IFR = fatalities/estimated cases. It doesn't have anything to do with percentage of the population.ATL Bear said:That number is used in the IFR.Sam Lowry said:So, not an IFR and not the comparable number.ATL Bear said:That's the percentage of people estimated to be infected by the flu as a percentage of US population.Sam Lowry said:I meant the link in the quote. Where do you get an IFR of 13% for the flu?ATL Bear said:I did and it's incomparable. Their IFR factor is .94% for Covid-19 whereas the flu in comparison is 13%. I also think they need to update their models with new data. Still using older China data.Sam Lowry said:Click the link:ATL Bear said:You're link. It has no corresponding estimates of non verified infections. It simply parrots something from the cruise ship and runs with 1%.Sam Lowry said:Basing an IFR only on known cases is a contradiction in terms. Can you show me where anyone is doing that?ATL Bear said:The confusion is there is no similar data point for IFR for C-19 as there is with the flu which has the biggest variant impact. The 38 million estimate for the flu is enormous from a ratio impact. The IFR on C-19 is only using known cases. But I think we'd all agree that there are thousands if not 10's or 100's of thousands of C-19 cases that haven't been found or were never logged as official as the symptoms were beaten/mild.Kyle said:I think maybe you flipped IFR and CFR, which led to the confusion.ATL Bear said:That's my exact point. If we measured flu mortality similar to how we're doing C-19, its mortality would look horrible. For example, if one were to estimate that right now 1 million people in the US have or had C-19, the current mortality rate would be .02%. If we took the number of hospitalizations from the flu and divided by deaths it would still have a high mortality. The missing number in this equation is an estimate of existing and past infections from C-19 since it was first confirmed in January.Kyle said:I'm not sure that's right. CDC estimates to-date there have been 38 million cases of the flu as well as 390K hospitalizations and 23K deaths. Wouldn't that 38M be the denominator of the IFR? The number of positive tests was much lower. Interestingly, if you look at the CDC site, its data shows 232,654 cases of positive flu tests for A&B with an estimated annual mortality of 23,000. So by CFR, that would put its mortality rate at 10%, which seems really high.ATL Bear said:That link still uses the 35 million for flu infections when there weren't 35 million confirmed cases of flu. Under the metrics they advocate for (IFR), flu would still have a higher death rate by a multiple factor. No one has an assessment of how many actual cases of C-19 are out there, from South Korea to the US. One thing that is being borne out is fortunately it's a comparatively mild virus (mortality) by any standard of measure.Sam Lowry said:
We are not basing our policy on different metrics for measuring death rates. Approximately 1% is an estimate of the total infection fatality rate, not the diagnosed case fatality rate.
http://theconversation.com/the-coronavirus-looks-less-deadly-than-first-reported-but-its-definitely-not-just-a-flu-133526
Even today due to the lack of available testing most people are being told to quarantine and only elevate if serious symptoms start to manifest. Many never experience them or require hospitalization so they never become a "stat" or are tested.
EDIT: For Covid-19 that is.Quote:
This statistic is harder to calculate, as it requires estimating the number of undetected infections. The Infection Fatality Rate (IFR) is the number of deaths divided by the true number of infections (including both confirmed and undiagnosed cases). One estimate of the IFR for COVID-19 puts this figure at 1%, and some new data suggests this is credible.

