Author Topic: How long can we wait while flattening the curve?  (Read 248245 times)

beltim

  • Magnum Stache
  • ******
  • Posts: 2964
Re: How long can we wait while flattening the curve?
« Reply #3050 on: July 12, 2020, 05:27:23 AM »
Not uncontested research. 

The arithmetic doesn't change much either way. If there are 3 cases among 165,000 people, you're unlikely to be a person spreading it, and unlikely to meet a person spreading it. If there are 10,000 among those 165,000, you're likely to be and to meet one. Somewhere in between is where it gets foggy.

You know how you go from 3 to 10,000 cases?  By not wearing masks.

Waiting to take measures we know are effective results in more people getting sick and dying.

Kyle Schuant

  • Handlebar Stache
  • *****
  • Posts: 1297
  • Location: Melbourne, Australia
Re: How long can we wait while flattening the curve?
« Reply #3051 on: July 12, 2020, 08:14:37 AM »
I will bet you $100 that my council area of 165,000 people will not get 10,000 cases - shall we say by the end of the year? I'm posting under my real name, so it's easy to hold me to the bet - whereas you as the typical anonymous drongo online, it's a lot harder. So you've got good odds here.

RetiredAt63

  • Senior Mustachian
  • ********
  • Posts: 13915
  • Location: Eastern Ontario, Canada
Re: How long can we wait while flattening the curve?
« Reply #3052 on: July 12, 2020, 08:31:26 AM »
I will bet you $100 that my council area of 165,000 people will not get 10,000 cases - shall we say by the end of the year? I'm posting under my real name, so it's easy to hold me to the bet - whereas you as the typical anonymous drongo online, it's a lot harder. So you've got good odds here.

You are in Australia, where governments are being sensible.  So that is a no go.  Now, if you were in one of the US states that are being stupid . . . . .

beltim

  • Magnum Stache
  • ******
  • Posts: 2964
Re: How long can we wait while flattening the curve?
« Reply #3053 on: July 12, 2020, 09:41:04 AM »
I will bet you $100 that my council area of 165,000 people will not get 10,000 cases - shall we say by the end of the year? I'm posting under my real name, so it's easy to hold me to the bet - whereas you as the typical anonymous drongo online, it's a lot harder. So you've got good odds here.

Violation of forum rules aside, the bet doesnít make sense because Melbourne will act rationally to prevent the spread of Covid. If you find me some similarly infected place somewhere that is discouraging mask use, I will take you up on that bet.

mathlete

  • Handlebar Stache
  • *****
  • Posts: 1638
Re: How long can we wait while flattening the curve?
« Reply #3054 on: July 12, 2020, 09:52:25 AM »
I'd like another week of data that isn't polluted by a holiday weekend in order to confirm, but it really looks likes US deaths are on the rise again, after falling/leveling out for about 2.5 months.

After a few more days, the trend is clearer. Deaths are going up in the US. Itís early on Sunday but itís already clear that today is on pace for the worst Sunday in a month. The uptick in deaths has lagged the uptick in cases by about two or three weeks, and cases growth hasnít slowed down or reversed. Weíre probably looking at a month of growing daily deaths. Or if Iím being optimistic, sustained higher deaths. Itís getting nasty again. Stay safe.

Kyle Schuant

  • Handlebar Stache
  • *****
  • Posts: 1297
  • Location: Melbourne, Australia
Re: How long can we wait while flattening the curve?
« Reply #3055 on: July 12, 2020, 06:33:51 PM »
You are in Australia, where governments are being sensible.  So that is a no go.  Now, if you were in one of the US states that are being stupid . . . . .
You didn't say that, you said: "You know how you go from 3 to 10,000 cases?  By not wearing masks."

So your assertion was that regardless of what government does, if we do not wear masks we'll go from 3 to 10,000 cases. Which is abject nonsense. Almost as much nonsense as our government, which has -

- closed retail which is out in the open with constant fresh air, while keeping open retail inside shopping centres with recycled air, as you can read here,  and
- insisted that we face having a healthcare system overwhelmed and so need to lock everyone down again, while at the same time keeping medical centres closed and cancelling contracts for ICU gear, which you can read about here.

Our country does not face the same problems as yours. This is partly because of the irresolute inadequacies of your government, but also because of your culture. Your country has twice the lethal car accident rate, and workplace deaths rate, and many multiples the accidental firearms deaths rate. America is quite simply a more careless culture than comparable Western countries.

Here people who have tested positive or who are awaiting tests mostly isolate themselves. I would expect fewer to do so in the US, not only from carelessness but also from necessity, as they lack a social safety net, so those minimum wage workers in customer service roles have to go to work, or be evicted or not eat.

In America, I would wear a mask, and probably a kevlar vest, too. Here I don't have to.

Culture matters. Australian culture has the flaw of an instinctive turn to authoritarianism. It comes from our convict heritage, the government can't help but think of themselves as prison wardens. American culture is careless and fails to look after its most vulnerable. Here the greatest danger to our health is the government, in your country it's your fellow citizens being idiots.

Abe

  • Handlebar Stache
  • *****
  • Posts: 1648
Re: How long can we wait while flattening the curve?
« Reply #3056 on: July 12, 2020, 10:41:49 PM »
Latest update on cases and deaths per state:

Some states are seeing a clear uptick in deaths per day:

Arizona, California, Florida, South Carolina, Tennessee, Texas. Overall deaths per day remain low in each state (California and Texas have the most at 100 & 80 yesterday - also the two most populous states).

Others have not had major loss of life (either no increase in deaths or less than 10 per day).

Overall the fatality rate remains much lower than the case counts, which is encouraging. However, several states (Texas, Florida, Arizona) are at or near max ICU capacity, and are starting to show the effects of this breakdown.  I am concerned we will not be off this first wave before the fall one (COVID + Influenza) hits. As I've told several colleagues, this fall we're all COVIDologists.

RetiredAt63

  • Senior Mustachian
  • ********
  • Posts: 13915
  • Location: Eastern Ontario, Canada
Re: How long can we wait while flattening the curve?
« Reply #3057 on: July 13, 2020, 04:46:52 AM »
You are in Australia, where governments are being sensible.  So that is a no go.  Now, if you were in one of the US states that are being stupid . . . . .
You didn't say that, you said: "You know how you go from 3 to 10,000 cases?  By not wearing masks."

So your assertion was that regardless of what government does, if we do not wear masks we'll go from 3 to 10,000 cases. Which is abject nonsense. Almost as much nonsense as our government, which has -

- closed retail which is out in the open with constant fresh air, while keeping open retail inside shopping centres with recycled air, as you can read here,  and
- insisted that we face having a healthcare system overwhelmed and so need to lock everyone down again, while at the same time keeping medical centres closed and cancelling contracts for ICU gear, which you can read about here.

Our country does not face the same problems as yours. This is partly because of the irresolute inadequacies of your government, but also because of your culture. Your country has twice the lethal car accident rate, and workplace deaths rate, and many multiples the accidental firearms deaths rate. America is quite simply a more careless culture than comparable Western countries.

Here people who have tested positive or who are awaiting tests mostly isolate themselves. I would expect fewer to do so in the US, not only from carelessness but also from necessity, as they lack a social safety net, so those minimum wage workers in customer service roles have to go to work, or be evicted or not eat.

In America, I would wear a mask, and probably a kevlar vest, too. Here I don't have to.

Culture matters. Australian culture has the flaw of an instinctive turn to authoritarianism. It comes from our convict heritage, the government can't help but think of themselves as prison wardens. American culture is careless and fails to look after its most vulnerable. Here the greatest danger to our health is the government, in your country it's your fellow citizens being idiots.

Hey Kyle, be careful who you quote, eh?  You are replying to beltim, not me.

Canada and my part, Ontario, may not be doing as well as Australia but we are doing a hell of a lot better than our southern neighbour.  Part of Australia's advantage is being an island continent.

ROF Expat

  • Bristles
  • ***
  • Posts: 289
Re: How long can we wait while flattening the curve?
« Reply #3058 on: July 13, 2020, 06:45:35 AM »


Our country does not face the same problems as yours. This is partly because of the irresolute inadequacies of your government

Kyle,

I object most strenuously to your use of the term "irresolute inadequacies."  I think my government, from the beginning and from the top, has been unwaveringly resolute in its inadequacies.   

But I do thank you for teaching me a new use for the word "drongo" which was previously just a bird for me. 

HBFIRE

  • Handlebar Stache
  • *****
  • Posts: 1249
  • Age: 42
  • Location: Huntington Beach, CA
Re: How long can we wait while flattening the curve?
« Reply #3059 on: July 13, 2020, 12:29:49 PM »
Doctors in Northern California say they have seen more deaths from suicide than theyíve seen from the coronavirus during the pandemic.


ďThe numbers are unprecedented,Ē Dr. Michael deBoisblanc of John Muir Medical Center in Walnut Creek, California, told ABC 7 News about the increase of deaths by suicide, adding that heís seen a ďyearís worth of suicidesĒ in the last four weeks alone.

DeBoisblanc said he believes itís time for California officials to end the stay-at-home order and let people back out into their communities.

"Personally, I think it's time," he said. "I think, originally, this was put in place to flatten the curve and to make sure hospitals have the resources to take care of COVID patients. We have the current resources to do that, and our other community health is suffering."

By late March, more people had died in just one Tennessee county from suicide than had died in the entire state directly from the virus.

MudPuppy

  • Pencil Stache
  • ****
  • Posts: 561
Re: How long can we wait while flattening the curve?
« Reply #3060 on: July 13, 2020, 12:51:20 PM »
Iím going to stop you right there on the Tennessee example. We didnít have any cases at all
In Tennessee until early March. Our first death wasnít until March 20th, 9 days before this article was published. We werenít issued safer at home orders as a state until March 31st. Notice I said safer at home order not a mandated shelter in place. That article is just cherry picking useless data to make itís point.

frugalnacho

  • Magnum Stache
  • ******
  • Posts: 4217
  • Age: 37
  • Location: Madison Heights, Michigan
Re: How long can we wait while flattening the curve?
« Reply #3061 on: July 13, 2020, 01:18:37 PM »
That article is two months old.  Disingenuous at best.  No actual numbers provided either.  I am skeptical that their claims were even true when the article was published, and I definitely don't believe it now.

I just looked up the stats for suicide vs coronavirus deaths in california for 2019, and the ratio is infinite.  The same is true for the entire USA.  Checkmate scaremongers.

mathlete

  • Handlebar Stache
  • *****
  • Posts: 1638
Re: How long can we wait while flattening the curve?
« Reply #3062 on: July 13, 2020, 04:02:19 PM »
https://www.buzzfeednews.com/article/salvadorhernandez/doctor-california-coronavirus-suicide-lockdown

Good article examining the claim (which was wrong and not backed by data) and how it caught fire.

Anyway, the comparison shouldn't be suicide deaths vs. COVID deaths. It should be suicide deaths vs. COVID deaths that would have occurred without the stay at home mandate. It's frustrating when we have to go down the rabbit whole debunking something that isn't even a slam dunk against cautionary measures anyway.

HBFIRE

  • Handlebar Stache
  • *****
  • Posts: 1249
  • Age: 42
  • Location: Huntington Beach, CA
Re: How long can we wait while flattening the curve?
« Reply #3063 on: July 13, 2020, 04:44:59 PM »
Unfortunately, the long term suicide toll will take many years to calculate.  There is also the factor of age here.  This might sound bad, but imo age matters when it comes to the value of a life.  Many of those being hit hardest by this event are younger in their careers (younger than my generation).  The poor, young, and marginalized are being hit the hardest.  For me, I don't mind if we lock down for 10 years, as I can withstand it financially and as an introvert who lives on the beach to stay sane.  But I'm not in the demographic suffering.  It's mostly those who are privileged and able to withstand this economically that are in support of lock downs.
« Last Edit: July 13, 2020, 05:08:23 PM by HBFIRE »

MudPuppy

  • Pencil Stache
  • ****
  • Posts: 561
Re: How long can we wait while flattening the curve?
« Reply #3064 on: July 13, 2020, 04:52:42 PM »
Iím not going down the old folks are less valuable road with anyone.

HBFIRE

  • Handlebar Stache
  • *****
  • Posts: 1249
  • Age: 42
  • Location: Huntington Beach, CA
Re: How long can we wait while flattening the curve?
« Reply #3065 on: July 13, 2020, 04:54:26 PM »
Iím not going down the old folks are less valuable road with anyone.

My opinion is that someone who has not had the benefit of a full life yet is both ethically of higher importance to protect and carries more economic value on average.  This is just my opinion.

MudPuppy

  • Pencil Stache
  • ****
  • Posts: 561
Re: How long can we wait while flattening the curve?
« Reply #3066 on: July 13, 2020, 04:56:52 PM »
Iím not going down the old folks are less valuable road with anyone.

HBFIRE

  • Handlebar Stache
  • *****
  • Posts: 1249
  • Age: 42
  • Location: Huntington Beach, CA
Re: How long can we wait while flattening the curve?
« Reply #3067 on: July 13, 2020, 04:59:58 PM »

Iím not going down the old folks are less valuable road with anyone.


Not expecting a response, was just clarifying my statement.

deborah

  • Senior Mustachian
  • ********
  • Posts: 10814
  • Location: Australia or another awesome area
Re: How long can we wait while flattening the curve?
« Reply #3068 on: July 13, 2020, 05:38:10 PM »
More younger people than older people are getting covid19. Sure, more older people are dying, but a lot of the people who donít are suffering severe complications. Possibly lifelong complications. It is important that we protect everyone from such a nasty disease.

HBFIRE

  • Handlebar Stache
  • *****
  • Posts: 1249
  • Age: 42
  • Location: Huntington Beach, CA
Re: How long can we wait while flattening the curve?
« Reply #3069 on: July 13, 2020, 06:28:58 PM »
This is interesting.

Once the US death data reported by the states is adjusted to put the dates into the weeks the deaths actually occurred, the trends look much differrent than something like worldometer. The states report the deaths as if they all happened this past week. The CDC includes the actual date of these deaths in their data. Per the CDC's data, more than half of the 6,617 deaths reported during the week ending July 11th actually occurred the week ending May 2 or earlier.

https://twitter.com/kylamb8/status/1281778023935737858

July 11: 2,340
July 4: 2,868
June 27: 3,064
June 20: 5,226
June 13: 5,665
June 6: 6,809
May 30: 7,769
May 23: 10,960
May 16: 11,545
May 9: 9,387
May 2: 12,590
« Last Edit: July 13, 2020, 06:31:20 PM by HBFIRE »

Gremlin

  • Bristles
  • ***
  • Posts: 314
Re: How long can we wait while flattening the curve?
« Reply #3070 on: July 13, 2020, 07:19:30 PM »
This is interesting.

Once the US death data reported by the states is adjusted to put the dates into the weeks the deaths actually occurred, the trends look much differrent than something like worldometer. The states report the deaths as if they all happened this past week. The CDC includes the actual date of these deaths in their data. Per the CDC's data, more than half of the 6,617 deaths reported during the week ending July 11th actually occurred the week ending May 2 or earlier.

https://twitter.com/kylamb8/status/1281778023935737858

July 11: 2,340
July 4: 2,868
June 27: 3,064
June 20: 5,226
June 13: 5,665
June 6: 6,809
May 30: 7,769
May 23: 10,960
May 16: 11,545
May 9: 9,387
May 2: 12,590
This looks suspiciously like a concept that actuaries refer to as a development triangle.  I'm surprised it takes as long as it has to report this sort of volume of deaths, but it is what it is.  This suggests to me that the lag between reporting of case numbers and the reporting of deaths is actually going to be significantly LONGER in the US than in other Western nations.  Since if half the COVID deaths reported last week actually occurred 10 weeks or more ago and that pattern is relatively constant, then it stands to reason that half the COVID deaths occurring today won't be reported until 10 or more weeks have elapsed from now.  That implies a very, very ugly scenario is playing out as we speak.

marty998

  • Walrus Stache
  • *******
  • Posts: 7039
  • Location: Sydney, Oz
Re: How long can we wait while flattening the curve?
« Reply #3071 on: July 13, 2020, 07:24:51 PM »
Great. Can we shut this thread down for 10 weeks to avoid all the nonsense claims in the meantime?

lost_in_the_endless_aisle

  • Handlebar Stache
  • *****
  • Posts: 1231
Re: How long can we wait while flattening the curve?
« Reply #3072 on: July 13, 2020, 07:30:07 PM »
This is interesting.

Once the US death data reported by the states is adjusted to put the dates into the weeks the deaths actually occurred, the trends look much differrent than something like worldometer. The states report the deaths as if they all happened this past week. The CDC includes the actual date of these deaths in their data. Per the CDC's data, more than half of the 6,617 deaths reported during the week ending July 11th actually occurred the week ending May 2 or earlier.

https://twitter.com/kylamb8/status/1281778023935737858

July 11: 2,340
July 4: 2,868
June 27: 3,064
June 20: 5,226
June 13: 5,665
June 6: 6,809
May 30: 7,769
May 23: 10,960
May 16: 11,545
May 9: 9,387
May 2: 12,590
This looks suspiciously like a concept that actuaries refer to as a development triangle.  I'm surprised it takes as long as it has to report this sort of volume of deaths, but it is what it is.  This suggests to me that the lag between reporting of case numbers and the reporting of deaths is actually going to be significantly LONGER in the US than in other Western nations.  Since if half the COVID deaths reported last week actually occurred 10 weeks or more ago and that pattern is relatively constant, then it stands to reason that half the COVID deaths occurring today won't be reported until 10 or more weeks have elapsed from now.  That implies a very, very ugly scenario is playing out as we speak.
According to that data, only the latest week was massively impacted by back-dated deaths from a number of weeks ago. The technique used to adjust them out of the weekly reporting was to instead just take the sum of deaths (based on attributed week of death) over the last four weeks. Why 4 weeks? I don't know but it looks like it won't make a big difference to use 3 or 5 or 6 to the conclusion. For most weeks, more than half of deaths are reported within 3 weeks so we will probably not need to wait 10 weeks to determine what was actually happening last week. The other conclusion is that (taking all of this analysis at face value) it's still too early to conclude that CFR is trending down as drastically as supposed thus far.

Bloop Bloop

  • Handlebar Stache
  • *****
  • Posts: 2140
  • Location: Melbourne, Australia
Re: How long can we wait while flattening the curve?
« Reply #3073 on: July 13, 2020, 07:58:03 PM »
Iím not going down the old folks are less valuable road with anyone.

My opinion is that someone who has not had the benefit of a full life yet is both ethically of higher importance to protect and carries more economic value on average.  This is just my opinion.

I struggle to see how anyone could disagree with this.

Kris

  • Walrus Stache
  • *******
  • Posts: 5664
Re: How long can we wait while flattening the curve?
« Reply #3074 on: July 13, 2020, 08:07:18 PM »
Iím not going down the old folks are less valuable road with anyone.

My opinion is that someone who has not had the benefit of a full life yet is both ethically of higher importance to protect and carries more economic value on average.  This is just my opinion.

I struggle to see how anyone could disagree with this.

So, okay, are we saying that thereís an equation of worth of life? So, like, a one-year-old is worth 100%, and a two-year-old is worth 99%, and a three-year-old is worth 98%, and so on...? Is that how we determine life worth?

Because it sounds like yíall are really committed to quantifying it. Is this correct?

If itís not this equation, could you tell me what your math looks like?

MudPuppy

  • Pencil Stache
  • ****
  • Posts: 561
Re: How long can we wait while flattening the curve?
« Reply #3075 on: July 13, 2020, 08:08:32 PM »
I believe the math goes


They wanna > they need ta

Bloop Bloop

  • Handlebar Stache
  • *****
  • Posts: 2140
  • Location: Melbourne, Australia
Re: How long can we wait while flattening the curve?
« Reply #3076 on: July 13, 2020, 08:13:48 PM »
Iím not going down the old folks are less valuable road with anyone.

My opinion is that someone who has not had the benefit of a full life yet is both ethically of higher importance to protect and carries more economic value on average.  This is just my opinion.

I struggle to see how anyone could disagree with this.

So, okay, are we saying that thereís an equation of worth of life? So, like, a one-year-old is worth 100%, and a two-year-old is worth 99%, and a three-year-old is worth 98%, and so on...? Is that how we determine life worth?

Because it sounds like yíall are really committed to quantifying it. Is this correct?

If itís not this equation, could you tell me what your math looks like?

https://en.wikipedia.org/wiki/Quality-adjusted_life_year

"Critics point out that the QALY figures oversimplify how actual patients would assess risks and outcomes. Proponents of the approach acknowledge its imperfections but consider the QALY better than alternative measures."

People who spout that "all lives are equal" and that public policy cannot compare the value of one life to another life obviously have never thought about the fact that we sacrifice myriad lives on the roads every year as a blood gift to the gods of convenience and speedy commutes.

HBFIRE

  • Handlebar Stache
  • *****
  • Posts: 1249
  • Age: 42
  • Location: Huntington Beach, CA
Re: How long can we wait while flattening the curve?
« Reply #3077 on: July 13, 2020, 08:13:57 PM »

So, okay, are we saying that thereís an equation of worth of life? So, like, a one-year-old is worth 100%, and a two-year-old is worth 99%, and a three-year-old is worth 98%, and so on...? Is that how we determine life worth?

Because it sounds like yíall are really committed to quantifying it. Is this correct?

If itís not this equation, could you tell me what your math looks like?

I wont try to quantify it as I think we were speaking in general principle -  yes I think if you must make the tough comparison, a 20 year old with her life ahead of her and hasn't had the benefit of living a full life is more important to protect than say a 90 year old -- from both an ethical and an economic standpoint. This is precisely the type of decision that Italian doctors faced.  I'm sure people will disagree with this opinion.  I'm open to hearing the flaws in this reasoning.  It would make an interesting ethics discussion, but would also derail this thread.
« Last Edit: July 13, 2020, 08:18:18 PM by HBFIRE »

SunnyDays

  • Handlebar Stache
  • *****
  • Posts: 1315
Re: How long can we wait while flattening the curve?
« Reply #3078 on: July 13, 2020, 08:18:00 PM »
One could also make the argument that an older person has made contributions to society and so deserves as much support as possible, whereas a child has done nothing yet so deserves little.  Not saying this is any more logical than valuing the young more.  Either one has merit.  How does a person decide?

HBFIRE

  • Handlebar Stache
  • *****
  • Posts: 1249
  • Age: 42
  • Location: Huntington Beach, CA
Re: How long can we wait while flattening the curve?
« Reply #3079 on: July 13, 2020, 08:20:47 PM »

This looks suspiciously like a concept that actuaries refer to as a development triangle.  I'm surprised it takes as long as it has to report this sort of volume of deaths, but it is what it is.  This suggests to me that the lag between reporting of case numbers and the reporting of deaths is actually going to be significantly LONGER in the US than in other Western nations.  Since if half the COVID deaths reported last week actually occurred 10 weeks or more ago and that pattern is relatively constant, then it stands to reason that half the COVID deaths occurring today won't be reported until 10 or more weeks have elapsed from now.  That implies a very, very ugly scenario is playing out as we speak.

Look more closely at the previous weeks in the data. Last week was an extreme outlier in delayed reporting, not the normal pattern. It appears to simply be an artificial data dump not based on anything about the virus.

Kris

  • Walrus Stache
  • *******
  • Posts: 5664
Re: How long can we wait while flattening the curve?
« Reply #3080 on: July 13, 2020, 08:21:19 PM »

So, okay, are we saying that thereís an equation of worth of life? So, like, a one-year-old is worth 100%, and a two-year-old is worth 99%, and a three-year-old is worth 98%, and so on...? Is that how we determine life worth?

Because it sounds like yíall are really committed to quantifying it. Is this correct?

If itís not this equation, could you tell me what your math looks like?

I wont try to quantify it as I think we were speaking in general principle -  yes I think if you must make the tough comparison, a 20 year old with her life ahead of her and hasn't had the benefit of living a full life is more important to protect than say a 90 year old -- from both an ethical and an economic standpoint. This is precisely the type of decision that Italian doctors faced.  I'm sure people will disagree with this opinion.  I'm open to hearing the flaws in this reasoning.  It would make an interesting ethics discussion, but would also derail this thread.

I am sort of okay with this, but the thing is, Covid isnít the trolley problem. It doesnít give you a 20-year-old on one track and a 90-year-old on the other, and you get to pull the lever.

Honestly, this sort of discussion seems kind of silly to me, especially because itís the kind of discussion that is usually had by people who theoretically think of themselves as low-risk, and therefore donít put themselves on the ďtrackĒ.

AnnaGrowsAMustache

  • Handlebar Stache
  • *****
  • Posts: 1919
  • Location: Noo Zilind
Re: How long can we wait while flattening the curve?
« Reply #3081 on: July 13, 2020, 08:34:52 PM »
This is interesting.

Once the US death data reported by the states is adjusted to put the dates into the weeks the deaths actually occurred, the trends look much differrent than something like worldometer. The states report the deaths as if they all happened this past week. The CDC includes the actual date of these deaths in their data. Per the CDC's data, more than half of the 6,617 deaths reported during the week ending July 11th actually occurred the week ending May 2 or earlier.

https://twitter.com/kylamb8/status/1281778023935737858

July 11: 2,340
July 4: 2,868
June 27: 3,064
June 20: 5,226
June 13: 5,665
June 6: 6,809
May 30: 7,769
May 23: 10,960
May 16: 11,545
May 9: 9,387
May 2: 12,590
This looks suspiciously like a concept that actuaries refer to as a development triangle.  I'm surprised it takes as long as it has to report this sort of volume of deaths, but it is what it is.  This suggests to me that the lag between reporting of case numbers and the reporting of deaths is actually going to be significantly LONGER in the US than in other Western nations.  Since if half the COVID deaths reported last week actually occurred 10 weeks or more ago and that pattern is relatively constant, then it stands to reason that half the COVID deaths occurring today won't be reported until 10 or more weeks have elapsed from now.  That implies a very, very ugly scenario is playing out as we speak.

And one that people seem incapable of seeing. There are going to be some mightily surprised folk in the next month or two. But it's too late to stop the rot now, and it will be two months past two late then. Sigh

Gremlin

  • Bristles
  • ***
  • Posts: 314
Re: How long can we wait while flattening the curve?
« Reply #3082 on: July 13, 2020, 09:03:53 PM »

This looks suspiciously like a concept that actuaries refer to as a development triangle.  I'm surprised it takes as long as it has to report this sort of volume of deaths, but it is what it is.  This suggests to me that the lag between reporting of case numbers and the reporting of deaths is actually going to be significantly LONGER in the US than in other Western nations.  Since if half the COVID deaths reported last week actually occurred 10 weeks or more ago and that pattern is relatively constant, then it stands to reason that half the COVID deaths occurring today won't be reported until 10 or more weeks have elapsed from now.  That implies a very, very ugly scenario is playing out as we speak.

Look more closely at the previous weeks in the data. Last week was an extreme outlier in delayed reporting, not the normal pattern. It appears to simply be an artificial data dump not based on anything about the virus.

...maybe.  But the numbers also don't add to the totals.  According to the figures, there's a further 7,126 deaths to be allocated somewhere in that table.  Top row shows 126,444 deaths, but the triangle only shows 119,318 deaths.  Do we just assume those 7,126 haven't died until they're allocated at some point in the future.  But that's still 10,958 short of the 137,402 reported deaths in total to 11 July.  Again, they have to fit in somewhere. 

So I went and looked at the original CDC source - not the Twitter account.  And this is what they said...

Quote
Provisional death counts are based on death certificate data received and coded by the National Center for Health Statistics ... It is important to note that it can take several weeks for death records to be submitted to National Center for Health Statistics (NCHS), processed, coded, and tabulated. Therefore, the data shown on this page may be incomplete, and will likely not include all deaths that occurred during a given time period, especially for the more recent time periods. Death counts for earlier weeks are continually revised and may increase or decrease as new and updated death certificate data are received from the states by NCHS. COVID-19 death counts shown here may differ from other published sources, as data currently are lagged

So not a record of deaths at all, but simply an indication of the timing of processing of death certificate data.  That makes a fair bit more sense.  Not sure that supports the original assertion that
Quote
more than half of the 6,617 deaths reported during the week ending July 11th actually occurred the week ending May 2 or earlier.

Instead this shows that more than half of the 6,617 death certificates processed that week are from people dying May 2 more earlier.

Abe

  • Handlebar Stache
  • *****
  • Posts: 1648
Re: How long can we wait while flattening the curve?
« Reply #3083 on: July 13, 2020, 10:26:23 PM »
The data I've been using for my graphs comes from NY Times. They are collating records from county and state health departments, so is more up-to-date than the CDC data. I'd hesitate to use national data at this point due to the processing lag noted. County departments in general are getting a daily update of COVID deaths, ICU status and admissions in order to manage patient volumes and distribute amongst hospitals if necessary. Thus those data can be considered accurate on at least a weekly basis, if not daily. Reports to the CDC, since they are not essential for clinical planning, are likely considerably delayed.

Updates for specific hotspots:

Houston - admissions to regular and ICU rooms is starting to slow down. Currently above max normal capacity, but within surge capacity using ERs and ORs.

California - cases continuing to rise across all southern california counties, except San Diego. Two counties have maxed their ICU capacity and are transferring patients to LA, Orange & San Diego. LA is at 95% capacity for ICU, Orange & San Diego ~70%

Florida - 85-90% ICU capacity overall across the state. Several major systems in overflow status.

Arizona - 90% capacity over last two weeks, stabilized with transfers to California and Nevada.

I'll add North Carolina due to reports from friends there - several counties have maxed ICU capacity, overall 75% usage across the state and climbing.

Overall pattern is that ICU capacity is manageable with hospital shutdowns, if no further waves occur. Survival is improving for a variety of factors, but mortality remains around 20-30% in ICU.
« Last Edit: July 13, 2020, 10:44:48 PM by Abe »

former player

  • Walrus Stache
  • *******
  • Posts: 5796
  • Location: Avalon
Re: How long can we wait while flattening the curve?
« Reply #3084 on: July 14, 2020, 12:37:06 AM »
Iím not going down the old folks are less valuable road with anyone.

My opinion is that someone who has not had the benefit of a full life yet is both ethically of higher importance to protect and carries more economic value on average.  This is just my opinion.

I struggle to see how anyone could disagree with this.
Economists do.  Y'all are apparently calculating monetary value using the human capital approach (a rather old fashioned method which only measures the value of production lost) rather than the individual/social willingness-to-pay approach which is more common these days.

Interestingly in relation to road safety, it seems that countries where road safety is better put a higher ceiling on their willingness to pay.  Correlating this to pandemics might explain why the EU has done so much better than the USA.  There's a good explanation here -

https://ec.europa.eu/transport/road_safety/specialist/knowledge/measures/monetary_valuation_of_road_safety_en

former player

  • Walrus Stache
  • *******
  • Posts: 5796
  • Location: Avalon
Re: How long can we wait while flattening the curve?
« Reply #3085 on: July 14, 2020, 12:42:14 AM »
Abe: does it make a difference to ICU stats that the preferred treatment these days seems to be CPAP equivalent rather than ventilators, with the result that more people might be treated outside an ICU than would have been thought appropriate earlier on in the pandemic?  If so then stats on ICU usage might be a less consistent marker over time.  Would percentage of hospital deaths outside ICU be a marker for this?  Although I guess as treatments get better/affected populations get younger that also is subject to change.

Disclaimer: not a statistician.

Spud

  • Stubble
  • **
  • Posts: 161
  • Location: Southwest England, UK.
Re: How long can we wait while flattening the curve?
« Reply #3086 on: July 14, 2020, 12:51:19 AM »
In the UK, masks will become mandatory in all shops from Friday July 24. If you don't comply, you'll be fined £100. This will give rise to a culture of wearing masks in almost all indoor public spaces.

marty998

  • Walrus Stache
  • *******
  • Posts: 7039
  • Location: Sydney, Oz
Re: How long can we wait while flattening the curve?
« Reply #3087 on: July 14, 2020, 03:32:31 AM »

So, okay, are we saying that thereís an equation of worth of life? So, like, a one-year-old is worth 100%, and a two-year-old is worth 99%, and a three-year-old is worth 98%, and so on...? Is that how we determine life worth?

Because it sounds like yíall are really committed to quantifying it. Is this correct?

If itís not this equation, could you tell me what your math looks like?

I wont try to quantify it as I think we were speaking in general principle -  yes I think if you must make the tough comparison, a 20 year old with her life ahead of her and hasn't had the benefit of living a full life is more important to protect than say a 90 year old -- from both an ethical and an economic standpoint. This is precisely the type of decision that Italian doctors faced.  I'm sure people will disagree with this opinion.  I'm open to hearing the flaws in this reasoning.  It would make an interesting ethics discussion, but would also derail this thread.

I am sort of okay with this, but the thing is, Covid isnít the trolley problem. It doesnít give you a 20-year-old on one track and a 90-year-old on the other, and you get to pull the lever.

Honestly, this sort of discussion seems kind of silly to me, especially because itís the kind of discussion that is usually had by people who theoretically think of themselves as low-risk, and therefore donít put themselves on the ďtrackĒ.

@Kris you wonder if those pushing for the QALY assessment to be done volunteer to bump their own parents from ICU so a random 20 year old stranger gets the bed in their place.


Kris

  • Walrus Stache
  • *******
  • Posts: 5664
Re: How long can we wait while flattening the curve?
« Reply #3088 on: July 14, 2020, 07:00:20 AM »

So, okay, are we saying that thereís an equation of worth of life? So, like, a one-year-old is worth 100%, and a two-year-old is worth 99%, and a three-year-old is worth 98%, and so on...? Is that how we determine life worth?

Because it sounds like yíall are really committed to quantifying it. Is this correct?

If itís not this equation, could you tell me what your math looks like?

I wont try to quantify it as I think we were speaking in general principle -  yes I think if you must make the tough comparison, a 20 year old with her life ahead of her and hasn't had the benefit of living a full life is more important to protect than say a 90 year old -- from both an ethical and an economic standpoint. This is precisely the type of decision that Italian doctors faced.  I'm sure people will disagree with this opinion.  I'm open to hearing the flaws in this reasoning.  It would make an interesting ethics discussion, but would also derail this thread.

I am sort of okay with this, but the thing is, Covid isnít the trolley problem. It doesnít give you a 20-year-old on one track and a 90-year-old on the other, and you get to pull the lever.

Honestly, this sort of discussion seems kind of silly to me, especially because itís the kind of discussion that is usually had by people who theoretically think of themselves as low-risk, and therefore donít put themselves on the ďtrackĒ.

@Kris you wonder if those pushing for the QALY assessment to be done volunteer to bump their own parents from ICU so a random 20 year old stranger gets the bed in their place.

Or bumping themselves.
« Last Edit: July 14, 2020, 07:58:06 AM by Kris »

LightTripper

  • Pencil Stache
  • ****
  • Posts: 804
  • Location: London, UK
  • Rural Londoner. Lazy workaholic. Confused.
Re: How long can we wait while flattening the curve?
« Reply #3089 on: July 14, 2020, 07:54:21 AM »
That's really not what QALYs should be about.

I am a fan of QALYs because they force you to face very starkly the consequences of your decisions, and to understand where additional resources should be focused to do the most good.

This does not mean I am relaxed about pulling the plug on my parents in order to give priority to a 20 year old.  I agree that saving the 20 year old will create more expected "QALYs", but I want there to be sufficient resources to ensure that everybody gets a fighting chance, and nobody is being triaged into a corridor. 

For me, that is what flattening the curve was all about: and unless we get rather lucky on one or more fronts, exactly what we need to be getting ready for this autumn/winter too.  But yes, it had a cost.  And if we end up living with Covid for a long time, there will be some very difficult decisions to be made about the trade-offs involved. 

Simply saying "save all lives at all costs" sounds nice but in the end most of us will draw the line somewhere (e.g. however much we want to help everybody and minimise loss of life, we may as a society decide it's not a good use of resources to build a safety barrier on a road in the middle of nowhere that may save one life every thousand years).  QALYs force us to really think about these uncomfortable realities.

To take another example, just because I may accept that a domestic violence centre (say) may generate more QALYs than a particular cancer treatment (for example), doesn't mean I shouldn't do both.  But understanding the good that those two things can do can help me decide how much tax I need to raise (and hopefully help me convince voters that both are worth doing).  And if voters disagree, and instead elect somebody who promises to cut taxes, hopefully QALYs can help inform those cuts so they do as little harm as possible.

former player

  • Walrus Stache
  • *******
  • Posts: 5796
  • Location: Avalon
Re: How long can we wait while flattening the curve?
« Reply #3090 on: July 14, 2020, 09:21:46 AM »
Right. But QALY only really works for medical decisions.  You use a different calculation for road traffic safety costs.  (Which apparently Bloop for one doesn't understand as he keeps trying to bring road safety into a conversation on health costs.)

LightTripper

  • Pencil Stache
  • ****
  • Posts: 804
  • Location: London, UK
  • Rural Londoner. Lazy workaholic. Confused.
Re: How long can we wait while flattening the curve?
« Reply #3091 on: July 14, 2020, 11:10:15 AM »
Personally I suspect we would make better decisions if we used QALYs for both (or CBA).  Certainly studies done in the UK have suggested that the implied thresholds for the cost of saving a life are not equal across the two, which means we are not saving the most lives we can with our taxes.

Anyway, I am not really aligning with Bloop (sorry Bloop).  I am all for being cautious.  There are just too many unknowns about this virus and this autumn/winter could be horrific.  I don't want to live in that world either, and certainly don't want to worry that my personal decisions have contributed to it in any way if it happens.

But recognising that there are costs as well as benefits to shutting down?  Yup.

And that whatever policy you put in place, you have to take people with you?  Also yes.  In fact one of the reasons I have been so frustrated by the UK response isn't so much what we've done as the way they've done it - with all this bluster and bravado and bullshitting about hitting meaningless targets - which I think has really missed the mark in building a sense of community and reliance on one another, and trust in one another, that is invaluable in the countries that have managed to foster that).

Gremlin

  • Bristles
  • ***
  • Posts: 314
Re: How long can we wait while flattening the curve?
« Reply #3092 on: July 14, 2020, 04:12:57 PM »

Economists do.  Y'all are apparently calculating monetary value using the human capital approach (a rather old fashioned method which only measures the value of production lost) rather than the individual/social willingness-to-pay approach which is more common these days.

I've been wondering for the past few weeks whether someone was going to bring this up.  This has always struck me as completely hypocritical that on a early retirement website - a concept which wouldn't exist if not for ideals underpinning willingness-to-pay - that there are a number of posters who, in this one very specific instance, are unwilling to consider any alternative to a human capital model. 

Rosy

  • Handlebar Stache
  • *****
  • Posts: 2352
  • Location: Florida
Re: How long can we wait while flattening the curve?
« Reply #3093 on: July 14, 2020, 04:50:24 PM »
This is interesting.

Once the US death data reported by the states is adjusted to put the dates into the weeks the deaths actually occurred, the trends look much differrent than something like worldometer. The states report the deaths as if they all happened this past week. The CDC includes the actual date of these deaths in their data. Per the CDC's data, more than half of the 6,617 deaths reported during the week ending July 11th actually occurred the week ending May 2 or earlier.

https://twitter.com/kylamb8/status/1281778023935737858

July 11: 2,340
July 4: 2,868
June 27: 3,064
June 20: 5,226
June 13: 5,665
June 6: 6,809
May 30: 7,769
May 23: 10,960
May 16: 11,545
May 9: 9,387
May 2: 12,590

... and just why would that be surprising since we all know that reporting deaths has a huge lag time?
Sometimes we see dumps of thousands in one day.

HBFIRE

  • Handlebar Stache
  • *****
  • Posts: 1249
  • Age: 42
  • Location: Huntington Beach, CA
Re: How long can we wait while flattening the curve?
« Reply #3094 on: July 14, 2020, 05:03:51 PM »

... and just why would that be surprising since we all know that reporting deaths has a huge lag time?
Sometimes we see dumps of thousands in one day.

This last week was the most significant one by a big margin.  It's just something to keep in mind when looking at trending data.

Bloop Bloop

  • Handlebar Stache
  • *****
  • Posts: 2140
  • Location: Melbourne, Australia
Re: How long can we wait while flattening the curve?
« Reply #3095 on: July 14, 2020, 07:19:40 PM »

Economists do.  Y'all are apparently calculating monetary value using the human capital approach (a rather old fashioned method which only measures the value of production lost) rather than the individual/social willingness-to-pay approach which is more common these days.

I've been wondering for the past few weeks whether someone was going to bring this up.  This has always struck me as completely hypocritical that on a early retirement website - a concept which wouldn't exist if not for ideals underpinning willingness-to-pay - that there are a number of posters who, in this one very specific instance, are unwilling to consider any alternative to a human capital model.

Can you elaborate on this? Willingness to pay in a private goods / property context is different, in my eyes, from allocation of scarce public resources. For example I might well be willing to pay $100k for a sports car, or convert that money into time in order to fund an early retirement, because that is what I want for myself. It would be stupid for public money to be spent on either purchase, because it's not 'for the greater good'. But that doesn't mean it's hypocritical to advocate that citizens each have the freedom to do what they like with their own money versus the public purse.

Right. But QALY only really works for medical decisions.  You use a different calculation for road traffic safety costs.  (Which apparently Bloop for one doesn't understand as he keeps trying to bring road safety into a conversation on health costs.)

No, I don't follow your reasoning. Traffic safety may well be funded on a "how much can we get funding for" basis. But it should follow the same assessment of relative cost vs relative reward. As LightTripper suggested, when considering things like road upgrades, speed limit changes, traffic calming devices etc, the most fair outcomes are gotten when we prioritise changes that will save a lot of lives. And remember that in choosing to set a speed limit at, say, 80km/h instead of 40km/h, we are accepting a certain annual loss of life for mere convenience.


deborah

  • Senior Mustachian
  • ********
  • Posts: 10814
  • Location: Australia or another awesome area
Re: How long can we wait while flattening the curve?
« Reply #3096 on: July 14, 2020, 08:04:08 PM »
It is noteworthy that it's been reported today that in Australia we have managed to have only 15% of people admitted to the ICU with Covid19 die. The UK and China recorded 40% and 44%, while in the USA it's about 70%. If we can all get the death rate down significantly, the debate changes somewhat.

Shane

  • Handlebar Stache
  • *****
  • Posts: 1295
  • Location: PA
Re: How long can we wait while flattening the curve?
« Reply #3097 on: July 14, 2020, 08:43:42 PM »
Just saw this article and wondering what the significance is, if any, to the Trump Administration's changing of the reporting procedures for Covid data:

White House tells hospitals to bypass CDC on COVID-19 data reporting

GuitarStv

  • Senior Mustachian
  • ********
  • Posts: 16600
  • Age: 39
  • Location: Toronto, Ontario, Canada
Re: How long can we wait while flattening the curve?
« Reply #3098 on: July 14, 2020, 09:05:09 PM »
Trump didn't have the level of control over the CDC he needed to sufficiently hide the problem, so this step makes a lot of sense.  CDC wasn't staying on brand.

habanero

  • Pencil Stache
  • ****
  • Posts: 668
Re: How long can we wait while flattening the curve?
« Reply #3099 on: July 14, 2020, 11:38:44 PM »
It is noteworthy that it's been reported today that in Australia we have managed to have only 15% of people admitted to the ICU with Covid19 die. The UK and China recorded 40% and 44%, while in the USA it's about 70%. If we can all get the death rate down significantly, the debate changes somewhat.
70% of US ICU patients die? Do you have a source for that? The figure seems absurdly high.