From a quick and dirty skim of the linked article, it looks like the 10-20% number may not be based directly on the Singapore data—but possibly it is based on China data. Quote in context:
Fifth, the medical community needs to collectively find better ways to communicate and engage the public in the social media era. The public is understandably anxious about COVID-19, given how rapidly the epidemic has spread with 10% to 20% of hospitalized patients becoming severely ill.
According to the data here (https://www.worldometers.info/coronavirus/) only 6⁄108 = 5.6% are in serious or critical condition. That’s about the same as on Diamond Princess (36/699=5.1%).
So 5.2% of cases in serious/critical condition, plus 0.9% deaths in the sum of these two especially relevant populations.
Hmm, I had noticed that dividing the number of serious or critical cases by the number of total cases gives less than 10%, but assumed that’s because not all cases had enough time to progress to where they might become serious or critical yet, and the 10-20% was the authors adjusting for that. But I guess you’re right that maybe they just based it on China’s data.
So extrapolating from the current Singapore+Diamond Princess numbers, assuming 50% worldwide infection rate and 0-100% dead among severe/critical cases—and no hospital care—about 0.5%-3% of the world population will die.
So the graph above implies that every demographic has approximately equal hospitalization rates, which other sources suggest are 15% to 20%.
This is a weird pattern – why are so many young people getting hospitalized if almost none of them die? Either the medical system is serving these people really well (ie they would die if they didn’t go the hospital, but everyone does make it to the hospital, and the hospital saves everyone who goes there), they are being hospitalized unnecessarily (ie they would live even if they didn’t go the hospital, but they do anyway), or it’s statistical shenanigans (eg most statistics are collected at the hospital, so it looks like everybody goes to the hospital).
Are these an overestimate? Maybe most cases never come to the government’s attention? There’s some evidence for this.
The young people who get hospitalized and recover are mostly not needing ICU or mechanical ventilation (~2% of total young adult cases do, i.e. 10% of the hospitalized young adults I guess), but I can’t find data on what interventions they are getting and whether they’re lifesaving.
From a quick and dirty skim of the linked article, it looks like the 10-20% number may not be based directly on the Singapore data—but possibly it is based on China data. Quote in context:
According to the data here (https://www.worldometers.info/coronavirus/) only 6⁄108 = 5.6% are in serious or critical condition. That’s about the same as on Diamond Princess (36/699=5.1%).
So 5.2% of cases in serious/critical condition, plus 0.9% deaths in the sum of these two especially relevant populations.
Hmm, I had noticed that dividing the number of serious or critical cases by the number of total cases gives less than 10%, but assumed that’s because not all cases had enough time to progress to where they might become serious or critical yet, and the 10-20% was the authors adjusting for that. But I guess you’re right that maybe they just based it on China’s data.
So extrapolating from the current Singapore+Diamond Princess numbers, assuming 50% worldwide infection rate and 0-100% dead among severe/critical cases—and no hospital care—about 0.5%-3% of the world population will die.
In other words, a CFR of 1-6%, with the lowest value overlapping with estimates being put out by governments right now.
EDIT: I just read Scotty’s new post on the subject and he’s confused by that 10-20% figure as well
The young people who get hospitalized and recover are mostly not needing ICU or mechanical ventilation (~2% of total young adult cases do, i.e. 10% of the hospitalized young adults I guess), but I can’t find data on what interventions they are getting and whether they’re lifesaving.
With regard to the timing of things, the recent WHO-China report has some interesting charts.