Indian scientists build model to estimate impact of interventions on COVID-19

The model gives an idea of how to achieve control over the trajectory of the disease and match availability of health care

The DBT said a multifaceted approach is being adopted to ensure that vaccine companies utilising different platforms and at different stages of development are fast tracked.

The present lockdown has closed many scientific research labs across the world, but it has not shut down research, especially on COVID-19 . Scientists from Chennai, Pune, Bengaluru, Delhi and Goa, who are interested in epidemiology and modelling, have come together to build a model that can help predict outcomes of the disease under different scenarios.

This model, named INDSCI-SIM, is the first detailed, State-specific, epidemiological, compartmental model for COVID-19 made in India.

The model can compare effects of multiple non-pharmaceutical interventions — such as different types of lockdowns, quarantines and expanded testing — in altering the trajectory of the pandemic.

Coronavirus India lockdown Day 27 updates

“The secret of a good model is to capture all the details that are relevant to understanding the spread of the disease. INDSCI-SIM accomplishes that,” said Gautam Menon of The Institute of Mathematical Sciences, Chennai, and Asoka University, Delhi, who is one of the people who developed the model.

Nine compartments

In the model, each State or City is represented by a population, every one of which is described by nine compartments representing various stages of the COVID-19 disease progression. This extends the four-compartment SEIR model that is usually used (SEIR stands for Susceptible, Exposed, Infected, Recovered).

As an illustrative example, the scientists have used the data for the population from the 2011 census. They showed that when there were fewer infected people, one can have a good control over the number of hospitalised people at any time by ramping up quarantine-testing measures in a calculated way. This should include the asymptomatic people being tested too. For example, given that the supply of ventilators is limited, if Tamil Nadu has about 3,600 ventilators available, if 60% of the population could be tested over a period of 240 days (and quarantined), they show that the curve is flattened significantly so as to match the availability of ventilators.

Important feature

“Our results underline the need for more broad-based testing as the only long-term solution to stopping the disease in its tracks. The ability to estimate hospital bed and critical care requirements is one important feature of our model,” said Professor Menon.

COVID-19 | State-wise tracker for coronavirus cases, deaths and testing rates

“The work was very intensive. Given that we had to work under lockdown conditions, there were emails and frequent virtual meetings,” said Pinaki Chaudhuri from the Institute of Mathematical Sciences, Chennai, one of the nine people who collaborated on the effort. “This model can be used not just for COVID-19, but for any future epidemics, too. We plan to have a district projection and age specific projections out by a week or so,” he added.

Published - April 20, 2020 09:46 pm IST