The continued upward trajectory of COVID-19 infections in India is a matter of concern. After the United States and Brazil, India has the third largest number of confirmed cases in the world. On August 3, India’s daily COVID-19 count surpassed that of the U.S. and Brazil. Even though the fatalities as a percentage of the confirmed cases are low, the alarmingly high positivity rate is an added cause of concern. The lockdown period was an opportunity to aggressively test, track and treat and fill the gaps in public health infrastructure, but that opportunity seems to have been frittered away. Professor K. Srinath Reddy, a public health expert and member of the Indian Council of Medical Research’s (ICMR) high-level technical expert committee on COVID-19, spoke to Frontline on India’s contribution to the global pool of knowledge on COVID-19, the reasons for the virus extending to new territories and the lessons learnt. He underscored the need for doubling India’s health expenditure and for a drastic reduction in out-of-pocket expenditure from its current levels. He also said there was a serious shortage of personnel at all levels of care and that investing in expansion of the health workforce was both a health system imperative and an economic opportunity for job creation. Excerpts from the interview:
What explains that even after a long period, India continues to have an upward trajectory of confirmed and active cases? Why is it that India has not able to reverse this trend as has happened in several other countries?
Our story has been one of the pendulum swinging too widely from lockdown to laxity. The COVID-19 virus exploits every weakness in containment measures. It is active again even in countries that had achieved good control earlier. This is now evident from Spain to Germany and Hong Kong to Japan. This sends a clear message that we cannot take this virus for granted and must maintain vigil for several months more.
The weakness of our public health systems came to the fore after we opened up. Early detection of cases, through symptom-based syndromic surveillance of households, prompt testing and isolation of suspected cases and close contacts, accompanied by vigorous contact tracing, are measures that are essential components of a sound surveillance and containment strategy. These have not been implemented adequately as a full package. The virus was also given ample opportunities to spread. Large gatherings were allowed for social or religious reasons. Masks came in late and have not always been worn properly, failing to prevent effective transmission.
However, it must also be noted that India is a vast country with a large population. The roll-out of the epidemic here has geographic and time dimensions that differ from [those of] a small country. If each of our States had acted with alacrity and efficiency from early on, rather than being dependent on Central directives, we may have seen more vigorous control. Within the States, district-level decision- making would have been needed from the very beginning, given the size of our districts. Even now, that is the level at which the response must be shaped, with monitoring at the State level and support from the Central level.
When is this likely to change and what would be the additional measures needed to achieve this?
The virus is now extending its presence to new territories even as those affected earlier are battling it out. So we will see different calendars for the epidemic in different parts of the country, with a landscape of many peaks and undulating hills portraying the infection.
We need to vigorously contain the transmission within the already affected urban zones while we energetically block passages of entry into villages and small towns. We have to step up the frequency of primary healthcare personnel-led household surveillance of symptomatic persons and close contacts for early testing and isolation, followed by energetic and extensive tracing of all named or indicated recent contacts. Citizen volunteers and elected local bodies can add to the strength of the formal primary healthcare system in performing these functions. Mild cases may be isolated, cared for and monitored at home, while persons with moderate or severe illness must be hospitalised without delay.
We need to develop integrated data systems that combine socio-demographic profile, emerging epidemiologic information, health workforce availability, healthcare facility capacity and readiness, supply chain status and transport data to quickly profile local area challenges and resources for a swift and contextualised response. While such data may be transmitted to higher levels of district, State and Central administration, village and ward-level data must be readily available to local implementers for responding without undue delay. So, people-partnered public health and decentralised data-driven decision-making must become the main engines of our epidemic response from now on.
Spread in south India
Within India, the number of cases were low in the southern States (except for Tamil Nadu) earlier, but are now accelerating. What explains this trend, especially as the lockdown and the easing of lockdown happened uniformly across the country.
More travel, into and within, these States, is a likely cause. The virus hitch-hikes with asymptomatic or pre-symptomatic travellers and spreads to more people in new areas. It is also possible that reports of good control in these States gave both the administrators and the public a false sense of assurance that the epidemic has ended locally, leading to a lowering of the guard. Bengaluru is a classic example of how public health failure has been snatched from the jaws of victory. If you give this virus an inch, it will take a yard. I do hope, however, that the well-earned reputation of the southern States for efficient administration will help them to quickly course-correct and regain control over the virus.
Are we clear about the reasons for the spread to new areas?
The virus moves with people and celebrates with crowds. Initially it entered via the international airports. Now it moves through road, rail and air to other parts of the country. If asymptomatic and pre-symptomatic persons are carrying the virus as they travel, neither do they know this nor can thermal screening detect it as they start their journeys. Physical distancing and masks can help cut the transmission but they are not universally followed. During travel, physical distancing becomes difficult. However, crowded events should certainly be avoided. Unfortunately, they have not been prevented in several areas.
Given the Indian experience, what have we contributed to the global pool of knowledge on the virus, treatment and so on?
We could have done more by way of organised research on clinical manifestations, outcomes and their determinants in different age groups. Our vaccine development efforts have been quite successful and clinical trials have begun to assess safety and efficacy. Characterisation of the virus strains and some mutations have been reported by Indian scientists. India is part of the World Health Organisation’s ongoing Solidarity trial on treatments, and other treatment trials, too, are under way in different parts of India.
This experience reinforces the need to build good epidemiological and clinical research capabilities in medical colleges across India. Presently, we depend only on a few elite medical colleges and research institutes. Many government and private hospitals are out of this circuit. The ICMR, State Health Departments and the National Board of Examinations (which affiliates private hospitals providing post graduate medical education) must build countrywide capacity for collaborative research and develop protocols which can be quickly implemented.
Among COVID-19 patients who have died, there are those who have not had any co-morbidity. What does this imply?
There are several possible explanations for this. First, several of them may have been repetitively exposed to a very high viral load under stressful conditions which sap immunity. This applies to healthcare providers and other frontline workers who come into contact with many patients and also to family members who have been in close and unprotected contact with a sick person. Second, several persons with co-morbidities may not have been diagnosed earlier. It is well recognised that many persons with diabetes, hypertension or coronary heart disease in population surveys are unaware of their condition. Third, risk factors such as high blood sugar and high blood pressure have a continuous relationship with blood-vessel damage and cause some harm even below the cut-off levels used for clinical diagnosis. This results in sub-clinical co-morbidity. Pre-diabetes, for example, has been found to be widely prevalent and carries the risk of vascular and renal disease. Fourth, high levels of air pollution may have already damaged many lungs. Fifth, malnutrition may have compromised immune status. Sixth, delays in diagnosis, transport, admission and treatment contribute to preventable deaths. The contribution of each of these factors would probably vary across different locations in India.
Acquired immunity
Earlier we were told that once infected, a person was not likely to contract it again. What is the current status of evidence from India about how long does immunity last among people who have recovered from the infection?
Acquired immunity, arising from infection, has two components. “Humoral immunity” is conferred by antibodies produced in response to viral invasion. “Cellular immunity” is conferred by thymus-derived T lymphocytes which mount additional defence. It has been recently reported that the anti-COVID-19 antibodies produced by an infected person decline by three months. However, it is believed that T cell-mediated immunity lasts longer. Since this is a novel virus, the extent and duration of immunity conferred by each of these pathways are still under investigation. Some stray cases of clinically manifest re-infection have been reported internationally but these appear to be very few. Most of the reports of positive viral tests in recovered persons have been attributed to “dead viruses”. It is possible that persons with low immune status could get reinfected. The Indian experience of such cases is very limited.
The ICMR sero-surveillance in May had suggested that 0.73 per cent of the population had been exposed to the infection. Since then the number of cases has multiplied manifold. Delhi’s surveillance data show 23 per cent exposure. What percentage of the population now is likely to have been exposed?
The ICMR report was on district-level surveillance data, reflecting rural and small-town populations surveyed in mid May. The Delhi survey was from June to July and covered a large city. So, differences are bound to be there. The sampling methods of each survey will have to be carefully examined to assess both internal validity and comparability. In general, the rates of viral exposure will vary across the country, being the highest in the big cities and the lowest in the villages as of now.
It must be recognised, though often not adequately publicised, that antibody tests can yield “false positive” test results too. This is because other coronaviruses, including those which cause common cold, can contribute cross-reactive antibodies. We do not know to what extent they are prevalent in our population at different times of the year. For statistical reasons, these false positive results get amplified when the test moves from a laboratory or hospital setting, where its accuracy was assessed in clinically proven cases, to the field setting, where the prevalence of infection is lower. So the 23 per cent positivity rate in Delhi is likely to be an overestimate of the true prevalence. Even if we think it will be 15 per cent instead of 23 per cent, that is a high number. So the virus has spread with ease. The good news is that most of the persons found positive were asymptomatic. The sobering news is that even at 20 per cent, Delhi will be below the herd immunity threshold, variably estimated to be at 50-70 per cent. The need to observe public health measures for containment is still very high.
India’s testing rate
Has India’s testing rate grown too slowly? Has it ended up chasing the spread of the infection or helped us to get ahead of the infection and help control its spread?
Our testing rates have varied over time. Low testing rates in the beginning were both due to operational constraints of testing kit availability and low numbers of people meeting the criteria for test eligibility. The testing rate picked up as kits, labs and eligible persons rose in number. They were still considered inadequate by international comparison, though there is no correlation between different testing rates and mortality rates per million population in any geographic zone of the world. Otherwise, the U.S. should have had far fewer deaths than most countries.
The need for increased testing is for quickly identifying cases and contacts for isolation. Since the real-time polymerase chain reaction (RT-PCR) test has only around 60 per cent sensitivity, it should be complemented with clinical and contact information for making decisions on isolation. It should not be assumed that a negative RT-PCR test rules out infection. Such expanded case definitions were not followed in practice. As criticism of low testing rates mounted, antigen tests were introduced. They had even lower sensitivity than RT-PCR and would miss more than half the infected persons. While testing rates went up and test positivity rates came down with these tests, the numbers of missed cases rose. We need to judiciously combine information from clinical data, contact history and RT-PCR or antigen tests to guide our strategy for identifying infected persons for isolation. Case identification must also lead to efficient contact tracing. The media, the public and even some policymakers are too fixated on testing numbers alone to recognise this composite template. That blinkered view, too, is as unhelpful as low testing rates.
Should the lockdown period not have been used to ramp up testing faster than we did? Did we lose an opportunity there?
During the full lockdown period, testing numbers were less important than preparing for larger scale testing as the lockdown ended. This is because the principal purpose of testing is to identify cases and isolate them as well as their close contacts. In the first 21 days of strict lockdown, this purpose was served even at low levels of testing as all potential cases and contacts were already isolated at home and symptomatic cases reached healthcare facilities. It is when that phase ended that testing and contact tracing should have been ramped up.
There were shortages of testing kits initially, in India and even globally, but the lockdown period gave time to fill those gaps. If symptom-based syndromic surveillance of households was routinely conducted by primary healthcare teams and citizen volunteers, more suspected cases would have been identified for testing.
Stigma and fear also kept people reporting for testing. Efficient and empathetic primary healthcare services would have countered that hesitancy too.
It must be recognised that a testing strategy must not merely chase numbers and do haphazard testing. It must be based on clear criteria. Once those are defined, all those who meet those criteria must be tested through active search strategies. Lag times in these could have been cut down through stronger primary healthcare services. Lamentably, our urban primary healthcare services are mostly absent or very feeble. Different States pursued different strategies for identifying persons to be tested, leading to missed cases and contacts.
Given that there is no specific “cure” for the virus, and many therapies are listed as investigational therapies, what is your opinion on whether there should be a definite treatment protocol or whether it should be left to the discretion of the individual clinician?
It bears repeating that this is a new virus and large clinical trials are still under way to identify which drugs can effectively reduce deaths in infected persons with different levels of clinical severity. So far, dexamethasone has been shown to reduce deaths in patients who require oxygen or mechanical ventilation. Others are still being evaluated for their impact on mortality. Some drugs have shown effects on the viral load or the duration of hospital stay, but not on mortality. Until we are better informed by ongoing clinical trials, clinicians will use their best judgement to manage patients under their care, by choosing among the various options available. It is too early to lay down a single evidence-based protocol for all clinical situations. Research in this area is still in a fluid state and science has not yet crystallised in the form of definitive recommendations.
As the number of cases are growing and the demand on the health system is increasing, do we not have a problem of inadequate personnel? While beds and infrastructure can be ramped up, are there not constraints even now regarding the availability of healthcare professionals?
We do have a serious shortage of trained personnel at all levels of care. From contact tracers in the field to intensive-care doctors, nurses and technicians who can confidently provide ventilatory support, we have the challenge of low numbers and low skill levels. This is where our long neglect of the need to build a sizeable, multi-layered and multi-skilled workforce is biting us now. Investing in expansion of the health workforce is both a health system imperative and an economic opportunity for job creation. We should learn this lesson at least now and start investing in creating an adequate health workforce which is also well distributed across the country.
From the COVID-19 experience in India, do you think that it is appropriate that such a large proportion of its healthcare facilities should lie in the private sector and that there should be significantly greater public investment in health?
I believe that a strong public sector should lead the way, even in a mixed health system that has grown by default rather than by design. This applies to all levels of care but is especially essential in primary and secondary care where most of the healthcare needs of the population must be met. Such transformation calls for higher levels of public financing for health, by both Central and State governments.
Whether for effectively combating public health emergencies like COVID-19 or for efficiently delivering universal health coverage, a strong public sector has to lead the way. It will not happen with only 1.2 per cent of the gross domestic product being spent on health. We need a doubling of that figure in three years and a further steady annual rise until we can reduce out-of-pocket expenditure on health to less than 20 per cent from the present 62 per cent. These additional resources must be used for strengthening both rural and urban primary healthcare, district hospitals, medical college hospitals and government laboratories.