Coronavirus crisis in India has not exploded yet: Bhramar Mukherjee

Bhramar Mukherjee, professor and chair of biostatistics at the University of Michigan, is leading a team of researchers that is using epidemiological modelling to do a situational assessment of the coronavirus crisis in India. In a conversation with Ruchika Chitravanshi, the biostatistician says the peak of virus may hit India by mid-May and a 42- to 56-day lockdown can help in flattening the curve.
 

We crossed 14,000 positive cases and several health workers are also infected. India has maintained we are not in the community transmission stage. What is your assessment?
 

Even if there was community transmission, you have already taken the hardest measure of social distancing. What you really need to work on is testing in high-contact and high-density areas so that you can easily track the pulse of the epidemic. This is not just an investment in the short-term tactics. The lockdown is not going to be an on-and-off discrete transition. Even if you come out of the lockdown, this is key to containment. The goal of the lockdown is that you have a manageable number of cases even if community transmission happens. And you can actually do cluster testing and really contain those. The only way to do that would be to come up with clever strategies of national testing. And there are alternatives. Instead of testing, say in rural areas, people can report their temperatures every day. Health status surveillance is going to be key, regardless of whether community transmission has happened or not. If I see this as a public health policymaker, this will help us to prepare our health care system. Until a vaccine is there you have to keep chasing.
 

How long do you reckon it will take to flatten the curve?
 

According to our calculations, a lockdown for a period between 42 and 56 days is going to really flatten the curve. There are concerns of collateral damage such as people can die of hunger or there could be the non Covid-19 health care issues. If the government can sustain the food, essentials and health care supply chain, I think it will take one or two months of staying at home — hard, but still doable. The right metric to decide if you can open is whether you are prepared enough to be open — psychosocially, health care capacity and testing wise. Do people recognise that there will be some form of suppression for quite some time? Because as soon as you lift it, there will be new cases.
 

What are the trends you are noticing in the new data in the light of extended lockdown?
 

There is a high uncertainty associated with all these numbers. None of our projections have actually exceeded the upper confidence interval. You do see a lag... if the lockdown started on March 25, you at least need two to three weeks to see the peak.
 

When do you think we will see the peak in India?
 

All our calculations (her group had initially predicted in March that India would have confirmed cases in the range of 4,836 to 28,925 by April 15) were based on 21-day lockdown. Now that the lockdown has been extended, the peak will be delayed further. We are waiting for the new models to run. You need about three weeks to see a decline in the number of new cases, and then another 10 days to see the decline in the number of deaths. The second half of May is when we expect to see it but in terms of reported number of cases. The thing is that as cases rise, you’re also increasing testing. So even if the true number of cases is going down, it’s the result of testing. The problem with India data I have is that I'm doing blindfolded modelling. One does not know whether it's the number of cases which are rising, or that you are doing more testing. But then, there is no way to decouple the two.
 

Our lockdown is till May 3 as of now…
 

You should see it turning the corner in the middle of May. First you start seeing a gradual decrease in the rate at which the curve is progressing. There will be a slowing down before you see the fall in the curve which we call turning the corner. About 14 days of constant decline in the number of new cases will serve as evidence of that, in a definitive way.
 

How do you reach a figure when it comes to the ideal number of tests per day for a country like India?
 

This is a critical question. Because maybe there are places where the community spread is happening, but it's silent. In the US, they are doing 100-2,000 tests. India is around four times bigger in population. So the reverse question is what is your maximum testing capacity and how should you distribute them. Data is actually now peaking in India but at a slower rate.
 

How can we extrapolate to measure the true incidence of cases in India?
 

I think by making an estimation on the basis of the number of deaths. Many epidemiologists have actually agreed that fatalities, if they are reported correctly, are a better benchmark to extrapolate that because mortality calculations are simpler than testing calculations. So, for example, in South Korea, we saw mortality of about 1.7 per cent. And in China and Italy, we saw a mortality rate of about 4 to 5 per cent. That keeps on changing, because this is not the peak period. But from the mortality data, you get a range that helps you back calculate.
 

What do you think of the correlation of Covid-19 with the climate and temperature?
 

There is some suggestive evidence of declining incidence with temperature, but it's not as strong as to say that something is going to really rescue us. This summer belt is confounded with so many other things. There are countries with different vaccinations, so there have been several hypotheses about each structure of a community. The use of the BCG or the anti-malarial vaccines is also correlated with temperature.
 

But I think the ultimate question is, are we ready? When do we come out of it? Do I have all the ducks in a row, in order to chase contact cases? For example, in our paper we have suggested syndromic surveillance in every hospital. We have asked them to monitor the case for the last two years for respiratory illness, high temperature and see how it is changing. Then immediately if you are suspicious, you can start some testing. These antennas have to be built which should go up at the sign of trouble.
 

If there is some form of post lockdown suppression, in terms of people maintaining social distancing guidelines, practising proper hygiene, that will maintain a long-term effect of the lockdown regardless of its length. So I think it is important that you know everyone does their part to make this work.
 

What is your view on India’s current strategy towards combating and containing Covid?
 

One thing which is reassuring is the testing data in India. In many other countries, as the pandemic is exploding, the fraction of positive tests goes up as a fraction of total tests. So, the positive figure in India has stayed relatively flat. I have not seen a sharp increase in the number of those tested positive. It tells me that the situation has not exploded yet. And all of the right public intervention measures were taken. India did a very good job, and unlike the US, closed its borders early on. If anything is going to save us this strategy will. I do think that things are going to get better.