A third round of SARS-CoV-2 antibody testing by researchers at Geneva’s HUG hospital suggests 9.7% of the canton might have had Covid-19.
Geneva currently has a crude fatality rate of 5.3%, based on the total number of people testing positive (5,031) and the recorded number of Covid-19 related deaths (269).
The antibody testing programme, which started in March 2020 and runs until June 2020, aims to understand the true scale of infection in the canton. SARS-CoV-2 testing has been rationed in Switzerland. Typically, only those with significant symptoms are tested. This means many of those infected are not included in the official case number. The question is: how many?
By 11 May 2020, Geneva had recorded 5,031 Covid-19 cases, 1% of the canton’s population, the highest rate across Switzerland’s 26 cantons. If this recent round of antibody testing is accurate, then it suggests only 1 out of every 9.7 cases made their way into Geneva’s official case number, suggesting it is significantly understated.
Some SARS-CoV-2 antibody tests have limited reliability and yield a problematic number of false positives – people testing positive for antibodies when they don’t have them. The Euroimmun IgG ELISA test, the primary test used in this study, has a specificity of 96%, according to one study. This means if 8% of a sample tested positive for antibodies, half could be false positives (without SARS-CoV-2 antibodies) if the real rate of infection was 4%. However, the Geneva study made certain adjustments designed to correct for this.
The antibody testing programme also looked at infection rates by age group. The results suggest that infection could be more prevalent among younger people than the official RT-PCR test case numbers suggest. This makes sense given older people are more likely to have severe symptoms and are therefore more likely to get a RT-PCR test.
Recorded cases testing positive using RT-PCR tests so far in Geneva represent 0.15% of 5-19 year-olds, 1.27% of those aged 20-49 and 1.44% of those 50 and over. Antibody testing suggests 6.0% of 5-19 year-olds in Geneva have had Covid-19, 8.5% of 20-49 year olds, but only 3.7% of those 50 and older, a result that suggests efforts to reduce social mixing of elderly people with others may have succeeded, according to the report’s authors. However, it is also possible the elderly develop a lower immune response, something requiring further investigation.
So far, 99.3% of those dying from Covid-19 in Geneva are 50 or over. An estimated 3.7% infection rate among this age group would still leave 96.3% in this high-mortality group that haven’t had it.
The authors of the report point out that assuming antibodies provide meaningful immunity, even at an infection rate of 9.7%, Geneva still has a long way to go to reach herd immunity.
HUG report (in French) – Take a 5 minute French test now
For more stories like this on Switzerland follow us on Facebook and Twitter.
Michael Schaefer says
In my view, the critical words about possible flaws with reagrd to the test performance seem not justified since in the cited study the authors have corrected the raw test results for these uncertainties. Maybe the author of this commentary has not read the full paper? Otherwise, he/she should have recognized this important point. Also, I would like to point out that the value of 3.7% seroprevalence in the 50+ age group is aggregated data over all three test time points and does not solely stem from the last round as this commentary imples. The actual seroprevalence in the 50+ age group within the last round of testing has not been stated in the paper. Thus, the 3.7% seroprevalence should not be compared to the cumulated number at the end of April, but with somewhat earlier death tolls.
Le News says
Thank you for your comments. We have made the clarifications you have suggested.