Chasing the virus

While COVID-19’s onslaught has laid low the mighty United States and much of the developed world, in India there are differing perceptions on the 21-day lockdown strategy, which includes a low rate of testing, and its effectiveness in containing the infection.

Migrant workers along with their families waiting to board buses to return to their villages, in Ghaziabad, on the outskirts New Delhi, on March 29.

At the time of writing, the totalnumber of COVID-19 cases across the globe had crossed the 1 million mark and the epicentre of thepandemic, four months after it emerged in Wuhan, China, in December 2019, has moved from Europe (Italy) to the United States. The biggest economic power in the world is reeling under the impact of the disease with 3,11,544 confirmed cases and 8,488 deaths (as on April 5) even with stay-at-home advisories and physical distancing measures in place. It is feared that 100,000 to 200,000 could die from the disease in the U.S. In comparison, India’s case load and death toll have been pretty much on the lower side, with 3,374 cases and 77 deaths (as on April 5).

On March 24, following an announcement by Prime Minister Narendra Modi the whole country went into a 21-day lockdown period beginning March 25. This was beyond the various non-pharmaceutical interventions (NPIs)— such as closure of schools and other educational institutions and selected workplaces, restrictive social measures such as physical distancing, stay-at-home/work-from-home advisories, avoiding mass gatherings, specific hygienic practices and a two-week self-quarantine by people with fever and cough—which were put in place gradually fairly early on since March 5. Obviously, preventing close contact among people through such measures without resorting to lockdown, slows down human-to-human transmission of the virus and as such the rate of increase in the case load would have come down, as was achieved in South Korea, Taiwan, Singapore and Japan. (In the wake of a consistent increase in the number of cases for two weeks, the Singapore Prime Minister announced a month-long partial lockdown, or what he calls “circuit-breaker”, to break transmission chains.)

It would seem, however, that a slight hike in the rate of increase in the number of confirmed cases from 12-15 per cent in early March up to March 17/18 to consistently over 20 per cent for the following six days (Figure 1) may have made the government press the panic button without probably looking at the real cause of that hike. Had there been a wider and broad-based diagnostic testing approach, the epidemiological cause may have shown up but despite the World Health Organisation’s (WHO) advice to affected countries to ramp up testing, testing remains woefully sub-optimal, at around 20 per million population. Had the level of testing been adequate, appropriate measures could have been taken to contain the infection and prevent the large increase in the number of cases needing hospitalisation and intensive care. So, with a strategy that was not anchored to widespread testing, lockdown was perhaps seen as the only viable alternative to prevent the rapid spread of the contagion. The Indian Council of Medical Research (ICMR) presumably had rational medical arguments in the Indian context to advice the government thus.

T. Jacob John, the noted virologist who was formerly with the Christian Medical College (CMC), Vellore, told Anadolu Agency, a Turkey-based international news agency: “India lost a good lead time when we knew that this was a pandemic in the making. The government had been warned in advance. A pandemic needed three months to go around the world, and see where we are now. We would have been better off if we had taken it seriously in early January. Most of the precautionary measures should have been taken in January itself when the government was warned. We are still mainly in a reactive mode and fighting the enemy from wherever it crops up, instead of being proactive. India has no war-room thinking, just heroic fighting.”

Even if such a country-wide lockdown was indeed deemed necessary, unlike other countries that have taken this drastic measure including China, this was not a planned exercise whose wide-ranging implications had been thought through. As in the case of the demonetisation decision on November 8, 2016, this too was a sledgehammer. A gradual and selective lockdown of regions and economic sectors in a phased manner would have enabled better preparedness and mitigated the impact, especially on incomes and livelihoods of the working class and daily wage earners, particularly migrant labourers, with food, shelter and cash compensations through the lockdown period. Such emergency measures to fight epidemics are taken essentially to “flatten” the exponential curve so that the case load increase is stretched over a longer time period and the peak case number does not become overwhelmingly high. It gives the government and its administrative arms, in health and other sectors, time to evolve treatment and associated strategies.

Jacob John recently told The Hindu that with community transmission of infection established (notwithstanding the government’s stand to the contrary), it was important to effect a shift in strategy and get the act together to direct resources on the health infrastructure front, which was bound to get overwhelmed in the coming months. “This is not a hotspot transmitting virus but it is a virus that transmits everywhere…. That phase is more or less over…. Chasing the virus must be phased down and checking the virus in sick people must be stepped up. The strategy from now onwards should be lesser and lesser efforts paid for travel-related search and more emphasis on illness-related testing…. Anybody with fever and cough must be checked.” He said containment measures should be given only 30 per cent priority; health care testing should get 70 per cent priority.

It is pertinent, therefore, to ask if the lockdown strategy achieved its objective of containing the infection, at least for now, notwithstanding the huge collateral damage to the economy and hardships to millions of poor people. Simple logic tells us that it should have by slowing down the virus transmission drastically. Many modelling studies have shown how the widespread but planned and phased lockdown in China imposed on January 23 succeeded in first preventing the spread of the virus from Ground Zero at Wuhan in Hubei province to other provinces and then in bringing down the overall case rate in the country.

The lockdowns across China have been eased with Hubei being the last province to lift restrictions in a phased manner from March 22. The number of new cases in Hubei has dropped to practically zero a day. This is true of other provinces, too, where easing of restrictions began in February itself, as Imperial College epidemiologists observed in a recent study, though it is too early to judge the course of case load in the months to come. It remains to be seen if easing of extreme restrictive measures results in the emergence of a second wave of infections.

Second wave

“[W]e need to be alert for a potential second wave of infections,” Nature has quoted Ben Cowling, epidemiologist at the University of Hong Kong (HKU), as saying. According to him, if a second wave does come, it should happen by April end. One modelling study in the United Kingdom suggests that physical distancing measures, such as closure of schools and other educational institutions, may be needed for extended periods to keep the number of severe COVID-19 infections low. But China has kept its extensive testing and contact tracing in place, and if that strategy proves the model wrong and is able to avoid the need to prolong such measures, it could be a lesson for other countries to follow.

But, given the high infectivity of SARS-COV-2, the risk of new outbreaks in China is not low and many cases are out there undetected, Gabriel Leung of HKU told Nature . He said one lockdown might not be enough; severe measures to suppress the second wave might be needed. The virus, he said, would have difficulty in re-establishing itself if 50-70 per cent of the people had been infected and were now immune. According to him, even in Wuhan this number is probably only around 10 per cent, which means that a large chunk of the population is still susceptible to the infection.

In India, too, for six days after the lockdown (which seems to have been, however, greatly unplanned), the rate of increase in the case load did drop from 20-odd per cent to 15-16 per cent until a cluster of cases began to emerge from different parts of the country of people who attended a religious congregation called Tablighi Jamaat at the Nizamuddin Markaz Mosque in South Delhi during March 13-15, and their contacts. The surge of cases linked to the Tablighi Jamaat seems to mask the decline in the rate of increase in the number of cases after March 30 owing to the lockdown and has, in fact, resulted in an increase in the rate.

About 2,000 people, including200 foreigners, many of whom had come to India as early as March 8, attended the Jamaat congregation. The cases linked to it started showing up after the known average incubation period of about twoweeks. The sudden increase in the number of cases in Telangana, Tamil Nadu and Delhi were traced to the attendees of this meeting. The fact that the meeting was allowed to take place points to a gross failure on the part of the law-enforcing authorities in the capital in implementing the Delhi government’s March 13 order against large gatherings (of more than 200 people) in the wake of the pandemic.

Also, the directive from the Ministry of Health and Family Welfare Ministry (MoHFW) for 14-day home isolation and self-quarantine for all travellers coming into India was issued only on March 18, three days after the congregation. The earlier order for 14-day home/self-quarantine, issued on March 10, was only for visitors and tourists from select countries with a significant number of COVID-19 cases. While some of the passengers did indeed come from these specific countries, a large number of them came from countries such as Indonesia, which were not among the specified countries. Had the government been more aggressive and proactive in containing the spread by enforcing the 14-day quarantine on all passengers arriving in India from the beginning of March itself, and not on March 18, the Nizamuddin mosque would not have emerged as the coronavirus hotspot. From this perspective, initiating criminal proceedings against all foreigners who attended the gathering would seem highly misplaced.

More lockdowns may help

A recent modelling study by Ronojoy Adhikari of the Institute of Mathematical Sciences (IMSc) and Rajesh Singh of Cambridge University looks at the impact of measures restricting physical and social contacts, including the 21-day countrywide lockdown. Unlike similar modelling studies on other countries, notably China, this study takes into account the particular age and social contact structure one obtains in India. The latter includes such aspects as three-generation households where children and grandparents live together, which gives rise to the possibility of substantial transmission between the first and the third generation given that the virus impacts the elderly severely. The paper points out that models that do not resolve age and social contact structure cannot capture the differential impact of each one of the measures enforced in India. The model makes the idealistic assumption of absolute absence of all social contacts and strict compartmentalisation, save household ones, and lockdowns are incorporated by total “switching off” of infection spread because of which the case load will decline immediately following a lockdown.

The study concludes that the 21-day lockdown is of insufficient duration to prevent resurgence. “Alternative protocols of sustained lockdown with periodic relaxation can reduce the infection to levels where social contact tracing and quarantining may become effective,” they write. The duo’s analysis further shows that a second lockdown of 28 days with a five-day gap cannot prevent the exponential resurgence after the 54-day period. Three successive 21-day, 28-day and 18-day lockdowns spaced by five-day gaps between them can bring down the case load per day to under 10 by June 10. But a more sustained single 49-day lockdown can bring down the case number to that level by May 13 itself.

But as Gautam Menon of IMSc and Ashoka University has pointed out in his article in The Wire , all models make assumptions that one does not obtain in the real world; in this case, continued exposure to infection due to departures from the strict no-contact assumption. In an email response to a query from Frontline , Adhikari said: “Our model is general, but has been used to construct a best-case scenario: social distancing is assumed to be instantaneous and complete in all spheres of contact other than the home. This provides an epidemiological lower bound [to the morbidity and mortality numbers], just as the unmitigated epidemic scenario… provides upper bounds…. Mathematical models provide if-then scenarios, but policy decisions must take into account, amongst others, economic, medical, social, and ethical considerations.”

A containment model with leaky compartments would have predicted that the numbers would continue to increase before beginning to drop, as ground data show. Since leakages of different magnitude at different strata of society—such as slum clusters and migrant labourers—would exist given the social realities in India, predictions of this and other models, on the basis of idealised assumptions as they invariably are, can only be guiding posts for NPIs at different points of time. Another problem with this model, Gautam Menon points out, is that it ignores asymptomatic infections, evidence for which is available in different parts of the world. Therefore, we need to watch the progression in the spread of infection in the coming days up to April 15, and how the government plans the exit strategy, about which there has been no word so far.

But what is puzzling is that, despite the low level of testing and the consequent non-reliability of case load data, why do the numbers in India continue to be low with the slope of the exponential only marginally steeper than the linear growth compared with the steep growth rates observed in other countries? According to worldwide data, particularly from extensive studies in China, we know that nearly 85 per cent of infections cause only mild symptoms from which patients recover soon by isolation and quarantine. About 15 per cent require hospitalisation and intensive care, a third of which die, most of whom belong to the age bracket of 60 and above. Of course, given the demographics in India with only 8.5 per cent in that age bracket, the fraction of deaths could be lower than a third but not significantly because China too has a low proportion (10.8 per cent) of elderly.

While much of the 85 per cent of the actual case load (not to mention the asymptomatic carriers at any given point of time) are being missed because of inadequate testing in the population and, therefore, would be below the “very low aperture testing radar”, the 15 per cent requiring hospitalisation, with a significant number requiring intensive care, would “not be sitting at home”, as the public health professional N. Devadasan observed in his April 1 article in Scroll.in , but would present themselves in hospitals even if they had been missed by the narrow testing criteria of travel abroad and/or links identified by contact tracing from a known case.

For a country with a large population such as India, 15 per cent of patients with severe respiratory problems will not be a small number given the projections of huge numbers. For example, modelling by scientists at Johns Hopkins University, U.S., in association with the U.S.-based Centre for Disease Dynamics, Economics & Policy (CDDEP) led by Ramanan Laxminarayan, with the so-called IndiaSim model predict 1-2.5 crore infections between March and August, and some 90,000 cases requiring hospitalisation in Uttar Pradesh alone.

A third of the latter figure (or marginally lower) implies a huge number of deaths, which cannot go unnoticed. But there is no evidence of any noticeable surge in the number of hospitalisations or deaths. Given this situation, Devadasan argued that perhaps India overreacted to the emerging scenario without paying attention to what data indicated. But his argument of overreaction is in hindsight. Severe NPIs would be needed given the world scenario, if not complete lockdowns. Nevertheless, with the missing higher than normal rate of hospitalisation and deaths with respiratory complications, the question remains whether there is any uniqueness about the Indian population that the numbers have remained low.

Extrapolating from the past experience of low incidence of SARS (caused by SARS-COV-1 virus) in India in 2003 compared with other nations, it has been argued by media commentators that perhaps the innate immunity of the Indian population against coronaviruses arising out of its unique genetic make-up may be the reason for that. Of course, arguments such as anti-correlation between malaria prevalence and COVID-19 incidence and between level of BCG vaccination and COVID-19 incidence have been making the rounds in social media. But these are not based on hard data or analysis.

Unique mutation

However, scientists from the International Centre for Genetic Engineering and Biotechnology (ICGEB) in a recent paper posted on the preprint repository biorXiv.org make an interesting observation. By comparing the published genome sequencing details of the novel coronavirus from its isolates in India, Italy, the U.S., China and Nepal as well as the genome data of the related viruses, SARS-COV-1 and MERS, they find that of these seven genomes only the Indian virus isolate has a unique mutation in the structure of the spike surface glycoprotein S that binds to a particular receptor molecule in the human cells and helps the virus evade the host immune system to proliferate and cause disease. More interestingly, of the 51 microRNAs that one finds in the human host system, one particular microRNA termed hsa-miR-27b has the capacity to recognise this particular mutation in the viral genome and trigger an immune response.

MicroRNAs are small non-coding stretches of RNA molecule (about 22 nucleotides long) found in plants, animals (including humans) and viruses that are now increasingly understood as having an important role in host-virus interaction. Specifically, they can alter the viral gene expression and have the potential to control the pathogenesis of an invading virus if there are appropriate molecular targets that these microRNAs can recognise and latch on to and interfere with the viral replication process. The ICGEB scientists have found that the unique mutation in the Indian isolate of SARS-COV-2 is a potential target to which the human host’s microRNA hsa-miR-27b can bind. Whether this particular host-virus interaction does offer any immune protection remains to be seen by studying many isolates of the Indian strain.

Rahila Sardar, one of the authors of the paper, is quick to point out that neither the paper nor the authors are claiming that the Indian virus is of low virulence and the Indian population can mount an effective immune response to it. “From the analysis of genome data of a single isolate, we cannot make any such claim,” she said. “We analysed only one genome sequence because, of the two genome data of Indian isolates available in the public database GIASID (Global Initiative on Sharing All Influenza Data) only one has better coverage of nucleotide sequences. We need more sample isolates to see if this mutation is common to all their genomes,” she said.

RNA viruses, such as SARS-COV-2, mutate often because the process they employ for replication is highly error-prone and their genomes accumulate these mutations during every replication cycle as the virus proliferates into millions of copies in the host system.

So one cannot conclude anything from a unique mutation observed in a single isolate. It would make sense only if the mutation turns out to be stable and is seen in a large number of isolates of a particular strain of the virus. Of course, this raises the question why more isolates of the Indian SARS-COV-2 strain were not studied and their genome data published. This would be of importance for evolving appropriate therapeutic intervention strategies in the Indian context.

So, such “feel good” arguments are of not much use during a pandemic caused by a highly infective virus, although there is certainly a need to epidemiologically understand the apparent low incidence of the disease in India. Is it because of the NPIs put in place early on or are there other factors at play? We do not know yet. Meanwhile, at an individual and societal level we need to observe all advised precautions to keep the virus at bay.