COVID-19 fatality rate may be much lower than previous estimates

[athena53]

I think the likely disparity in the recovered number is due to the people who are tested positive but not admitted to the hospital. There are simply not enough resources to follow up on them. I would think the vast majority will be recovered, because if they were likely to die, they would probably have been admitted to the hospital in the first instance.

Thanks- that's my suspicion, too.

 

[ivinsfan]

I worked with numbers for a living and have seen some cringe-worthy misinterpretations of the data. Ironically, I try not to dwell on numbers knowing all the pitfalls, apples-to-oranges comparisons by country, etc.

But- one thing I'm finding interesting is the percentage of "unresolved" cases. The most current numbers I have for the US show 764,265 reported cases, 40,565 deaths and 71,012 recoveries. So- 5% have died and 9% have recovered. That leaves 86% unresolved. I'm sure there's some lag in reporting, especially deaths since the data would come from death certificates- but what's happening to the rest? Are they dying at home or recovering at home and escaping notice?

The answers could have a huge effect on our assessment of the ultimate death rate. Right now the numerator and denominator are both moving.

In my state I think they used recovered to mean still alive and past the 14 days quarantine, it's a data point and with huge numbers testing positive every day the recovered is certainly a lagging indicator.

 

[Masquernom]

Check out the wisdom of Dr. Bhattacharya...

Using his 4% assumption for % of population infected, and Kentucky population of 4,500,000. I get the following numbers;

KY Population 4,500,000

Assume 4% have virus

Total infected would be 180,000
number tested so far 32,319
9.2% test positive 2,973

5.0% who die 149

total % of infected that die 0.08%

Kentucky has tested 32,319 people with 2,973 testing positive. As of today, 148 have died.

Assuming you need 70% infection rate for herd immunity, we have a long way to go.

 

[ivinsfan]

Using his 4% assumption for % of population infected, and Kentucky population of 4,500,000. I get the following numbers;

KY Population 4,500,000

Assume 4% have virus

Total infected would be 180,000
number tested so far 32,319
9.2% test positive 2,973

5.0% who die 149

total % of infected that die 0.08%

Kentucky has tested 32,319 people with 2,973 testing positive. As of today, 148 have died.

Assuming you need 70% infection rate for herd immunity, we have a long way to go.

sadly these numbers aren't really useful to anybody...that 4% number is a WAG...even in your state there have to be areas with more infections and some with almost zero infections so statewide the datea isn't really that helpful as to what comes next...you are not ever going to get 70% herd immunity with lockdowns in place, the vaccine, which doesn't exist yet would be the only way this is possible..

 

[Masquernom]

sadly these numbers aren't really useful to anybody...that 4% number is a WAG...even in your state there have to be areas with more infections and some with almost zero infections so statewide the datea isn't really that helpful as to what comes next...you are not ever going to get 70% herd immunity with lockdowns in place, the vaccine, which doesn't exist yet would be the only way this is possible..

Yes, just wanted to do a ball park calc to see if the numbers from the presentation make sense according to what's been published here in Kentucky. Based on this small sample so far, it looks like his numbers may be in the ball park. Here they have been mostly only testing people that show up sick. It looks like the virus is much more prevalent than that of course.
Will be interesting to see if when they reduce the "stay at home" rule, infections increase. You'd think they would from these numbers.

 

[Jeffman52]

Infection rates are going to be nearly impossible to nail down. Even with testing. You can test negative today, contract the virus that very day, and go on to infect dozens of others. Eventually we will get some form of herd immunity but we probably over a year from there.

I ran some projections on deaths in the US, a couple of weeks ago and so far I have been close. Obviously the further out the calculations run, the less reliable. We are at 41K deaths right now.

April 30 = 57K deaths.

Reduced deaths will lag the peak by a number of weeks.

Mid May 80K deaths,

Assuming the number of deaths takes a significant turn to the better during May.

May 31 = 90K deaths

That's on the optimistic side. I actually think it will be higher.

June 10th - 100K

 

[Chuckanut]

Interesting data point number 2- on the USS Roosevelt they have tested 94% of the sailors, 600 are positive, 350 have no symptoms, 7 ended up in hospital of these 3 to ICU one of whom has died. That is in young healthy males.

Just males on the carrier? I thought they had women on them in these modern times. Or did no women contract the virus? If so, perhaps we should man our military vessels with women for the duration of the pandemic.

 

[Ronstar]

Just males on the carrier? I thought they had women on them in these modern times. Or did no women contract the virus? If so, perhaps we should man our military vessels with women for the duration of the pandemic.

I suspect that there are females aboard - Friend’s son is on the Roosevelt and has trained female pilots.

 

[Gumby]

I suspect that there are females aboard - Friend’s son is on the Roosevelt and has trained female pilots.

It has been over 25 years since women first joined the crews of combatant ships (March 10, 1994).
https://www.history.navy.mil/conten...ersity/women-in-the-navy/women-in-combat.html

 

[stepford]

Back on topic...


Just released results from a study in LA County using serological tests show 2.8-5.6% of those from a representative group of 863 residents have antibodies indicating COVID-19 exposure. This is comparable to the 2.5-4.2% rate in the Santa Clara County study released on Friday.

The methodology of the two studies is a bit different. Participants in the Santa Clara study came from Facebook volunteers while those from LA were obtained using a market research firm to choose a representative population sample. While neither is a true large scale random sample the similarity of the two results certainly seems to indicate that the infection rate in urban portions of California is at least "a few" percent.

 

[donheff]

Back on topic...


Just released results from a study in LA County using serological tests show 2.8-5.6% of those from a representative group of 863 residents have antibodies indicating COVID-19 exposure. This is comparable to the 2.5-4.2% rate in the Santa Clara County study released on Friday.

The methodology of the two studies is a bit different. Participants in the Santa Clara study came from Facebook volunteers while those from LA were obtained using a market research firm to choose a representative population sample. While neither is a true large scale random sample the similarity of the two results certainly seems to indicate that the infection rate in urban portions of California is at least "a few" percent.

Did they have data for the specificity of the test? It needs to be very high - 99%+ to be reliable for finding the true positives if the prevalence in the population is low. I hope these assessments are accurate because that would mean the fatality rate is lower than we thought. But Google why most published research findings are false.

 

[ejman]

It continues to baffle me that according to today's numbers at https://www.worldometers.info/coronavirus/#countries the total death to cases ration has now hit 7%. It has been constantly climbing from about 4% in March. This makes no sense to me as the death rate should be declining as more and more people are tested as shown at the worldmeter site. I have read many statements that the true death rate is well below 1% perhaps below 0.1% and I certainly hope this is so but damn it, it would be nice if this stat started showing this down trend.

 

[ivinsfan]

It continues to baffle me that according to today's numbers at https://www.worldometers.info/coronavirus/#countries the total death to cases ration has now hit 7%. It has been constantly climbing from about 4% in March. This makes no sense to me as the death rate should be declining as more and more people are tested as shown at the worldmeter site. I have read many statements that the true death rate is well below 1% perhaps below 0.1% and I certainly hope this is so but damn it, it would be nice if this stat started showing this down trend.

Death rate is a number but doesn't explain everything. Our particular state MN is having horrible death rates ATM..last I saw something 88% of deaths were of people in group living and the vast majority of them in nursing homes..even though the homes have been locked down for at least 5 weeks the number of NH with infections continues to rise rapidly...this is a big issue does it mean more staff are asymptomatic carriers because the nursing homes are screening staff daily..

 

[Aerides]

It continues to baffle me that according to today's numbers at https://www.worldometers.info/coronavirus/#countries the total death to cases ration has now hit 7%. It has been constantly climbing from about 4% in March. This makes no sense to me as the death rate should be declining as more and more people are tested as shown at the worldmeter site. I have read many statements that the true death rate is well below 1% perhaps below 0.1% and I certainly hope this is so but damn it, it would be nice if this stat started showing this down trend.

This is another reason to aggressively expand (triple+) testing. Still, it appears difficult in most cases to get a test without a high number of symptoms. If a lot of low-symptom/asymptomatic folks are still being missed, then the Cases denominator is stacked with sicker people to begin with.

 

[LOL!]

I suppose this has already been mentioned in this thread:

The death rate has to go lower because those folks with known co-morbidity problems are more likely to do everything they can to not get infected in the first place.

So say originally the infected stats were 80% recovered at home, 20% hospitalized, and 2% died. Nowadays folks know that a large fraction of the 2% deceased were seniors with diabetes, COPD, heart problems. That doesn't mean that ALL of the 2% deceased were seniors with co-morbidity issues.

I know that I have no co-morbidity issues, so I will go grocery shopping, but I know quite a few 70+ year-old individuals with diabetes and COPD that are not stepping out of their homes for awhile. They are relying on others to bring them food and supplies.

 

[Sojourner]

It continues to baffle me that according to today's numbers at https://www.worldometers.info/coronavirus/#countries the total death to cases ration has now hit 7%. It has been constantly climbing from about 4% in March. This makes no sense to me as the death rate should be declining as more and more people are tested as shown at the worldmeter site. I have read many statements that the true death rate is well below 1% perhaps below 0.1% and I certainly hope this is so but damn it, it would be nice if this stat started showing this down trend.

When you hear things like the "death rate is well below 1%", they aren't referring to the case fatality rate (CFR). They are referring to the infection fatality rate (IFR). The IFR measures the percentage of all infected people who die from the disease. CFR only counts people who felt sick enough to go get tested, so this number will always be higher (usually, much higher) than the IFR, since a substantial percentage of infected people never get tested.

From what I'm hearing and reading, the overall IFR for Covid-19 is very likely under 1%, while the CFR could be the more widely reported 3%, 4%, or even higher. And, of course, these numbers aren't "one size fits all". The IFR for a healthy 20-year-old is probably close to zero, whereas the IFR for a diabetic 85-year-old could be as high as, say, 15%.

 

[copyright1997reloaded]

I generally agree with your statements but I would like to point out two areas of concern:
1) the ratio of deaths to total cases seems to be going up over time not down as one would expect given that
2) the number of tests is increasing substantially - almost 4 million in the US, of which 3/4 of million were confirmed cases (this data from today's worldmeter results.


So, as more and more tests are performed the ratio should start going down hopefully soon but I wish it would do so already!

Your observation is correct. I've been tracking the data for about six weeks in a spreadsheet. Here is the Deaths as a % of positively confirmed cases:
https://docs.google.com/spreadsheets/d/e/2PACX-1vQbIt2d1xYnfxfYpwgwNLIbuap5bfK396ukGFipDe6sKCnV21qqVglkA2q2WCEipspXZa6HqYlQqxAN/pubhtml?gid=1734147132&single=true

Deaths as a % of all tests (both positive and negative) has also trended up, currently sitting at 0.94% (That is a slight tick down from 0.96%, surprisingly the first down tick since March 14th.) What we obviously don't know if the multiplier effect, i.e. people in the population who have had (or have) the virus but haven't been tested.

ETA: There is likely significant bias - that is, the people who tend to get tested are "sick enough" to go through the hassle of the test. (Also likely to test other types, e.g. those who think they get every disease, people in positions w/exposure who get repeated tests, ...)

 

[ivinsfan]

Your observation is correct. I've been tracking the data for about six weeks in a spreadsheet. Here is the Deaths as a % of positively confirmed cases:
https://docs.google.com/spreadsheets/d/e/2PACX-1vQbIt2d1xYnfxfYpwgwNLIbuap5bfK396ukGFipDe6sKCnV21qqVglkA2q2WCEipspXZa6HqYlQqxAN/pubhtml?gid=1734147132&single=true

Deaths as a % of all tests (both positive and negative) has also trended up, currently sitting at 0.94% (That is a slight tick down from 0.96%, surprisingly the first down tick since March 14th.) What we obviously don't know if the multiplier effect, i.e. people in the population who have had (or have) the virus but haven't been tested.

ETA: There is likely significant bias - that is, the people who tend to get tested are "sick enough" to go through the hassle of the test. (Also likely to test other types, e.g. those who think they get every disease, people in positions w/exposure who get repeated tests, ...)

the powers that be in my state like the throw numbers around. In the month they said there could 10 infected for every positive test number, in the last week or so I'm seeing them use the number there could 100 infected for every positive test. IOW nobody knows anything.

 

[donheff]

When you hear things like the "death rate is well below 1%", they aren't referring to the case fatality rate (CFR). They are referring to the infection fatality rate (IFR). The IFR measures the percentage of all infected people who die from the disease. CFR only counts people who felt sick enough to go get tested, so this number will always be higher (usually, much higher) than the IFR, since a substantial percentage of infected people never get tested.

From what I'm hearing and reading, the overall IFR for Covid-19 is very likely under 1%, while the CFR could be the more widely reported 3%, 4%, or even higher. And, of course, these numbers aren't "one size fits all". The IFR for a healthy 20-year-old is probably close to zero, whereas the IFR for a diabetic 85-year-old could be as high as, say, 15%.

+1 and under CDC guidance states have been correcting for Covid deaths outside of hospitals. That probably bumped up the rates.