cigar wrote: ↑Sun Mar 29, 2020 4:47 pm
Folks, here are the facts in the US:
1. About 15% of the tested are positive. To be tested you must be symptomatic (or have a lot of money and/or power) Note that the test has a roughly 30% false negative rate (you're positive, it misses it) so none of these cases wind up being considered for admission as Coronavirus cases. These presumably all recover without admission, plus all those who are true negatives.
2. Of that 15% which test positive 12% (CDC numbers) meet criteria to be hospitialized. Note that of these exactly zero require an actual hospital bed at that stage; oral medication and monitoring every few hours can be done anywhere, by anyone.
3. Of the 12% about one quarter wind up requiring an ICU. Now we're truly in the "hospital required" category. Invasive procedures (vent, whether positive pressure mask or intubation) or critical monitoring (e.g. heartbeat, SpO2 on an "always-on" rather than spot-check basis, etc) This is also where artificial feeding and hydration come into it because you literally have no choice; with a positive pressure vent required to breathe or a tube down your throat you can neither eat or drink so a tube for that goes SOMEWHERE.
4. Of those who went on vents in Wuhan 95% did not survive. That is, the "save" rate if you got there was five percent. We do not know what it is here, but the "base" save rate for a vent in a US hospital from all causes is right at 50% to discharge and 30% one-year survival (vents invariably produce fairly serious damage to the victim and you frequently die not long after that.) It is laughable that our hospitals can approach their "stock" 50% save rate - even if we're twice as good as the Chinese, which isn't implausible, our save rate is only 10%.
So of those who meet the criteria to be tested in the US today (by definition if you wind up in the hospital with KungFlu you're going to get tested) 0.45% of those people wind up in intensive care (0.15 * 0.12 * 0.25) = 0.45%.
Except.... that's not the real rate. Note that the test misses 30% of positives. So those people are positive, but don't end up in the numbers (they go home and recover.)
So among the positives that wind up in the ICU are (0.12 * 0.7) * 0.24 = 2.0%.
Current estimates on the worst-case end that nine out of ten people do not develop symptoms sufficient to meet current testing criteria. Therefore, of those who contract Coronavirus the percentage of those who wind up in an ICU is 0.2%, approximately.
By the way -- this clowns the "2% fatality rate" figure instantly. Fatality is some fraction of the 0.2%, in other words. But no, the media (obviously) cannot do multiplication -- we're talking grade school math here folks.
Incidentally that "9 out of 10" is probably low -- perhaps by a factor of 10 or even more There is reason to believe, specifically, the Oxford paper, that the true rate positives compared with those who are sick enough to get a test is 100:1, not 10:1. That's plausible given what we already know; among a very old population on the Diamond Princess they still only had 306 symptomatics out of 3,711 persons on board. It's implausible the others on board were not exposed, which means either (1) they previously had the virus -- possibly on the cruise -- and recovered asymptomatically, thus testing negative or (2) R0 is wildly wrong, particularly considering that some of the "pairs" include couples in the same cabin where only one of them got sick and the other tested negative.
That would make the fatality rate 0.02%.
Everyone who does not require ICU care requires no hospital at all; we can provide drugs, we can provide at-home monitoring of SpO2, we can provide at-home monitoring of BP and heart rate and temperature and we can provide at-home supplemental oxygen. All of these are trivial to do and cheap. Such data can be transmitted in or care-givers can be given thresholds where transport is actually required.
Keeping people out of the hospital when no actual intensive, "always on" care is required improves outcomes; you immediately remove all the instances of hospital-acquired infections, for one, and if you're already fighting a virus the last thing you need is an infection you acquire in the hospital on top of it. If said person lives alone with nobody willing and able to help (e.g. nobody seropositive and thus immune) then we may well need someone for them to be, but it doesn't have to be what you think of as a "hospital."
Now let's face facts: If you wind up on a vent in the ICU with this virus you are almost-certain to die. That probability is almost-certainly less than or equal to 95% but materially greater than 50%. The hospitals aren't publishing those numbers (gee, I wonder why?) but you can bet the odds suck. Let's call it 90% failure.
Ok, so now of the cases, assuming we have 100 who get infected but do not qualify for a test for everyone who does we can change the outcome of 0.002% of the cases by not overloading the medical system.
Because the outcome is so bad if you wind up on a vent we must be extremely aggressive in trying to prevent that. We have evidence, but not scientific proof, that a number of said therapies (e.g. hydroxychloroquine, etc) are effective. Exactly how beneficial is not known but avoiding the ICU, given the stats, is the key to not being dead. We have to use what we've got; any improvement is better than none, and while I'm sure there is a range of effectiveness for what is being tried until something proves better you use what you got.
Let's say that out of the 330 million people in the US 70% will eventually get the infection. I have no reason to doubt this. The idea that we can actually repress math is laughable. The idea that we're going to get a vaccine with persistent immunity is contrary to every piece of science on coronaviruses to date, both in humans and animals. The data in fact does not support any other conclusion; either this thing is not nearly as transmissible as we think it is (unlikely), the percentage of people who have or have gotten "silent" infections is outrageously high or the range of people who cannot get it for some other reason (e.g. cross or natural immunity) or some combination of those elements has to be true. Diamond Princess along with South Korea, Japan and Italy all demonstrate this conclusively. Anyone arguing otherwise is arguing against all of the existing data.
So we have 230,100,000 people who get this thing (immunity level is reached by whatever means) before it's over in the United States.
230,100,000 * 0.002% = 4,602.
Of those we can change the outcome from dead to not-dead by not overloading the hospitals in 4,602 instances. Maybe.
We just took a 30% stock market crash and destroyed the jobs of over 3 million people in the last week over what looks like 4,602 possible lives saved.
Now maybe that estimate is very low. But even on the higher end of reason for silent or minimally-symptomatic attack; that is, 1 in 10 people get ill enough to meet testing criteria rather than 1 in 100 we're talking less than a moderate flu season's excess mortality.
For reference approximately 8,000 people die every single day in the United States.
We took an intentional depression in what appears to be an attempt to save less than a day's mortality. Will suicides exceed this when you throw five to ten million, maybe even twenty million people out of work -- three million last week alone?
In other words for those who say "any sacrifice is worth it for even one life" you're the ******* because you're killing more people than you're saving.
Now we can easily figure social distancing is close to worthless.
NY's data makes this quite clear. So does Florida's.
Both slammed the door; SE Florida and NYC.
The bend should be evident in one viral generation time. The new case rate should collapse in two viral generation times. If Community Transmission via bars, restaurants and "social interaction" was more than 2/3rds of the total the effective R0 would go under 1.0 and community transmission would collapse. If it was half then R0 would be 1.5 and we'd have transmission approximately equal to a bad seasonal flu.
IF you actually bent the curve.
These measures did not bend it to any material degree. Enough time has passed to know this is true; at most they have lengthened a "turn time" by one day (in other words, R3.0 to R2.5.) That's effectively nothing!
Why not?
It's being spread in the medical environment -- specifically, in the hospitals -- not, in the main, on the beach or in the bar.
When Singapore and South Korea figured out that if as a medical provider you wash your damn hands before and after, without exception, every potential contact with an infected person or surface even if you didn't have a mask on for 30 minutes during casual conversations with others (e.g. neither of you is hacking) transmission to and between their medical providers stopped.
Note -- even if you didn't have a mask on and were not social distancing in the work environment, which of course is impossible if you're working with others in a hospital, you didn't get infected.
And guess what immediately happened after that? Their national case rate stabilized and fell.
The hypothesis that fits the facts is that a material part of transmission is actually happening in the hospital with the medical providers spreading it through the community both directly and indirectly.
Remember that all disease R0 is a composite of all the elements of transmission. If any material part of transmission is happening in hospitals and other medical settings stopping that will stop or greatly attenuate community transmission. Every medical provider goes home and interacts with the public.
Then the hospital fills up and guess what -- they call in more doctors, nurses, orderlies and other people. In fact they've done exactly that; in hard-hit places they're getting volunteers. Excellent, they need the workers, except every one of those new workers in the place is also a brand new vector to the rest of the community too unless they wash their damn hands before and after every contact with any item or person as well.
What's worse is that the data is that if you wind up on a vent you die nearly all the time. They had a doc on Tucker Carlson the other night confirming that we are not doing materially better than Wuhan in this regard.
We're wrong about how this thing is spreading and we're wrong about the silent attack rate. The step functions in the data here in the United States cannot be explained by ordinary community transmission but they are completely explained if the transmission is happening not among ordinary casual contact -- that is, not "social distancing", but rather through the medical system itself. That explains the step functions that are seen in places like Florida since it takes several days before you seek medical attention after infection and it also explains why NY, despite locking down the city and more than one viral generation time passing -- in fact two -- has seen no material decrease at all in their transmission rate.
In addition it further is supported by the fact that what we've seen here, in Italy, in Wuhan -- indeed everywhere is not an exponential curve. It's a step-function flat acceleration graph. Broad community transmission doesn't happen this way (you instead get a straight and continual exponential expansion until you start to obtain suppression via herd immunity) but if the spread happens as each "generation" gets driven to hospitals for testing and medical attention and the spread is largely happening there what we see here and in other nations in the case rate data is exactly the function you produce in terms of exposure rates.
In other words there should be no straight-line sections in the case rate graphs -- but there are.
Fix the protocols in the hospitals right damn now. PPE is not the answer if your hands, gloved or not, become contaminated and not immediately washed off. Hand-washing at an obsessive level -- before and after each patient interaction and before and after each contact with a piece of equipment that might be contaminated is. In other words the monster vector (remember, R0 is a composite, not a single number) is not oral droplets -- it's fecal.
This also correlates exactly with the explosive spread in nursing homes where many residents are incontinent.
Folks, by definition medical facilities concentrate sick people into small spaces. If what's wrong with them is not infectious this doesn't matter. But if it is you had better not transmit anything between them or between you and them or you instantly become one of the, if not the only vector that matters.
Then as the place fills up you have more people working and thus more vectors into the rest of the community. Even if you have gotten the virus as a nurse or doctor and recovered and thus are immune if you have it on your hands and go down the escalator to the subway you can still contaminate the railing and the grab-rail in the car unless you wash your damn hands before and after any contact with any thing or person!
The presence of step functions and apparent linear-fit line segments in what should be a clean parabolic curve says this is exactly what has happened.
That in turn explains why the lockdowns are not doing a damn thing -- except destroying the economy, that we must do everything in our power to keep people out of the hospital in the first place and that, in turn, means using even potentially-valid prophylaxis and promising (but not yet proved) treatments early in the course of the disease so as to keep people out of the damned hospital in the first place while fixing the protocols in the hospitals so they stop transmitting the bug.
Don't tell me about all the doctors and nurses doing this already. That's a lie. I've been in plenty of hospitals (and worse, in nursing homes) in my years and in exactly zero instances have I seen any evidence that before and after each contact, with zero exceptions, those hands go under a stream of water with soap.
And reopen the damned economy.
Now.