This NEJM article you referenced (https://www.nejm.org/doi/full/10.1056/NEJMe2002387) is a poor estimate of the Case Fatality Ratio. I'm perplexed it passed peer-review (I'm not even an expert, so why am I finding such glaring flaws?)
It's citing two references (ref. 4 and ref. 5) that estimate CFR at respectively 2% and 1.4%, however both use the naïve CFR formula (deaths/cases) that's known to underestimate the CFR due to deaths lagging cases in a rapidly growing outbreak. I wrote about this bias and provided 5 peer-reviewed epidemiological papers that also document it, see: https://blog.zorinaq.com/case-fatality-ratio-ncov/
Then it remarks that many mild cases are probably not detected (ie. high underascertainment) but strangely makes no effort at quantifying it, or referencing other papers that do quantify it.
The 4 best published studies that correct for these two aspects (naïve CFR bias + underascertainment) are:
Thank you for sharing this. I keep seeing people try to claim "but we don't know the true number of unreported cases therefore the true CFR is lower than reported" as if epidemiologists are somehow unaware of this factor.
Comments
This NEJM article you referenced (https://www.nejm.org/doi/full/10.1056/NEJMe2002387) is a poor estimate of the Case Fatality Ratio. I'm perplexed it passed peer-review (I'm not even an expert, so why am I finding such glaring flaws?)
It's citing two references (ref. 4 and ref. 5) that estimate CFR at respectively 2% and 1.4%, however both use the naïve CFR formula (deaths/cases) that's known to underestimate the CFR due to deaths lagging cases in a rapidly growing outbreak. I wrote about this bias and provided 5 peer-reviewed epidemiological papers that also document it, see: https://blog.zorinaq.com/case-fatality-ratio-ncov/
Then it remarks that many mild cases are probably not detected (ie. high underascertainment) but strangely makes no effort at quantifying it, or referencing other papers that do quantify it.
The 4 best published studies that correct for these two aspects (naïve CFR bias + underascertainment) are:
https://www.mdpi.com/2077-0383/9/2/523
https://www.imperial.ac.uk/media/imperial-college/medicine/s...
https://institutefordiseasemodeling.github.io/nCoV-public/an...
https://github.com/calthaus/ncov-cfr/blob/d30f02e1c20e06103a...
(3 of the 4 above studies are even referenced by the WHO in their Coronavirus disease situation report 30 (references 10, 11 and 12) and in situation report 31: https://www.who.int/docs/default-source/coronaviruse/situati... and https://www.who.int/docs/default-source/coronaviruse/situati... )
And they all estimate the CFR to be between 0.5% and 1.6%, or about 5x to 16x deadlier than the flu (0.1%)
Thank you for sharing this. I keep seeing people try to claim "but we don't know the true number of unreported cases therefore the true CFR is lower than reported" as if epidemiologists are somehow unaware of this factor.