A Wrong Message for a Tired Land

Should India Stop Worrying and Start Living with Covid-19?

India’s lockdown succeeded in slowing down the rate of growth of Covid-19 infections and so undoubtedly saved many lives. Estimates of lives saved range from 8,000–32,000 (by the distinguished epidemiologist Prof. Gautam Menon & his colleagues) to 120,000–210,000 by BCG (the Boston Consulting Group).

The lockdown also gave India time to stock up ammunition for the Covid War. In March the severe shortage of PPE (Personal Protective Equipment) was seriously endangering healthcare staff; now India produces 600,000 PPE kits a day and is thinking of exporting them. In mid March, India was testing less than 100 people per day and could test at most 8,000; now it is testing over 100,000 each day and has the capacity to test up to 300,000. The numbers of hospital beds, ventilators and isolation facilities have also increased.

However, the lockdown inflicted substantial pain, especially on the poor and migrant workers. It devastated the Indian economy which, Crisil the Indian ratings agency predicts, will contract by 5% this year. India had to end the lockdown and re-open the economy because of the terrible economic strain, even though its most important goal had not been achieved.

Source: https://www.worldometers.info/coronavirus/country/india/

The lockdown failed in its major objective of flattening India’s case growth curve. The number of new cases each day continues to increase. As long as this happens India is losing the Covid War as growth in healthcare capacity cannot outrun the exponential curve!

It is natural for Indian citizens to feel tired and emotionally drained when emerging from such an economically and psychologically stressful lockdown. Many wish that the virus would just go away. While all of us can certainly empathize with that feeling, such wishful thinking is potentially dangerous.

Dubious miracle cures for Covid-19 are likely to flourish in India given its current psyche. Patanjali, a multi-million dollar Indian company established by the Yoga Guru Baba Ramdev, claims to have developed a herbal remedy for Covid-19 that sells for approximately $ 8. (The relevant Indian ministry has refused Patanjali permission to advertise or sell the drug till more details are forthcoming about the clinical trials it conducted). In the southern Indian state of Tamil Nadu traditional Indian systems of medicine such as Siddha, Ayurveda, homeopathy and naturopathy are being tried out on patients in Covid Care Centres. Of course most of these patients presumably have mild or no symptoms -for otherwise they would have been in a hospital instead of a Covid Care Centre- and would probably have recovered anyway.

I believe that research into traditional system of medicines should be encouraged. Their use to treat Covid-19 or any other disease should also be permitted, provided that their efficacy and safety has been established through rigorous clinical trials, following established protocols and backed by publicly available data. Proponents of traditional medicine in India and elsewhere claim that the playing field is tilted heavily against them because of the large promotional budgets of multi-national drug companies. However, that can’t justify tilting it in their favor by relaxing the rules of evidence.

The main danger is not that these remedies will cause serious harm to the user. It is that their availability may lull people into a false sense of security so that they disregard safety norms of mask wearing, social distancing and avoiding large gatherings.

In their current tired state of mind Indians don’t like being told that they are in for a hard slog. That things will get worse before they get better; that it is important to be psychologically prepared for a long drawn struggle against Covid-19; that they can’t afford to let down their guard; that rules on social distancing, wearing masks (properly) and no large public gatherings would need to be strictly followed, possibly for several years. As Pascal, the great French mathematician & philosopher said “too much truth is paralyzing”.

It is much more pleasant to believe that Covid-19 is is no big deal. That it is just another one of the many small risks that we put up with everyday. That we can stop worrying about Covid-19 and carry on with our lives as usual.

Is the Risk of Covid-19 No More than that of Road Accidents?

In a recent article Ashraf & Karthik (link provided in an endnote) argue that Indians worry too much about the novel coronavirus, simply because it is novel. The authors seem to provide data that supports this view.

The argument is likely to resonate with many Indians in their current frame of mind, whether or not they understand the data analysis. Even if the article itself doesn’t attract much attention, the central memes - “the risk of Covid is lower than that of driving” or “you don’t worry about TB, so what’s the big deal about Covid” — are quite likely to spread.

The crux of the authors’ argument can be summarized easily using the table below.

Deaths in India from Different Causes (Sources: https://www.worldometers.info/coronavirus/country/india/, http://statisticstimes.com/demographics/population-of-india.php, https://economictimes.indiatimes.com/news/politics-and-nation/road-accidents-claimed-over-1-5-lakh-lives-in-2018-over-speeding-major-killer/articleshow/72127418.cms?from=mdr, http://www.stoptb.org/resources/cd/IND_Dashboard.html

The authors argue that the numbers show that the risk of death from Covid-19 in India is much lower than that from road accidents or tuberculosis (TB). Since the possibility of dying in a road accident or from TB does not bother Indians much it is irrational for them to worry excessively about the coronavirus. The article is intended to soothe Indians so that their immune systems are not stressed!

Let’s see why this pseudo-statistical argument is wrong.

Decision theorists — beginning with Frank Knight — distinguish between risk and uncertainty. The difference is succinctly explained in


risk is present when future events occur with measurable probability

uncertainty is present when the likelihood of future events is indefinite or incalculable”

Road accidents and tuberculosis deaths belong to the sphere of risk; Covid-19 to that of uncertainty.

Deaths by road accidents are not contagious and do not increase exponentially.

Source: https://www.statista.com/statistics/746887/india-number-of-fatalities-in-road-accidents/

As the above graph shows the number of road accidents in India increased steadily from 2005–2011 but seem to have reached a plateau in the last few years (2012–2018).

Now the number of road accident deaths in India in 2019 has apparently not yet been published. But we can estimate it reasonably well. The average number of road accident deaths between 2012–2018 was about 145,000. The standard deviation, which measures the variability of the numbers, was about 5,000. Till the number of deaths in 2019 is made public we can think of it as a random variable, which may take any one of a range of values, even though the deaths have occurred in the past. It doesn’t seem unreasonable to assume that this will follow an approximately normal distribution. With near certainty then, the number of deaths due to road accidents was somewhere between 130,000 and 160,000. The probability of dying in a road accident in India in 2019 was between 100 and 123 in a million.

The above calculations are only intended to be illustrative. Better statistical modeling of road accident deaths in India is certainly possible but not necessary here. The point is, that we can predict reasonably well the probability of a road accident death in India. Of course, like every prediction, this one too is uncertain but the range of uncertainty is fairly narrow.

Risk assessment should be a prelude to risk management. What conclusions should individuals and policy makers draw from this data? A very superficial analysis would suggest that driving in India is not too risky an activity and drivers should not be deterred by it. Policy makers should therefore consider the number of deaths due to road accident in India each year “acceptable”.

Any analysis along these lines would be not only superficial, but wrong. Deaths per capita is not the right metric to use while evaluating road safety. India has a low per capita death rate from road accidents simply because relatively few Indians own vehicles and therefore drive less.

A fairer metric to assess the risk of death by road accident is deaths per billion vehicle kilometers. This does not seem to be readily available for India. We consider then an useful proxy; deaths per 100,000 motor vehicles.

Source: https://en.wikipedia.org/wiki/List_of_countries_by_traffic-related_death_rate

It is clear from this table, and even clearer if you study the source data, that India’s road safety record is patchy compared to not only the developed world but several low and middle income countries. Risk averse drivers are advised to avoid Indian roads altogether. Others should drive with care. Policy makers and enforcement authorities definitely need to take steps to make Indian roads safer.

The good news is that several government and non-government initiatives are being taken to do just that. The stabilizing of the death from road accidents graph seems to show that they are having some effect.

And What About Tuberculosis?

Let’s turn now to tuberculosis (TB). It would seem fair to compare TB with Covid-19 since both are contagious diseases that affect the respiratory system. TB kills many more each year than Covid-19 has done to date. Why then does India-and indeed the world- worry so much more about Covid-19?

The first reason, Ashraf & Karthik argue is familiarity. Unlike the coronavirus, TB has been with us for several thousand years. In fact the disease was widespread in ancient Egypt and Mycobacterium tuberculosis complex DNA has been found in lung tissue and gall bladder samples of an Egyptian mummy! We fear Covid-19 more because it is unfamiliar and we don’t yet have a cure.

There’s truth in the argument. It is quite true that people tend to fear the unknown. It needs to be added that such fear is often far from irrational. The Incas should probably have been far more scared than they were, of the strangers who came to their land, riding horses and carrying unfamiliar weapons.

Inca-Spanish Confrontation at Cajamarca (Source: https://en.wikipedia.org/wiki/Battle_of_Cajamarca )

A novel pathogen, that we understand very little yet, deserves to be treated with extreme caution.

The second reason, according to the authors is that “TB, despite its high death rate …has not become India’s obsession because the middle class …is relatively insulated from it.” Again, there is no doubt considerable truth in this claim too. The death of hundreds of thousands each year from a treatable disease evokes much less indignation in India than it should.

However Ashraf & Karthik miss the main reason why the world is obsessed with Covid-19 but not with TB. The diseases are significantly different.

We have all by now heard about the Reproduction Number R of an infectious disease. Suppose that John infects Mary. Then we call John a primary & Mary a secondary case. R is the average number of secondary cases (such as Mary), infected by a primary case (such as John).

The actual value of R cannot be observed. It is estimated from case data using various statistical techniques. According to most experts, the R for Covid-19 lies between 2 and 3, though some estimates are much higher.

What about TB? A meta-study of 56 papers finds widely ranging values of R from 0.24 to 4.3. Two Chinese studies which report R values of 3.5 and 4.3 should probably be excluded because they seem to consider all TB infections. Only about 5–10% of all infected cases become active, i.e., full-blown, symptomatic and infectious themselves. The remainder remain latent. A latent case may become active — sometimes after years- if the infected person’s immune system becomes weak as a result of other infections or aging.

For India, one study estimates R to be 0.92, though there is probably considerable variation among different segments of the population.

So the R for Covid-19 is generally higher than the R for TB. However, R is not the only number that determines the speed at which an epidemic spreads. Possibly an even more important one is serial interval. This is the time between two successive cases on the chain of transmission.

Serial Interval

The median serial interval for Covid-19 is estimated to be between 4–8 days. However, according the meta-study cited above, estimates of the median serial interval for TB vary from 0.57 to 1.65 years.

So, a person infected with TB typically infects fewer persons than one with Covid-19. Also, TB has a much longer incubation period. The upshot is that TB smoulders in a population but rarely — if ever- flares into an epidemic the way Covid did and Sars, Ebola or Plague might.

Covid-19 thrust itself on world’s consciousness with its brutal exponential growth, overwhelming healthcare systems and killings tens of thousands in weeks. Covid-19 received attention because it demanded immediate attention. You need to deal with it immediately, just as you would need to deal with a building on fire.

Borrowing an analogy from the Nobel Laureate Prof. Amartya Sen, the difference between Covid-19 & TB is like the one between famine and chronic malnutrition. Famine is sharp and can kill many in a short time-frame; chronic malnutrition, on the other hand, often stays below the radar though it kills more over decades.

We don’t yet know the risk of Covid-19, and can’t quantify the probability of dying from it because we don’t know how many it will eventually kill. Year-on-year or quarter-on-quarter comparisons are utterly meaningless when the deaths can grow exponentially and not linearly.

However, citizens, policy makers and governments all need to act even when they can’t quantify the risk. We don’t know how many Covid-19 will kill but we do know that it could kill many. Dr. Jayaprakash Muliyil, the former principal of Vellore Christian Medical College and a distinguished Indian epidemiologist, for expample, believes that Covid-19 could kill 2 million Indians. That is several years’ worth of driving and TB deaths combined.

Why can’t we know for sure exactly how many the pandemic will kill? One reason the trajectory of an epidemic cannot be predicted with as much accuracy as that of a spacecraft is because it depends on the actions of citizens & governments. An enormous number of deaths will be prevented if people wear masks, maintain social distance and avoid large gatherings.

Downplaying the seriousness of Covid-19 has had terrible consequences in the US, Brazil & several other countries. The world has painfully learned the impact of the meme “no worse than a flu”. Messages like “the risk of Covid is lower than that of driving” or “you don’t worry about TB, so what’s the big deal about Covid” could at this point have even more terrible consequences in India, which has a larger population and higher population density than the US and Brazil. While Indians are concerned about things being bad, they could get much, much worse if they allow themselves to become careless.


Disclosure: I am the Director of Smart Consulting Solutions Pte Ltd, incorporated in Singapore and its subsidiary Radix Analytics Pvt Ltd, incorporated in India. I am also a Visiting Faculty Member at the Indian Institute of Management, Udaipur. However, the opinions expressed in this post are solely mine and not necessarily shared by any company or institution with which I am affiliated.

I found Ashraf & Karthik’s article on Prof. Jagdeep Chhokar’s Facebook wall.

I am, as always, grateful to my friend Dr. Ashish Kumar Dawn for many insightful comments and assistance with the section on tuberculosis.





I have spent over 30 years in academia and industry exploring how to use mathematical methods to solve real world problems