Print Friendly, PDF & Email

Insights into Editorial: What are the implications of the Delhi sero-survey findings?





To estimate the extent of infection in the community, scientists at the National Centre for Disease Control, a Health Ministry body tasked with monitoring epidemics, conducted a district-wise serological survey across Delhi.

Preliminary findings showed that 23.48% of Delhi’s population had been exposed to the virus and that a large number of those infected were asymptomatic.

About Sero-Survey:

A serological survey is done to detect the presence of specific antibodies, and is used to assess the prevalence of a disease in the population.

The test indicates past infections (and which triggered an immune response), and is not used to detect active infections.

Results of two different serological surveys, one in Delhi and the other in Mumbai, were announced.

Both the surveys found the presence of novel Coronavirus-specific antibodies in slightly less than 25 per cent of those tested.

This could mean that in either of these cities, there were chances that about 25 per cent of the city’s population had already been infected by the virus, if we assume that the disease was uniformly spread in the population.

How was the Delhi sero-prevalence survey done?

  1. Unlike the real time RT-PCR (reverse transcription-polymerase chain reaction) tests or antigen tests that scan for the presence of the coronavirus in the body, the sero-prevalence survey was designed to look for the presence of antibodies, produced in response to the coronavirus, in the blood.
  2. On an average, antibodies can be detected five days after an infection sets in. The coronavirus, in general, diminishes after 21 days.
  3. Antibodies, for a host of other viral infections, are normally expected to linger for many months and confer protection against fresh infections by the same virus.
  4. However, it is still too early in the evolution of SARS-CoV-2 to know how long the antibodies will last.
  5. According to health experts, because the actual virus withers away quickly, many of those infected show mild symptoms.
  6. Antigen and PCR tests are often cumbersome to be widely deployed in the field and are not always the best way to estimate the spread of the virus.
  7. After the Indian Council of Medical Research led a multi-institutional study in May to estimate national sero-prevalence using the Kavach-ELISA antibody-detection kit, developed and tested in India, the government decided to do a similar study across Delhi.
  8. In June, the city was grappling with a high number of cases and several parts of the city were locked into containment zones.

What did the survey find?

  1. Nearly 22,000 samples were collected from 11 districts. It emerged that on average, 23.48% of Delhi had been exposed to the virus.
  2. A large number of those infected were asymptomatic. Like in all statistical surveys, extrapolating the results from a sliver of a population to an entire city comes with its own caveats.
  3. Generally, those who commission such surveys report their findings in a peer-reviewed journal and detail, for instance, how accurate the test was in picking up infections and the error margins in extrapolating the results to a population.
  4. No such details were made available, so the announcement that nearly a quarter of Delhi’s 20 million population had been exposed to the virus is at best only suggestive.
  5. A lot of the samples were collected from Delhi’s densest pockets so the extent of variation of infection in less populated parts of Delhi is not known.
  6. The Delhi government has said it will keep repeating these surveys at periodic intervals to gauge the spread of the infection.
  7. The survey found that nearly eight of the 11 districts had a sero-prevalence exceeding 20%.
  8. To the Delhi government, the fact that after six months of the global pandemic there was a 23% spread was indicative of the effect of lockdown and containment zone management. But the government also cautioned that the survey showed 77% of the population to be vulnerable to the infection; therefore, measures such as masks, hygiene, social distancing ought to be strictly adhered to.
  9. Another way to look at serology surveys is to compare the number of RT-PCR positive cases and the number of those who were infected as per their antibody profiles.
  10. Thus, assuming that 50 lakh (or 25% as per this survey) were infected and comparing that to only about 100,000 who were confirmed positive, it appears that a PCR test caught only about 1 in 50 who are infected by the virus.
  11. This undermines claims by the Delhi government of its ability to contain infection spread.
  12. However, missing a sizeable proportion of those infected by the coronavirus is par for the course given that it spreads rapidly and leaves a large proportion of those afflicted with only mild (and frequently unreported) symptoms.

 How reliable are serology surveys?

While it is routine in epidemiology to deploy antibody tests to estimate the prevalence of a disease, the sheer speed of the spread of SARS-CoV-2 and the global nature of the disease have complicated the matter.

Generally, they take a while to develop and are tested in the field before being deployed in large numbers.

This is done to rule out errors such as the test marking the wrong antibodies, or an inappropriate sample that does not represent a population being chosen.

Such problems hindered an early plan by the ICMR to use China-manufactured antibody kits in April and May to gauge infection spread.

The ICMR is still to publish details of its May sero-survey that estimated 0.73% of the population in 63 surveyed districts to be infected. For COVID-19, the reliability of such surveys still rests on nascent science.

The Ministry has attributed to the lowering of COVID numbers in Delhi to the proactive efforts taken by the government to prevent the spread of infection, including prompt lockdown, effective containment and surveillance measures, such as contact tracing and tracking, as well as citizen’s compliance to COVID appropriate behaviours.


High-quality sero-surveys allow explicit characterization of the distribution of immunity at a particular time point, but they also provide mangers with an opportunity to evaluate assumptions made about natural transmission dynamics and vaccine program performance.

The health ministry has so far maintained that there is no evidence yet of community transmission in the country.

There are large outbreaks in some clusters but the sharp exponential rise in cases as in community transmission has not happened.