NITI Aayog CEO Amitabh Kant. Photo: Ishan Saxena/Wikimedia Commons, CC B-SA 3.0.
India reported its first COVID-19 cases on January 29, 2020. Since then, according to the Union health ministry, the country has reported 108,923 cases and 3,435 deaths due to COVID-19 (as of 8:30 pm on May 21, 2020). This said, every other number pertaining to the Indian government’s response has been repeatedly contested, not least because the flow of information has been constipated at best.
First, almost all updates about the government’s efforts began to come only from bodies headquartered in Delhi, such as the health ministry and the Indian Council of Medical Research, or even the prime minister’s office. Next, data-collection mechanisms involving a nationwide network of laboratories and healthcare facilities withheld data that could have allowed independent experts to assess the effectiveness of the government’s response. Then, topical experts stopped attending the health ministry’s press briefings, leaving journalists’ technical questions to be answered by administrators. Finally, the health ministry’s briefings became less frequent (and simultaneously less useful).
All this together triggered uncertainty about the government’s plans, and therefore panic, as well as confusion about contradictory protocols between the Centre and state governments. The lack of data as well as the increasing amount of work required to properly contextualise each tidbit only worsened the situation. It is in this milieu that Amitabh Kant, the CEO of NITI Aayog, has made a string of claims about how well India is dealing with the COVID-19 epidemic. The data to accept or reject his claims isn’t readily available, but as it happens, a combination of news reports, some mathematics and basic epidemiology allows us to try.
1. May 19, 2020: “It is not the total number of positive cases but the mortality and recovery rates which are critical. We are faring well on both. Fatalities are 2 per million compared to 275 in the US and 591 in Spain. Our mortality rate is 3% compared to 16% in France. Our recovery rate is constantly improving and is now 38%.”
– India will always fare well on any metric that is decided on the basis of a value divided by the total population because India is the world’s second most populous country. The more meaningful metric here is the case fatality rate – i.e. the number of people who die due to COVID-19 divided by the total number of people who have COVID-19. This way, reports from India or any other country can’t be skewed by the size of its general population.
– The global average mortality rate, by which Kant presumably means the case fatality rate, on that day was around 3% as well. Comparing India’s figure with the one for France is convenient because it shows India in good light. But compare it to the global average and India is… average. France is the outlier here.
– A growing recovery rate is not special. It is the natural order of things. If COVID-19 has a case fatality rate of 3%, fully 97% of those who get the disease will recover. And as more and more cases are reported, more and more people will obviously recover in the future. By the time the pandemic concludes, the 38% will have to reach 97% or so (depending on what the final case fatality rate is).
2. May 18, 2020: “Success in the battle against COVID-19 will be determined by lives saved and not total cases. 1.3 billion Indians and yet COVID-19 deaths per million are just 2, against 511 in Spain. Mortality rate is 3% against 16% in France. These rates are among the lowest in the world! We must set an example for the world.”
– India is home to over 1.3 billion Indians but the novel coronavirus – in the absence of widespread community transmission – spread around the world from Hubei province by people on flights and ships. Unlike Spain, India’s rural population is 900 million; many of these people are quite unlikely to have recently caught a flight or ship, just as they are likely to have been more than a few degrees removed from people living and working in urban environments.
– As it happens, Russia has the world’s lowest case fatality rate at the moment, around 0.97% (which has been contested). Turkey, which is in the top 10 countries in terms of number of people with COVID-19, has a case fatality rate of 2.76%. As long as we are making arbitrary comparisons, why not include Russia and Turkey as well?
3. May 16, 2020: “India’s test-positivity rate (TPR), which is the ratio of positive cases to the total number of tests conducted, is just 4%. The US’s TPR is 14%. Since March 20, our testing has grown by 100-times. Also, India is one of only seven countries to have crossed 2 million tests. Focus on high case-load cities.”
– Only 4% of people tested turn out to be positive. However, this figure by itself is not very useful. For example, on April 24, India’s TPR was 4.76%, which was higher than South Korea’s 2.23% and lower than Germany’s 6.79%. Both these countries have effectively flattened their curves even though their TPR varies by 4.56%.
– The TPR is the day-wise weighted average, calculated by dividing the total number of people who have tested positive divided by the total number of people who have been tested. So when a government increases testing at the same time the epidemic gets going, the TPR will first rise, stagnate for a bit, and then begin to drop before stabilising at a certain number. South Korea’s TPR settled at 2% after it tested 1.2% of its population. As of May 20, India had conducted 2.5 million tests; if each test involved one individual and no one underwent repeat tests, the government has thus tested 0.1% of its population.
– Who India tests has been limited by the official case definition at different times. The most recent one does not allow India to discover asymptomatic cases of infection, at least on paper. Tests are restricted to those who recently came in contact with someone else who tested positive, and people with influenza-like illnesses whose symptoms can’t be explained by any other conditions. There is no way for people without any symptoms and no recent contact with someone else who tested positive to get tested, partly because the government insists India does not yet have community transmission. That is, the TPR and the likelihood of community transmission are connected, and the government has been celebrating the former while playing down the latter. This is a self-fulfilling prophecy.
– It is unclear how the government is both keen to rationalise the number of tests it conducts (for, among other reasons, the high cost of each test) in order to reveal the most people who may be infected and also wants to celebrate the fact that it conducted 2 million tests. It has to be one or the other.
Find out more: COVID-19 in India: A Curious Case of Missing Numbers
4. May 15, 2020: “The recovery rate of India is on a constant path of improvement. From just 9% on 18 March to 34% today. Our fatality rate is ~3% as compared to 14% in Italy and the UK. The key to winning the battle is to dedicate our efforts to keep the fatality rate low and improving our recovery rate.”
– As stated earlier, the recovery rate can only improve with the novel coronavirus. As more people fall sick today than yesterday, more people will also recover over the next two weeks than yesterday, assuming the case fatality rate remains more or less fixed. Celebrating the increasing recovery rate is like taking responsibility for gravity bringing a thrown ball down.
– It is important to keep the case fatality rate down, but Kant’s repeated association of this number with the recovery rate – instead of the quality of healthcare and living conditions, which have a strong influence – suggests that he believes the government’s actions are responsible for keeping the recovery rate going up.
– An increasing recovery rate does not translate to a decreasing fatality rate. The novel coronavirus’s case fatality rate in India as a whole has been hovering around the 3% mark. If the recovery rate is increasing, it only means we are registering more people beating COVID-19. As mentioned earlier, the 38% figure eventually has to reach 97%.
– Kant’s repeated deference to the all-India figure is misleading. Health is a state subject in India, so interstate differences matter, especially since Gujarat and West Bengal have case fatality rates of 5.97% and 8.15%, respectively. Focussing on the nationwide average hides these problem states.
5. May 8, 2020: “Incredible! India leads the world in leveraging technology to fight COVID-19. #AarogyaSetu: the most downloaded healthcare app and among the top 10 most-downloaded apps in the world for April 2020, in its first month itself. Never seen this before. We are united in our fight against COVID-19.”
– The government forced office-goers to download the app but later softened this requirement. The ruling party’s social media unit also mounted an intensive campaign to have people download the app and use it regularly. Finally, refusal to download and use the app was made a criminal offence before the government again relaxed its order. After all these measures, Kant’s tom-tomming the app’s manufactured popularity comes across as cheap.
– Kant claims the country is leveraging technology in a “lead the world” kind of way. However, Aarogya Setu was designed and advertised as a product that would keep people safe, and that is the basis on which most of those who downloaded the app did so, not because it had a great user interface. Comparing its success in terms of the number of downloads against those of TikTok and Instagram, which are both products of entrepreneurial success, is as misleading as claiming the app’s success is proof of “leveraging technology” is misguided.
6. May 1, 2020: “India’s share in global fatalities stands at 0.5%. This is in sharp contrast to the US, which stands at 27.2%. Italy accounts for 12% despite reporting their first positive case a day after India. Our five-day moving average of fatalities has also remained consistently lower.”
– The US failed to act in time, costing 36,000 lives according to one study, and together with its president’s serial mismanagement, the number of cases ballooned before Washington state, followed by New York, began to report a tsunami of patients at their hospitals. India was less callous, but here too the question can be asked: As late as March 13, the Union health ministry was saying publicly that the coronavirus was not a health emergency. If the Centre had alerted everyone earlier, couldn’t more lives have been saved?
– This said, the issue with Kant’s tweet is that many researchers and experts have noted that the novel coronavirus has impacted India much less than it has other countries, notably those to which Kant repeatedly draws attention. And Kant would have been well-placed to take credit for this good fortune if only the government could explain why exactly this is so. Without this explanation, which even epidemiologists don’t yet have an answer for, Kant’s rush is only opportunistic.
– The actions of the governments of West Bengal and Maharashtra suggested one possible explanation, a bleak one: that government hospitals and health surveillance facilities are underreporting deaths due to COVID-19, such as by listing a comorbid condition as the primary cause of death instead of COVID-19 itself.
7. May 1, 2020: “As we move into lockdown-3, India continues with the lowest rate of positive cases detected as a percentage of the total tested. It’s only 3.4%, despite ramping up testing. It was 4.7% on 15 April. In the US, this figure is 17%, and in France, it is 23%. We are now conducting 50,000 new tests per day.”
– It is good, for reasons indicated earlier, that India’s TPR is dropping. However, considering the government has refused to admit the presence of community transmission, even in the face of multiple sets of evidence that suggest community transmission is underway, the current TPR is largely a reflection of one portion of the infected population. If the case definition is expanded to include those without recent contact with infected individuals or recent foreign travel, the TPR will likely increase again.