A question that has been puzzlingmany people about the COVID-19 pandemic is why India’s death rate remains low relative to many other countries where the epidemic is widespread, and which are economically better off. Is it that deaths are not being reported or is there some protective factors at work?
This is not merely an academic question but an important one to draw lessons from evidence of the ‘first wave’, evaluate dominant understandings and recalibrate strategic planning for the ‘second wave’ as well as the subsequent endemic phase. We use available data for an inter-country comparison in an attempt to answer the question. Does demography, lifestyle, immunity, or the lockdown explain the paradox?
The Economic Survey 2020-21 (ES) of January 2021 narrates a congratulatory story of how India fared better than other countries in terms of COVID-19 cases and deaths. It claims that this was due to effective management of the pandemic and argues against the possibility that the low cases and case fatality rate (CFR) were due to some form of ‘natural immunity’. It has used mathematical modelling to estimate the likely number of COVID cases and deaths as well as the impact of ‘non-pharmaceutical interventions’ (NPIs) such as lockdowns. It corroborates the earlier modelling in March 2020 by Imperial College (United Kingdom) and Johns Hopkins (United States), of the NPIs ‘flattening the epidemic curve’ and thereby allowing time for health services to be geared up. This gives it scientific credibility, but only until one examines the data closely.
A closer reading of the ES reveals a reductionist approach that negates the complexities of both epidemiology and health service system dynamics. The March 2020 international analyses have been found to have overestimated the impact of the pandemic and of the NPIs’ effectiveness, primarily by assuming that no one will have any immunity against COVID-19 because it is a ‘novel’ virus and so the entire global population will be ‘susceptible’. The ES similarly applies universal models for calculating estimates of cases and deaths across countries without considering the possibility of diverse epidemic dynamics in different contexts, other than age or density of populations, testing rate and beds by population. These analyses treat the pandemic as some kind of new phenomenon appearing in isolation of other health conditions and their influences through ecological and epidemiological interactions.
Also read: COVID second wave: Clueless Centre cannot hold
While the ES’ analyses requires a separate article, this one examines the claims of the epidemiological effectiveness of the lockdown. The ES analyses reveal a reductionist medical technology and mathematical modelling-dominated mindset that prevents any thinking about other possible measures along with NPIs, hospital beds and ICUs, and vaccines. Negating the role of natural immunity in an epidemic is flawed understanding and belies data available before January 2021.
Comparison with three high-income countries
Over one year and two months into the epidemic in India, from the first case of COVID-19 detected on January 29, 2020, the country has reported 1,33,55,465 confirmed cases and 1,69,305 deaths. The numbers are, however, dependent on the population size. Therefore, we need to go beyond the plethora of horrific figures being put out daily, to compare rates, as in Table 1. It gives us some idea of the dynamics of COVID-19 in India compared to three high-income countries with open democracies.
Sero-surveillance, indicating the infection rate (IR), that is, the exposure levels to the virus in the population, is high in India compared with the sero-surveillance findings in the U.S., Germany and Sweden. It was way higher in all the four Indian cities—Delhi, Mumbai, Chennai and Ahmedabad—from where it was available by August 2020.
On the other hand, the case rate, that is, the number of cases per million population, is the lowest in India. Since this number is dependent on the rate of testing, we looked at the test positivity rate (TPR) , which should have been high if the low testing was missing out cases. We found that the TPR of India is way higher than that of Germany (which, like South Korea, has excelled in contact tracing and testing), but is of a similar order as in the U.S. and Sweden. So the low rate of testing is not likely to miss out cases much beyond those in these two countries. Similarly, the COVID death rate is lower in India compared with the three countries. The case and death rate per million could be low because the denominator is the whole Indian population while the spread (and testing) has largely been in the cities with later and lesser spread to rural areas. Therefore, more analysis of the reporting data and grounded studies are needed to assess the extent of under-reporting from outside the major cities to get firmer figures.
However, it is astonishing that India’s CFR is the lowest, that is, even among those who are symptomatic and tested to confirm the infection, relatively fewer people are dying. So, there does seem to be a real lower health impact of the pandemic compared with the other countries, and the reasons for this high asymptomatic IR and low CFR need to be understood.
Impact of Intervention Responses?
India had the most comprehensive, universally imposed lockdown in March 2020. The U.S. started lockdowns later, and not as stringently as in Indian cities. Sweden did not impose any lockdown. It only asked its elderly citizens to isolate themselves. Yet, as serosurveys found, COVID infection among Indian city dwellers was three times more than those in the U.S. and Sweden. For instance, it was already 21 per cent in Delhi by June-July 2020, soon after the lockdowns began to be relaxed, while it was 7 per cent in Sweden with no lockdown. Delhi’s IR escalated to 56 per cent by December 2020. A limited impact of the Indian lockdown is evident.
The lockdown and contact tracing-testing-isolation in India and the U.S. does not match up to Germany’s, so Germany’s 20 times lower IR is explainable. India’s medical services are not a patch on Germany’s and Sweden’s universally available services. The Global Index of Health Security (see Table 2) based on several parameters for health service availability and quality, emphasises that health services are poorer in India compared with high-income countries. The relatively higher CFR in the three countries cannot be attributed to medical care in India.
While the slowing down of the viral spread owing to the lockdown gave some preparation time to ramp up infrastructure and equipment on a war-footing, what is notable is that, according to the Union Health Minister, only 2.32 per cent of the COVID patients needed oxygen, 1.61 per cent ICU care and 0.28 per cent ventilator support. Despite this, the overwhelming of hospitals in Indian cities during the peak period of the epidemic revealed the unaddressed lack of human resource in hospitals, limiting the advantage gained by increasing bed capacity and equipment. The delay in salary payments to front line workers and poor working conditions took a further toll.
So, the low CFR is mainly due to the low proportion of serious cases, and less attributable to tertiary level medical interventions. This is not to undermine the essentiality of quality medical services in reducing suffering and loss of lives, but to recognise the overriding epidemiological processes during the pandemic. It is significant that there is lower conversion in India from ‘infected’ to ‘case’, and from ‘mild to moderately symptomatic’ becoming ‘a serious case’.
Also read: Second wave of COVID-19 in India
With this inter-country analysis, it is difficult to accept the claims of impact of official interventions as explanation for India’s lower CR and CFR. The lockdown certainly slowed the progress of the pandemic in India, as the rate of increase was higher until the first two weeks of April compared with mid-April onwards. However, the extent of slowing down is unverifiable since it can only be based on projections of ‘what would have been if the interventions had not taken place’, with such projections classically erring on the side of overestimation. It is clear that low CFR cannot be an outcome of medical interventions alone.
It must be noted that the effective control of the epidemic in Germany, South Korea and Thailand as well as in Indian States such as Kerala is not so much due to hospital services as to primary level public health services. The containment measures are conducted by the primary level staff in the communities. The first contact treatment of confirmed cases, too, happens through domiciliary care and COVID centres. Hospital services can save the lives of those who seek them in the serious stage. Indian cities generally suffer from a lack of primary level services in the public system. No wonder the IR is high and hospitals are overwhelmed.
Explaining the paradox
Indicators of the four countries show that India certainly has an advantage by virtue of the proportion of population above 65 years of age being markedly lower (as in Table 2). But this seems to have a caveat, since an early analysis by P. Mukhopadhyaya had shown that age-specific CFR for COVID cases was higher among the younger age groups in India as compared to Italy. Prevalence of non-communicable diseases (diabetes, heart disease, hypertension) is known to be high in the younger age groups in India. So this may be a contributory factor.
Population density seems an indisputable factor for COVID spread. Indian cities with high population densities and a high number of young asymptomatics could have contributed substantially to the high IR in India. In Swedish old-age homes, the density of susceptible persons being high could explain its higher CFR.
Equally relevant is the consideration of the generally higher occurrence of communicable diseases in India and other low-middle income countries compared with high-income countries, except in the case of some diseases such as influenza and COVID-19. This is important to consider because both influenza and COVID-19 have a similar mode of spread, affect those with lowered immunity and yet are markedly more serious public health problems in high-income countries.
Environmental factors, lifestyle and viral load
Could it be climatic factors, hot and dry weather as has been suggested, that have limited the incidence of influenza and COVID in India? The rising number of cases in the summer months overturns this claim.
However, the high population density in India could facilitate more rapid spread. The majority remaining asymptomatic could be due to a lower viral load resulting from the difference in lifestyle in Indian cities, where there is more open-air routine activity than in the high-income countries. Alternatively, or in addition, it could be a higher level of immune efficiency owing to habituation to a high infectious disease load.
Natural Immunity as a Factor
The lower conversion from IR to CR and from CR to CFR, both indicate some process of immunity. Contrary to assumptions of initial modelling of the pandemic, data from sero-surveillance in the general population and other studies indicate that there is some form of pre-existing ‘general immunity’ and ‘innate immunity’ in a proportion of the population, and that ‘specific immunity’ develops after exposure to the virus.
A systematic review by Shah and others at the Indian Institute of Public Health (IIPH) Gandhinagar showed that, across 14 countries, the secondary attack rate, that is, more cases in the household after one COVID case, is only about 5-50 per cent. Thus, despite living in close proximity to infective COVID cases, in more than half the households other family members had protective immunity.
The story of the classic Tuberculosis Prevention Trial for effectiveness of the BCG vaccine against pulmonary tuberculosis, conducted in the 1970s by the Indian Council for Medical Research (ICMR) in collaboration with the World Health Organisation (WHO) and the Centres for Disease Control (CDC), U.S., is worth remembering in this context. While similar trials in Britain showed up to 80 per cent effectiveness, the BCG trial in India found ‘zero’ effectiveness of the vaccine. This result was understood to be a consequence of pervasive exposure to other bacteria similar to the TB causing bacillus. Thereby, immunity had developed even before the vaccine bacillus entered the body, and thus the vaccine provided no advantage whatsoever. Could a similar process of innate or prior immunity be acting in the case of COVID-19 infection?
Thus, to some extent the younger age of the population could explain the low conversion from IR to CR, but more strongly, some lifestyle or population immunity-related factors seem to be responsible for India’s advantage. The high levels of sero-prevalence have coincided with the decline in occurrence of cases and deaths in all the cities surveyed, leading to a logical inference that some form of protection had developed with the high exposure. Alternatively or in addition, it could be that there was already some pre-existing innate immunity or protective response due to prior exposure to other organisms with similar antigenicity.
Since the problem in serious cases of COVID-19 arises from a hyper-immune response, is there some process keeping that in check? Could the innate immunity against viruses have been bolstered and regulated by the immune-modulatory role of culturally widespread use of herbal formulations (and region-specific Ministry of AYUSH preparations)? The phenomenon of population immunity needs to be studied more rigorously and understood in all its complexity and under diverse conditions: demographic, material, social, cultural, ecological, health care-related and epidemiological. This is a subject rather neglected in the past few decades owing to the overriding hope placed on vaccines.
Implications for Action
Our analysis has highlighted two phenomena that require reconsideration of the dominant approach, besides requiring the Central government to review its complacency reflected in the ES, which is costing people their lives in the second wave. One is that there was a small decline in the spread owing to the lockdown but that population-level virus spread continued even during the lockdown at a higher rate than in other countries. Restricted closures seem necessary when the surge seems to overwhelm hospitals and cremation/burial grounds, but in the long term they only delay cases until the next spike. Overwhelming of hospitals witnessed in several cities in the initial surge in May-June 2020 was not there in later months, not because cases declined but because fewer cases became serious and the medical system learnt better clinical management techniques. Hospitalisation declined due to clinical regimens followed as outpatients and through online consultations, requiring only primary level and general practitioner (GP) services. Recognition and valorisation of these service providers (and not only of hospitals and ICUs), with urgent training of GPs and other primary level health care providers in rational treatment regimens of mild and moderate COVID-19 cases, will allay anxieties and give confidence to patients to continue in home care.
Anecdotal evidence from health centres in India and other countries demonstrates the value of such care in allaying the panic of patients, providing adequate first-level care and identifying in time those who need to go to hospital. If we had ramped up such services in preparation for the second wave we may not have seen it overwhelming patients and hospitals.
Secondly, that there is development of population immunity faster in India than in the other countries due to either a faster spread because of higher population density but with lower viral load and/or some unknown phenomena such as repeated immune challenge that elevates protective immunity. Understanding this phenomenon and bolstering it may add cost-effective and sustainable measures to India’s epidemic control strategies, even while it struggles with administering vaccines.
While primary health care services play a decisive role in limiting the spread of the virus and, therefore, need to be strengthened, the high IR but lower CR and CFR relative to the other countries can be explained epidemiologically only by some other protective factors. However, it remains to be seen whether the low CR and CFR will persist when the virus spreads to rural areas and among the most undernourished communities, and result in a surge of cases.
Ritu Priya is a medical doctor and public health professor and Krishna Choudhary is a. researcher scholar at the Centre of Social Medicine and Community Health, Jawaharlal Nehru University, New Delhi.
