> From our data, about 10% of patients require ICU (Intensive Care Unit) or sub ICU assistance and about 5% of patients die.
I think that we really need to understand precisely what conclusions these represent when we see statements like this.
Exactly what does "10% of patients" mean? 10% of those who test positive? 10% of those who test positive and also show symptoms?
It seems that a majority of people who contract the virus and test positive (let alone those who are exposed) do not show any symptoms at all.[0]
Of those who do show symptoms, estimates regarding how many any seek out, receive, or require care vary widely, so much so that it's very difficult to draw any conclusions using any dataset that I've seen.
So I ask: When the authors here say "From our data", exactly what dataset are they using and what operations are they performing on this dataset to arrive at this number?
This is crucially important, because if this 10% is actually 10% of those who already seek treatment for respiratory illness, and also subsequently test positive for COVID-19, then the death rate is actually much, much lower than most of the published reports.
It seems to me that we need to know the answer to this before proceeding with any other restrictive public policy solutions.
I'm particularly concerned with the decision to close schools, which in many cases (at least here in the USA) is inevitably going to lead to kids being placed with grandparents for care, which is a terrible outcome if our goal is to reduce exposure to vulnerable populations.
It is OK to question the panic. It is OK to believe, as I do given the available data so far, that the panic is more harmful than the virus itself. I have started a subreddit to gather data and opinions inclined toward using this as an opportunity to appropriately prepare rather than race to drastic solutions likely to hurt vulnerable populations.[1]
0: https://eurosurveillance.org/content/10.2807/1560-7917.ES.20...