US COVID-19 death toll has been questioned.

The real number exceeds current estimates at best by 2, and at worst by 14 times, said epidemiologists who created a new methodology for assessing mortality from COVID-19.

The study was published at medRxiv, TASS reported.

According to scientists, results of their calculations are in good agreement with the latest results of tests for antibodies to a new type of COVID-19 in New York, according to which COVID-19 affected 10-20% of citizens. 

According to other models, the current number of deaths can only be explained by the fact that 100% of New Yorkers have become infected with the coronavirus, scientists write.

Since the beginning of the pandemic, about 4.7 million people have been infected, more than 315 thousand have died. Over 290 thousand infected were recorded in Russia, 70 thousand people recovered, about 3 thousand died.

As there is no complete statistical data on all COVID-19 cases, scientists cannot accurately estimate what proportion of recorded and unaccounted infected usually die from the infection. Experts estimate this share according to IFR (Infection Fatality Ratio).

Initially, experts from the World Health Organization (WHO) suggested that approximately 3.5% of COVID-19 carriers die from coronavirus infection, but subsequent observations have shown that this parameter is actually significantly lower and ranges from 0.1 to 1 for different countries and regions. 4%.

To evaluate IFR more precisely, a group of epidemiologists led by Yale University professor Douglas Rothman developed and tested a new methodology to calculate this indicator based on current statistics. Scientists have come to the conclusion that previous estimates were too optimistic.

Calculations showed that now the mortality rate from COVID-19 infection in these countries is from 0.6 to 5%, which is several times higher than the expected value of this indicator for the US, the UK, and other countries where a large number of infections are observed. Based on this data, epidemiologists calculated IFRs for the entire US and for individual outbreaks such as New York.

The results showed that specialists can significantly underestimate the number of COVID-19 victims in New York and other parts of the US. In reality, the IFR for them may be about 1.8%, that is, more than the calculations showed on the basis of previous epidemiological models. The fact is that these models relied on less relevant observations of the spread of infection.

As the researchers hope, their model and calculation results will help the medical services of the US, as well as other countries, to more accurately assess the extent of the infection, as well as more effectively contain it before the first vaccines and drugs for COVID-19.