Data, New Data, Different kinds of Data, and Covid 19: Bharati Jagannathan


“There are three kinds of lies: lies, damned lies and statistics”, quoth Mark Twain. We could add a fourth, pretence of statistics in the absence of it. So, there’s data, more data, and the immensely useful pretend data about COVID-19. And almost all of it liable to totally dissimilar interpretations. In fact, this has been the best lesson, for those who in general find statistics challenging and humbly retreat in the face of data-based proofs in any argument, that the same set of data can serve completely opposite ends. However, I digress.

There was speculation in early March that India had fewer cases of infection owing to 1) exposure to malaria and sometime ingestion of quinine (in medical formulations like hydroxychloroquine), or 2) BCG vaccinations in childhood, or 3) warm weather hindering the spread of COVID-19 like many other influenza viruses. Till we realized that it was the effect of abysmal levels of testing.

Then there’s the data on mortality rates. Some early studies suggesting a frightening 3-4% fatality were eventually rubbished when wider testing revealed that almost four-fifths of all infected persons were asymptomatic, only about 15% developed serious symptoms, a smaller set still needed hospitalisation, and that the actual mortality rate is probably about 1%, lower than the SARS epidemic that swept across several countries about a decade back.. This is not to suggest that there is no need to worry and that we can put on our dancing shoes again and go party; far from it. The possibility of one in every hundred infected persons dying, coupled with the probability of 60% of the world’s population getting infected still gives us a frighteningly high score of about 3 in every 500 people in the world. All the same, there is something quite peculiar about the way in which mortality is peaking in certain countries at certain points. Much of the current wisdom is built on the assumption that minimising physical contacts—which goes by the unfortunate label of social distancing, and takes the administrative form of enforced lockdowns—is key to minimising the spread of the virus.

India seems to have imposed one of the severest lockdowns anywhere in the world in response to the crisis. While the privileged participate in recipe challenges while ruing the absence of domestic help to clean up afterwards, migrant workers are corralled in make-shift quarantines with irregular food, and lakhs of wage earners who still have a job are staring at uncertain futures—with the economy in free fall, there is no telling what jobs will remain three months later, and if those that do not disappear will suffice to put food on the table. Despite the upbeat predictions of the prime minister that India is set to emerge as the new manufacturing hub of the world post this crisis, most specialists remain deeply anxious about the economic health of not just India but of the entire world. Nor are specialists agreed as to the solution: while economists seem to prefer a gradual reopening of the economy, the medical fraternity advises further caution and a longer period of isolation.

It is fairly clear that these measures of isolation or distancing do not even aim to defeat the virus—that is already acknowledged to be impossible. All that is sought is to spread the disease over a larger period so that hospital facilities are not overburdened with an excessive load of patients all at the same time. In other words, flattening of the curve will simply stretch the disease and resultant mortality over perhaps a year rather than have it all descend in a matter of weeks—the hope is that some lives at least may be saved if enough medical facilities are available, and perhaps, a vaccine may be found before the virus has gone through all of us.

It is not news that tuberculosis affects as well as kills more people in India each year than the novel coronavirus is projected to, and yet, neither the government nor civil society has ever considered it enough of a threat to plan action around it. It has also been hypothesised that if this state of even partial lockdown and depressed economic activity continues very much longer, we may see more deaths from hunger than from the virus; in fact, some such statistics, of 200 COVID deaths as against 300 hunger-related ones, were being publicised a fortnight back. Yet, no one really knows. Choosing the lesser evil has never been more difficult.

Officially, India has about 35,000 COVID-19 positive cases and about 1100 people have died so far. We are told everyday that the early imposition of lockdown has worked wonders in checking the spread. Clearly, the poor are beyond the visual range of both policy makers and analysts: how else does one understand this obliviousness to the effect of the lockdown on the migrant workers who first gathered in their lakhs for the few buses to get home, or proceeded to walk hundreds of kilometres across states, and queue up at food distribution centres? Secondly, among those of the working poor who have stayed on, how many can practice any sort of distancing? Families live crowded together in single room tenements with toilets typically shared by several families. This is in pucca buildings. Slums often have only a couple of water outlets at which women line up for their turns; this will only worsen as the summer intensifies. It is in such places that your chemist’s assistant and hospital nurse live, where your plumber returns to sleep at night. I sit in my comfortable house typing up this essay on my laptop, scrupulously maintaining distance from all except my immediate family, while a stream of sabziwalas bring a rather sorry collection of vegetables and fruits to my doorstep. “Where do you go to, my lovely, when you are alone in the night, dear rediwalas?” Oh, sorry, they are never alone, at least not physically.

What can any lockdown achieve when the vast majority of essential service providers live in such conditions that they are compulsorily in close physical contact with hundreds of others, any of whom might be a carrier of the virus? This is not a lament about my/ our lack of safety; far from it. My purpose is to point to the intense myopia in our middle-class citadels endorsing the lockdown. I am reminded of the Mahabharata story of Parikshit who, cursed by a sage to die of snakebite within a week, built himself a sealed fortress on a tower. But Takshaka, the king of snakes, transformed himself into a worm and entered a brahman’s gift of fruit for the king and, once within the stronghold, changed back to his original form and fulfilled the terms of the curse. If nothing else, the impossibility of creating an airtight bubble to exclude the virus should itself be an argument against extending the lockdown in Indian conditions.

The virus has definitely been around since February; was it then so enthralled by the wonders of India that it took an extended vacation from spreading till reaching the much-maligned markaz in Nizamuddin? Did the glories of Gujarat stun it into immobility during the Namaste Trump event? Did Vishnu’s Sudarshana chakra make it flee from the 50,000 daily devotees at Tirupati till the 18th of March? It is safe to assume that the only reason to claim the absence of community transmission is the absence of aggressive contact tracing except in the case of the Tablighi Jamaat attendees. There’s no doubt that the virus is spreading, and killing too, but is it killing at a rate comparable to the US, for instance, or even the UK?

I return now to the statistics. Official statistics on COVID mortality are likely to be skewed for the simple reason that only those who were positively identified as infected can be counted among its fatalities. Even in Italy, only hospital deaths were counted; those patients who died of the virus at home remain outside official data. Things are complicated further in India by sick people refusing to visit doctors for fear of being confined in notoriously unhygienic isolation facilities. However, there’s one set of data that I had assumed was basic, and am surprised to find completely missing. What is the average daily/ monthly number of deaths in India? Have the April figures shot up by a significant degree? Cremation and burial grounds require death certificates and, I presume, maintain records of the number of funerals each day.

This is a call to bona fide journalists to collect and compile this crucial data. Can we at least look at the urban areas? If, then, we observe an unprecedented spike, I expect we should bow to the advice of the medical experts and submit to an extended lockdown, despite the terrible cost to the marginalized and deprived. If, however, there hasn’t been such a spike after almost 45 days of acknowledged community transmission, we might want to analyse if we in India are comparatively immune, for whatever reason—genetic, exposure to numerous other viral ailments, tropical climate, BCG vaccinations in childhood or Iqbal’s, “kuch baat hai ki hasti mit ti nahin hamaari”…

It is crucial that this data is immediately created and examined and, if—as I hope—the dead are not piling up, it is time to start acting rationally.

Bharati Jagannathan teaches History at Miranda House, Delhi University.

One thought on “Data, New Data, Different kinds of Data, and Covid 19: Bharati Jagannathan”

  1. Excellently reasoned article. The whole political and policy response to this virus has been so pathetically ‘middle class’ i especially in its symbolism.
    And speaking of statistical confusion, I should mention one more – it makes no sense to announce daily ‘recovery rates’, improving or otherwise. Official recovery rates were low to begin and are still under 30% because most patients have not yet had time to recover. These rates are gradually improving as more and more patients have had this 2-3 week time since infection. But they are still so low (instead of the very likely 95 or so percent) because of all the newer cases that still need the 2-3 weeks after initial infection to test negative. Recovery rates need to be based on only those patients who have had the infection for at least two weeks. And given that there are probably many more undiagnosed positive cases, actual recovery rates may be even higher than such a calculation will produce.
    Again, this is not to downplay the disease in India (though let’s thank the gods for maybe, maybe sparing us the rich country levels of infection and/or death) but to try to focus simultaneously on all the other costs that come with treating a total lockdown as the only solution to this new risk in so many already unimaginably risky lives.


We look forward to your comments. Comments are subject to moderation as per our comments policy. They may take some time to appear.

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s