Cannot emphatically say India has passed the COVID-19 peak

Dr. Gautam Menon and Dr. Giridhara R. Babu discuss whether India has gone past the peak. Dr. Menon says in major cities in India and in some districts adjacent to those cities that already have sufficient numbers of people infected with novel coronavirus, the number of new infections is going down. But one can’t say that a bigger or an equivalent peak is not in store for India as the virus has not been spreading uniformly in India. Dr. Babu says that it is not a single wave at the national level which we should be looking at; instead, we should be looking at the State level. We should also be looking at the data from serological surveys, verbal autopsies and missed deaths from the States that are not reporting well, to understand the real nature of the peak.

Since mid-September, the number of daily new COVID-19 cases recorded has been reducing from the peak of 90,000-odd cases. A few days ago, the COVID-19 India National Supermodel Committee, constituted by the Department of Science and Technology, based on mathematical modelling studies, concluded that India had passed the COVID-19 peak in September. It also found that there will be fewer than 50,000 active cases from December and if proper safety protocols continue to be followed, the pandemic can be “controlled by early next year” with “minimal” active symptomatic infections by end-February.

In a conversation, Dr. Gautam Menon, Professor of Physics and Biology at Ashoka University and co-author of COVID-19 modelling studies, and Dr. Giridhara R. Babu, Professor of Epidemiology at the Public Health Foundation of India, Bengaluru, and a member of the Karnataka COVID-19 Technical Advisory Committee, discuss whether India has gone past the peak. Edited excerpts:

The number of daily new cases recorded has been reducing from the peak of about 90,000 cases since mid-September even as the number of tests done each day remain at about 10,00,000. Does this suggest that the pandemic has peaked in India?

Dr. Gautam Menon: I think what it suggests is that in the major cities, and in some districts adjacent to those cities that already have sufficient numbers of people infected, the number of new infections is going down, which is to be expected. I don’t think one can say that a bigger or an equivalent peak is not in store for us, because my feeling is that the virus has not been spreading uniformly in India. There are many districts where the number of cases remains small. So, it’s these cases that we will have to watch out for, especially after the festival season.

Dr. Giridhara Babu: The most dense areas might have already been affected, whether it’s the metros or the tier-2 and tier-3 cities. While the infection is spreading temporarily from one area to another, we cannot certainly say that the worst is over yet. Until we touch the baseline, which is rock bottom, we will not be able to know whether another peak is visible. In my experience with measles and polio, even a slight deviation from the cohort of susceptible people versus how the infection is spreading will ensure that a bigger peak comes much faster, especially when people have been hiding away from the virus. So, we will have to see how it plays out during the winter and afterwards, especially when people start mixing with young people who can spread the infection to the elderly and to those who have co-morbidities. It is not a single wave at the national level which we should be looking at; instead, we should be looking at the State level. We should also be looking at the data from serological surveys, verbal autopsies and missed deaths from the States that are not reporting well, to understand the real nature of the peak.

Except Telangana, the other southern States have been reporting a high number of daily new cases, while large populous States such as U.P., M.P., and Bihar have reported far fewer cases. How do you explain this enigma?

Dr. Babu: Absence of evidence is not evidence of absence. Just because the southern States or some other States, including some in the north, are reporting more cases does not mean that only these States have circulation going on and other States have controlled it. Saying that only a few States have higher transmission compared to others is a statistical fallacy. This is the biggest trap we have seen in many infectious diseases. Due to integrity of reporting cases, many States get classified as high-risk States. The same thing happened, for instance, in the HIV programme. We should not repeat that mistake. We should build strong and resilient systems of data collection and reporting.

There are many ways of assessing whether a State has already gone past the surge in cases or whether it is yet to surge. One of the key things I would like to see is State-level serological surveys throughout the State, and also pick some infections in the acute phase. Testing randomly throughout the State will help us know the case-to-infection ratio, and comparisons between reported cases versus those who have already developed antibodies. Knowing what proportion of people in the general population have positivity by doing random antigen tests and RT-PCR and verbal autopsies of deaths will tell us whether the State has gone past the surge in cases or is yet to surge.

Dr. Menon: Data are vital. What we need is much more granular data that will give us a better idea of whether it is just lack of reporting or insufficient testing that is governing these numbers or whether there is a genuine reason to be happy.

But some of these States, like U.P., have been doing a high number of daily tests, yet the cases are low. Is it because the tests are not directed because there is over-reliance on rapid antigen tests or under-reporting?

Dr. Menon: The ratio of RAT to PCR test is fairly high in both Bihar and U.P. The method of who you test and what level you test is certainly an issue. Also, whether the testing is done broadly enough in the general population to pick up certain numbers is a question. As a fraction of the absolute population of the State, those test numbers are not good on their own.

Dr. Babu: Also important is when the testing is done. If testing is increased at a time when the surge in cases has not happened already, or the surge in cases has already happened in the past, the number of people who are actually infected might be low. So, if the tests are increased when prevalence is low, you will obviously pick up fewer numbers. And we don’t know which districts are in which stage of transmission in several States.

The ICMR’s second sero survey says over 7% of the adult population has been infected and the virus has spread to the rural districts. Is testing really good in rural areas or will these areas see a surge later?

Dr. Babu: I feel that there is disparity in the way we test in metros and urban areas and in rural areas, which is a function of this trend of the health system as such. The ICMR survey gave a clear approximation that the infection has spread to rural areas, at least in the surveyed areas. The rural areas should not be ignored; we will have to do more testing in rural areas to corroborate that. We also need to understand whether the rural dynamics of transmission are similar to urban areas. And that can only be known by doing more sero surveys there. We still don’t know what proportion of people are acutely infected and yet not detected. So it’s a sum total of all these things that will give a clearer picture. Without that kind of granular data, we will not be able to say whether rural areas have missing transmission or not.

Dr. Menon: From informal reports it does seem like many more people in rural areas are ill than are actually going in to be tested. Also, the delays in getting tested after falling ill and in getting tested and getting a result are significant in the progression of the disease from person to person.

The expert committee has projected a minimal number of fresh cases by end-February. Is it correct based on mathematical modelling to project for a period of four months during the pandemic?

Dr. Menon: Well, I would not do that and I would not recommend that anyone does it either with any degree of confidence. I think all one can say is that this particular model with these particular numbers show that cases will come down for three or four months down the line. But we have no reason to believe that those numbers will remain the same, the parameters in that model will remain the same. A much better index is what is the fraction of the population of the country that you think is currently infected. And that will tell you how many more people are there to be infected.

Currently there are about 7.5 lakh active cases but the committee says it will drop to 50,000 in December. Do you think it’s feasible?

Dr. Menon: It could be unusual for that to happen. They believe that 300-400 million people in India are already infected, which seems a little too high to me. I think 200-million plus is probably a reasonable ballpark figure. There are lots of people yet to be infected, a lot of them are in areas that have not been exposed yet. So, at what rate they will get infected is the important question. I suspect that there will be a background of infection as we go towards the next couple of months. It will not dip very significantly at least for the next two or three months.

Dr. Babu: These predictions are good, as long as they serve the purpose of planning, ensuring that we are stepping up our resources. But then if they give some kind of robust confidence that all is well and we might be free from this disease beyond December, I don’t think any of us have that kind of optimism yet. I would want to look at seven days average, at the minimum, to understand what is going to happen in the next seven days.

Dr. Menon: Yes, I agree. I don’t see what the point is of saying that something’s going to happen four months later when we don’t have enough knowledge. A week, two weeks in advance is the best that one can do. I worry that putting such a long horizon serves a political purpose rather than a public health purpose.

The committee has not been able to provide any valid explanation for why India seems to have peaked. What could be the reasons considering that India has gone through various stages of unlock and movement of people has increased?

Dr. Babu: The data suggest that areas that were prone to increased population movement and where people were working even when the lockdown was there and who did not follow the public health regulations were the ones who got the earlier waves of circulation. With the most dense areas having been affected already, it’s only the lesser population density areas which are the next focus. Therefore, the speed at which it will spread is going to be definitely different from what it was earlier. The test positivity is also going down as testing numbers have not majorly decreased. Definitely there seems to be some decrease in cases but since we have not touched the baseline, we can’t be sure that this is over.

What will the combined effect of the festival season, winter, unrestricted movement of people, opening up of business establishments and non-adherence to non-pharmaceutical interventions be on the virus spread?

Dr. Babu: We don’t know what has happened before to actually know what is happening now and to predict what’s going to happen in the future. Let’s assume region A has already had a surge in cases and nearly 40%-50% of the people have been infected. So, even if festivals are celebrated and people congregate, one may not see that kind of a spike in cases in that region. Whereas, in region B, if the earlier surge was not there, the same kind of behaviour might result in big hotspots and many deaths. We don’t know what’s happening at the district level. So we need to make a beginning of trying to understand what is happening in each district to be able to predict which districts require more care. We are not using the data in the wisest manner. That’s my worry.

Published in The Hindu on October 23, 2020