Mumbai: The results of a seroprevalence survey in three Mumbai wards should not be taken as “a green signal”, and people should continue to follow physical distancing and wear masks, says Daksha Shah, deputy executive health officer of the Brihanmumbai Municipal Corporation (BMC), Mumbai’s civic body.

The survey, for which about 7,000 people were tested, found that 57% of the people tested in slum pockets had antibodies to COVID-19, as did 16% people in non-slum pockets. The SARS-CoV-2 seroprevalence study, conducted along with the Tata Institute of Fundamental Research (TIFR) and the NITI Aayog (the Centre’s policy think-tank), covered north and central Mumbai.

The BMC’s approach to handle slums and non-slums differently has paid off to some extent, says Ulhas Kolthur-Seetharam, professor at the biological sciences department, TIFR. Further, a second survey to be conducted later this month will “give us some clue about the level of immunity and how much antibodies are required to confer protection,” says Jayanthi Shastri, professor and head of microbiology at Nair Hospital, and head of the molecular diagnostic lab at Kasturba Hospital.

Edited excerpts from a video interview:

Dr Shah, how did you define the scope of the study, and how did you go about it?

DS: This study was planned in May and early June. We decided to take wards that represent [all] 24 wards of Mumbai--one each from the city, the eastern and the western suburbs; those reporting the highest, the moderate and the lowest number of cases; and cover both the slum and non-slum population. The age- and sex-wise startification was done. The study was planned basically to understand the extent of spread of infection, and going further, whether it is going [to rise] or reduce or whether it is [going to be] static, and the level of antibodies, which might have developed in the population.

Dr Kolthur-Seetharam, can you define the project and the scope from your perspective?

UKS: We wanted to capture whether population density--crowding--and socio-economic conditions would be a factor that could drive prevalence. This could be true not just in Mumbai but in many other complex metropolises. City populations are not homogenous. So it is important to address what factors could be contributing to this. If you do not have data, you cannot form hypotheses. So we wanted to do this study to get this kind of cross-sectional information in these three wards. As Daksha rightly pointed out, we wanted to do this in three wards which had high, low and median reported cases; therefore it would be, in some sense, representative of many wards.

DS: This [the survey] indicates that there are some asymptomatic infections already in the community, and this has given rise to the antibody response, which is good in a way. We now understand that there is a sort of protection in the community, especially in the slum areas. However, we need to look into whether this is going to be static, or reduce or increase. Right now, we are observing that more cases are coming from non-slum areas. This also corroborates our study. So definitely, there is a level of immunity that is being reached, but to what extent it will go to the herd immunity is yet to be known. There are a lot of expert opinions required on this.

Does this mean that these people--this 57% representative sample--have actually got this disease? Or are they just immune to it?

DS: They have had an exposure to the virus, and in response, their bodies have produced antibodies. Some of them might have suffered a very mild [disease], or they may have remained asymptomatic. But the essence is that they have got infected.

To what extent are we sure that the sample was exposed to this very virus?

JS: We ensured that we use the best quality tests to screen for the presence of antibodies. So, we have used the Abbott Chemiluminescence test, which is 100% specific. There is no cross-reactivity with other coronavirus infections as well as with other respiratory viruses.

So we are very sure that all the samples that are positive are true positives, and thereby we would like to believe that the prevalence that we have seen across slums and non-slums--which is 57% and 16%--[corresponds to] the people who have the antibodies. And that means they have been exposed to the infection. They could have been asymptomatic.

There was an ICMR survey in May, which talked about 30-40% sero-positivity in Mumbai slums. That figure has increased. Are we able to project in any way how the transition or the trajectory of the virus is moving?

UKS: Our study specifically excluded containment zones that were active during the period of the study, and is more representative in terms of population that has either been asymptomatic or mildly symptomatic and recovered. To get a true prevalence and the change, we are going to do a second round. We want to go back to the same population, in the same stratified manner because we want to get high confidence, statistically low-variance data. And we are going to do this in August, and that is when we will see in the slums and the non-slums--they are showing two different numbers--which of these and to what degree will you see a change. That will have a huge impact on our analysis and others who can think about herd immunity and all of that. So it will be useful data.

Dr Shah, are you able to extrapolate to the total population of slums by using this? Mumbai has a population of 18-20 million, maybe 40-50% are in slums. Therefore 47% or 50% of that figure are likely now immune to this disease?

DS: I would say not immune, but I would say that they have been exposed to the infection. And wards that have a similar population pattern and demography could have a similar type of antibody response. That is why we will also plan something in further many wards. This study of extrapolating into other wards and the modelling is under way. So we will definitely look into how it can be projected for the entire Mumbai.

How would you look at these data versus the fatality rate that we have seen so far in the city of Mumbai?

UKS: The idea of the study was not to necessarily get just a number that could be reflective of the entire city. There are benefits of doing that. And there are also benefits of actually understanding, like I mentioned earlier, the differences--because it is useful to know, if you want to bring in a policy change, would you want to apply it globally across the entire city or do you want to tailor it based on slum and non-slum. There, I would say that probably BMC’s approach to handle this both in the slum and the non-slum differently has paid off to some extent. So, what we are seeing as a lower infection fatality rate and all of that could be because BMC is actively isolating symptomatic people and getting them medical treatment. And that, I think, seems to be working. So the idea is [that] there are two approaches: One, you can get a number for all the wards, a number for the city; that has its value. But studies like this which have more cross-sectional information also have value, because then you can go back and ask what kind of parameters will be actually influencing the prevalence.

Dr Shah, having got this information or data, what are the likely next steps going forward for the city?

DS: This should not lower our guard. We are going to continue our preventive measures and our efforts in containing the spread of the virus with the same vigour and vitality. We should continue to do this. People should not take this as some kind of a green signal. There are still some susceptible populations who can get infected. So wearing the mask and social distancing would be very important measures that people have to keep in mind.

With the precautions and with all the necessary safety measures, people can do their necessary work as usual now. We need to tackle the fear and stigma, and basically focus on our mortality rates, on how we can protect our high-risk individuals and the populations that are most vulnerable.

Dr Shastri, how are you taking these data and how are you looking at things ahead? What are the new research angles that you would be possibly examining?

JS: I am looking forward to the second round because we would like to see how the prevalence varies between the first and the second round. And talking of immunity, there is a lot of data coming around the antibodies--whether they last, whether they confer protection. So basically, only time will tell us and as part of this study, we are also going to be looking for neutralising antibodies in these samples. That would also give us some clue about the level of immunity and how much antibodies are required to confer protection.

Do any of the people who are part of the sample include those who have formally recovered from the disease? DS: [they] could have been [a part of the sample], because this is anonymous.

JS: We did not capture that information in the survey form

Dr Kolthur-Seetharam, we have also seen a similar survey in Delhi. Are you looking at data from other places or centres in the world? How is this converging? Any other insights you can draw?

UK: This is very useful. What you think as prevalence today is continuously going to be continuously changing in a week’s time. So, I think this will provide a lot of information for people to compare both longitudinal studies and cross-sectional studies to understand exactly how the disease spreads, what are the manifestations, susceptibility of populations, and most importantly, this Holy Grail of if and when there will be herd immunity. But remember, this all depends on how long the immunity lasts. New data is emerging, and I think that is going to be really exciting to look forward to these things.

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