Last updated 26 April 2020

The currently evolving COVID-19 (coronavirus) pandemic is a very serious event. Thousands of people have already died, and many more may well die in the coming months. Moreover, there has to date been a lot of confusion about what is really going on, particularly regarding estimated Case Fatality Rates (CFRs), which vary massively across countries (0.14% to 15%, as of 21/4/20). In order to help reduce that confusion, I have written the statistical report below:

CLICK TO READ: “Sampling Bias, Case Fatality Rates and COVID-19 Testing”

This report analyses current CFRs across 60 countries, and finds that wide (260 fold) variations in sampling bias can account for much of the current wide range of CFR estimates, both across countries and across time. Countries that test more widely, relative to the number of deaths, have more reliable CFRs, and these countries have lower CFR estimates. It was also found that neither GDP per capita, nor percentage of population over 65, nor median age of population can account for the wide range in CFR estimates. This is likely to be because any effect from varying population age and GDP per capita is being masked by much larger variations in testing.

Key conclusions are as follows:

  1. The Infection Fatality Rate of COVID-19 is likely to be at the low end of the current CFR range (i.e. much closer to 0.14%, than 15%), and taking account of other recent studies it is concluded that the IFR will likely be in the range 0.28% – 0.68%. It should be noted, however, that COVID-19 likely has a higher basic reproduction number (R0) than was initially estimated, and so many people are still at risk of dying, especially the elderly and the vulnerable. The fatality rate (IFR) is likely to be less than some fear, but the infection rate (R0) will likely be much higher than many realise. Governments thus still need to instigate strict measures to manage the pandemic, and individuals need to abide by those measures, to avoid over-burdening health systems, as is already occurring in some countries.

  2. The actual number of people infected with the virus is likely to be many times more than is currently being reported, because most countries have not tested those with only mild, or no, symptoms. Assuming that currently reported deaths are accurate, an estimated IFR of 0.48% would imply that 37 million people globally are already infected, as of 21/4/20.

  3. More countries need to do more testing urgently. Beyond the obvious clinical priority of diagnosing the seriously ill, this needs to also include widespread testing of those with only mild, or no, symptoms. Only then will the true extent of the pandemic become clear. This is important, because countries that have done more testing to date (relative to number of deaths) also seem to be the ones better able to manage the pandemic. This is probably because countries that test more are able to do more robust quarantining of infected individuals, and more robust tracing and testing of recent contacts.

CLICK TO READ: “Sampling Bias, Case Fatality Rates and COVID-19 Testing”

Limitations: these findings are limited by the availability, accuracy and consistency of official reports. Many other factors may be affecting CFRs than could be studied in this report, and thus the IFR for COVID-19 could still turn out to lie outside the estimated range (i.e. 0.28% to 0.68%).

Disclaimer: The report has not been peer reviewed.