‹ First  < 31 32 33 34 > 
 
   
 

COVID-19: The existential choice

 
Skipshot
 
Avatar
 
 
Skipshot
Total Posts:  9986
Joined  20-10-2006
 
 
 
22 May 2020 11:25
 
Q - 22 May 2020 09:19 AM

He’s not taking it, he’s hawking it.

Yeah, he’s lying about taking it.  Just shows to go you how little he really cares about the “general welfare.”

 
MARTIN_UK
 
Avatar
 
 
MARTIN_UK
Total Posts:  4998
Joined  19-08-2010
 
 
 
22 May 2020 11:45
 

I grow weary of trying to work out in which ways out betters are trying to pull wool over our eyes at any given time. I’m unsure I would recognise honesty if it was to present itself in front of me these days.

 
Cheshire Cat
 
Avatar
 
 
Cheshire Cat
Total Posts:  1536
Joined  01-11-2014
 
 
 
22 May 2020 11:48
 
unsmoked - 22 May 2020 10:37 AM

Question:  If the Entertainer determines that he is going to lose in November, will he attempt to cancel the election?  Could he do it with Senate approval?  Supreme Court ruling?  (threat of Chinese interference?  Mail-in ballots?  Biden in cahoots with PBS and The New Yorker as part of a deep state plot to overthrow the government?)

Image Attachments
 
Cartoon.jpg
 
 
 
 
Avatar
 
 
Q
Total Posts:  40
Joined  04-05-2020
 
 
 
22 May 2020 15:19
 
MARTIN_UK - 22 May 2020 11:45 AM

I grow weary of trying to work out in which ways out betters are trying to pull wool over our eyes at any given time. I’m unsure I would recognise honesty if it was to present itself in front of me these days.

Two things MARTIN: 1) I learned a long time ago when someone tells you, “Don’t worry, everything is fine” without being able to provide the justification for that view, that’s when it’s time to start worrying. 2) Since we live on Planet Bizarro, surely, anything Trump says is a lie, so the truth is the opposite. He said he’s taking it, so he’s not. Never mind that he announced initially that he had purchased millions of tabs in a deal with his buddy with which they planned to saturate the state of New York, money and citizen manipulation their rewards.

 
TheAnal_lyticPhilosopher
 
Avatar
 
 
TheAnal_lyticPhilosopher
Total Posts:  1199
Joined  13-02-2017
 
 
 
23 May 2020 06:04
 

The cure and the disease

Trump has claimed we can’t let the cure be worse than the disease, and this has become a favorite talking point among conservatives, especially those in charge of loosening restrictions and “opening up the economy,” as it were.  But the data tells a different story.  Far from the cure closing the economy, and much further from it being worse than the disease affecting it, mobility data in the US—and therefore the correlative economic activity that can be drawn from that—indicates that the disease has not only been far worse than the cure: the cure has in fact led, counter-intuitively, to a lessening of the impact of the disease.  This can be seen by noting mobility among Americans prior to the first Federal recommendations (March 16), further mobility prior to the “lock down” orders (roughly April 1), and mobility again at the time these orders were lifted (~April 30), up to the present.

For instance, prior to President Trump’s March 16 “slow the spread” recommendations, mobility averaged across the US had already declined by 18% from the early month norm (range -8% to -35%, median -17%, SD 5.7%), indicating an impact of COVID-19 on the economy prior to mandated measures to combat it.  By the time the first “stay at home” orders were in place—or in lieu of that, closure of non-essential businesses—average mobility had already dropped an additional 25%, for a total decline of 43% (range -16% to -68%, median -44%, SD 11.9%).  So, absent stay-at-home orders and non-essential business closures, Americans were already staying at home and avoiding businesses and therefore limiting their economic activity because of the threat of COVID-19—and little else.  In fact, official stay-at-home or business closure orders only decreased mobility by another 7%, with all states reaching their mobility low point in the week from April 7 to April 14.  Clearly, then, the disease was having more impact on mobility—and therefore presumably on the economy—than “the cure,” as prior to the cure mobility was already down by 43%, and with the cure it was only down an additional 7%.

What’s even more noteworthy than this early impact of the epidemic prior to mandated attempts to fight it, during the cure mobility universally increased across the states, not just decreased initially; and in virtually all cases it increased dramatically, with only five states (Maryland, Massachusetts, Connecticut, New Jersey, and New York) and DC increasing 6% or less (average increase was 12%, with a median increase of 15%).  So, while the stay-at-home and business closures had a small effect early on—basically the first two weeks of April, at -7% mobility—from that early point on the effect of these orders was to increase mobility, and with that mobility, correlative economic activity.  Indeed, by April 30 the average mobility in the US was only down to -38%, or 12% higher than the April low, and 5% higher than at the beginning of the lockdowns.

And what’s even more noteworthy yet—and what’s even more dispositive for dispelling ’the cure is worse than the disease’ mantra—the average increase in mobility since “opening up” (call this April 30, the day the Federal guidelines expired and most states starting loosing restrictions) has only been 2%.  That’s 2%.  Since opening up average mobility across the US has only risen marginally in the two weeks from the point at which states were closed, or more than three times less than the impact of the cure, indicating again that the “cure” has had less an impact on mobility than one might at first expect.  Far less, in fact, otherwise we’d expect a rise in mobility approximating March 16 levels once the restrictions were loosened, but so far we are seeing nothing close to that.  Indeed, this is probably because the restrictions didn’t cause the initial fall in mobility (and therefore in economic activity); fear did.  Therefore relaxing them shouldn’t be expected to offset it, so long as that fear is present.

Data for the third week during “opening up” is still coming in, but prior to its arrival the picture seems clear: prior to “the cure” for COVID-19, Americans were already so afraid of becoming infected that they were already staying home and avoiding businesses, and now that they are free to start moving about, they are only marginally more comfortable doing so.  In fact, one of the most interesting lessons of “the cure” can be found in the well-known concept of risk homeostasis: once mitigating COVID-19 was in place, people felt safer in going about their daily activities, therefore their daily activities increased.  While this does speak against how effective mitigation might be in the long run, it speaks even more to how necessary it is that people feel safe before resuming activity, and how much safer “the cure” made them feel—a feeling of safety at odds with the feeling of threat they felt prior to mitigation measures being put in place.  The rather large effect of this risk homeostasis and the comparatively miniscule increase in activity since the lockdowns have been relaxed should speak to how necessary the cure is, and how relative a threat it is to the disease it’s meant to fight.  For the data clearly indicates that people are far more afraid of COVID-19 than they are in favor of opening the economy prematurely—a fact also borne out by every poll on the subject so far conducted.

So, let’s set aside this nonsense about the cure being worse than the disease and focus instead on the cure being effective enough for people to feel safe in going about their business.  In short, stop relaxing the lockdowns and don’t let the reckless indifference of some make the situation worse for everyone (and even more so, let’s not sacrifice public safety on the re-election hopes of an idiot).  Instead let the lockdowns do their work in suppressing the epidemic, mitigate the economic impact as we are currently doing, this way people won’t just feel safer as economic activity proper resumes: they will in fact be safer.  For as the data shows, absent that feeling of safety the damage to the economy is a foregone conclusion, the “cure” for the disease notwithstanding.

 
Jefe
 
Avatar
 
 
Jefe
Total Posts:  7242
Joined  15-02-2007
 
 
 
23 May 2020 11:39
 
Q - 22 May 2020 09:19 AM
unsmoked - 19 May 2020 10:03 AM

By the way, do you think Trump is taking hydroxychloroquine in order to get his fans to use up the 29,000,000 doses the U.S. bought from India?

He’s not taking it, he’s hawking it. Here’s an example: 6:20 into the video, UNHINGED MAGA-HEADS BERATE REPORTER, the guy with the black Trump tee-shirt keeps approaching the reporter attempting to violate his space saying, “No, I got hydroxychloroquine, I’m fine.”

https://www.bing.com/videos/search?q=The+Young++Turks&&view;=detail&mid=2BA500CA3B53377E1BD62BA500CA3B53377E1BD6&&FORM;=VRDGAR&ru;=/videos/search?q=The+Young++Turks&qpvt=The+Young++Turks&FORM=VDRE

https://arstechnica.com/science/2020/05/hydroxychloroquine-linked-to-increase-in-covid-19-deaths-heart-risks/#p3

Comparing the treatment groups to controls and adjusting each patient’s risk factors, such as congestive heart failure, the researchers found the following:

COVID-19 patients given hydroxychloroquine alone had a 34-percent increased risk of dying in the hospital and a 137-percent increased risk of developing a serious arrhythmia.
Those given hydroxychloroquine with a macrolide had a 45-percent increased risk of dying in the hospital and a 411-percent increased risk of developing a serious arrhythmia.
Those given chloroquine had a 37-percent increased risk of dying in the hospital and a 256-percent increased risk of developing a serious arrhythmia.
Those given chloroquine and a macrolide had a 37-percent increased risk of dying in the hospital and a 301-percent increased risk of developing a serious arrhythmia.

But profits on the drug may be up.

 
 
Jefe
 
Avatar
 
 
Jefe
Total Posts:  7242
Joined  15-02-2007
 
 
 
23 May 2020 11:45
 
TheAnal_lyticPhilosopher - 23 May 2020 06:04 AM

So, let’s set aside this nonsense about the cure being worse than the disease and focus instead on the cure being effective enough for people to feel safe in going about their business.  In short, stop relaxing the lockdowns and don’t let the reckless indifference of some make the situation worse for everyone (and even more so, let’s not sacrifice public safety on the re-election hopes of an idiot).  Instead let the lockdowns do their work in suppressing the epidemic, mitigate the economic impact as we are currently doing, this way people won’t just feel safer as economic activity proper resumes: they will in fact be safer.  For as the data shows, absent that feeling of safety the damage to the economy is a foregone conclusion, the “cure” for the disease notwithstanding.

Republicans and Conservatives are pathologically incapable of allowing the lower classes ‘less work’, and cringe at any sort of ‘support pay’.  The ‘worse than the cure’ factor is their own manufactured aversion to sensible, economy preserving measures of temporary support for society and systems that they themselves had a hand in designing to be so brittle.

 
 
unsmoked
 
Avatar
 
 
unsmoked
Total Posts:  9140
Joined  20-02-2006
 
 
 
23 May 2020 12:31
 

Trump is encouraging Christians to go back to church.  Last night on the PBS newshour, David Brooks, a conservative NY Times columnist correctly noted that this is the same ploy to capture evangelical votes as his 2017 ‘We’re going to put Christ back in Christmas,” ploy.

https://www.youtube.com/watch?v=4wYTmVrVvsw

Well, guess what!  We’re going back to church to be with God, and to pray in public with a show of piety, just like Jesus told us to do, even though being in church is a very dangerous thing to do right now . . . because we’re free Americans and we don’t have to listen to scientists and their hoax about global warming or to Democrats who want to take your machine guns away from you and make it illegal for good people to wear swastikas and. . . wait a minute!  Who wrote this speech?  Bring him to me!  No, wait.  Just bring me his head! 

quote:  “And when you pray, do not be like the hypocrites, for they love to pray standing in the synagogues and on the street corners to be seen by others. Truly I tell you, they have received their reward in full.”  -  Jesus

ploy  noun

a cunning plan or action designed to turn a situation to one’s own advantage.

[ Edited: 23 May 2020 12:35 by unsmoked]
 
 
Skipshot
 
Avatar
 
 
Skipshot
Total Posts:  9986
Joined  20-10-2006
 
 
 
23 May 2020 14:16
 

Thanks for the detailed analysis, Anal.  The cure was to get everyone on board with slowing the spread of the virus, not eliminating it, and so far the cure is working.  How easily this virus spreads is evident by its spread with strong efforts to slow it, so it is wise to take it seriously.

Of course, with a great majority of people following health department guidelines which slow the spread, this gives cover to the crackpots to feel safer in saying the disease is overhyped because hardly anyone is getting it, which is exactly the point of the guidelines.  Giving the crackpots their way and opening up society increases the risk of spread and death, but the crackpots are usually faceless nobody’s with no accountability, so if they are wrong then they can pretend they were not calling for opening society, and the public crackpots will simply deflect blame and stick to their guns, because contrarianism is their religion.

 
Cheshire Cat
 
Avatar
 
 
Cheshire Cat
Total Posts:  1536
Joined  01-11-2014
 
 
 
27 May 2020 10:18
 

This is a rather shocking statistic:

”... as of May 22, in the 39 states that currently report such figures, an astounding 43% of all COVID-19 deaths have taken place in nursing homes and assisted living facilities.”

There is one silver lining—or perhaps bronze lining—to the COVID long-term care tragedy. The fact that nearly half of all COVID-19 deaths have occurred in long-term care facilities means that the 99.4 percent of the country that doesn’t live in those places is roughly half as likely to die of the disease.

https://tinyurl.com/yba77yaz

 
 
unsmoked
 
Avatar
 
 
unsmoked
Total Posts:  9140
Joined  20-02-2006
 
 
 
27 May 2020 11:11
 
Cheshire Cat - 27 May 2020 10:18 AM

This is a rather shocking statistic:

”... as of May 22, in the 39 states that currently report such figures, an astounding 43% of all COVID-19 deaths have taken place in nursing homes and assisted living facilities.”

There is one silver lining—or perhaps bronze lining—to the COVID long-term care tragedy. The fact that nearly half of all COVID-19 deaths have occurred in long-term care facilities means that the 99.4 percent of the country that doesn’t live in those places is roughly half as likely to die of the disease.

https://tinyurl.com/yba77yaz

Speaking of assisted living, there’s the 60 million at the other end of the age spectrum.  https://www.art.com/products/p15063694438-sa-i6855931/pat-byrnes-i-d-rather-be-working-new-yorker-cartoon.htm

 
 
TheAnal_lyticPhilosopher
 
Avatar
 
 
TheAnal_lyticPhilosopher
Total Posts:  1199
Joined  13-02-2017
 
 
 
28 May 2020 07:10
 

Our current course

At the time states began re-opening—or if not yet opened, then currently—35 states had the epidemic under control, while 16 did not (including D.C. as a state).  Since re-opening, 12 states have lost control of the epidemic, so 28 states now have an out of control epidemic, while 23 do not.  These estimates come from the latest research on state-by-state transmission, found here and here.

It should be noted the estimates are derived from current mobility and case data, not projected courses of the epidemic.

For those states with the epidemic under control, the average effective reproduction number is 0.78.  For those where it is out of control, the average reproduction number is 1.12.  So, where it is under control, cases are declining pretty quickly, and where it is out of control, they are rising slowly, with infections on average doubling every month to month and a half. 

This doubling does not account for a rising reproduction number, which is the current trend in most states.  If the rising reproduction number trend continues, doubling time will decrease—dramatically (a difference of 1.1 to 1.2 means a doubling time reduction of 50%)

Of the states discussed here and here (Colorado, Georgia, Minnesota, Mississippi, Oklahoma, South Carolina, Tennessee, and Texas), all but one (Oklahoma) either had the epidemic under control at the time of opening, then lost control (Colorado, Minnesota, Tennessee), or it did not have the epidemic under control at the time of opening, and the epidemic got worse (Georgia, Mississippi, South Carolina, and Texas).  Oklahoma had the epidemic under control at opening, and still does (though marginally, with RT rising from 0.81 to 0.98). 

On average “opening up” increased the effective reproduction number in these early-opening states by 19%, with Tennessee and Texas experiencing the largest increases (29% and 27% respectively).  This translates to a reproduction number slightly higher than the national average (1.16 versus 1.12), with Texas at the highest (1.29).

The average doubling time of infections for these seven out-of-control states ranges from approximately 15 to 30 days—so again, worse than the national average.

It also bears noting that since opening mobility in these states has increased well above the national average of 2% over April 30—on average a 6% increase—and this increase comes on top of an average 12% increase in mobility from the time of their initial closures.  Taken together these two increases indicate that not just “opening” per se predicts disease transmission, but rather increased mobility—a causality further supported by the fact that in 19 states the effective reproduction number was already below 1 when the stay-at-home or business closures officially went into effect (as noted here, by then mobility on average had declined by 43% across states).  Also, it bears noting that at this same time an additional 13 states had effective reproduction numbers below the current average of the above seven states, meaning these seven states now are worse off on May 26 than were two-thirds of states at the beginning of the lockdowns, roughly March 24-April 1.

As predictable as these trends were four weeks ago, and as disconcerting as they are now, it is not too late to reverse course.  Although most conservatives consistently misrepresent the role of the lockdowns (either intentionally or out of ignorance—see here and here), their purpose was both to save lives and suppress the epidemic, such that containment—and with that near-full economic re-opening—would be possible.  In none of the states where the epidemic has either slipped out of control or has not yet been controlled is it too late to reverse that trend and get back to suppression, and thus back to the best chance to save both lives and the economy.  For absent that, on our present course we will lose more of both.

In any case, let’s hope planners don’t need signals much more obvious than the trends here before enacting more sensible policy, because like at the beginning of the pandemic weeks of denial now means months of pain and tens of thousands of deaths down the road (pdf).

[ Edited: 28 May 2020 15:58 by TheAnal_lyticPhilosopher]
 
TheAnal_lyticPhilosopher
 
Avatar
 
 
TheAnal_lyticPhilosopher
Total Posts:  1199
Joined  13-02-2017
 
 
 
28 May 2020 07:16
 

That’s a good article on nursing homes, but for the statistical illiteracy at the end (alas, the quoted text).

If 43% of COVID-19 deaths are in nursing homes, and nursing homes have 1.5 million people and the US has 328 million, at 102,000 deaths the chances of dying in if in a nursing home (all else equal) are 2.9%, while in the general population it’s .018%—so about 160 times less likely to die if in that “99.4%”, not “roughly half as likely.”

And never mind the moral non-sequitur of this “silver lining”…

 
TheAnal_lyticPhilosopher
 
Avatar
 
 
TheAnal_lyticPhilosopher
Total Posts:  1199
Joined  13-02-2017
 
 
 
28 May 2020 07:16
 
Skipshot - 23 May 2020 02:16 PM

Thanks for the detailed analysis, Anal.  The cure was to get everyone on board with slowing the spread of the virus, not eliminating it, and so far the cure is working.  How easily this virus spreads is evident by its spread with strong efforts to slow it, so it is wise to take it seriously.

Of course, with a great majority of people following health department guidelines which slow the spread, this gives cover to the crackpots to feel safer in saying the disease is overhyped because hardly anyone is getting it, which is exactly the point of the guidelines.  Giving the crackpots their way and opening up society increases the risk of spread and death, but the crackpots are usually faceless nobody’s with no accountability, so if they are wrong then they can pretend they were not calling for opening society, and the public crackpots will simply deflect blame and stick to their guns, because contrarianism is their religion.

!

 
TheAnal_lyticPhilosopher
 
Avatar
 
 
TheAnal_lyticPhilosopher
Total Posts:  1199
Joined  13-02-2017
 
 
 
28 May 2020 14:52
 

A guy on our COVID response team brought up this JAMA letter today and asked for the group’s opinion.  He tends to lean conservative (the only one), but he’s been swayed considerably by the rest of us non-political types. 

The study uses an antibody test to estimate the percent of the population infected in Los Angeles County by April 10.  It is the sister study to the one discussed here.

Below is a modified version of what I suggested to the group. It’s directly related to a discussion in this thread (this article was even obliquely anticipated in that discussion), and that discussion mirrors a tendency in the national conversation—to its detriment, I might add. 

Anyway….

1. The article not only fails to account for the positive predictive value of its antibody test; based on the stated sensitivity and specificity it adjusts the infection prevalence estimate in the wrong direction.

Specifically, with a sensitivity of 82.7% and a specificity of 99.5%, given a reasonable estimate of the population prevalence on April 10 (0.8%), the positive predictive value of the test is only 57%, meaning only 20 of the 35 positive results are likely true positives.  Not only do the authors fail to account for this limit of any test, they adjust their calculated prevalence estimate upwards for sensitivity and specificity, when the proper adjustment should be downwards.  Calculated correctly the percent of the population that’s been infected based on this testing is 2.3% (from 4.06% in the sample), not 4.65%.

For a CDC warning about positive predictive value and antibody tests, see the guidelines here.

2. The infection fatality rate for this inflated result of 4.65% is nonsense in light of what we now know about COVID-19.  Not what we estimate, but what we know.

First, its implied infection fatality rate of 0.2, if generalizable, would mean the entire population of New York City has already been infected (8.4 million *0.002 = 16,800 deaths—and currently 16,400 have died).  We know this is not the case, and serology estimates indicate it’s more like 20% infected.  So an infection fatality rate of 0.2 can’t be right—or if it is, it’s due to extraordinary circumstances that don’t generally obtain.

Second, the authors indicate selection bias is likely a limitation of the study because symptomatic persons might be more likely to participate.  They are right.  Perhaps 30% of their sample (240 participants of 863 test results) could be symptomatic people seeking a test, a far cry from what one would expect in random sampling, where the prevalence of reported symptoms would be less than the study estimate—using their inflated number—of 4.65% infected (not all infected are symptomatic).  Even after adjusting for non-COVID caused symptoms (like flu and pneumonia), 30% symptomatic individuals indicates strong self-selection bias.

Third, calculated correctly the IFR for the study should be 0.4, which could indicate that either Los Angeles is demographically young, or it is protecting its elderly far better than other localities (or both).  But more likely it indicates the selection bias the authors point to as a likely study confound.  For instance, the IFR for New York City based on both a better sampled serology study and other estimates is around 0.9 to 1.0%.  Why expect LA, a similar metropolitan area, to be so different than New York City?  It’s possible, of course, but at less than half the estimated value it’s highly suspicious.  The better explanation is selection bias biasing the estimate downward.

3.  This letter shouldn’t have been published, as it contains statistical errors that should have been caught by peer review.  In any case, on top of that, these errors add legitimacy to a politicized debate that already shouldn’t be political in the first place, as it’s based on wishful thinking, not careful research using our best scientific approximations.  In fact, given how basic these errors are, it’s hard not to read this study as an attempt at providing “scientific cover” for this politicization.  That one of the authors to its sister study related the same findings in an Op Ed for The Wall Street Journal without disclosing his authorship only adds to this impression. 

But, the damage is done.  Now this “error” is out there, and the waters are even muddier than they need to be in estimation that is already uncertain.

 
‹ First  < 31 32 33 34 >