General Discussion
Related: Editorials & Other Articles, Issue Forums, Alliance Forums, Region ForumsLeghorn21
(13,520 posts)Link to tweet
And in Washington state:
https://www.democraticunderground.com/?com=view_post&forum=1082&pid=11807
The_jackalope
(1,660 posts)12% of all "Closed" cases have died, while have 88% have recovered. It will be interesting to watch what happens to that 12% number as the majority of the cases move from the "Active" to the "Closed" category - as they must.
I still think the final death rate (the number of closed cases that die) will be much higher than the interim 2.67% that is currently reported by calculating the percentage of the total number of cases (i.e. closed+active) that have died. It won't be the 20% that it looked like it could be a while ago, but even 12% still puts it in Spanish Flu territory. We'll just have to wait and see.
dewsgirl
(14,961 posts)higher and you don't put draconian measures in places, like locking down 760 million people over the flu.
PoindexterOglethorpe
(25,746 posts)and don't seek medical attention, in reality the death rate is going to be much lower.
But what no one is telling us is just when these two who died sought medical help, how sick they were at that point, and what was actually done for them.
The_jackalope
(1,660 posts)Lower than what? Lower than 2.5%? lower than 12%? Remember that some unknown number of that unknown number of unreported cases with mild symptoms will progress to become major cases.
I suspect we'll see the final death rate settle down between 8% and 12%. Of how many cases? Who the fuck knows.
PoindexterOglethorpe
(25,746 posts)By a lot.
Again, most people have very mild symptoms and don't seek out medical treatment at all. Where in the world are you getting a final death rate of between 8% and 12%?
The_jackalope
(1,660 posts)With a big assumption that the reports of of case categories are relatively accurate. My arithmetic technique is in the post. The formula is (Deaths/(Active+Recovered+Deaths). When the disease has run its course, Active trends toward 0, leaving the final death rate at (Deaths/(Recovered+Deaths). The latter number is given on the stats page as 12% right now: Deaths/Closed cases, where Closed = Recovered+Deaths.
I think it will trend lower as the disease progresses, but not enough to put it all the way down to ~2%.
Blues Heron
(5,898 posts)Are you sure everyone that gets better ends up on the closed list? So far the CDC of China is reporting 2.3 death rate (still 23 times the flu rate of 0.1)
The_jackalope
(1,660 posts)And someone with a "mild" case can infect and kill someone else. It doesn't depend on the severity of the transmitting person's illness, but on the infected host's response to the virus.
Having a large reservoir of unreported mild cases could bias the total case numbers down a bit, but it would also tend to raise the apparent R_nought of the virus, meaning it would appear to be more contagious than it "really" is, as new cases would appear as contagion from a reservoir of infection with fewer apparent sources.
Blues Heron
(5,898 posts)Thanks again for staying on top of the latest developments as this nightmare unfolds.
PoindexterOglethorpe
(25,746 posts)the death rate there so far is zero percent.
And again, a lot of people aren't being diagnosed, have mild cases, and are recovering. So far, at least 75,000 people have been infected, and so far about 2,000 have died. That's a bit over 2%. And if, as is claimed, many, many more are sick, again, the percentage of deaths is even lower.
Ms. Toad
(33,915 posts)Last edited Wed Feb 19, 2020, 08:59 PM - Edit history (1)
That means you don't compare the # dead to the currently diagnosed. You compare the # dead to the # diagnosed 2 weeks ago.
That's 28266. That puts the death rate at a bit over 11% (2125/28266)
PoindexterOglethorpe
(25,746 posts)then the 11% would be correct. We haven't a clue how many people have really contracted the virus and not sought medical help. I keep on seeing that the majority of people simply don't get very sick.
I'd also like to know more about those who have died. Supposedly, as is the case with flu, it's mainly the elderly and those who already have underlying health issues. Also, smokers don't have the healthiest lungs in the world.
Ms. Toad
(33,915 posts)I suspect the same is true for influenza (having no record of the milder cases). So we will likely be comparing apples to apples. There will be infections missing from both total infection tallies. I have no reason to think we are any better (or worse) at tracking mild cases of COVID-19 than we are at tracking mild cases of the flu - so the ratio of death rates should be about the same with the dust settles.
As to age - these are likely low (since they are based on # deaths / total infected):
80+ years old 14.8%
70-79 years old 8.0%
60-69 years old 3.6%
50-59 years old 1.3%
40-49 years old 0.4%
30-39 years old 0.2%
20-29 years old 0.2%
10-19 years old 0.2%
0-9 years old no fatalities
No information about smoking, specifically. Chronic respiratory distress is the closest - 6.9% of people with chronic respiratory distress (out of the the total infections - so again, low)
https://www.worldometers.info/coronavirus/coronavirus-age-sex-demographics/
Demonaut
(8,909 posts)defacto7
(13,485 posts)they will infect more people unknowingly. The number of deaths by any measurment would be higher. The death rate may effect our understanding of the deadliness of the virus but the number of deaths depends a great deal on the number of exposures.
PoindexterOglethorpe
(25,746 posts)And not seek medical attention.
So far, the death rate, based only on those cases where people seek medical attention, is still hanging in at about 2%. And even though some absurd number of people got infected on board the Diamond Princess, thanks to an ill-thought-out and poorly managed quarantined, I don't believe any of them have died from the corona virus. In fact, a quick google search shows that about 621 people on that ship have tested positive for the corona virus. So even a death rate of 2%, 12 people should have died. So far, no one has. I know it's still possible that some will die, but again, the fear mongering is rather out of proportion to the danger.
Ms. Toad
(33,915 posts)so your 12 people dead figure is off by quite a bit, since those it takes about 2 weeks to die, if you're going to die (so those just diagnosed would not yet have died.)
It's not fear-mongering. It is having a bit of respect for an illness that appears to be about 20 times as infective as the flu, and have a death rate that is higher than any seasonal flu other than the 1919 influenza or the 2009 influenza.
If we treat it as trivial, it will spread, and the number of deaths could well exceed the 1919 influenza.
PoindexterOglethorpe
(25,746 posts)And given how the quarantine itself was a complete cluster-fuck, it's astonishing that at least so far, so few people have tested positive for it.
Oh, and that 621 people infected is the most recent numbers. Not from two weeks ago.
Ms. Toad
(33,915 posts)People don't die instantaneously. I takes time. Approximately 2 weeks.
So when you determine how many should be dead today, you multiply the presumed death rate (2%) by the number who have been sick long enough for the virus to have killed them. That's the 20 people who were infected on February 5.
That means the number we might expect to be dead not, with a death rate of 2% is .02 x {the number infected 2 weeks ago} - NOT the number infected today.
Two weeks ago there were 20 people infected. 2% of 20 is .4 people, not 12 people. So it is not surprising that the current death toll on the ship is 0.
PoindexterOglethorpe
(25,746 posts)And even though the quarantine was apparently more than sloppy, only about 20% of those on board got infected. The incubation period for the corona virus is from 3 to 12 days.
Meanwhile, in China, where most of the cases are, the death rate is still 2%, so I cannot understand why people are breathlessly predicting a far higher death toll. Yes, that is significantly higher than the death rate from flu but not the numbers people keep on predicting. And at the risk of repeating myself, many people, perhaps most, don't have symptoms or have very mild symptoms.
It also appears that the high rates of smokers in China may be a factor in the preponderance of deaths there. Apparently, how the corona virus kills is mainly by causing acute respiratory distress syndrome. And here's a fascinating link about ARDS and putting people on ventilators.
Ms. Toad
(33,915 posts)Incubation period is not what is relevant for death rate.
The time from infection to death is what is relevant.
The time from first symptoms to death = 14 days (on average)
That means people dying today started showing symptoms on February 5th. People who started showing symptoms on the 6th will not die until tomorrow, on the 7th will not die until Friday.
From the worldometers.com site:
The correct formula, therefore, would appear to be:
CFR = deaths at day.x / cases at day.x-{T}
(where T = average time period from case confirmation to death)
This would constitute a fair attempt to use values for cases and deaths belonging to the same group of patients.
So in order to use the number of dead today to calculate a death rate, you have to know how many people were showing symptoms 14 (or more) days ago. Because those starting to show symptoms today will still be alive today - even though they may be dead 14 days from now.
We dont' have a means of identifying when individuals on the boat started showing symptoms, so using the confirmed date of infection as a surrogate, there were only 20 people on board who could potentially be dead by today (the other 600-ish have not had enough time yet to die).
So the calculation for how many you would expect to see dead is not 621 (the number of confirmed cases today) x .02, it is 20 (the number of confirmed cases 14 days ago) x. .02.
As for predicting a higher death rate - you need to do a similar calculation.
Those "eligible" to be dead are those who have lived with the disease for 14 days (the average time from symptoms to death). That means the potential dead include only those with confirmed infections on February 5. There were only 28,266 people with infections then. Of those people who have lived with the disease for 14 days as of today, there are 2128 deaths. That puts the death rate at 7.5%
There are going to be other factors, and this is just a rough estimate becase the number of people for whom we know the final resolution is relatively small. There is more information at the site I linked to above about reasonable ways to predict the death rate. FWIW, the rantes they are calcualted are between 2.2% and 11% (15 - 85 times as deadly as the seasonal flu)
A further FWIW - I was 3.5 days off on my January 29 prediction of when the deaths from this virus would exceed the deaths from SARS (I predicted it would be 3.5 days earlier than that threshold was crossed). As unpredictable as this is with the relativley small numbers on 1/29, that's pretty darn close.
PoindexterOglethorpe
(25,746 posts)Ms. Toad
(33,915 posts)While it would be nice to have absolutely complete data, we never have that about any illness.
And - since the death rate for the flu was determined without being able to count those who didn't seek medical help, it is reasonable to compare that rate to the death rate for COVID-19 (which similarly undercounts the people who didn't seek treatment).
We're comparing apples to apples, so the comparison is valid, even though there is not complete information to be able to provide absolute numbers for either one.
defacto7
(13,485 posts)The outcome is unknowable. Projections are mostly futile.
ecstatic
(32,566 posts)Especially concerned due to trips coming up next week!
defacto7
(13,485 posts)The experts are inputing info every day. My personal opinion is that there is no reason for alarm here and now. Staying informed of where the outbreaks are is important. I wouldn't go to China at this point but most places, there's nothing to worry about.
No matter what the situation, washing your hand often is always a good idea when you're travaling. Enjoy your trip.