Mumbai: Seven in 10 COVID-19 cases did not lead to another case, pointing to what looks like super-spreading, according to a study of 575,071 tested contacts of 84,965 confirmed cases in Andhra Pradesh and Tamil Nadu--the largest study of its kind, published on September 30.

Children also transmit the virus, mostly among the same age-group, the study, led by Ramanan Laxminarayan, founder and director of the Centre for Disease Dynamics, Economics and Policy, found. This “will have implications for whether we want to reopen schools”, he said in this TEDx Gateway webinar on August 29, where he presented preliminary findings from the study. About half the COVID-19 deaths among the elderly (85 years and above) happened within seven days of testing, he said. Andhra Pradesh and Tamil Nadu conduct as many tests on a per-capita basis as many developed countries do, he added.

The risk of zoonotic disease is concentrated in south and southeast Asia, and the rapid rise in consumption of poultry and other animals in India means that we are capable of breeding the same kind of infectious diseases, he said, adding that we need to put systems in place to catch these viruses at an earlier stage.

Much of what constitutes public health has nothing to do with doctors, vaccines or antibiotics, but it really has to do with water, sanitation and hygiene, Laxminarayan said. We need to invest far more in systems to be able to respond to such outbreaks, invest enough time in enhancing the population’s understanding of health--which drives behaviour--and use technology to help in early detection of outbreaks, similar to a weather-warning system, he suggested.

Edited excerpts from the discussion:

The numbers of COVID-19 cases and deaths in India continue to rise, even as they decline down in some parts of the country. What is this telling us in terms of our ability to forecast the life-cycle of this disease at this point?

It's a really tough question to answer, simply because the response is completely different, and [because of] the reactive lockdowns going on throughout the country. Looking forward, the only certainties we have are the same certainties that we had back in March. One is, everyone is susceptible until they've had an infection--and even then, we don't know how long [they are immune]. And number two, the reality is we have a large country. A lot of our population is still uninfected and a lot of them are in rural areas, in places without healthcare facilities. I think we still have some way to go. We might not necessarily visually see this in the way that we saw it in Delhi or Mumbai or Chennai, but that pain of the disease is going to continue because the disease is spreading all over the place.

Is there anything from the data, including in previous pandemics, to suggest that at some point it fades out? And if it faded out with much less medical intervention and systems intervention in the past, then is it likely to fade out in a similar timeframe now?

It's interesting. The 1918 pandemic was a flu pandemic: People who had been exposed to a milder form of that similar flu the previous season were not as badly infected. It was a new strain, but of an influenza virus to which people had some prior experience. In that sense, we don't actually have an experience with a widely transmitted virus, just through respiratory contact, in our living memory. There's nobody around who's ever experienced that.

So hypothetically, the cap [for the susceptible population] for 1918 flu was not the entire world--it was always going to be something like 25% [of the population], because a lot of people had been exposed before, whereas hypothetically, the cap here [with COVID-19] is the entire planet--until we get some sense of herd immunity. That's why we're in a very different bucket now.

What are comparable mortality and morbidity rates for H1N1 in India and hence, are we overreacting to the COVID pandemic?

H1N1 was a more transmissible disease, but had a lower mortality rate. I don't think we're overreacting. Let me put it this way: There are many things we could have done differently. If we had great data, we might not have had the need for the initial lockdown, for instance. So this is not to justify the way in which we have approached the pandemic, but it is to justify the seriousness with which the pandemic has been taken in India, which I think is entirely appropriate.

You've been studying the likely spread of the pandemic in India. As days pass and the number of cases continue to rise, what's your assessment of the number of people infected now?

I'm triangulating between sero-prevalence studies, and the numbers of infections reported at that point in time. Depending on place and time, we're undercounting infections by a factor of anywhere between 50 and 150, which means that for every infection that's actually reported, there are probably 50 to 150 infections that are missing. So I would say, as a rough bet, that there are probably about 200 million infections out there [in India] right now--which is when you have 2.5-3 million [reported] cases, and we're under counting by a factor of, say, 80 to 120. It is definitely not 3 million infections right now. I hope everyone understands that. [Editor’s note: The interview was conducted on August 29, by when India had reported about 3.5 million cases.]

What do you make of modelling studies--which have either underestimated or overestimated the spread of COVID-19?

Modelling studies are meant to give you a sense of what will happen if you do nothing. There is no modelling study that can tell you what will happen if you do various things; how can I predict what all you will do? So, the earliest modelling studies--and I would point to two of them, one by ICMR, and one by our group--both gave some predictions back in March of what would happen by July, which was mostly that something like 20-25% of India could be infected if there was no response measure whatsoever.

Obviously, in response to the modelling, a lot of things like the lockdowns have happened. So naturally, the disease goes on a different trajectory. But that doesn't take away the value of the modelling or of the intervention strategy. This is like a doctor telling you that if you don't lose weight, you will have a heart attack by the age of 65. If you then decide to lose weight and do not die by age of 65, that's a good thing. But it doesn't take away the value of that prediction. These are built on a lot of science.

You're saying that they seem to have more or less rightly estimated, at least from what we know today. Is that correct?

Well, today, I'm telling you that we have probably about 200 million infections as far as the sero-prevalence surveys go. You can take a range, say 100 to 300 million, because we really don't know [the exact numbers]. That is in the range of what the models predicted. And that's in spite of having what is described as the harshest lockdown in the world. So I don't think the modelling studies were off by very much at all.

You've emphasised the zoonotic nature of the source of the disease. Even in India, there are cases of wild meat markets and potential sources. Is that something that we can fundamentally change, because you seem to be, in a way, making a case for vegetarianism.

I'm not making a case for vegetarianism, but I'm certainly saying that meat-eating at the scale that is happening on the planet does have environmental consequences. It has climate consequences: Something like one in six tonnes of carbon emissions actually come from animals. And it does have this disease consequence as well. It is like all other risks.

But I'll say two things: One is, I think that eating wild animals is probably a bad idea. I think that is universal anywhere, given the risk that you now pose to other people. With respect to domesticated animals, I think that in fact, the Chinese government is advising people to eat a little less meat. You don’t have to eat meat at every single meal. I think meat plays an important role in people's diets without question, but we need to be mindful of the fact that there probably won't be enough meat for everybody to have meat every single day, twice a day.

You have also said that we should catch the viruses early--before they jump to humans. What can we specifically do there, from the lessons so far, if we can start acting even at this point?

If people really remember, the Indian Council of Medical Research (ICMR) is actually set up to do exactly that, not respond to outbreaks--which is actually the job of a different agency called the National Centre for Disease Control. ICMR has now taken over this role of responding to the outbreak. But ICMR’s job is to actually help look for and predict and warn us about these outbreaks.

We spend abysmally low on medical research. We spend something like $200 million. This is chump change for a country the size of India. So I think, given the size of our economy, where we stand in the world, and the risks that are faced by our own people, we need to invest a lot more in that kind of research. Otherwise, we're going to face another Nipah virus outbreak or Zika might show up, etc. But I think that cadre of biomedical research just does not exist at the scale that a country of India’s aspirations really deserves.

To what extent is the research on epidemiology in India at par with the developed world?

Unfortunately, this is a place we've not spent much effort in training people. Though we have a lot of the world's population, we don't have the epidemiologists to keep up with that quantity of people. It is a skill that is sadly missing in the numbers that are needed in India. We need far more.

How can each country contribute in a way to solving the larger problem of the pandemic?

At this stage, realistically, a vaccine is the only solution. Now, we might get to a stage where most people are infected and therefore have some level of immunity, and so we might end up in a different place. But a vaccine is really the only solution. Here, my plea would be to not screw it up. It is much better to get a right vaccine which is safe, effective, trusted and widely available--all four of those things [are important]--before we get it out there, rather than rush to market with some vaccine that has not even gone through phase 3 trials and risk destroying people's confidence in vaccines. We're playing with people's expectations here and telling them that we have a serious problem and a vaccine is what is going to save you. We cannot afford to put a vaccine out that has side effects or something that you discover will have some long-term health consequences 10-15 years down. We need a fast vaccine, but not at the cost of safety.

You wrote that the US spent $5.4 billion fighting Ebola, from which only one American died. The money was spent and, presumably, a lot of preparedness happened, but that hasn't helped the US nor has it helped the world.

In the case of the US, obviously, some things have changed since Ebola. They have a different president. So that makes a bit of a difference. But I think it's quite telling that they were willing to spend that kind of money to prevent this disease from entering the US. And the response this time has been so different.

It's hard to second guess. But I'm second guessing because these are things that I said and others said back in March; if we had tested at the rate of 200,000 tests a day back in March, we could have identified the hotspots of COVID-19, and stamped out COVID way early. The lack of testing at the outset could not be compensated by the scale up of testing that has happened much more recently.

The UK, Italy and a few other European nations also suffered. They are developed nations with a much larger budget on research as well as healthcare. Why couldn't they prevent this?

This question keeps coming up because I think we don't remember that we got the epidemic later than they did. Italy got it and they were caught on the upswing, because they didn't even know that they had a pandemic. The US was in a slightly better situation, but then they botched it up in a massive way. We were very lucky to have had time compared to those two countries. But it also points to the fact that having wealth alone is not a predictor of whether you can actually tackle a pandemic. You need public health and you need leadership. And that's what the lesson from the US really should be, that wealth doesn't protect you; investment in public health does.

What is your opinion on the Swedish model of response to the pandemic?

People don't actually understand the Swedish model very well. The Swedish model was not that everyone was roaming around and interacting with each other. Swedes are very compliant. So if you suggest ‘stay at home’, they all stayed at home. The part where they didn't do as well was in preventing the disease from getting to their senior citizens in elderly homes where it caused a lot of death and damage. So they had an idea of a policy, but I don't think it was implemented in a way that protected the elderly. And our lesson from there is, let's keep a close eye on our elderly in reducing mortality.

Is there a good method of data planning, collection and analysis that could be an adaptive template for rapid response teams?

We have one. It was put out by the WHO after SARS, and India follows a version of it, which is run by the National Centre for Disease Control. Our problem is that we don't provide the data back to researchers to study. We're happy to work with states which might help. But that system exists.

In a country like India, how reliable is the data collection, especially from villages and particularly now, given factors like floods in many parts?

I would say India doesn't do that badly when it comes to data collection. With respect to things like contact tracing, all of these we've done well. The place we don't do well is really in mortality assessment. We don't assign a clinical cause of death even in the best of times for four out of five deaths. And that's where we lose a lot of information.

We have very little data transparency nationally on the COVID pandemic. I think with the researchers that we have and the data, we should be doing far more in helping clinicians with evidence that helps them treat their patients. And I feel that the amount of clinical research and epidemiological research that has come out of India is not commensurate either with our capacity for that research, or with the size of the country and the datasets we have.

Your study said half the COVID-19 deaths among the elderly happen within seven days of testing. Any insights on why this is happening?

We see this in other parts of the world as well, that a lot of people show up and then, within the first 72 hours, they're gone. It's possibly because they have severe respiratory distress that they came in with. It's also a function of when they were found to be needing hospitalisation, and when they actually came in. It points to a lot of people who are arriving possibly at hospitalisation too late. The silent respiratory distress [happy hypoxia, where oxygen saturation drops without necessarily a person feeling uncomfortable or breathless] is possibly also behind it. So if we can push that window a little bit forward and bring people in a little earlier, I guess we might be able to save some of these lives.

How can I predict which elders I should be taking care of?

The big predictor is comorbidity: If you're able to tell who has diabetes, hypertension. Particularly in India, we find in these data, something like 80% of people who die actually have these comorbidities. So it's a good predictor.

Are Tamil Nadu and Andhra Pradesh doing anything correctly to handle this situation?

Both states have an early reach-out programme. They hand out pulse oximeters at a wide scale. These are all quite helpful.

India has a crude death rate of under eight per 1,000 population. How is this year's mortality compared to previous years? Are we seeing a scale of deaths way beyond the normal number of deaths in the country?

The point is well taken--which is, is this really going to be such a massive blip on our overall mortality? It could have been, if we had not responded. If there had been no lockdown, etc, that could have been the case. But I think that the response has been good. And that's held the mortality rate down significantly, which again, is a point that deserves repeating.

Coming back to your point about the need to invest in public health, how do we acknowledge this and put more emphasis on this going forward?

I think the main takeaway is that what is public health in times without the pandemic is good pandemic response--that is the sum of it, which means that if we're able to prevent infections from transmitting during normal times--whether it's tuberculosis or malaria or whatever else--then we will do far better during a pandemic.

As a country, we don't understand public health. It has very little to do with medicine. Public health has to do with things like engineering--having clean sewers. Most of us don't know who the sanitation engineer is, in our area. That person is the one who controls our lives and our destiny to a level that we don't actually understand. We don't give importance to the things that actually make a difference in our quality of life, such as water, sanitation, hygiene and air quality--that's really what makes our quality of life better. That's what makes developed countries what they are, that's what differentiates us from them. And I think it's time we took public health in that way really seriously. It is time for us to put the attention on reducing infectious diseases even during normal times. And if we did that, we would have a much better system to deal with the next pandemic when it goes wrong.

Update: This story was updated on October 1 to include the findings from the interviewee’s study published on September 30.

We welcome feedback. Please write to We reserve the right to edit responses for language and grammar.