A government-appointed scientific panel’s claim that the Covid-19 epidemic will run its course by February 2021 in India has come under criticism from multiple scientists who questioned the method used by the panel to come to the conclusion.
“I am pretty sceptical of long-term predictions even from more detailed models, and this one is pretty crude,” said a professor at the Institute of Mathematical Sciences, Chennai.
Because of the involvement of the Department of Science and Technology and the office of the Principal Scientific Adviser to the Prime Minister in the exercise, several scientists DH spoke to, wished to remain anonymous.
The critics are of the opinion that the method used was too simplistic even in March when there was barely any data. In October, it is inexcusable not to have state-level estimates derived from real data.
Manindra Agarwal, a professor of mathematics at Indian Institute of Technology, Kanpur and a member of the government panel, said the model had 10% error probability on a national scale. This would increase further if it is used to predict the situation in a state or district.
The panel uses publicly available data to create its “supermodel” rather than relying on the Indian Council of Medical Research’s official data due to the limitations of ICMR data policy.
Within the academic circle, there are growing murmurs of how the panel used a simplistic model to come out with conclusions justifying the actions already taken by the central government.
“I find it surprising that respected scientists should put their names on a very flawed piece of work. It just seems a political exercise,” said another IMSc professor.
Agarwal countered the charges. “I can’t help if the conclusions support the government policy, but there was no political inclination,” he said.
“We wrote to all the scientists and received 30 proposals from which we picked up two models. Subsequently, the panel decided to work together to improve upon the models and the scientists who originally contributed were duly acknowledged,” he said explaining the process.
A second key conclusion of the government panel is that the lockdown decision announced by the Narendra Modi government in March was perfectly timed. This too has been challenged by the academicians.
“The delay in the migrant population moving back (to their home state) did help to spread the disease, unlike what they claimed,” said a scientist.
The conclusions that movement of migrants brought about by the lockdown didn't significantly alter the numbers, were made without any basis in the model, which was not designed to test such a hypothesis at all, said a researcher involved in computer modelling.
Agarwal said the panel modelled the migration trend in Uttar Pradesh and Bihar but left out West Bengal, which did not allow migrant workers in for nearly a month.
“What is new in the model that we had not assessed earlier except the name of the model – Supermodel!” wondered a professor at the Indian Institute of Science, Education and Research, Kolkata.