1 coronavirus-positive person infected 3-4 people, reveals analysis by Tata Institute of Fundamental Research scientist

One Covid-19 positive individual in Mumbai is likely to have infected three to four people during the first 15 days of the nationwide lockdown, according to an analysis by a scientist at the Tata Institute of Fundamental Research (TIFR) based on a susceptible-exposed-infected-recovery (SEIR) statistical model.

While the lockdown was put in place to reduce the number of people that one Covid-19 positive person can infect (denoted by R0), the analysis attributes a rise in cases to areas of high population density, where it is not possible to maintain physical or social distancing. R0 close to 1 means one individual will infect one or less than one person thereby containing the spread of the virus.

Sourendu Gupta, theoretical physicist at TIFR, who undertook this analysis, said if Covid-19 cases are confined to certain geographical structures within Mumbai, the possibility of a “second wave” surge of cases in areas where housing density is lower cannot be ruled out. “Even when the lockdown ends, we have to be careful because infection may have unevenly spread through the city with some parts highly affected and some others which are not. Care therefore needs to be taken to ensure infection doesn’t spread uncontrollably into less affected areas. For that more testing is necessary,” said Gupta. “What happens in the second half of April will determine how safe we are in lifting the lockdown.”

Inference drawn from SEIR on the number of infected Mumbaiites from fatality counts between March 31 and April 14 has projected that the doubling rate of the infection was 5.7 days. This means if Mumbai recorded 100 cases in one day, that number rose to 200 in five days and 1000 in 15 days.

Similarly, the doubling time for fatalities during the same period is 5.3 days which according to the analysis can be compared to the doubling time during the early days of the novel coronavirus outbreak in Wuhan before the imposition of a quarantine.

Gupta said it will take two weeks for the toll to jump ten-fold if the current doubling time continues for fatalities. That’s 100 or more deaths every day - similar to the peak number of Covid-19 deaths in New York City.

“In the hypothetical cases discussed here, Covid-19 wards in the city would have to deal with several hundred admissions a day with care periods for each case ranging up to two months. Again, using the estimates given above, these rates would be reached in two to four weeks if the rate of growth of the epidemic does not come down,” states the manuscript of the study published in ArXiv, an open-access archive by Cornell University.

Currently, the number of Covid-19 positive tests is likely to be a small fraction of the total number of infected – one out of 750 cases of infection. “More tests have to be done especially when the government is thinking of unlocking the city. In the first two weeks after the lockdown ends, we need to be sure the infection is not spreading without knowledge.”

If the doubling rate of the infection doesn’t change, the statistical model puts the likelihood of the first Covid-19 cases in Mumbai as early as December 20, 2019 and February 1, 2020 – with January 15 as the median point for ‘seeding of the epidemic’ – affecting individuals with a travel history to China where the outbreak had already begun.

“The first fatalities (in Mumbai) due to this cryptic spread of COVID-19 would not have happened before mid-February. It is certainly possible to check municipal records of deaths in February to see whether any excesses over the same month in previous years are visible. If there is a statistically significant signal, it would support the idea of early cryptic seeding of COVID-19 in Mumbai,” states the study. Shankar R, retired theoretical physicist, Institute of Mathematical Sciences, Chennai, who was not involved in the study, said the only way to estimate the number of infections is via the death toll. “In a large population, huge number of people who have not been tested carry the infection and we don’t know the actual numbers,” he said.

Meanwhile, the national growth rate of confirmed cases of those infected with the novel coronavirus has been on the decline since the nationwide lockdown on March 25, finds the data analysis team of the Indian Scientists’ Response to COVID-19, a voluntary group of more than 400 scientists. The team stated that fresh outbreaks can significantly alter the scenario and therefore cautioned against any extrapolation based on current data.