Covid-19: What you need to know today
Some states start off by not testing enough; others start off well, but as cases continue to rise, the message goes out to go easy on the testing. Tamil Nadu, Delhi, and Maharashtra, then, deserve credit for actually continuing to test more.
It isn’t a competition, yet most states have started treating their response to the coronavirus disease as one.
Competitions have metrics, and this one has two — the number of cases, and the number of deaths. Metrics, though, have the ability of provoking undesirable behaviour.
Click here for full Covid-19 coverage
For some time in the late 1990s and the 2000s, EVA or economic value added was the hottest buzzword among financially inclined management types. It was a concept created by Stern Stewart, and in one of my previous avatars, I helped bring the study to India (for years, the annual EVA computation of top Indian companies was published in a magazine for which I used to work). The problem with EVA was that it encouraged short-termism. Executive compensation was usually linked to EVA targets in companies that believed in the metric. Sure, Stern Stewart came up with another concept, MVA or market value added, and several other measures to prevent such behaviour, but that’s another story.
In this case, because state governments, and the politicians who head them, see low Covid-19 infections and deaths as a manifestation of their abilities as administrators, and vice versa, they have started to game the system.
The simplest form of this is just not testing enough. Some states start off by not testing enough; others start off well, but as cases continue to rise, the message goes out to go easy on the testing. Tamil Nadu, Delhi, and Maharashtra, then, deserve credit for actually continuing to test more.
Any state, which, according to publicly available data, tests fewer people than the national average per million is definitely doing badly on this score. The national average, on the basis of Friday night’s data, was 2,096 per million. States such as Uttar Pradesh, West Bengal (although it seems churlish to criticise the state for testing at a time when it is recovering from the fallout of Amphan), and even media-darling Kerala are testing fewer people per million than the national average. It doesn’t matter how good the public health system in a state is — if it isn’t testing enough, it isn’t doing enough. Period.
But the testing data is also misleading. No state breaks up the number of tests carried out so far into new tests and tests on those people who already tested positive for Covid-19 (and who can be declared Covid-19-free only after a negative test). India isn’t alone in this. The US CDC, it emerges, doesn’t do this either. This means calculations of the number of people per million of population that have been tested could be far off the mark.
If there is one thing the coronavirus disease has taught us, it is that larger policy and public health decisions related to the pandemic have to be based on data, and science. Unfortunately, few governments around the world seem to recognise that.
Right now, as countries around the world, including India, open up, the most important data is the extent of the disease’s prevalence — has it infected 10% of the population or 30%? (Another way of asking the same question would be, is 30% of the population now immune, or is it only 10%?). Widespread testing, perhaps using rapid antibody tests, may provide the answer — and also shed light on the actual fatality rate of the disease. Current data shows it is around 3% in India, but I am willing to wager that the actual rate will be much lower, perhaps even lower than 1%, a level that may warrant the removal of more restrictions on movement and activity.
But a country that wants to make decisions on the basis of data will also have to ensure that it comes in quickly. In India, data on cases as well as deaths is reported with a lag, making it all but impossible to know the real trend. I am told the number of cases reported every day is based on the results of tests of varying vintages. It’s getting measured for sure — but not in a way that will help us manage it.
