks, in months. What challenges are posed by accelerating this timetable so hugely? 

Richard: When Tony Fauci was giving evidence to the United States Congress recently, he was asked about the usual time it took, takes to produce a vaccine.

And the evidence that he gave, he said seven years. And he was asked, what's the fastest time a vaccine has ever been produced? And he gave the example of a Zika vaccine, which took about 18 months, although the Zika outbreak, the outbreak had pretty much ended by the time there was a vaccine, so it never was fully brought to fruition.

But let's say that 18 months has been the shortest time so far. I think we're looking at roughly the same time course here. And there's a real trade off because what you're trying to do is you're trying to get a vaccine that works and also a vaccine that is safe and it's very important that one is not going to rush a vaccine into you.

clinical use until you can be as sure as you must be that the vaccine is safe. Now inevitably, no vaccine is 100 percent effective or 100% safe and it won't be taken by 100 percent of people. So you're trying to make sure that you've got all of the evidence available to show that it is maximally effective in as many populations and high risk populations as possible and that it's maximally safe.

So we do have to go through these very important Tests before a vaccine can be released for general use if we do reach that point. Because we do need to be sure that it is as safe as possible as well as effective. There are also other interesting twists. In one of the studies that we published the study from China they found that the immune response in older people wasn't as effective, wasn't as strong as it was in younger people.

And so as they're moving on to do phase three trials, they're now going to introduce a second dose, a booster dose of the vaccine for older people that will hopefully stimulate their immune systems more because it looks like they need to be stimulated more to get a good response to the vaccine. So these twists in the vaccine story are going to take place.

over the course of clinical testing. And that's why I think we should be optimistic about the direction of travel we're going in, but not so ambitious that we think we're going to get a vaccine for example, by the end of the year. If we have a vaccine by the end of 2021, we will have done incredibly well.

Gavin: So you mentioned there the twists and turns, of course, in In the story, of these vaccine developments, I think one thing is worth talking about, when we think about these twists and turns, is the global state of affairs and the state of global cooperation between different countries, between different actors.

in this storyline. So why do you think that global cooperation currently seems so fragmented when we're looking at the kind of overall global tackling of this virus? 

Richard: Yeah, it's very fragmented. And in some ways, it's understandable. Governments have a first duty to protect their own publics, their own people, and so they're going to do the best they can to make sure they've got enough vaccine for if it, if and when it comes, to protect their people.

But the danger of that is that, that many countries will lose out, and only the strongest country the country with the most money will win. I think if you take a global perspective then you, I think most people would agree that the those who should have access to the vaccine first are those who are most at risk of developing.

severe, even fatal disease if they become infected. And we know enough about this virus now and the disease it causes to know what those risk groups are. Older people who are living with chronic disease black and minority ethnic communities, people who are on the front line of the epidemic.

That's to say people who are working in health care settings, people who might be working in supermarkets, people who are working on mass transit systems, people who are coming into contact with other people all the time. Now, those are the groups that need to have access to the vaccine first before the rest of us.

And if we could agree that through some sort of global Convention or understanding resolution passed to the World Health Assembly, some mechanism that takes place to agree that those are the groups that gets the vaccine first. That would be a huge step forward right now. there is a real danger that those groups most at risk will not get fairly or equitably treated if a vaccine does become available.

And that should be a cause of not just global concern, but actually global shame if we're not able to find an agreement to protect those most vulnerable groups first. 

Gavin: Really hopeful results there, Jess. There's still a really long way to go, obviously. But it's incredible the progress that's been made in such a short amount of time on these two vaccines.

Jessamy: Yeah, exactly. And there's so much other work going on. So this is just one approach that's underway. So the other approaches that people are using are to use the virus itself in some form of either being inactivated or live virus. Examples that we have in this sort of day to day now live attenuated vaccines include the MMR chickenpox and then inactivated examples of polio.

So that's a very tried and tested method, but it does take a little bit longer for the most part because you have to be absolutely sure that those viruses aren't going to, be harmful to people. So there's a lot of trials underway using part of the virus. And then there's this form of platform, which is the sort of adenovirus or another virus.

So they essentially get some genetically modified virus and put in some of the SARS CoV 2 genes inside it. And then the virus shuttles it into our cells and hopefully our cells read it. And respond to it with some kind of immune response. So that's what this particular, both these trials are using that type of approach.

And then there's some other work going on which is using DNA or mRNA and also protein subunits and virus like particles. So there's an awful lot of work going on in this space. 

Gavin: I think one of the interesting things about this, these two sets of vaccines, is that not only, obviously the timetable is extremely accelerated, we're doing what, as we were saying, we're doing in months and weeks what we would be doing in years normally, but actually these approaches to the creation of a vaccine are themselves unusual as well, right?

Jessamy: Yeah, so what's really interesting about this type of approach is that, they are promising, they are a modern kind of platform. But actually there aren't many other vaccines that use this particular approach. So the only one that's been approved, although not fully proved in that it protects against the disease, is the Ebola vaccine.

Otherwise there's a Raley's vaccine which is used for live animals. to immunize them against rabies. But there aren't other vaccines that we use this particular method at the moment. 

Gavin: I think since we put these papers out on Monday, and obviously there was a slight leak the week before, I think it's really shown the level of interest, obviously, in these vaccines is absolutely massive.

It's huge. As we're recording this on Monday, I think every single newspaper has, even the Daily Star has the vaccines on the front page. But I think it's important as well to sound notes of caution about it. The timetable at the moment, if the Oxford vaccine is successful, looks like it might be around the end of 2020, start of 2021 that it can actually start being.

rolled out and of course the government's already bought 100 million or is it 300 million doses of this vaccine But it might be the case that this is a vaccine that is partially effective. It might only take the edge off some of the symptoms, for example. And while that would be great and very useful and fantastic, especially in case of a second wave, it still might be the case that we could be years down the line before we get something that's actually fully immunogenic when it comes to COVID 19.

Jessamy: Yeah, there are so many things to think about. This particular Vaccine, as Richard said, what it shows at the moment is that it is immunogenic, that it does make your cells respond to something, but seeing as we have so little understanding of the correlates of protection or how we're meant to protect ourselves from this virus, we really don't know what that means.

We hope that it means that it gives you immunity to it. Whether that's true or not is something that we still need to do a proper trial on. And then there's the sort of other aspects to think about with this particular, type of platform. So using these adenoviruses has got a slightly contentious past in that this was the approach that was used in the early 2000s to try and develop an HIV vaccine.

And what happened there was that actually they ended up having to stop those trials in 2007. Because there was a concern about the type of adenovirus, this AD5, that causes the common cold. And people having some kind of a response to that, which then affected whether they were more or less likely to get HIV.

So You know, there is still a long way to go, there are still things that we might find out. In those particular cases, those HIV trials were then started again with a different population who had already got existing immunity to that virus that they were using to transport the HIV genes into.

And that proved to be safe, but it didn't work. There's that kind of technical side to think about. But then there's also the side of, whether this is going to work in the elderly. We've spoken before on this program about the elderly having a poorer sort of immune response to most things, and being able to get that kind of immune response with a vaccine is very difficult.

The hopeful things about these two studies is that, in the elderly, it did make some kind of response, although it wasn't as much as with young patients. And then again, there's also this sort of issue of whether any kind of immunity would last for a long time and how we might do boosters.

Gavin: Of course, and that's something that really comes out when you have these long term trials of these vaccines. Normally, these years long trials, you come out with at least an awareness of how long protection, if it's there, would last. But obviously, the timetable acceleration in this case doesn't allow for us to monitor anyone for that long.

Jessamy: Yeah. So, it's really hopeful. But as you say, there is still quite a long way for us to go to be sure that this is going to really make a difference to how we're all living and whether we're able to get back to the things that we were doing before the virus.

Gavin: One of the many unanswered questions about virus transmission is to what extent public transport is a risk. Jess and me spoke with Manu Sasidharan and Giorgio Hadjimitriou from the Department of Engineering at Cambridge University. And they talked about the impact public transport has on COVID 19 transmission.

Jessamy: Manu, perhaps you could just tell me where we are at the moment in terms of public transport and safety for COVID 19. 

Speaker 4: Given the limited understanding that we have of how things work around COVID 19, and of course the lack of a vaccine, the transport sector can contribute greatly to tailoring public health interventions of human mobility reduction and social distancing, there's of course a trade off that needs to be achieved between the potential public health benefits of curtailing public transport during a pandemic like COVID 19 and thereby delaying the community spread against the socio economic impacts of curtailing or reducing that human mobility.

So that trade off needs to be achieved so that we have the near normal functioning of our society. 

Jessamy: And I guess that trade off is particularly pertinent when we see the figures that, the London Transport, I think that they're they lost 90 percent of their income during lockdown.

What plans are underway to try and make public transport safer and so that, they can fill up that capacity? I read an amazing, statistic of the 9 million people that are within London, 60 percent of the commuters use public transport, which essentially equates to 325, 000 people during rush hour getting onto trains every 15 minutes.

How, from an infrastructure point of view, is it possible to deal with that while also making it safe? 

Speaker 5: It's good to mention that during the lockdown we had an 80 percent decreased use of public transport compared to normal times. At the moment walking and driving is back to normal levels, while public transport use is still 40 percent lower compared to 2019.

So there are many things that should be done to go back to normal. We observe a significant swift from public transport to driving and walking. And If people are going back to work, then there should definitely be some travel restrictions and social distancing measures. They should continue. And since we should keep the social distancing measures, so there are many things that need to be done.

In the short future, I believe that governments should definitely see in improving the infrastructure for walking and cycling. See in the cities. It's ideal for many reasons, like exercising and environment and avoiding transmission of viruses. So our focus should go to this perspective in the near future.

Jessamy: Yeah, that makes complete sense, just instead of trying to make public transport very safe, we need to also focus our efforts on these other aspects of transport, which are just inherently much safer for people. 

Speaker 4: And if I may add the general, the linear transportation networks of roads and railways are, and it's associated infrastructure.

They're considered as the largest public owned assets in most of the countries, and it's not that different in the UK as well. But what we're seeing is that in a post COVID 19 world, where the economy of many countries are likely to suffer, we need to identify and implement. cost effective and user centric strategies for managing our transport infrastructure.

The case of the fact that the current world, much of these infrastructure is managed or maintained under constrained budgets. And when the future economic uncertainties we would have to find the best way of going forward. And at the same time, it is also imperative to recognize that. We have a once in a generation opportunity to deliver a transformative change in decarbonizing transport.

If we look at COVID 19 as a disruptive event that has brought about a range of changes in our travel behavior. So as a human mobility choice, as what my colleague said, as human mobility choice and post COVID 19 world might incline towards public transport, we also need to make sure that it doesn't bring about huge significant risk of increased greenhouse gas emissions and environmental.

problems. 

Jessamy: Might those discussions start just on a kind of practical level. You talk about the sort of transport infrastructure being, for the most part, some of the largest aspects of government owned economies. How can we use this point to move away from things? 

Speaker 4: As I mentioned earlier, the current way by which we manage or maintain our transport infrastructure is often in what we say is a silo management system, where we look at a particular infrastructure and take decisions on that area.

But once we take a step back and think we need at the end of the day transport and its infrastructure as a public service. So it needs to work efficiently for the betterment of our citizens. So if we are to tailor our strategies in a way that we look at that. whole system perspective or a whole transport system perspective where we consider strategies, which is a win for every single stakeholder, the transportation system, including the environment.

That would be a decision. That would be a starting point. And there are currently. plans to bring in that whole system approach. Even discussions has been happening and even pushes from the Office of Rail and Road Regulator talking about the advantages of bringing such a whole system approach into picture.

Jessamy: Thanks Manu and just Getting back to COVID 19 and public transport, what evidence is there so far for transmission of COVID 19 on public transport that you're aware of? 

Speaker 5: We investigated the correlation between, first between human mobility and COVID 19 related deaths. So we started by seeing how the government measures affect human mobility, so we divided time into some periods where until the from the beginning of 2020 until the 8th of March, we had an increased use of transport and of watching and driving, but then from the 8th of March and until the 23rd of March, where we had the general lockdown.

There was a gradual decrease of mobility after the lockdown, after the 23rd of March, public transport use was decreased at the level of 80%. And then from the beginning of May, when some of the measures were lifted, we gradually increased the use of public transport. How is this related to COVID 19 related deaths?

The daily increase of deaths of a pandemic can be modeled by a generalized logistic growth and we know that based on previous studies and based on previous pandemics. As I said, we divided time into some periods until the 8th of March, until the 23rd of March, and we moved on 18 days later since the meantime between exposure to COVID-19 virus and the reported death is 18 days.

So we use this. to examine how the increasing number of deaths data until these dates can show how affects deaths. So our conclusions were that with some predictions, our conclusions were that if there was no human mobility reduction at all, then the deaths would be several times higher than the current number we have.

If there was only advice and recommendations for avoiding nonessential traveling, and then the number of deaths would be again higher than the current number, but lower compared to no reduction of human mobility at all. 

Jessamy: There are other studies as well, but I suppose what we're saying is that the assumption is that because we've reduced the number of people traveling on public transport, and we've therefore reduced the deaths, that there is some kind of causal association there, which obviously Can't be proved, but can be modeled, as you say, and I just wondered, now that we've got a, this potential of aerosol transmission, what's the ventilation like on public transport, on different types of public transport, say on the tube, on trains?

Speaker 4: So WHO has very recently recognized the fact that there is a risk of airborne transmission in specific indoor locations such as crowded and inadequately ventilated spaces over a prolonged period of time with infected persons. So we're of course not sure of how long that prolonged period of time should be and all that but There needs to be enough social distancing measures and at the same time safety options of wearing masks while traveling on the public transport.

And in terms of the ventilation side of it especially in the London underground where you can imagine that. It is a challenge in itself, but there are efforts being made to increase the ventilation within the London Underground, and not just due to the pandemic, or generally it has been happening over the period of time.

Jessamy: So I think one of the most interesting things that came through to me from that interview was just how difficult it is to really know. How much COVID 19 transmission is going on public transport, what are the studies, what are the things that we have to do to try and be able to quantify that?

And also, just from a logistical point of view, how difficult it is for people to start getting back to work? 

Gavin: One of the main questions for me is, we're still months in, and we still don't really seem to have a lot of reliable information on COVID 19, the virus itself, living on surfaces, whether it transmits through the air, how long it hangs in the air for.

concentrations in the air with a certain amount of people in a certain amount of space. It would seem, from what we've seen so far, that the more people you get into a more tightly enclosed space, the worse the outcome is. But I think that's about all we can semi reliably say by this point.

Now, obviously, public transport is one of the places that you see people crammed into the most of a morning on their way to work. So that makes it, you would think that would make it one of the problem areas for COVID 19 transmission. But yes, I think what came out of that interview is that we don't have any proof of that.

It's one of those things that sounds right, but we still don't really know. 

Jessamy: Yeah. And this sort of twist about aerialization puts that added sort of emphasis on ventilation, which. It's such an issue in so much of our public transport here in London. And, everywhere, really.

And how we might start to think about improving that is, is a real sort of minefield.

Over the last few months, you'll have been aware of a lot of future predictions taking place. And we thought it was an interesting question as to how these actually get done. How do you predict the future scientifically? How do models account for chaos? To find out more, Gavin chatted with Kathleen O'Reilly of the London School of Hygiene and Tropical Medicine about her work.

Gavin: And so I'm joined today by Dr. Kathleen O'Reilly, who is an assistant professor at the London School of Hygiene and Tropical Medicine. And there she specializes in the use of mathematical and statistical models to inform things like infectious disease control and eradication. So Dr. O'Reilly, thank you so much for joining me today.

Kathleen: Thank you for inviting me. 

Gavin: So let's start the real basics then. So you decide to make a new model to predict some aspect of the future say obviously in your case infectious disease control or eradication. Where do you start? 

Kathleen: Yeah, I guess this sort of comes from thinking about what the problem is.

And obviously we're interested in public health and infectious disease control. So as a biologist, I always come at it from asking, what's the biological question that I want to first of all, ask and ideally answer using a model. From that. It's all about developing a framework but that framework might just be as simple as what, what is a particular question that I want to ask and what information do I need to try and answer that question.

Gavin: So how do you, decide which kind of variables and information to use and exclude generally when you're making a model? What's the decision process? 

Kathleen: There's a number of things that you want to be thinking about. Once you've got your framework, you might have some kind of a a mechanism behind the process.

So thinking quite simply about the incidence of an infectious disease you might be thinking first question, is this a vaccine preventable disease? Are individuals vaccinated in the population? One of the key things that we would want to know is What kind of data do we have?

So is it incidence of disease, incidence of infection, and consequently, what the level of immunity might exist in the population? And this might relate to immunity from the infectious disease, or alternatively, immunity from something such as vaccination. So it's all of these kind of components, but.

Of course, one of the issues with this, especially when you think about it from a biological perspective, there's many different variables that are potentially going to impact on the thing that you care about. So then it's about trying to think big, okay, what are all of these different variables?

And then start to think small, which of these variables are the ones that matter? And you might know that answer ahead of time, but alternatively you might want to test this, and you might go about doing this using a model. 

Gavin: Yeah, so that leads me on to my next question. How do you assess the validity of a model once you have one?

Kathleen: It's it's part of this kind of thinking process is. Thinking in relation to research in general, it's part of the sort of research cycle, so to speak. So you've got an idea, you start to think about it a little bit more, you might discover, you gather and analyze your data, you have some kind of sort of end of chapter.

So you might be writing about your initial results and then you'll be like sharing and providing some impact of your research. But thinking about this as a series of little cycles that there's going to be an iterative And that would be it would include things such as validation.

So there would be some internal validation in relation to it just really simple tests perhaps initially. So if you had a model and you had some process for vaccination and you removed vaccination from the model, you might. expect to see an increase in cases. You would perhaps test that in your model and do some sort of sense checking initially.

And I think, any modeler doesn't work in silo by themselves. They always work in collaboration with experts in that particular infectious disease, for example. So there would be lots of. collaboration with other experts to establish, first of all, whether things make sense. And if things don't make sense, is there a problem of some sort, or is it a counterintuitive finding, for example it would be relatively circular process until you become confident and you've been able to illustrate it through validation and then you're starting to use the results from a particular type of model. 

Gavin: It seems to me anyway a lot of what models are used to predict seems quite inherently chaotic like for example inputting human behavior it is the kind of a way of accounting for this kind of Chaos, for want of a better word.

Kathleen: It it does depend that there are particular sort of properties. I guess the first thing to emphasize is that there, there are different types of models. For largely statistical models that there might not be chaotic properties. But if you're thinking about sort of mathematical models of infectious diseases, where you've got importantly dynamic property of previous infections affecting the instance of future infections chaos is part of that process.

So one of the interesting parts one of the things that I think of when I'm trying to explain chaotic theory is the application of infectious diseases to seasonality. So there, what you have in many settings is a very moderate change in, for example, temperature or humidity of a pathogen, which can result in quite, large changes in the incidence of infectious diseases.

And there, there's the properties of that model are reasonably well understood, but it can result in quite unusual behavior. For example, the work that Brian Grenfell has led over the last sort of 20, 30 years has looked at the seasonality and the cycles of measles epidemics, especially in unvaccinated populations.

And There, the sort of the non linear properties of seasonality, but also the effect of the proportion susceptible in the population is very striking, actually, and it's one of the good examples of chaotic properties of infectious diseases. 

Gavin: So I guess a very general question and you might be quite biased as somebody who works in modeling anyway But what makes modeling so useful?

How generally important is it when we will talk about scientifically thinking about the future. 

Kathleen: So I was really struck by mathematical modeling during my undergraduate degree. So at this time, it was when Creutzfeldt Jakob disease had emerged in the UK. And, at the time, the turn the early 2000s there was only a small handful of cases.

But because of the modeling that had been done by several groups, largely in the UK, and, an increased understanding of what the cause was, so the BSE epidemic in cattle in the UK. What was becoming quite apparent was that the cases that you saw at that moment in time were a function of what had happened, for example, I think it was about, between five and ten years ago.

And the consequence of that was that the expectation was that there would be many more cases to come. And, that's one of the reasons why a lot of cattle in the UK were destroyed because of the BSE epidemic, because of the the impact on both cattle and the potential impact in humans.

And for me, that was, a real sort of eye opener of what additional observations Modeling can use in addition to what you can see, in terms of current cases being reported and you can see that this kind of effect again and again, but it's the sort of the link between what you see now and what you can potentially see in the future by having a model of the underlying process as well as just what you're seeing now obviously another example of this, and it's obviously quite pertinent at the moment is the COVID cases that are reported in one week have an impact several weeks later because of the, the timing that individuals have in terms of their infectiousness, the subsequent acquisition of infection of other individuals and the timing at which they're likely to, for example, report to healthcare services or hospitals.

Gavin: Yeah. How has modeling evolved over the last few years in the field? Have there been any noticeable jumps forward, 

Kathleen: interestingly enough myself and other people at the London School of Hygiene and Tropical Medicine and at Imperial College we put together a sort of a special edition actually on in a, in another journal on how methods in modeling have changed.

And I think We started to think about this in quite a lot of detail and we were thinking prior to the 2000s, it was really quite a theoretical approach, but the use of models and the increasing use of models by infectious disease epidemiologists and also ecologists meant that the variety of models out there has really increased.

What has More recently, so from sort of 2010 or so onwards, has been the rapid change in how we use data. So I think prior to this time we were, modelers were always complaining that we didn't have the right data to put together a informative model. Now, it's almost like we've got so much data, we don't always know what to do with it.

And I think this is where fields such as data science, statistics, increased collaboration with the people who are collecting the data becomes really important so that we understand the inherent biases in how different data sets are collected, and how we should, as modelers, use that data inside of our models.

I think one of the other things that has happened. has been that, in a very good way, there are actually many different groups which have the expertise in infectious disease modeling. So COVID being a very good example, actually, there are multiple groups looking at pretty similar questions using a multitude of different models, sometimes getting the same results.

And hopefully most of the time getting the same results, but sometimes getting different results. And actually that needs. That as a field in itself in terms of multi modal comparison, how we communicate to stakeholders, that's also changed quite a lot. And it is quite an interesting area because, you start to think, OK, so how are we communicating the results of our models to people that care about the results?

And that, that has also really changed over the past couple of years. 

Gavin: It was a really fascinating chat to have with Kathleen to talk about scientific modeling. Now, obviously, modeling as in predicting the future, looking at trends has always been important. It's something that people have been doing for however long we've been making predictions about what will happen.

But this particular method of scientific modeling of creating these incredibly complex models make predictions about the future. As Kathleen said at the end of that interview, it's really progressed over the last decade or so and it's it's really become a vital tool and obviously it's become something really central to to our predictions of how decisions we make during the COVID 19 crisis will affect outcomes.

Jessamy: Whether you're a sort of big modelling fan or not, it certainly has a place in our scientific understanding. And I think we've seen that a lot through this sort of whole pandemic. Although there have been a lot, there's been a lot of modelling and a lot of varying results, for the most part, it's been valuable.

And it's allowed us, even if at times it hasn't been completely accurate, it's allowed us to, start a discussion and to start thinking about things in a scientific and logical way, even in the absence of having proper data. But as Kathleen said there, a model is only as good as the data that you put in.

So if you put in bad data, then you'll get a bad result. And the more assumptions that you make, then the more unlikely it is that it's not going to be accurate. So I think as long as, when we're reading these modelling studies, we bear that in mind. Then you can navigate your way through how much emphasis you should be putting on it and how much trust you should be putting on the result.

Gavin: Yeah, I think the saying isn't it, all models are wrong and some models are useful. None of them are these completely cast iron, this is absolutely what will happen in the future. It's a best guess, but our method of making the best guess has gotten a lot more complex and involved. And generally those outcomes, in terms of the guesses we've made, have improved.

And it's, it's a really fascinating area, I think, to to think about, because you see these, obviously, these headline figures coming in. There's going to be X amount of this thing happening over the next 10, 20 years. It's really interesting to think about the People in the lab somewhere inputting those predictions into a model and then seeing what comes out the far end and then testing it and revising and coming back with more Conclusions, it's a really fascinating area and I just wanted to talk about as well before we went the GBD population forecast paper that we put out recently so I spoke to Kathleen For some of the coverage of that paper as well, and it's got some really fascinating predictions that have gone against a lot of other models, I guess recently over the last few years, so You know, I think the kind of accepted wisdom of population is that it will just keep growing and growing and growing Population numbers around the world, but actually this global burden of disease Paper that we put out recently suggests that it's going to peak around 2050 and then actually world population will decline Up until 2100 and that's going to be driven by some really interesting demographic trends a surge in population in sub saharan Africa and a decline in India and China, which is a new addition to the literature almost 

Jessamy: yeah, I think also what's so interesting that came out of those papers is this concept of the changing demographics of our population.

I thought particularly what was interesting is that they still kept a working age of 18 to 64. And with that working age, they said that actually by the time we get past 2064, when population in the Western world starts declining, we're going to need immigration to fill our workspace up. And I completely agree with that.

And obviously I would, promote that thinking. But I do think it's interesting That we would still keep working age at 64. I think the accepted sort of Wisdom for the most part is that we're all going to be working until we're much older than 64 and from that point of view, I think societally we're gonna have to change, to allow that kind of huge increase in life expectancy Otherwise, what are we all gonna be doing?

Gavin: We can't end, we can't end a podcast on that kind of downer. It's still decades until retirement, Jessamy, come on. 

Jessamy: No, I don't want to end on that kind of, I'm just talking, I'm just, what do you, what do you think about that? 

Gavin: I think for sure. And I think as well in future podcasts, we'll be talking with Cassandra Coburn, who is the new editor of the Lancet's shiny, exciting new journal, The Lancet Healthy Longevity.

And we can chat those sorts of trends then for sure. But, Yeah, definitely. I think with the increased life expectancies comes actually the increased ability to work until later in life where with no real serious side effect at all. So absolutely, that's something I'm sure that will be that will be looked at and addressed as we go on and as this potential.

Working age versus pension age crisis unfolds as was predicted in in the global burden of disease paper. The The fact that the model which currently has several people of working age supporting one person of pensionable age is going to be massively overturned in the next few decades. Is something that we will need for sure to have policies in place to deal with.

Jessamy: Yeah, it's such an interesting change that's going to happen in our working lives. And, the kind of concept that we might all have education and then work and then retire and then having to, adapt to having different stages in our lives where we're going to have to retrain and, maybe take time out and all of those things becoming more accepted, I think is.

It's such a kind of interesting place to start a conversation from this sort of in depth and really fascinating modelling study. 

Gavin: Thanks for listening to this episode of The Lancet Voice. You can contact us with any feedback on podcasts at lancet. com and we'll see you again next time.