The infection fatality rate for Covid-19 is higher than 0.1%

In a now-deleted tweet, writer and commentator Toby Young claimed that a study reported in the Daily Mail shows that Covid-19’s infection fatality rate (IFR) is 0.1%, a figure comparable to seasonal flu. 

This tweet incorrectly cited the study and the Daily Mail article, as well as incorrectly calculating an IFR rate (the proportion of all infected people who die from Covid-19). Mr Young has subsequently issued a correction acknowledging the error.

As many people on Twitter had pointed out, and Mr Young subsequently acknowledged, his calculation for the IFR of Covid-19 was incorrect. If over 5 million people had caught the virus in the UK by August, as argued by the study, and using a current death figure in the UK of around 50,000 (the current figure for deaths within 28 days of a positive test is over 52,000) then you would get an IFR of around 1%, not 0.1%.

However, it’s worth noting that both the study and the Mail article that was cited already assume an infection fatality rate. In fact, that’s how the study calculated its figure for how many people in the UK have been infected—by “backcasting” from death figures up to the end of August and using an estimated IFR to produce an estimate of how many infections would have led to those deaths.

The MailOnline article states “Fatality rate changes based on a patient's age and any comorbidities, but by reviewing 26 studies the researchers settled on a percentage of 0.76 per cent”.

Similarly, the research paper devotes a paragraph to explaining this age variation and differing reports, before saying “When the meta-analysis was restricted to the six studies that were considered to have a low risk of bias, a best estimate of 0.76% and a 95% confidence interval of 0.37–1.15% was obtained.”

This results in an estimate that 8.23% of the UK population were infected, with a 95% confidence interval of 5.73% to 14.43%. Applying this percentage (8.23%) to the UK population is around 5.5 million people.

As we’ve written before, the true infection fatality rate can vary depending on factors such as a population’s characteristics and the quality of medical care, but a figure as low as 0.1% is not credible. Most studies have suggested that in reality, Covid-19 kills between 0.5% and 1% of people who catch it, which fits with the estimate of 0.76% used in this study.