As more data is collected, the true death rate of Covid-19 will decrease
The problem the previous article did not address is infection fatality (death) rate, i.e the amount of people who die for the amount of those infected by the virus. The UK currently has a 16% death rate if you divide fatalities by the official number of cases – a number which is nearing the rate of Ebola. This number for ‘normal’ influenza (flu) is widely reported to be around 0.1%.
However, the rates reported in the media for Covid-19 are highly innaccurate because most testing is on people who have been affected severely enough to be admitted to hospital, and very small proportions of the population have been included in calculations. People who have recovered after catching the virus without needing hospital treatment or those mildly affected or asymptomatic will not have been tested. Therefore, just like with any other new virus, the ‘official’ death rates are likely to keep on decreasing as time goes on and more people are tested. It strikes me as un-transparent, for these limitations in ‘official’ statistics not to be noted, especially from sources who are expected to uphold high values of scientific rigour.
‘True’ fatality rate likely to be between 0.05% and 0.6%
So what is the likely death rate? There is still a large amount of uncertainty, and only large amounts of data in retrospect will fully solve it. An early government estimate was 0.1%, but to investigate, I used numbers from multiple different sources. At the end of March, a model from Oxford University said that half the UK could have been infected by the virus, which would result now in a death rate of around 0.065%. Around the same time, the same scientists from Imperial College London who released the study saying 500,000 people could die from Covid-19, estimated that ‘1 in 40’, or 1.75 million people in the UK had contracted it, which would be a death rate of 1%. This is at odds with new data which suggests a minimum of 8 million people could have been infected. The same team estimated a number of 0.66% for the infection fatality rate in China.
The percentage of people who died after the Diamond Princess Cruise Ship outbreak, compared to those who had been infected, was found in late March to be 1.1%. However, as 78% of the people on the ship were over 60 years old, this number would be much larger than the actual value for populations with a normal age distribution. This suggests that numbers near 1% are over-estimations of the danger posed.
It’s no surprise, therefore, that as time goes on, the number is falling. New data from tests in New York suggest that 1 in 5 people in the city have been infected with coronavirus, which led to a new death rate estimate of 0.79. London and New York are similar in many respects; they both have very extensive transport links to the outside world and are incredibly popular as tourist destinations, dense populations, and huge public transport systems which see billions of people squashed together in close proximity each year, not to forget high levels of air pollution. Therefore, extrapolating these findings to London, if a fifth of its population have caught the virus, that results in a death rate of 0.27%.
Other cities followed suit and similar studies were carried out in Helsinki, Finland; Stockholm, Sweden; and Gangelt, Germany, where the death rate was found to be lower than New York. In Helsinki the number was estimated to be 0.19%, in Gangelt 0.37% and in Stockholm 0.4%.
A study by a team at Stanford University found that the virus had infected 50 to 85 times as many people as the official figures suggested in California. They concluded that the fatality rate of coronavirus was between 0.12% and 0.2%. When extrapolated to the UK, it would result in 8-14 million people infected, with a death rate of 0.16% to 0.27%.
An average for the data comes out as ~0.4%, which is decreasing the longer the pandemic has gone on. When extrapolated to the UK, the range of figures gives a range of 4-33 million people infected with the virus. While these figures are estimates, they can give us a far better idea of the upper and lower limits of the death rate in the UK. Even the very unlikely upper limit of 1% is far lower than the figure that was described at the beginning of march, by the WHO director general Tedros Adhanom Ghebreyesus, as being 3.4% worldwide.
The fatality rate of ‘common’ flu is reported at being around 0.1% and while the scientific community is regularly moving the coronavirus estimate towards this number, it’s worth noting that even a number of 0.2% would mean it poses twice the danger. It’s also important to understand that influenza vaccines do exist, which would significantly lower the death rate.
What do you think? Read the next installment in the Covid-19 Review here! It addresses the problem of ‘mild or asymptomatic’ cases which could mean we have severely underestimated the amount of humans coronavirus has spread to.