User Power
Value/Post Ratio
605%
- Oct 16, 2014
- 801
- 4,843
Wow, the fact that they got a very simple, but very important part of the mathematical equation for "death rate" completely wrong is very concerning...
From the article:
This is completely inaccurate. Here's an analogy:
Imagine there's a test being given to 100 people. And imagine that after 10 people finish the test, turn it in and get graded, we find that 2 of the 10 have passed the test.
Since 2 of 100 people have passed, can we now say that 2% of the people who took the test will pass it?
Of course not. We still have 90 people who haven't yet completed their test and turned it in to be graded. Most likely our pass ratio will be closer to 20% (the 2 in 10 number) than 2%.
This is what that "Death Rate" statistic above is doing. Basically, it is dividing the number of people who have died by the number of people infected. But, that denominator -- the number of people infected -- include a LOT of people who we don't have an outcome for. They haven't died, but they also haven't recovered.
They are the ones who -- based on this analogy -- are still taking the test!
That Death Rate is accurate if -- and only if -- every single person who has gotten the virus, but who hasn't yet died, ends up not dying. And that's obviously not going to be the case.
In reality, the "Death Rate" should be calculated as:
Where the number of RESOLVED CASES is the total number of people who have either died or fully recovered.
Btw, if you do *that* math, we find that the Death Rate would be between 18-20%.
Though, in reality, that's the upper bound, and it's probably lower than that, simply because the number of deaths will be much more accurately reported than the number of recoveries (many people will recover without ever getting diagnosed).
So, 2% is the lower bound; 20% is the upper bound; and the real number is somewhere in-between.
Unfortunately, with the majority of people "still taking the test," it's really hard to know exactly what the Death Rate is right now.
All great points. I didn’t read the methodology, and missed that when looking at the article. Thanks for pointing it out.
However we can still glean from this that whatever the risk, by far the greatest risk is for people over 60 and people with preexisting medical conditions.
This doesn’t mean we shouldn’t take precautions, I certainly am. Even a 2-3% death rate for under 60 would be horrendous. Whatever the outcome, I up my emergency preparedness in general
Dislike ads? Remove them and support the forum:
Subscribe to Fastlane Insiders.