Scientific uncertainties, weak institutions, and nationalism led to COVID-19 devastation

People at a wholesale shopping market ahead of the Holi festival in Delhi on 26 March 2021. Rajat Gupta/EPA
14 May, 2021

By the time 2020 drew to a close, several countries around the world—and indeed cities in India—had seen multiple COVID-19 surges. Some of these surges continued into 2021 and were proving catastrophic. New and more dangerous variants of SARS-CoV-2, the virus responsible for the disease, were emerging, including in the United Kingdom, Brazil and South Africa. Globally, the pandemic was far from over. India, however, had been seeing a steady decline in cases. The decline was not uniform: Delhi had surged in November, and cases in Kerala were still high at the end of the year; but the national outlook seemed fairly bright. 

By February, though, storm clouds were gathering. I wrote that a new wave could be developing, but a swift response could still prevent it from going national. Writing in IndiaSpend, the journalist Rukmini S discussed Maharashtra, the international situation, and new variants of the virus, concluding that “new evidence is casting doubts over India's pathway through and out of the pandemic.” By early March, the risks were even more evident. Based on the data they had gathered, the Indian SARS-CoV-2 Genetics Consortium or INSACOG reportedly warned top officials of the looming dangers posed by new home-grown variants whose properties were not yet clear. We now have plenty of evidence that INSACOG was right.  

But there is no evidence that government was listening, or watching the data, or tracking the international situation. The government’s own scientific taskforce failed to meet during February and March, and individual members of the taskforce who were concerned about a new wave were unable to make their voices heard. Instead, the first four months of 2021 saw a slow vaccine  roll-outbacking for a full-fledged Kumbh , and long drawn-out elections with vast rallies. On 7 March, Harsh Vardhan, the health minister, declared that India had reached the “endgame of the COVID-19 pandemic.” Two days later, Manindra Agrawal, a member of a panel formed by the department of science and technology to model the COVID-19 trajectory and author of the government-backed “Indian supermodel,” confidently tweeted that there would be no “second wave.” Aside from political considerations, it appears government inaction on COVID-19 may have been driven by flawed scientific advice and narratives which had become popular over the course of the pandemic. What were these narratives?

Knowledge of COVID-19 has evolved rapidly, and is still evolving. There have been competing theories about the origins of the disease, the ways it spreads, the nature of immunity, fatality rates, and the properties of new variants. Some theories are speculative, but have high-profile backers. It is not easy navigating this changing landscape, and public-health bodies have made mistakes. Consider the early recommendations by the WHO against face masks, and the painfully long time it took for the United States Centres for Disease Control and Prevention to acknowledge that COVID-19 is airborne. 

Although the scientific uncertainties are global, other factors have helped shape an understanding of COVID-19 in India: advisory bodies lacking independence; limited and sometimes manipulated data; and, above all, a tendency to view all problems through a lens of nationalism and exceptionalism. 

The latter tendency is well illustrated in an infamous editorial by Shekhar Gupta in The Print from April 2020 titled “Covid hasn’t gone viral in India yet, but some in the world & at home can’t accept the truth.” It includes the line, “Our crematoriums and graveyards are not out of wood or space.” We see a trend which is now familiar: an obsession not with the epidemic itself, but with perception—particularly international—of government handling of the epidemic.

 If the main goal of government and its cheerleaders is to manage the narrative, then gaps in knowledge and data are not a worry. On the contrary, weak data provides opportunities to construct convenient stories. 

As the first wave of disease wound down, there were some unknowns. The first was how widely the disease had spread in different parts of the country. Highly variable testing and death recording meant that case and fatality data were of little use in gauging this. Data from seroprevalence surveys provided insights, but was too patchy to create a complete national picture. 

The second unknown was how many had died of the disease in different parts of the country. Poor surveillance and deliberate manipulation led to a situation where the true death toll was hard to estimate. This remains truer than ever today.

The third unknown was what led to the winding down of the first wave. What were the contributions of continuing mitigation and population immunity? Were some parts of rural India largely spared as a consequence of lockdown? Did high levels of infection drive the slow-down in city slums, while reductions in mobility were key in non-slum areas? 

The breathing space between India’s COVID-19 waves could have been used to try and shed light on some of these questions. Instead, there was self-congratulation and spurious analysis. 

The first chapter of this year’s Economic Survey, released in January 2021, is about government handling of the COVID-19 crisis. It uses misleading analysis to claim that “India has been able to effectively manage both the spread of COVID-19 and the fatalities.” States are naively ranked according to their cases and fatalities with Maharashtra, in particular, singled out as a failure. So, limited data was convenient when it came to managing the narrative. 

But official narratives were also changing in important ways. By the end of June 2020, government spokespeople talked less and less about limiting disease spread, and more and more about limiting fatalities. In early briefings “no community transmission” was a frequently repeated claim. The fact that it was repeated suggested that the aim was to keep infections low. But as time progressed the official line seemed to become that the country has seen relatively few deaths, even though the disease has spread with some limited consequences. Officials were hinting that disease containment may not be so crucial any more. 

The emphasis on mortality is problematic when death data is so incomplete. One particular danger is that if fatalities become a key measure of success or failure, the pressure on states and local authorities to under-report COVID-19 deaths is likely to grow. 

It was not just the government now emphasising fatalities. In August 2020, public health experts including members of the Indian Public Health Association issued a statement arguing that the focus should be on preventing deaths, rather than containing the disease. The statement made some important points about the harsh side-effects of lockdown, but failed to explain how deaths were to be limited without disease containment. 

It seems that both government and some public health bodies were suggesting that mitigation needs rethinking. Such lines of thought were not limited to India. Shortly after the IPHA statement, a number of scientists signed an open letter called the Great Barrington Declaration, in which they argued for allowing disease to spread amongst “those who are not vulnerable” while adopting measures to “protect the vulnerable.” A right-wing US think-tank sponsored the widely-criticised declaration, which implied that the risks of disease-spread had been overstated. 

The Great Barrington Declaration reflected a growing backlash against mitigation. But with no vaccines available at that stage, what was going to end the pandemic? The authors answered clearly: herd immunity through natural infection. India’s COVID-19 response, particularly in the months leading up to the second wave, appears to have been guided in part by misplaced beliefs about herd immunity.

The basic idea of herd immunity is that once a sufficient fraction of people become immune to a disease, either through infection or vaccination, it winds down naturally. The fraction who need to be immune before we see outbreaks dying away is termed the herd immunity threshold, or HIT for short. Several studies have estimated that COVID-19 has a HIT of around 60 percent. But this value is not fixed: it depends on the virus and its environment. More transmissible variants would drive the threshold up, as would more crowded housing. 

Herd immunity is an idealised mathematical notion, to be applied with abundant caution. In its simplest form, the theory relies on populations interacting in simple ways. Treating India, with its slums, its high-rises, and its various patterns of rural development, as a single entity to apply this theory is risky. 

The simple theory also assumes that reinfections can be ignored. This was a popularly held view. For example, in October last year, the New York Times ran a piece with the headline “Coronavirus Reinfections Are Real but Very, Very Rare.” A number of studies on reinfection indicate that reinfection is not that rare, and this headline is, in fact, misleading. 

Several groups have attempted to address some of the limitations of basic herd-immunity theory. Some extensions to the theory conclude that herd immunity is harder to achieve than at first thought, for instance because of more transmissible variants or reinfections. But there were also modifications to the theory that came to optimistic conclusions: some papers suggested that a population may effectively reach herd immunity at relatively low levels of infection, perhaps even as low as a mere 10-20 percent.

The real-world relevance of this analysis was unclear. Some serosurveys in slums in India were reporting high levels of infection at odds with a low HIT. But the notion that the HIT in India could be considerably lower than originally presumed became widespread. In September 2020, two public health experts wrote an opinion piece in The Hinduclaiming that “Since about 30% herd immunity is sufficient to reach the peak of the epidemic curve, we can be confident that India indeed has reached the peak of the COVID-19 epidemic.” A piece in Dainik Bhaskar the following month claimed that once 40 percent of a population have antibodies to the disease the dangers diminish. 

In fact, a paper on the second national serosurvey, co-authored by the head of the Indian Council of Medical Research, cited two preprints that made the argument for a low HIT. Both preprints included authors who were signatories to the Great Barrington Declaration. Of course, these citations do not imply that “India’s HIT is lower than expected” was an official line. But this literature was certainly on the radar. 

By the time the first wave had wound down, a number of groups were claiming that India was at or close to herd immunity, either because the HIT was lower than assumed, or because spread had been very wide. In January, VK Paul, NITI Aayog member and chairperson of the national committee on vaccine strategy, stated that, "Most of our highly populated districts and cities have had their run of the pandemic by now ... and may have what you like to call herd immunity, to an extent." Agrawal, the researcher with the government-backed modelling project, went further and claimed in February that herd immunity had been reached nationally. This claim was based on a calculation  that 60 percent had been infected nationwide at this point. But the third national serosurvey had estimated that only 20 percent of adults in India had so far been infected. 

Given that some of these reckless claims of herd immunity were coming from members of a panel formed by the department of science and technology, it seems likely they contributed to official complacency. As Gautam Menon, a biophysicist, wrote presciently in October 2020, “Basing public health policies on models that are flawed is dangerous, since the lives of people are at stake and false optimism carries risks with it. Science should not serve political ends.”

In summary, it seems the belief that India was close to herd immunity was widespread as the first wave wound down. At the same time, low recorded fatalities were used as evidence that you could get away with allowing disease to spread with limited consequences, a dangerous view also proposed by various international right-wing groups. 

The possibility that death undercounting might account for the low recorded mortality was ignored. So was the concern that wide spread of the disease increases the risk that more dangerous variants of the virus will arise. Even though government and its backers praised national lockdown uncritically, the role of further mitigation was downplayed.

In the lull after the first wave, government-aligned scientists could have called for a countrywide survey of COVID-19 spread and fatalities. We might have found out whether mortality really had been low, why the wave wound down, and what were the main vulnerabilities. They could also have called for rapid vaccination and ramping up of health infrastructure to cushion the blow from possible future waves. But there were no such calls. 

All said and done, weak science and compromised scientific institutions were at least partly responsible for the complacency and mistakes that allowed the second wave to take off so dramatically and tragically. Independent and expert scientific advisers might have been able to push for early action to avert the crisis. Such action could, at least, have bought some time and lessened the toll. 

 Murad Banaji is a mathematician with an interest in disease modelling.