On 16 September, India reported 97,894 new cases – 11,36,613 tests were done that day, which meant the positivity rate was still a relatively high 8.6 per cent. On 17 September, the country had 10,17,754 active COVID-19 cases – the highest ever. The numbers seemed to be on an exponential growth curve – and then suddenly, without any explanation, the daily COVID-19 tally pulled back from the brink of a lakh, much to the disbelief of many people.
India had crossed one COVID-19 peak – a peak in an infectious disease is the point from which the numbers start coming down. India’s numbers started coming down from around mid-September. By the end of the year, statistical projections were bandied around to claim that India would be one of those rare countries that would escape with just one COVID-19 peak.
It was only when an expert committee formed by the Government of India’s department of science and technology, consisting of mathematicians, statisticians and computer science experts, came out with their final report more than a month later that there was an official confirmation that September had, indeed, been a COVID- 19 peak in India.
“Based on the mathematical model developed by the Committee, the following conclusions have emerged about the course of the pandemic. The number of active cases has peaked at around 10 lakhs. However, this number will start rising again if proper practices of masking, disinfecting, tracing, and quarantine are not followed. The downward trend will continue only if we continue with above practices. In addition, based on temporal profiles of analyses done for Bihar and UP, the Committee concluded that the impact of labour migration on the total number of infections in these states was minimal. This observation indicates the success of quarantine strategies adopted for the returning migrants,” the committee wrote.
Among its members were faculty from some of India’s most elite institutes, including the Indian Institutes of Technology (IIT), the Indian Statistical Institute (ISI) and the Indian Institute of Sciences (IIS).
The mathematical “supermodel” was tailor-made for India, or so its authors claimed. They arrived at an estimate of the population-level spread of the infection as against the cases that indicated the number of people who had actually tested positive for the virus, by looking at the number of daily positives.
When I spoke to him a day after the official press conference where the September peak was announced, Professor Manindra Agrawal, professor of computer science at IIT Kanpur told me, “The unique feature of our estimate model is that we take into account the fact that there are large numbers of asymptomatic people in the population that have not even been tested for the virus but they continue to infect others. Earlier models had not been able to accurately capture the disease dynamics because they had failed to explicitly account for this. We do and we do so by trying to calculate backwards from the daily positives every day to reach an estimate of how many people would have had to be carriers for that many people to be detected.”
Apart from Prof Agrawal, the committee of experts included Prof M Vidyasagar of IIT Hyderabad, Dr Gagandeep Kang of Christian Medical College Vellore, Prof Biman Bagchi of IIS Bangalore, Profs Arup Bose and Sankar Paul of ISI Kolkata and Lt Gen Madhuri Kanitkar from the Ministry of Defence. The committee was tasked with arriving at a “COVID-19 India National Supermodel”.
According to the supermodel, India had 60-65 asymptomatic undetected infections for every one of the 75 lakh-odd cases that had been detected in the country till then. This was almost double the estimate drawn up by the ICMR from the results of its August sero survey that stated that the country had about 26-32 undetected infections for every positive case.
This, in turn, meant that by October, when the committee’s report was submitted to the government, close to 50 crore Indians would be infected by the SARS-CoV2 virus – a number that accounted for about 38 per cent of the country’s population. It also indicated that the testing, despite the massive push, had actually missed about 45-49 crore of those infections. The ICMR estimate, on the other hand, indicated only about 9-11 crore infections had been missed.
However, like all government-appointed panels since that fateful day when we were promised a “victory over Corona in 21 days”, this one also faithfully defended the lockdown, its timing, and the way it was done. Just like with demonetisation in 2016, when the entire government machinery had focussed only on justifying that draconian disaster for months, a lot of the Indian government’s resources and mind space during the pandemic was utilised in justifying the timing of the lockdown and also its length.
Clearly, the government was, by no means, oblivious to the severe criticism that had dogged it since March about the announcement and execution of the countrywide shuttering exercise. So, in language that was unusual but unambiguous, the group of experts, in one voice, hailed the lockdown, said that the hardships inflicted upon the migrant workers was unavoidable, and in general said everything that the government wanted to hear.
“...This observation indicates the success of quarantine strategies adopted for the returning migrants,” the report said.
The Committee also simulated what would have happened via hypothetical alternative scenarios with regard to the timing of the lockdown regime.The following conclusions emerged:
- With no lockdown, the pandemic would have hit India very hard, with a peak load of 140+ lakh cases arriving in June. Given our lack of preparedness back then, the healthcare system would have been overwhelmed, leading to many additional deaths.
- Had India waited until May to impose the lockdown, the peak load of active cases would have been around 50 lakhs by June.
- The peak of active cases came in late September at around 10 lakhs. By this time, we were far better equipped to handle the pandemic in terms of diagnostics and vital equipment inventories.
The “supermodel” was criticised by health experts for the absence of epidemiologists or virologists in the group – certainly not the first or the last time during the pandemic that the Indian government chose to treat “expertise” as an undifferentiated monolith. This systematic sidelining of domain experts, the general opacity about data and decisions and the compulsion to lavish praise on the government must have been some of the factors that led to a scathing editorial in the medical journal The Lancet on 26 September.
It was about India’s COVID-19 management strategy in general and the role played by the ICMR in particular. The editorial stated, in black and white, the things that had only been spoken of in hushed whispers till then. It also made it very clear that the criticism was not just cosmetic. It discussed how the compulsive “positive” spin in government communication could end up impacting public compliance on measures such as masks and social distancing, which could ultimately prove to be counter-productive.
The worst of the editorial was, unfortunately, reserved for the ICMR – largely due to Dr Bhargava’s mysterious letter on 2 July commanding scientists to prepare a vaccine by 15 August.
“According to news reports, hours before announcing the national lockdown, Prime Minister Narendra Modi told owners and editors from India’s largest media organisations that it was important to tackle the spread of pessimism, negativity, and rumour. This pressure to avoid negative news, and to offer reassurance, appears to have been felt by several professional scientific organisations in India. The Indian Council of Medical Research (ICMR) has been singled out by experts for straying from scientific evidence, appearing at worst politically motivated and at best overly optimistic. A letter from the Director General of the ICMR, Balram Bhargava, said that the ICMR envisaged launching a coronavirus vaccine on Aug 15 (Indian Independence Day; a deadline considered unrealistic by most medical experts); ICMR has supported treatment with hydroxychloroquine despite insufficient evidence; and news reports claim that data on coronavirus infection were removed from a scientific paper,” The Lancet wrote in that editorial.
The searing disapproval though failed to dent the sangfroid of the powers that be in the organisation that is the torchbearer of medical research in India. Five days after The Lancet piece was published, the ICMR embarked on a narcissistic and slightly incongruent journey of self- applause as it unveiled an elaborate commemoration of its own history. It was interesting, even if slightly ill-timed. The ICMR was chronicling its own history while the country was fighting a pandemic.
The international journal – among the most prestigious in the world of medicine – even raised questions about the repeated “lowest in the world” claims questioning the veracity of the data on which those claims were based.
“Transparency of the data on COVID-19 cases and deaths, especially those underpinning the case fatality rate, has also been questioned, as detailed in a recent World Report. The Indian Government reports a case fatality rate of 1.8 per cent, much lower than the reported rate in other countries, but it is difficult to know if the numbers are comparable. Hope is important, and recognising successes is vital, especially during a pandemic. But presenting the current situation in India with a too positive spin not only clouds reality but also hampers vital public health initiatives. Perpetuating unrealistic claims or failing to honestly report negative news creates uncertainty among the public and health-care professionals, discouraging people from taking preventive action or taking public health messages seriously. India has the expertise in medicine, public health, research, and manufacturing to lead the nation through the COVID-19 pandemic. To capitalise on these attributes, the country’s leaders must respect scientific evidence, expert commentary, and academic freedom, and not provide false optimism,” read the editorial.
The “supermodel” was also criticised because an epidemic, after all, is more than a set of statistical analyses. It requires knowledge of how a virus will behave in vitro and over a period of time; its behaviour in vivo and how the human body will respond to it – will it mount a defence or give in without a fight; and, how the public at large will behave. In other words, it is a complex interplay of virology, public health, epidemiology, and immunology, to name just a few of the disciplines.
The elements of statistics and mathematics, many would say, play only second fiddle to these sciences. A pandemic almost by definition, has scary numbers, but those numbers alone do not tell the full story. The research paper eventually appeared in the IJMR, the in-house journal of the ICMR that was middling at the best of times – with an impact factor of 1.5.
Public perception about it slid rapidly during the trying pandemic times to become one of the most distrusted and ridiculed medical journals in the country. Impact factor looks at a variety of aspects, such as how often research from a particular journal is quoted, how many views it gets, etc to estimate how seriously a journal is taken. The Lancet, for example, has an impact factor of over 60. The IJMR, unfortunately, was widely viewed as an outlet for the “subpar” research done by the ICMR.
This was a stunning fall from grace for an organisation that began its life as the Indian Research Fund Association (IRFA), established by Sir Harcourt Butler on 15 November 1911. The IJMR was born two years later. The IRFA was renamed the ICMR in 1949, under former Prime Minister Nehru. It is the apex body in India for the formulation, co-ordination, and promotion of biomedical research, and is one of the oldest medical research bodies in the world.
The “supermodel” paper published in the IJMR had three authors – Prof Agrawal from IIT Kanpur, Dr Kanitkar, Dy Chief Integrated Defence Staff (Medical), HQ Integrated Defence Staff, and M Vidyasagar of the Department of Artificial Intelligence, IIT Hyderabad. Epidemiologists had reason to be miffed; many felt it was a political exercise designed, among other things, as yet another validation of the controversial and crippling lockdown that India had been through. But speaking out is not easy in India.
Gautam Menon, Professor of Physics and Biology at Ashoka University, Sonepat, and at the Institute of Mathematical Sciences, Chennai wrote on his Facebook page: “DST funded a bunch of people to make a ‘supermodel’ in the full knowledge that this was going to be a political exercise. This expectation is certainly vindicated – if you tell people that a ‘model’ shows that the lock-down save many lives and keeping migrant workers on for many weeks before allowing them to return home was an unqualified success, none of them are going to look into the innards of the discussion to see how that was done. The usual cheerers will cheer and perhaps that is what was intended in the first place.”
The model, according to Menon, does not even cross the minimum threshold of acceptability. In the same post, he suggested a “simple test” to check: “Just send it to the top 10 epidemiologists in the world and ask them what they think of it, if you don’t trust me. The authors on the final paper are a computer scientist, a control engineer and a retired senior army officer – none of them have any epidemiological interests or expertise as far as I can see, although Dr Kanitkar is a medical doctor. The genuine epidemiologists on the committee no doubt declined as gracefully as they could, not wanting to be associated with this train wreck. The IJMR, a vehicle for ICMR, reduced its reputation even further by rushing a paper into publication that no referee in their right mind would consider acceptable. This is, put simply, a tragedy for Indian science,” he wrote.
The fact that the government had entrusted the job of plotting a pandemic to mathematicians and statisticians was clearly a disturbing notion for people who had spent their lifetime tracking infectious disease outbreaks. Interestingly, while the “supermodel” was primarily criticised for validating the Government of India’s COVID-19 management strategy and lockdown, some within the government were not happy with the timing of this model being made public.
This was mainly because it coincided with the period wherein the numbers had just started coming down from the September “peak”, but the festival season and the dreaded winter were right around the corner. That is why, one of the most important parts of the weekly COVID-19 press briefing was a reiteration of the need to keep using masks and be mindful of physical distance. For a government-appointed committee to choose this time to say that the worst was essentially over and that the pandemic would draw to a close by February 2021 ran slightly afoul of the government’s strategy of keeping people on tenterhooks to ensure COVID-19 appropriate norms continue to be followed in toto.
Excerpted with permission from Billions Under Lockdown: The Inside Story of India’s Fight Against Covid-19, Abantika Ghosh, Bloomsbury India.