Author Topic: COVID-19 Data Model Updated Daily  (Read 20175 times)

Beasho

  • Cortez Bank Status
  • *****
  • Posts: 2579
    • View Profile
Re: COVID-19 Data Model Updated Daily
« Reply #90 on: March 28, 2020, 11:28:42 AM »
PDF Print out of Model attached.

I am getting the On-Line version updated with this latest Model.

Beasho

  • Cortez Bank Status
  • *****
  • Posts: 2579
    • View Profile
Re: COVID-19 Data Model Updated Daily
« Reply #91 on: March 28, 2020, 12:02:02 PM »
From a high school friend who is now a Doctor:

"Stories from the front line are alarming but represent only a fraction of cases. The unknown seems to be why some are minimally symptomatic and why some go into respiratory failure (even healthy people). Hot zone statistics seem to suggest 81% get mild symptoms, 15% get pretty sick and may wind up in the hospital with supplemental oxygen but survive. 4% become critical and of those the death rate is is 70-80%. In other words, if you wind up on a ventilator you better have your affairs already in order. This is the story out of Seattle and NOLA anyway."

--> 4% X 70% = 3.2% Death Rate (now I am running numbers).  This is HIGHER than reported but within the realm of believably.

Takeaway:  YOU DO NOT WANT TO BE ON A VENTILATOR

surfcowboy

  • Cortez Bank Status
  • *****
  • Posts: 3860
    • View Profile
Re: COVID-19 Data Model Updated Daily
« Reply #92 on: March 28, 2020, 12:14:15 PM »
First, thanks Beasho. This is pretty solid and although I’m not the best numbers wonk, I have a pretty solid track record with trend analysis in my biz and I agree with your peak date estimate. (I picked it for later next week but the average will win, which I didn’t take into account.)

2 thoughts. The data sources would need an insane amount of collusion between completely disparate agencies to pull off faking these numbers.

Second, I work in “The Média” and while there is bias, it’s mostly like your jobs. A bunch of people trying to pull shit off with mixed results. As a business owner I can’t even get my team to turn in timesheets half the time, let alone pull off a global scam. The bias that exists is normally a few producers or writers and they get to do just enough to not get fired and in true journalism (make your own call as to what constitutes that) if you can’t make up a convincing fairy tale to back you stuff, you get to go home.

tarquin

  • Malibu Status
  • **
  • Posts: 87
    • View Profile
Re: COVID-19 Data Model Updated Daily
« Reply #93 on: March 28, 2020, 12:31:31 PM »
Yes you only have to look at what's been going on in Italy, Spain,France and all the other countries. People get put on ventilators for  weeks on end. Then dont normally make it. This is one of the many reasons it's so hard on the hospitals and staff.
 These people are doing an amazing job and all deserve medals as far as I'm concerned. So anything we can do to help them should be done. What ever country you are in.
 Make masks etc. But it looks like stay home and try to slow the spread is the best option now that govts all over the world have been too slow to react.
 Look at the positive side. We can make a difference environmentally. In a very short time.Something to think about.

Admin

  • Administrator
  • Cortez Bank Status
  • *****
  • Posts: 5258
    • View Profile
    • StandUpZone
    • Email
Re: COVID-19 Data Model Updated Daily
« Reply #94 on: March 29, 2020, 02:28:57 AM »
Has anyone seen any data sources showing logarithmic scale charts, or "rate of doubling", instead of linear charts?

Hi Fishman,

I saw that Johns Hopkins added a Logarithmic tab to their chart (bottom right): https://coronavirus.jhu.edu/map.html

LaPerouseBay

  • Teahupoo Status
  • ******
  • Posts: 1890
  • downwind dilettante
    • View Profile
    • Email
Re: COVID-19 Data Model Updated Daily
« Reply #95 on: March 29, 2020, 11:29:37 AM »
No log scale on this one yet, but it zooms in to counties. 

https://www.statnews.com/2020/03/26/covid-19-tracker/
Support your local shaper

Beasho

  • Cortez Bank Status
  • *****
  • Posts: 2579
    • View Profile
Re: COVID-19 Data Model Updated Daily
« Reply #96 on: March 29, 2020, 11:40:49 AM »
Report for the day.  United States Fatality is LOW.

In Line with Germany and Japan.  Korea has spiked.

Admin

  • Administrator
  • Cortez Bank Status
  • *****
  • Posts: 5258
    • View Profile
    • StandUpZone
    • Email
Re: COVID-19 Data Model Updated Daily
« Reply #97 on: March 30, 2020, 04:47:14 AM »
Watching Dr. Fauci (and Dr. Birx) estimate 100,000 to 200,000 deaths in the US in yesterday's briefing was disorienting (the whole briefing was bizarre). 

"Whenever the models come in, they give a worst-case scenario and a best-case scenario. Generally, the reality is somewhere in the middle. I've never seen a model of the diseases that I've dealt with where the worst case actually came out. They always overshoot.  I mean, looking at what we're seeing now, you know, I would say between 100 and 200,000 (deaths). But I don't want to be held to that," he said, adding that the US is going to have "millions of cases."

He is seeing a peak in deaths in 2-3 weeks.  If correct, those would be death numbers in the range of our current infected numbers.  A commentator mentioned this being like thirty to sixty 911 events. Trump is now saying that 100,000 deaths would be a great success and based on a warning he was given (2.2 Million dead) he extended social distancing measures to April 30.

Here is a good article: https://www.ft.com/content/f3796baf-e4f0-4862-8887-d09c7f706553

John Ioannidis, a professor of epidemiology at Stanford University, has branded the data we have about the epidemic “utterly unreliable”.  “We don’t know if we are failing to capture infections by a factor of three or 300,” he wrote last week. If thousands more people are surviving than we know about, then current mortality rate estimates are too high — perhaps by a large margin.

Researchers at the University of Hong Kong have estimated that, in Wuhan, where the pandemic began, the likely death rate was 1.4 per cent — much lower than the previous estimate of 4.5 per cent, which was calculated using official statistics on the region’s cases and deaths.
« Last Edit: March 30, 2020, 06:20:05 AM by Admin »

Beasho

  • Cortez Bank Status
  • *****
  • Posts: 2579
    • View Profile
Re: COVID-19 Data Model Updated Daily
« Reply #98 on: March 30, 2020, 08:43:10 AM »
Watching Dr. Fauci (and Dr. Birx) estimate 100,000 to 200,000 deaths in the US in yesterday's briefing was disorienting (the whole briefing was bizarre). . . .

He is seeing a peak in deaths in 2-3 weeks.  If correct, those would be death numbers in the range of our current infected numbers.  A commentator mentioned this being like thirty to sixty 911 events. Trump is now saying that 100,000 deaths would be a great success and based on a warning he was given (2.2 Million dead) he extended social distancing measures to April 30. . . .

Researchers at the University of Hong Kong have estimated that, in Wuhan, where the pandemic began, the likely death rate was 1.4 per cent — much lower than the previous estimate of 4.5 per cent, which was calculated using official statistics on the region’s cases and deaths.[/b]

I am estimating numbers BELOW Fauci.  This model shows 82,000 Deaths in the United States but requires the Exponential Growth of Cases Positive Cases to Reach 4.7 Million in the United States.

It will not.  There is not enough Dry Powder in the Keg due to
1) CHANGE in Social Behavior and
2) Death Rates are likely LOWER than currently recorded because of Many, Many more benign infections than we are currently measuring per the Wuhan data above. 

My Conclusion:The death rate in the US is likely to be BELOW 82,000.

Days to Double in the US has INCREASED to 5.6.  Italy the United States and the Remainder of Europe reported FEWER cases yesterday than the day before.  Looks like a top but still too early to call it.

Full output attached as PDF:
« Last Edit: March 30, 2020, 09:12:50 AM by Beasho »

Beasho

  • Cortez Bank Status
  • *****
  • Posts: 2579
    • View Profile
Re: COVID-19 Data Model Updated Daily
« Reply #99 on: March 30, 2020, 08:52:26 AM »
With regards to Comorbidity Corona Virus may just be accelerating the leading causes of Death in the United States. 

This model shows what 82,000 deaths would look like IF the COVID-19 numbers simply displaced the HIGH Risk causes of Death from 2017. 

It would result in a 5% drop in the leading causes with NO NET CHANGE in Total Deaths.

Here is an article suggesting that Deaths as measured in TOTAL Numbers are in FACT DOWN Across all age groups.  Lets hope it not FAKE news.  Could it be possible:

WILL MORE PEOPLE WILL SURVIVE THIS YEAR THAN IN PAST YEARS BECAUSE OF THE CORONA VIRUS?

https://www.dropbox.com/s/2y2gvh2q7axdzz2/COVID-19_UnintendedConsequences.pdf?dl=0&fbclid=IwAR3eHyr0bLrB0zQBmol11P7_xrDJRhXRYz5oLaT3o7jGGbgusYMs12SxvHw
« Last Edit: March 30, 2020, 09:10:43 AM by Beasho »

Admin

  • Administrator
  • Cortez Bank Status
  • *****
  • Posts: 5258
    • View Profile
    • StandUpZone
    • Email
Re: COVID-19 Data Model Updated Daily
« Reply #100 on: March 30, 2020, 09:12:33 AM »
Of note: the CDC just moved the USA as a whole from the initiation phase to the acceleration phase on the 26th noting that much of the country is still in the initiation phase.  At this point we have to believe that they have working estimates to adjust for the failure to sufficiently test and for the huge variances in testing by location and over time.
« Last Edit: March 30, 2020, 09:18:34 AM by Admin »

Beasho

  • Cortez Bank Status
  • *****
  • Posts: 2579
    • View Profile
Re: COVID-19 Data Model Updated Daily
« Reply #101 on: March 30, 2020, 09:30:59 AM »
OK - Plan for Denominator to GO UP!

Can we assume Death Count is more Solid?  Aka not a bunch of hidden bodies somewhere.

US Profile ONLY. 

PonoBill

  • Cortez Bank Status
  • *****
  • Posts: 23268
    • View Profile
Re: COVID-19 Data Model Updated Daily
« Reply #102 on: March 30, 2020, 09:32:50 AM »
Actually, you never want to be on a ventilator in any case. My daughter is a Respiratory Therapist, and during one of our conversations early in the outbreak, she said she heard that caregivers in Italy were putting people on ventilators when their O2 level dropped beyond a certain point. She was outraged, she said something like: "you put a ventilator tube in someone's lung and you just took 20 percent of their lifespan away. A lot of people are never the same. We view it as a last-ditch effort." I'm sure she's realized now that the people they're putting on ventilators are at the last-ditch.

The April 30th date is probably a good one. Most of the predictive models I've seen for the effect of shelter in place show a peak at around April 21. The peak for Oregon is estimated on April 24.

One of the new problems with testing is that the most intensive testing is happening in places that are already a hot spot. What pretty much everyone expected--that testing only sick people vastly under-reports the infection rate--is obviously true. The huge jump in infected people in New York is probably as much an artifact of greatly increased general testing as it is an indication of infection growth. Increased testing in areas not identified as hot spots would enable those infection loci that are under the radar to be detected and appropriate action taken. Then again, the large number of infections reported in New York will probably get people to comply more with social distancing requirement. New Yorkers are notoriously hard to push to comply with anything.
« Last Edit: March 30, 2020, 09:39:59 AM by PonoBill »
Foote 10'4X34", SIC 17.5 V1 hollow and an EPS one in Hood River. Foote 9'0" x 31", L41 8'8", 18' Speedboard, etc. etc.

Admin

  • Administrator
  • Cortez Bank Status
  • *****
  • Posts: 5258
    • View Profile
    • StandUpZone
    • Email
Re: COVID-19 Data Model Updated Daily
« Reply #103 on: March 30, 2020, 09:51:34 AM »
Yes, there have been a group of interesting JAMA articles about this.  This one was about Italy.  With a mid-stream change in testing protocol the case fatality rate jumped from 3.1 to 7.2%.  Equally interesting is the large variances in start dates which of course determines the significance of death count.  https://jamanetwork.com/journals/jama/fullarticle/2763667

After an initial, extensive testing strategy of both symptomatic and asymptomatic contacts of infected patients in a very early phase of the epidemic, on February 25, the Italian Ministry of Health issued more stringent testing policies. This recommendation prioritized testing for patients with more severe clinical symptoms who were suspected of having COVID-19 and required hospitalization. Testing was limited for asymptomatic people or those who had limited, mild symptoms. This testing strategy resulted in a high proportion of positive results, ie, 19.3% (positive cases, 21 157 of 109 170 tested as of March 14, 2020), and an apparent increase in the case-fatality rate because patients who presented with less severe clinical disease (and therefore with lower fatality rate) were no longer tested (case-fatality rate changed from 3.1% on February 24 to 7.2% on March 17). These more mild cases, with low fatality rate, were thus no longer counted in the denominator.

On the basis of the number of cases and of the advanced stage of the disease it was hypothesized that the virus had been circulating within the population since January.

SUP Leave

  • Sunset Status
  • ****
  • Posts: 477
    • View Profile
Re: COVID-19 Data Model Updated Daily
« Reply #104 on: March 30, 2020, 11:18:48 AM »
Beasho, thanks again. I can't tell you how much I value your math on this thread.

As just an observer of the human condition, this outbreak is fascinating. I have friends and virtual friends social media who will not (or cannot) allow themselves to look at this virus objectively. Everything is viewed emotionally. It is almost like people *enjoy* staying at home and virtue signalling about how good they are at social distancing. I'm crawling out of my skin being restricted from outdoor activities. I have almost worn through my golf net. I broke my mtn bike on Saturday.

If statistics tell us anything it is that the number one cause of dying from CV-19 is the number one cause of dying any other time. Being old and/or unhealthy wins every time. Sure there are going to be outliers, just like in every day life. In the WA state statistics as of Friday we have 195 total deaths, 58% of which are 80+ years old. Is CV something to fear more than heart disease?

The biggest take away from this, when it is all over is that we should dedicate far more time to our health and our families' health than anything else. Living in fear of a virus while sitting on your ass eating foods with HFCS and complex carbohydrates is simply bewildering.

I certainly have been giving Dad lectures to my girls about it (captive audience). They will likely have ocular issues moving ahead with the number of eye rolls they have given me these past weeks.
Make paddleboarding great again!