Human behaviour biggest uncertainty for coronavirus models

India hasn't gathered enough data to gauge the scale of community transmission, say clinical scientist Gagandeep Kang and virologist Shahid Jameel.

Subhra Priyadarshini

doi:10.1038/nindia.2020.53 Published online 30 March 2020

[Edited excerpts from a webinar hosted by Nature India, Translational Health Science and Technology Institute, The Wellcome Trust DBT India Alliance and the IAVI on 24 March 2020. Gagandeep Kang is a clinician-scientist and the executive director of the Translational Health Science and Technology Institute, Faridabad and  Shahid Jameel is a virologist and the chief executive officer of Wellcome Trust DBT India Alliance.]

A home quarantine notice in front of a house in north Delhi (name and address smudged to protect the identity of individuals).
Q: There's a conspiracy theory still doing the rounds that the COVID-19 virus escaped from a bioweapons lab in the United States or China.

Gagandeep Kang: The conspiracy theory is quite widely circulated. People do say the virus was released from the Wuhan Institute of Virology. But I have to tell you that the Wuhan Institute of Virology has outstanding virologists. The quality of the work that they do is very good.

The other conspiracy theory is about the US actually taking the virus to China and releasing it at the military games held in Wuhan in October 2019. We now have evidence from a lot of sequencing studies. One of the good things about this particular outbreak has been how ready people were to share sequences from viruses around the globe. And those sequences have been analysed by people who really know what they are doing. Now, the nice thing is that, if a virus is modified, you can tell because there will be signatures, the ways of modifying a virus are understood by virologists. So you will be looking for evidence of modification. And people have done that in multiple groups now. And they have said that this is not a modified virus.

This virus has evolved over time, as RNA viruses do, and its link to viruses that are found in animals is evident, but its link to being an engineered virus are not. This is based on the best analytic techniques available today from reputed credible biologists from multiple laboratories. There's no evidence that this is actually a bioweapon.

Q: There's another Whatsapp message floating around that pandemics hit in the 20s every hundred years — the plague of Marseille happened in 1720, the great cholera outbreak in South Asia and Southeast Asia happened in 1820, the Spanish Flu in 1920. And now we have Covid in 2020. Much as this sounds like a fascinating coincidence, would you like to shed some light on this?

Shahid Jameel: Well, I don’t remember about 1720, 1820 or even 1920. But I have read a little bit about 1920. And most of you would have read that too, that Spanish Flu didn’t actually start in 1920. It started in 1918. So it lasted from 1918 to 1920. And it was soldiers returning from World War I that took the virus around the world. So if we go by this whole theory about a pandemic happening once every 100 years, I think we should have been ready. We should have prepared our hospital capacity, we should have prepared everything.

Having said that, I think it’s important to understand that when you disturb the ecosystem, these things will happen. When you keep destroying forests, there will be more animal-human contact. And because of that there will be a higher probability of viruses in animals getting into humans. We have evidence that glaciers are melting, the permafrost is melting. What will come out of these glaciers, we don’t know. We don't know what’s hidden under them. So I think if there is any lesson in this, it is that we should be taking care of our ecosystem, we should really be taking care of our world.

Q: People are curious about when we will see the peak of the outbreak in India? What are the models prevalent now? Which are the ones we should believe? How many people in India would get infected and how many would likely die of the pandemic?

SJ: You have to understand that models are exactly that: they are models and they are based on certain assumptions that are valid at that given time. Models always fail where there are uncertainties. Human behaviour is the biggest uncertainty. The virus exploits human behaviour and, therefore, that brings in a big uncertainty.

I have heard extremes. One model says about 30 to 50 crore Indians are going to get infected and there will be 10-12 lakh deaths, which essentially means that anywhere from about 25% to 40% of our population will get infected. I really feel that is a bit extreme. I have seen a reasonable model by a group called COV-IND-19 Study Group which is made up of some biostatistics scholars based in the US, mainly Indian students and faculty. Their predictions are that in the best case scenario, there will be roughly 60,000 to 65,000 infections, roughly 4,000 deaths. In the worst case scenario, it will be about 21 lakh infections and about 13,000 deaths. But again, I will say models are models and they are based on assumptions.

GK: I have been involved with some modelling groups and I have spent a fair amount of time sitting on an advisory committee at the World Health Organisation that views the models. One of the things to remember is that models are models. And in India, we actually don’t have a lot of data. Because all of the data that we have from patients is really based from the testing we have done so far and that testing has been limited to imported cases and their contacts until last week. Last week, we had reports of a few, very few cases where there was no known contact. 

When you look at building a model, you need to know what is your population, who lives where, what is the structure of that population, who interacts with whom. Now in terms of structures, occupations, general nutritional profiling, we have a lot of surveys that have been done in the country that inform the building of models like that. You can look at the National Family Health Survey. You can look at Census data. But if the only information that you have on disease is based on incoming travellers, actually what you are modelling is what is happening outside the country.

If you look at the models that are currently available, three models have been highlighted. It is great that these models have been built. But what they are using is assumptions based on data coming from other parts of the world. Given that this is a viral infection, you don’t expect it to behave differently in different populations. [But] the one piece of information that is important for transmission models that we don’t have in any shape or form from this country is what is called social mixing. Who interacts with whom for how long, which would be important to refine these data. Most of the social mixing that is in these models is derived from other populations. And we know that we have much greater density in our populations, which means transmission is likely to be high.

India should be looking now to generate data that will better inform the existing models. 

Q: What do we know about the behaviour of the virus and how is it behaving in different age groups of people?

GK: The initial thinking was that children don’t get infected. But now we know that children do get infected at the same rate as other age groups and seem to see milder disease. As age goes up, you see more severe disease in individuals. We know that as you get over 60 (or 70 or 80), when you get severe disease, mortality is also higher. But the virus doesn’t behave any differently. It is really the response of the host or the ability of the infected person to handle the virus that determines the severity of the outcomes. And what we are seeing in the most severe disease – respiratory distress, multi organ failure – is a factor of age.

For some reason, when you have a lot of chemical modulators being released during infection, you see that younger people seem to tide through it better, not all of them because we know that a smaller proportion of young people can also get severe disease or severe symptoms and die, but the proportion is higher as you get older.

SJ: The age effect that we are seeing is not something that is unique to this virus. There are many other viruses that show very mild or asymptomatic disease in children. But as you grow up and you get that infection, you get severe disease. The two examples that I can think of immediately are chickenpox and Hepatitis A. You get chickenpox as a child, you get over with it fairly quickly. You get chickenpox as an adult, you have a very bad course of infection. 

The second is hepatitis A. People who are growing in India, almost everyone by the age of 10 has had a Hepatitis A infection, and the antibodies they get protect them for life. But if you grow up in an environment where you drink bottled water, and you do get this infection when you are an adult, Hepatitis A can be a very bad infection. 

Now as far as old age is concerned, you have to realise that older people have accumulated a lot of other diseases, they have high blood pressure, they have bad lungs, either because of growing up in polluted cities or smoking, and this virus affects the lungs. So co-morbidities accumulate as you grow older and that’s probably why you are getting far higher mortality. 

Q: How effective has the Indian government’s response been so far? And what more needs to be done to curb the spread of the virus here?

GK: In terms of the government response, I think the thing to understand is how quickly you need to move in a pandemic. And we have seen over the last few days, a rapid acceleration in how the Indian government is responding. If you look at what governments have done in, say Singapore or South Korea, they have not shut the country down. They have focused on an approach where they essentially tested lots and lots and lots of people. They identified and isolated the infected people, and they made sure that their contacts were followed up. 

If you look at what our government started to do, it was exactly the same thing, identify infected people coming in, follow the contacts and make sure that things are going well in terms of isolating the contacts and following them up. But very quickly after that, we started to see stories about so-and-so escaping from quarantine, people hiding their travel histories, a lot of things happened that made us very insecure about the ability to identify all cases that are coming into the country and the ability to trace all contacts. 

So, in response to that, we are now in a situation where we are trying as much as possible to limit people interacting with other people. The sources of transmission are largely young people and working people: it is the young in schools and in colleges who have a lot of interaction with each other, and working people, whether it is formal or informal sectors. The idea of the shutdown, the idea of limiting travel is to stay in place. Stay in place for as long as it takes for disease to manifest itself. 

And then we need to be ready to test. For two weeks, if nobody is going anywhere, anybody who developed symptoms will be tested, we will find whether they have coronavirus or not. And because they have stayed in place, the number of contacts that they have is relatively small. So then you will be able to quarantine their contacts. If you let everybody move around as much as they want, contact people as much as they want, this will rapidly reach a stage where we won’t be able to test people. When people get sick, we won’t have hospital facilities available, we won’t have enough ICU beds available. This is the big worry. 

Right now, because we don’t have data, we don’t know what we are dealing with. So, the lockdown that the government is doing is exactly the right thing. What we need to get out is the message to people everywhere, as far as possible, treat your own self and your family as a potential source of infection. Try not to infect other people. If we can do this initially for a short period of time, that will buy us enough time to build the systems that allow us to test and to treat. 

My biggest worry right now is what it means for people who live in places where isolation is difficult. Where all the instructions around hand washing are difficult, where they may not have a toilet. How do you communicate with them? What kind of messages do you give to them? And I think [it is important to be] thinking through what advice we can give that is practical, that is actionable at a very, very granular level. If you don’t have water, what can you think about doing? You could think about designating paper towels or a piece of cloth that you use to touch frequently touched objects. If you can’t wash your hands 10 times a day just because you don’t have enough water, wash the cloth and then be careful about what you are touching. 

Watch the full webinar here: 

Q: In terms of research, what should be the main priority for India right now? Should we focus on diagnostics first or vaccines or building capacity in ventilators?

SJ: We are in the middle of a pandemic. So the most urgent thing is to do capacity building. We have to ensure that we have enough hospitals and enough personal protective equipment for our health workers, without which they are at very grave risk. We need to get ICUs ready, making sure there are enough ventilators. There is no doubt in my mind that in this situation, we need to build capacity to treat people. That should be our top priority. 

For research, I think the most important thing is cooperation, collaboration. This is an opportunity to do some really good research and there is a low hanging fruit out there. The world has now put over 950 different strains of the COVID-19 virus in the public database. A country of India’s size and capability has put only two. We have no idea about how the virus is evolving in the country. And it is not because there is not enough expertise in the country to quickly sequence and do the bioinformatics. We have the capacity. It is just that nobody has access to the virus. The virus is there with the NIV (National Institute of Virology) and ICMR (Indian Council of Medical Research) and nobody else has access to it to be able to sequence it.

For example, there are labs with containment facilities that can grow the virus. And this is, by the way, a fairly easy virus to grow in culture. Now, imagine if 10-20 labs in India, have the culture going. Each one of them could be testing a whole bunch of molecules and somebody may come up with a very nice hit. Somebody may come up with an existing drug that works against COVID-19.These are opportunities we always lose, because there is reluctance in the country to share.

Q: Is there a high probability of community transmission already having started in India?

GK: I think that it is likely that we have community transmission given that we have cases without known contact history. The extent of that community transmission, we don’t know yet. Speaking from just looking at whether hospitals are admitting large numbers of cases that have severe respiratory infection patients who are on ventilators, I think every hospital always functions at capacity. So it is a little difficult to test (more widely) until we start to get data from testing patients who are currently admitted with severe respiratory infection. 

We must remember that the testing regime was liberalised only last week, and laboratories were unable to test until the notifications went out over the weekend. So in the next week or so we are going to have a lot of laboratories ramping up capacity to test. I hope that they will be testing all of their severe acute respiratory illness patients that will give us a sense of how much is out there.

But after we get going in hospitals, we also need to be looking at what testing needs to be done in the community. This will give us a sense of how big the problem is. And as we keep doing it over time, we will get a sense of trajectories as well. That will better inform models and allow us to make predictions about where we need to be emphasising stronger responses.

SJ: I do believe that community transmission is happening. We are right now unable to measure it because we are not testing at scale. And it is really unreasonable to believe that in an outbreak like this, community transmission won’t be happening. If you simply look at the numbers that are being reported on a daily basis, and see how steep the curve is becoming, it’s hard to believe that community transmission is not happening. But we shouldn’t be alarmed about it, we should gather evidence and we should build proper contingencies for it.

Now that more labs are going to be testing, one hopes that there will be more reliable data in terms of numbers, because for a country of our size, the testing really has been limited.

[Nature India's latest coverage on the novel coronavirus and COVID-19 pandemic here. More updates on the global crisis here.]