New Delhi: Around 1 million people have died of Covid-19 in India so far, The Economist has estimated in a modeling report to gauge excess deaths worldwide caused by the pandemic.
“On the basis of the model it would appear that around 1 million people have died of Covid-19 in India so far,” the report said.
According to the report, India is witnessing between 6,000 and 31,000 excess deaths per day. This estimate is well above the daily official figures that are hovering around the 4,000-mark.
Earlier this month, an editorial in the Lancet journal had said that India could potentially record 1 million deaths due to Covid by 1 August. The editorial had quoted projections made by The Institute for Health Metrics and Evaluation, a global health research organisation.
The Economist report published on 15 May tried to model “the level of excess mortality over the course of the pandemic in countries that do not report it”.
According to the US Centers for Disease Control and Prevention, “Excess deaths are typically defined as the difference between the observed number of deaths in specific time periods and expected number of deaths in the same time periods.”
The Economist’s estimates suggest there is a 95 per cent probability that the Covid death toll worldwide is between 7.1 million and 12.1 million, with a central estimate of 10.2 million.
“The official numbers represent at best, a bit less than the true toll, and at worst only about a quarter of it,” the report said.
How the model was conceived
For the purpose of the model, The Economist collected data from more than 200 countries and territories based on 121 indicators.
Apart from taking into account official Covid cases and deaths reported by countries, it also used indicators such as reports on tests which were found to be positive since they provided an estimate into the magnitude of under-counted cases.
Further, sero-prevalence surveys were also used to estimate the part of the population that had antibodies — this provided a measure of past infections.
Other indicators used for the model included: steps taken by the government for pandemic control; extent to which citizens could move around; media freedom; geographic location; and systems of government.
In certain scenarios where countries didn’t publish relevant data, the absence of data was also taken into account.
After this data was collected via several indicators, a machine-learning model was trained, which used a ‘gradient boosting’ process to “find relationships between these indicators and data on excess deaths in places where they were available”.
The model used these relationships to make estimates of excess mortality, especially for locations that didn’t have this data available.
Toll difference in other parts of the world
According to the estimates, there was a 95 per cent probability that the coronavirus pandemic led to 2.4 million and 7.1 million excess deaths in Asia by 10 May. The official figure for Covid deaths in Asia is 0.6 million.
The model suggests 1.5 million-1.8 million excess deaths were seen in Latin America and the Caribbean even as the official figure is 0.6 million. The toll in Africa is pegged at 2.1 million excess deaths as opposed to the official figure of 0.1 million.
Around 1.5 million-1.6 million excess deaths were gauged in Europe while the official figure is 1 million.
The US and Canada have reported 0.6 million fatalities due to Covid, close to the estimated 0.6 million-0.7 million deaths in these countries.
In Oceania, 1,218 official deaths have been reported, but the model predicted around 12,000 to 13,000 deaths.