Recently stumbled upon a report by the "Centre d’épidémiologie et de santé publique des armées" about this outbreak - "Document n°929/ARM/SSA/CESPA du 4 mai 2020" - report in French
Back when it was an ongoing story in the headlines, news reports read like "many infected, most without any symptoms, no dead". Time to look at it again.
Short summary
Out of 1767 people on board, at least 1064 were confirmed as infected (60%) and 29 developed serious disease requiring oxygen.
Of those 1568 participated in a subsequent phone survey (study population), age range 18-60, median age 29 years. 1001 of the study population were confirmed PCR positive, 546 negative and 21 results were unknown.
Out of those merely 13% were asymptomatic. 8.1% reported atypical symptoms (neither loss of smell or taste, no fever nor cough), 78.9% reported typical symptoms, see bellow.
26 of the confirmed cases became seriously ill requiring oxygen - that is 2.4% of the total infected or 1.5% of the total population of a cohort with a median age of 29 years and presumably way above average fitness. Further 3 cases required oxygen but their PCR test was negative or not available at the time the report was written.
The symptoms considered "typical" were: loss of smell 57%, headaches 56,7%, loss of taste 46,4%, weakness 46,3%, muscle pain (myalgia) 45,2%, fever 44,8%.
Those seriously ill reported different prevalence of symptoms: fever 90%, muscle pain 90%, cough 86%, headaches 76%, dyspnoe 72%, weakness 66%.
In univariate analysis those aged 36-46 had a 6fold risk to develop serious disease, those 46-60 14.5 fold. Obesity (defined as BMI>25) carried an approximately 3.5 fold risk.
According to the authors loss of smell and/or taste as symptom had a sensitivity of 67% and a specificity of 80%
Comment and secondary analysis
This numbers leave very little room to speculations that there might be a high percentage of asymptomatic cases and only a very small fraction of those infected become seriously ill. Also this leaves very little room to speculations that some substantial share of the population might be magically immune to COVID-19.
We can safely assume that a little more than just the confirmed 1064 were infected: the PCR tests used had a sensitivity of around 96% under optimal circumstances back than, in reality the testing started a little late so many infections might have cleared before testing for COVID-19 was done. The specificity of the PCR teste otoh was well over 99% even back then so the number of false positives should have been 18 or less.
To illustrate the disaster which an uncontrolled spread could cause in a western society consider this numbers: in the 36-45 age group 5.2% of the infected or 3.4% of the total population would require oxygen. In the 46-60 age group this would be 11.6% resp. 8.7%. Even in the less affected 18-25 and 26-35 age groups still around 1% of all infected require oxygen. Those numbers are certainly extremely optimistic - comparing the top fit sailors with the same age group of the general population, assuming that the epidemy would stop after infecteing 60% of the population and not even considering those over 60.
Did something change since the outbreak? I think there is some progress reducing mortality, however this is at the cost of treating patients earlier with oxygen which would cause even higher demands on the health system to provide optimal treatment.
