Sweden's coronavirus fatality rate 0.6 percent, according to new report

The virus, according to the report, proved fatal in approximately 4.3 percent of cases among the over 70-year-olds, compared with only 0.1 percent of cases for those under 70.

These figures suggest it is around six times as deadly as the seasonal flu, Public Health Agency researcher Martin Berlin told The Local. 

"Just as for covid-19, it’s difficult to estimate [fatality rate] for seasonal influenza due to the problem of estimating the total number of infections," Berlin said, "but a typical estimate [of the fatality rate of flu] is 0.1 percent".

The new study was based on the number of (both reported and unreported) coronavirus infections and (reported) deaths in the Stockholm region, Sweden's worst affected, during March 21st - 30th. 

Using only the cases confirmed by a diagnostic test would give a distorted mortality rate; Sweden has been testing sparsely, reserving many tests to the groups most at risk of infection and of serious illness as a result. That means a much larger share of the population is thought to have been infected with Covid-19 than those who have been tested positively for the virus, with the vast majority only experiencing mild symptoms.

In recent weeks, Sweden has more than doubled the number of coronavirus tests carried out and has expanded testing to include anyone with symptoms and a medical referral. 

• Sweden announces major overhaul of coronavirus testing strategy
   
• Has Sweden passed its coronavirus peak yet?

So instead of using the cases which have been confirmed by testing, the estimated number of infections used in the study was based, in part, on a survey in which a random sample of the population in the Stockholm region was tested for Covid-19.

The researchers concede, however, that there is still uncertainty concerning the relationship between the total number of infections and the number of confirmed infections. The study’s estimation is that, for every confirmed Covid-19 case, there have been around 44 ‘invisible’ or unconfirmed coronavirus cases. But this number could later prove to be either higher or lower.

While the study did try to account for the unreported infections, researchers only included deaths from confirmed coronavirus cases in their analysis. This is the same method the Public Health Agency has used to compile its figures of the death toll; the agency includes all deaths that occur where the person tested positive for the virus within the past 30 days.

"This is likely to be an underestimate of the total number of deaths, however, since we observe excess mortality during this period that cannot be completely accounted for by the confirmed deaths. We cannot incorporate these excess deaths formally in our estimations, but I wouldn’t say that the difference would be drastically different if we could include these," Berlin explained.

Excess mortality points to the difference between the usual or average mortality rate during a certain period of time, in a specific geographical area, and the current death rate. Excess mortality numbers have been used in other (international) studies to account for fatalities that are likely to be largely related to coronavirus, but have not been reported as such.