Absence of evidence is not evidence of absence
18 March 2020YOGESH JAIN
I have been working as a community doctor in Chhattisgarh for over two decades. As one of the founders of the Jan Swasthya Sahyog—People’s Health Movement—an NGO that provides healthcare in rural Chhattisgarh, I understand the infrastructural problems of the state. There are only around 150 ventilators in the state’s public health infrastructure, and even fewer doctors who know how to run them. Given this state of the healthcare system, as confirmed COVID-19 cases rise well over a hundred in India, I worry about the community transmission of the disease, and dread whether it will overwhelm our weak health systems—both public and private. If a sharp surge of cases as occurred in Italy takes place in India, I worry about our capacity to serve the thirty-two million residents of the state.
The union health ministry’s current testing criteria is limited to international travelers and their direct contacts, and only if they show symptoms such as fever, cough and breathlessness. The tests are not available for people who have severe respiratory illnesses, but have not travelled abroad in the last 14 days and have not been in contact with a person confirmed to have been infected with COVID-19. As a result, none of the patients in my hospital would be eligible for the test presently. The explanation offered by the Indian Council of Medical Research, the nodal agency framing India’s policy for COVID testing, is that there is no evidence of community transmission in the country. This rationale—or lack thereof—reminds me of an old aphorism: absence of evidence is not evidence of absence.
Technical decision makers have claimed that there is no evidence of community transmission. But the source of this evidence is thin and inconclusive. Between 15 and 29 February, the ICMR conducted random tests on 20 samples of patients with respiratory illnesses to determine whether India had community transmission of COVID-19. The samples were sent to the various Viral Research and Diagnostic Laboratories across the country to test whether they were infected with COVID-19 by the RT-PCR—a molecular biology technique that looks at the nucleic acid core of the virus.
The ICMR stated that it did not find any evidence of COVID-19 in those samples, and on that basis, concluded that there is no community transmission in India. If we assume the median date of these samples to be 22 February, over three weeks have lapsed since we concluded that there is no evidence of community transmission. But there is a fundamental flaw in this conclusion. For COVID-19, three weeks means a lifetime of evolution of the epidemic—for instance, Italy has shown the exponential increase that can take place in a short time span.
On 15 March, the Economic Times reported that the ICMR was beginning the next set of tests. According to the report, each ICMR laboratory would test twenty random samples every week for community transmission. Two days later, the ICMR stated that it had conducted tests on 1,020 random samples, and that preliminary tests on 500 of them did not yield any positive results for the COVID-19 virus. Balram Bhargava, the ICMR’s director general, had accordingly concluded, “There have been no signs of community transmission.”
Yet, the ICMR has so far not disclosed crucial information about these samples, without which it is difficult to accept its conclusion at face value. For instance, there is no clarity on the nature of these samples taken by the ICMR. It is unclear whether they represent an adequate sampling of cases that were likely to be the COVID disease. Similarly, it is also unclear whether the samples comprise an adequate sample size. As of 10 am on 18 March, India had tested 11,461 individuals for COVID-19, and confirmed 145 positive cases. According to Our World in Data—a statistical research initiative working on issues of global concern—India had conducted 4,058 tests as of 6 March, which made it the country with the lowest number of tests per million persons, at just three tests per million.
By 9 March, South Korea was reportedly conducting 4,100 tests per million persons. Our World in Data reported that by 13 March, the country had conducted 248,647 tests. Given these statistics, it is clear that India’s testing so far has been abysmally low, and a fortnightly or weekly testing is untenable. The procedure must change towards more frequent, or in fact, daily testing and assessment. As a result, the ICMR’s conclusions on the absence of community transmission based on 500 tests does not inspire much confidence and raises pressing concerns about whether the nodal body is accurately assessing the rapidly escalating scale of the disease in the country.
If we look at the chronology of the COVID-19 diagnoses in India, it has seen a significant surge since March 2. Since then, the number of confirmed cases has risen from five to 145, as of the morning of 18 March. The Indian government’s failure to track the possibility of community transmission on a daily basis is telling of its inadequate response to the evolving crisis. Meanwhile, the central and state governments have shown incongruous enthusiasm about putting in place containment and social-distancing strategies, such as the closure of schools, malls and other public spaces. But these decisions do not reflect any sincere efforts to identify the actual burden of the pandemic within the country.
The paramount importance to test, and not wait, was marked by Michael Ryan, the executive director of the World Health Organisation’s Health Emergencies Programme, during a press briefing on 13 March. When asked what lessons Ryan had learnt from his experience dealing with the Ebola outbreak, he responded, “You must be the first mover. The virus will always get you if you don’t move quickly.” Ryan added, “If you need to be right before you move, you will never win. Perfection is the enemy of the good when it comes to emergency management. Speed trumps perfection. The problem we have in society at the moment is that everyone is afraid of making a mistake, everyone is afraid of the consequence of error. But the greatest error is not to move.” The Indian government has failed in its responsibility to implement forward-thinking measures to mitigate the ongoing public-health crisis, and instead, continues to frame reactive policies.
Meanwhile, there is a debate between bureaucrats and health experts over parsing the difference between “local” and “community” transmission. The union health ministry has defined “local” as transmission through known, local community contacts, and different from the larger community, where a link could not be established with an individual known to be infected. This differentiation between local and community transmission is misleading, given that locals with whom an individual is in contact are in turn a part of a larger community with their own sets of contacts. Moreover, there does not appear to be any need for this debate, considering that the WHO has only two criteria for the modes of transmission prevalent in a country—imported cases or local transmission—and it has classified India as a country with local transmission.
Instead, the need of the hour is swifter and wider testing for COVID-19 that would not be limited to those who travelled abroad or their direct contacts. This would be important for determining if there is an actual burden of community transmission in India, and the scale of the same, which would then allow us to implement informed containment strategies. Moreover, it would also be necessary to make the strategic shift from a predominantly containment-oriented strategy of controlling COVID-19 to a mitigation strategy. Moving towards a mitigation strategy is not tantamount to accepting defeat of the containment strategy, but a realistic and balanced long-term approach that minimises the impact on the health, survival, social life and economic well-being of people at large.
This expanded testing would entail, for instance, testing those who are admitted with respiratory distress due to pneumonia in public medical colleges and other district hospitals. A quicker and more frequent measurement of the community transmission of the virus would enable the government to take the right decisions regarding our response going forward. Once those decisions are taken at the national level, states can be advised to follow suit, so that patients in smaller cities and rural areas can be tested. By doing this, areas that are likely to be affected would become better prepared to deal with the COVID-19 pandemic.
YOGESH JAIN is a community doctor and founder of the Jan Swasthya Sahyog, a public-health NGO based in rural Chhattisgarh.