As the brouhaha over the supposed dangers of the coronavirus reaches fever pitch, it is necessary to put things into perspective. There are too many alarmist global projections about Indian deaths due to Covid-19. But these ignore several factors that are unique to India.
My intention is to use information, particularly through data, to produce some reassuring insights with respect to India. I am propelled by the fact that many people, especially those on social media, have begun to collectively express a great deal of anxiety, centered on the rising number of infected cases and even deaths.
Model vs model
First and foremost we need to debunk much of what the mathematical models are telling us. Models often do not agree with each other even when it comes to a small-sized country such as the UK where some experts at Oxford University are challenging the predictions of the Imperial College model.
In any case, Imperial College had to revise its model within two weeks of making it public. If this is what happens in the context of the UK, a country with a comparatively very small population, I am not sure as to how we can be so confident about a complex scenario like the one that exists in India.
The Center For Disease Dynamics, or the CDDEP model seems to predict that in a worst-case scenario, India may be headed for at least millions of infections, and a large number of deaths. Even if a model were to somehow take into account the need to incorporate corrections along the way, it shall necessarily fail–at least in the Indian context–for some glaring reasons.
Hence, I am staking my reputation as a mathematician, howsoever modest it be, when I state that no model can predict the outcome in India or the US in accurate fashion.
Why no prediction qualifies for India
In the Indian context, the number of variables is so large, that a one-size-fits-all model shall not work.
For instance, I have not seen a single model that could anticipate the huge outbreak of migrant movements that have suddenly taken place in the last few days in India. I cannot see how this migration shall necessarily lead to an exponential growth in the number of infections. By all accounts, these migrants are young and have moved out from largely rural areas or slums where the carriers of the infection, were not present in any significant numbers.
This could be due to the fact that the first wave of infections into India happened through travellers returning from abroad. They would not be frequenting habitats that are peculiar to the migrants of Uttar Pradesh and Bihar. So, for all that we know, the migrants may not even have been infected much. But, if they are carrying infected numbers in a significant way, it may lead to herd immunity amongst them. This, simply put, means that a significant majority of the migrants could become immune to the virus altogether. This possible result will then be due to them having been infected in very large numbers during the stage when they were congregating in such close proximity to each other. In that event they shall reach a stage of immunity in the days to come.
In any case it is not clear how their sudden and unhindered intermingling in large numbers shall affect the rate of infection of the general population.
Demographic dividend for India
Another reason why a western model shall likely fail in the Indian context is due to India’s demographic distribution. All data worldwide clearly indicates that the coronavirus infection hits the elderly in a truly virulent manner. The young, too, have been infected in Italy where 25.7 per cent of the infected population is in the 19-50 years age bracket, and in France, where 30 per cent of the infected population is in the 15-44 years age group.
Yet, the deaths in both these badly-affected nations have occurred in overwhelming numbers in the elderly. In Italy, 74.2 per cent of the deaths are in the age group of 70 years and above and in France, 79 per cent of the deaths are in the age group of 75 and above. In other words, even if the young get infected, their survival rate is much higher.
India’s lessons from Iran
Let us look at some data on Iran. I find that none of the so-called mathematical models have even ventured to model the situation on Iran. Although Iran has suffered much, there are several valuable lessons and insights from Iran for India.
Let me first assert that Iran, which has been badly handicapped because of the sanctions, has actually fared far better than France, Italy and Spain. This is despite the fact that Iran’s population is a little larger than that of the above three European nations and health facilities in the country by far much poorer than in any of these three nations. The number of deaths in Iran is way below these nations. At the time of writing, Iran has had a death rate of 6.5 per cent as opposed to 11.3 per cent for Italy and 6.8 per cent for France.
An even more important measure of Iran’s situation not being as grim as is being made out is the fact that the death curve has been quite similar–for comparable timelines–to that of China, and it is beginning to reach the point where China had begun to control the deaths. Actually, for the last four days, the number of deaths in Iran has been steadily declining.
Let me also add that Iran had not implemented the severe lockdown measures that China had put in place. It has consistently allowed its citizens more freedom for social movements and has only recently started debating a lockdown.
This brings us to the important: what is it that is working for Iran? And it is here that I venture to make a claim. Iran’s demographic distribution is working in its favour. It has a fairly young population. Its median age is 32 years and 38 per cent of its population is below 24 years of age. Actually, almost 49 per cent of its population is in the age group of 25-54 years. This has worked to the advantage of Iran. I must add that this does not take away from the heroic work that health care workers in Iran, Italy, France and other nations are doing.
Age is a key factor
Of course, I am at no stage implying that age is the sole factor. However, from all of the above data for France, Italy and Iran, I do believe that the age distribution of a population can be a very important factor if not the most important factor.
And this brings me to India. As things stand, India has a very young population. Its median age is 28.4 years. The actual data runs somewhat like this: 44.7 per cent of the population is below 25 years of age and 41.24 per cent of the population is in the age group of 25-54 years. So altogether, more than 85 per cent of India’s population is below 54 years of age.
Based on the data as analysed for France, Italy and Iran, India’s young are likely to act as a bulwark against the virus. It seems likely that this shall keep the death rates very, very, low indeed. India, with its young population coupled with the lockdown, shall come out of the crisis sooner and better than many European nations.
This does not mean that the infection rates shall necessarily be low. We may get fairly high infection rates. I must also mention that there is no reliable data for infection rates at the global level.
What seems clear is that the death rates for India shall be very low and to my mind that is the most important parameter that should be used to gauge the situation.
However, I am compelled to add, based on the considerable data that I have—a small but well-distributed random, anecdotal sample—that even the infection rates are, as yet, way below most nations.
I urge the Modi government to keep testing small, random samples through well-designed experiments to keep getting regular inputs on what is happening in the nation. The government ask everyone to wear masks whenever they step out. If high quality masks are not available, let people wear homemade masks. There are many reasons for that and one of them is that people may stop spitting.
I end with the fond hope and fervent prayer that India shall do well.
The author is the former vice-chancellor of University of Delhi, a distinguished mathematician and an educationist. Views are personal.