COVID-19: 21% of the people in Chennai have been have been infected. What does this tell us?

 The government of Tamil Nadu along with the Indian Council of Medical Research (ICMR) conducted a sero-survey on the number of people infected by the coronavirus. Though the result was released on 1st September, the samples for the survey were taken during the end of July. It concluded that around 15.26 lakh people amounting to 21% of the population had been infected by the virus.

On 28th July, the official number of confirmed cases was still under a lakh. Roughly 14 lakh (93 %) of the cases had been undetected in spite of the city having tested a large number of people.

Chennai is way above any other city in testing people for the virus. By 28th July, the city had conducted 5.47 lakh tests (6.66% of the population). At that time, no other city in India had conducted so many tests.

Chennai has done aggressive testing. Does the sero-survey show that it didn't have much effect?

No, in fact aggressive testing has helped. On comparing this sero-survey with other sero-surveys one can get an idea of the COVID-19 situation. Back in June, the ICMR conducted a sero-survey taking samples from 83 districts. They found that India probably had around one crore cases by then itself. Unfortunately, details were not released and the report was silenced.

However, other city-based sero-surveys have been conducted. Delhi reported that 22.86% of its population had been infected by early July. Pune and Mumbai showed the percentage of people infected even higher.

*According to the sero-survey

Cities used as scapegoats:

All the cities mentioned above are showed by the media as hotspots. These cities are densely populated so the virus spreads quicker. What's more, these are cities with a lot of international travel, so the virus would have reached them first. Sero-surveys confirm that these cities have had many more cases. But what about other cities?

A sero-survey was conducted in Berhampur, Odisha. Having reported under 2,500 by the middle of August, this city had been nowhere in the picture with respect to COVID-19 cases. However, the sero-survey shows that as much as 31% of the city has been infected by the virus! Imagine how many cases there might be in big cities alone, not to forget smaller cities, towns and rural areas.

Mumbai, Chennai etc. have actually done a good job by detecting at least some number of cases.

Certain small cities (specially when the state/UT is small) have contained the virus better. For example in Puducherry the sero-survey showed that only 5% of the people had been infected by the end of July. However, one can't judge this based on the number of confirmed cases.

Why is it that so many cases go undetected?

The survey was conducted to see if people have been infected within the last three months (using rapid antigen tests). On the other hand, the tests which are normally used (RT-PCR tests), are to check if people are currently infected. The virus lasts for around two weeks. Therefore, if the virus has infected 21% of the population in a span of three months, the average number of people infected at a given time will be just 3.5% (This is just a very rough estimation). So when the test-positivity rate in Chennai was 28%, the tracing was being done very efficiently. However, still more tests need to be conducted. Ever though there were comparatively few confirmed cases, there were many people in quarantine. By the middle of August, as much as 20% of the people in Chennai had gone through quarantine.

In India, (and in most other countries), there has always been more room for testing. Sometimes certain statements are shown through media like:

More testing is necessary only up to a point. After that more testing is useless

Such impressions are deliberately spread to cover up failures of not being able to conduct tests. What's worse is, sometimes fewer tests are conducted so that there are very few known positive cases. The reality is that if more tests are conducted, more confirmed cases can be found and positive cases can be isolated and treated.

The average percent of cases in Chennai is 21%, but if you break it up into the 15 zones, the cases are ranging from 44% to 7%.

Awareness:

Another reason why it helps to find cases is there will be more awareness. As long as the cases are far away, it is difficult to wrap one's mind around the reality of the virus. However, as it comes closer, one starts being more careful.

Herd immunity:

Herd immunity is unlikely to work because antibodies for the virus last for around three months. Those who get infected, can get reinfected once there are no more antibodies.

Hope in death-rate:

Viruses are bound to mutate. Sometimes they become less deadly.

Severely affected people are more likely be tested, than those just mildly affected. So if you compare the confirmed cases with the confirmed deaths, the death-rate will appear high. Unless random testing is done, it is difficult to find the real death-rate. Scientists have not been able to find it yet.

On 28th July, the number of COVID-19 confirmed deaths in Chennai was 2,053 (2.12%). Just as most cases were not detected, some deaths definitely would not have been detected. In smaller cities and rural areas there would have definitely been many undetected deaths. However, within Chennai it is likely that a reasonable percentage of deaths have been detected. The difficulty in tracking the virus, is that most people are asymptomatic. However, once the testing increased, most people with severe symptoms would have been tested. In early May, the government started running fever clinics for symptomatic people irrespective of contact/travel history. Those with severe symptoms would be tested. By mid July, 10 lakh people had been screened in these fever clinics. The city was conducting around 10,000 RT-PCR tests a day. There were a total of 550 fever clinics. It is likely that almost all the people with very severe symptoms in the city are being tested. So deaths wouldn't have been undetected unless the patient died suddenly.

We can't say how low the death-rate is. It definitely can't be 15 times lower than what it appears (2.12%), but it is likely to be lower way lower considering there were actually around 15.26 lakh cases by then. Maybe the virus is mutating, or maybe the death-rate has always been low.

However, the death-rate will remain lower if the spread is slow enough that the hospitals can cope with it. What's more, even though 21% of Chennai was infected, a lot of the infected people must have had a low viral load. So it is important to wear masks and maintain physical distancing.

Though the cases are rising, everything is opening up for economic activities; it remains to be seen what happens as things open up.