It is likely that around 15% of Delhi’s population has already been infected by the Sars-CoV-2 virus which causes the coronavirus disease (Covid-19).

That is one of the preliminary findings of antibody tests administered to 22,823 people across all 11 districts of the Capital between June 27 and July 5. The Delhi government and the National Centre for Disease Control (NCDC), which conducted the study, are yet to publish the results.

Hindustan Times reported the preliminary findings on Friday.

Even at 15%-plus or 15%-minus (the final number may be a few percentage points higher or lower) this would suggest that there is community transmission in Delhi, though I sometimes wonder why we continue to be obsessed with this. In a press briefing by the health ministry, the first in 28 days — once, they were as regular as this column — the issue came up and the ministry once again danced around it. We should simply stop asking the ministry about this; it’s like repeatedly asking Shahid Afridi his age.

It was on the basis of antibody tests such as this that the US Centers for Disease Control and Prevention said late last month that only one in 10 cases in that country was being reported. Delhi, based on a similar extrapolation, would then be reporting only one in 30 cases. Assuming that the actual proportion isn’t 15% but 10%, it would still be reporting only one in 20 cases. That may sound excessive, but it is probably correct.

We already know many people who are infected remain asymptomatic. How many? A study by the Indian Council of Medical Research based on around 40,000 people infected by Sars-CoV-2 puts the proportion at around 28% in India. The actual proportion may be much higher — I have seen international studies with numbers ranging from 45% to 78%.

We also know that many infected people suffer mild symptoms, do not require hospitalisation, and recover, often even without any sort of medical intervention. A June study by the US CDC put the proportion of hospitalised patients at 14% of those infected (it covered 1.3 million Covid-19 patients), although this is more a measure of how many of the infected were admitted to a hospital as opposed to the number that needed to be.

Given this, it isn’t surprising that 15% of Delhi’s population may have been exposed to the virus.

On Friday, the New York Times reported, based on data from tester CityMD, which “administered 314,000 antibody tests in New York City… 26% of the tests came back positive”. In some neighbourhoods, some clinics turned in positive results as high as 56% and 68%, the report added. Both numbers are around the 60% that is widely accepted as the proportion of the population that needs to be infected for so-called herd immunity — a phenomenon where a virus or any other pathogen can’t infect too many people because it soon runs into someone who has already been infected, and is therefore immune, breaking the chain of infections.

In one particular area of Delhi, HT reported, the proportion in the early sero-survey findings was as high as 25%. Still, at 15%, Delhi is far away from 60%, or even the 43% defined in an article in Science by researchers from the universities of Stockholm and Nottingham — but the number should also be seen as 15% of Delhi’s population now being immune (we still do not know how long this immunity will last).

These people are potentially safe from a second wave, if and when one breaks, and also in a position to go out and do their thing without the fear of getting infected or passing on the infection to someone. Purely mathematically, it also means the Capital’s death rate is actually much lower (a little over 3%, based on the current numbers).

Interestingly, the 15% number is higher than the 6% a similar study in France came up with and the 5% a study from Spain (reported earlier this week in The Lancet) did. The Spanish study used the low number to highlight the difficulty of achieving herd immunity — which means this is not going to be a disease that runs out of steam in the foreseeable future, but one that needs a vaccine.

Delhi should do wider antibody testing (or sero survey, or serological study, as these are called) to better understand the real prevalence of the coronavirus disease. So should every other state. A high prevalence isn’t necessarily a bad thing.