Started By
Message

re: Daily COVID Updated as of 11/2/20 8:00 PM

Posted on 4/5/20 at 10:42 pm to
Posted by Taxing Authority
Houston
Member since Feb 2010
57090 posts
Posted on 4/5/20 at 10:42 pm to
quote:

Looks like the total number of recoveries will be double the total numbers of deaths by the end of day tomorrow.
Given how many recoveries we're not counting... that is impressive. Our healthcare workers are doing an amazing job.
Posted by Taxing Authority
Houston
Member since Feb 2010
57090 posts
Posted on 4/5/20 at 10:46 pm to
quote:

I don’t know why you’re so focused on recoveries since as long as it’s a virus that people recover from and in a somewhat short period of time (weeks, months at most), then eventually recovers will just be the # of cases - # of deaths.
The trick here is how "recoveries" are counted. We are counting almost none of them unless they've been hospitalized or were serious cases. Most of the "recoveries" that get counted are going to be people that were sick enough to enter the medical system, and were there long enough to have 2 negative tests.

Posted by buckeye_vol
Member since Jul 2014
35236 posts
Posted on 4/5/20 at 10:55 pm to
quote:

The potential is there for some hinky revisions through Tuesday, but this feels different than the last couple of weekends, if you want a totally subjective, unscientific take on this.
One great thing that I just noticed above chrome’s reporting is that the snapshots of his daily updates (specifically the final updates) show the pre-revision data. So unlike the updated data (on worldometers and his spreadsheet), these show the extent to which they’re revised and allow for an apples to apples comparison to the current pre-revised data.

Take a look at the differences in the current figures and the initial final updates from the past 2 Sundays.

The things that stand out to me are that: (1) weekend data has much more error; (2) there have been some major revisions to the data; and (3) and the weekend data (at least 1 of the 2 days) often shows more promising trend than the more reliable long-term data.

Now that we’re getting a large enough sample, I’m thinking about downloading his data and running some time-series analyses to better detect the trends, seasonalities (e.g., days of the week in this case), and cycles. In addition, I want to see if we can find the best lag period between case data and death data and test that as a regressor itself.

Current Sunday:

Last Sunday:

2 Sundays ago
Posted by buckeye_vol
Member since Jul 2014
35236 posts
Posted on 4/5/20 at 10:58 pm to
quote:

The trick here is how "recoveries" are counted. We are counting almost none of them unless they've been hospitalized or were serious cases.
In that case, I think the pertinent data would be length to recovery (and/or deaths for that matter) from various dates (diagnosis, hospitalization admit, and/or ICU admit) even is it’s just those that are hospitalized and admitted into ICU.
Posted by tiger91
In my own little world
Member since Nov 2005
36703 posts
Posted on 4/5/20 at 11:02 pm to
I don’t know that you’ll ever get time from diagnosis to recovery or death. Lol. They can’t even report positives on the date of the test but rather when it’s available. .
Posted by Chromdome35
NW Arkansas
Member since Nov 2010
6823 posts
Posted on 4/5/20 at 11:04 pm to
If you email me, I will share the data page where I'm cataloging the data.

chromdome351@gmail.com
Posted by Taxing Authority
Houston
Member since Feb 2010
57090 posts
Posted on 4/5/20 at 11:31 pm to
quote:

In that case, I think the pertinent data would be length to recovery (and/or deaths for that matter) from various dates (diagnosis, hospitalization admit, and/or ICU admit) even is it’s just those that are hospitalized and admitted into ICU.
For sure.
Posted by imjustafatkid
Alabama
Member since Dec 2011
50304 posts
Posted on 4/5/20 at 11:43 pm to
quote:

I don’t know why you’re so focused on recoveries


Because they're being underreported.
Posted by Taxing Authority
Houston
Member since Feb 2010
57090 posts
Posted on 4/5/20 at 11:43 pm to
quote:

Chromdome35
Sent.
Posted by buckeye_vol
Member since Jul 2014
35236 posts
Posted on 4/6/20 at 12:06 am to
quote:

Because they're being underreported.
Because it’s not practical or even possible to track them at this point, since the only way to track those that aren’t hospitalized are to give them tests (and multiple tests) to confirm they’re negative and that would be a waste of a limited resource at this time.

In 4 to 6 weeks, when things start opening up and we’re well past the peak then maybe we’ll be able to start getting a better sense of recoveries.
Posted by the808bass
The Lou
Member since Oct 2012
111496 posts
Posted on 4/6/20 at 12:25 am to
Here’s an anecdotal timeline:

Day -4: exposure
Day 1: symptoms progress to ER visit with fever and kidney pain, discharged home without testing for Covid 19 (ER said, correctly, they thought it was viral - lol)
Day 5: doctor’s office visit, test administered
Day 6: positive result back from state lab, staying at home with pulse ox
Day 10: symptoms worsening, oxygen levels dipping below 90
Day 11: oxygen drops to 87 in midday, admitted to ICU, on oxygen
Day 12: HCQ/Zpack administered, it is expected that he will he intubated
Day 16: first day of marked, sustained improvement - normal food, able to ambulate, never intubated
Day 17 - prognosis looks good, lasix started for fluid on lungs
Day 20 - moved from ICU to normal room
Day 22 - oxygen discontinued
Day 23 - discharge

Posted by Tigerbait357
Member since Jun 2011
67865 posts
Posted on 4/6/20 at 12:26 am to
quote:

I don’t know that you’ll ever get time from diagnosis to recovery or death. Lol. They can’t even report positives on the date of the test but rather when it’s available. .




Which is why the numbers are very F'd up at times.
Posted by the808bass
The Lou
Member since Oct 2012
111496 posts
Posted on 4/6/20 at 12:28 am to
LINK

quote:

This report describes the epidemiologic and clinical features of the first case of 2019-nCoV infection confirmed in the United States.


This was interesting because it seems like a pretty garden-variety Coronavirus hospital admission.
Posted by dafif
Member since Jan 2019
5536 posts
Posted on 4/6/20 at 12:46 am to
BuckPshrink

quote:

what's lost in this is the preponderance of the target demographic in fla - senior citizens. fla should be getting destroyed right now a la n. italy, especially given the fact that people really aren't practicing social measures


Agreed...and yet they are not...with all the cases really in S Fla. death rate much lower than NY.
Posted by tigersownall
Thibodaux
Member since Sep 2011
15294 posts
Posted on 4/6/20 at 12:59 am to
Can someone explain to me the daily growth rate percentage. It seems like the number of cases by day grows but the percentage has been dropping since March 22nd
Posted by buckeye_vol
Member since Jul 2014
35236 posts
Posted on 4/6/20 at 1:30 am to
quote:

Can someone explain to me the daily growth rate percentage. It seems like the number of cases by day grows but the percentage has been dropping since March 22nd
It’s total cases (or deaths) for the day divided by total cases (or deaths) the day before.

It’s the same calculation as say an investment account or anything that grows exponentially. So if you had $100 yesterday (or 10 cases or deaths) and it grew by $10 today (or 10 cases or deaths) the growth would be 10%.

But if you started at 200 yesterday, you would need to add 20 to achieve the same growth rate.
Posted by trinidadtiger
Member since Jun 2017
13292 posts
Posted on 4/6/20 at 2:45 am to
quote:

Agreed...and yet they are not...with all the cases really in S Fla. death rate much lower than NY.


Betcha if you checked the socio economic indicators of the two groups you would find South Florida is more white (less overweight/diabetes) and better income (hence better medical care).
Posted by Jinglebob
Member since Jan 2020
948 posts
Posted on 4/6/20 at 5:35 am to
It looks like they decided to start documenting the data for number of people tested.

US 1.7 million so far, Spain #2 with 900k.

Is there any way to add that data into your spreadsheet.

The sky is falling crowd won't see any value but the grass is greener crowd probably would.

And it would give a little context to the # cases data.

Pretty sure we tested around a million people this week (and had 200k new cases) compared to testing 10k a week in mid March (and 3k new cases). Obviously there is a lag in there but hopefully you are picking up what I am throwing down.

Posted by AMS
Member since Apr 2016
6495 posts
Posted on 4/6/20 at 5:56 am to
quote:


Agreed...and yet they are not...with all the cases really in S Fla. death rate much lower than NY.


Actual quantitative viral exposure is a factor in severity of the infection. NY being much closer quarters likely means larger ‘doses’ of virus.
being a less densely packed is a form of Social distancing With/without other specific behavior modifications.

More folks densely packed shedding higher loads is partially why it hits regions like wuhan, ny, and ital with greater severity.
Posted by AMS
Member since Apr 2016
6495 posts
Posted on 4/6/20 at 5:59 am to
quote:

Can someone explain to me the daily growth rate percentage. It seems like the number of cases by day grows but the percentage has been dropping since March 22nd


Sampling effort plays a role. With the more severe cases getting priority testing first. As tests are more available and used more frequently this dilutes the initial rate growth. Better sampling gives you a more realistic picture.

first pageprev pagePage 140 of 331Next pagelast page

Back to top
logoFollow TigerDroppings for LSU Football News
Follow us on Twitter, Facebook and Instagram to get the latest updates on LSU Football and Recruiting.

FacebookTwitterInstagram