B/CS number of cases update? 11-17-20 Staff Edit on OP

673,647 Views | 4711 Replies | Last: 6 hrs ago by Nosmo
jeffk
How long do you want to ignore this user?
AG
Multi-talented you are.
cavscout96
How long do you want to ignore this user?
AG
theNetSmith said:

jeffk said:

Fair enough. Really all of the answers to that question (including mine and yours) were speculative. The constant policing of responses by a couple of posters in particular generates way more derailments than any "doom and gloom" posters.
I pride myself on both my thread derailments and my "doom and gloom" posts.
I wasn't referring to your post, but your D&G is top-notch!

**Note the "winky"-face this is truly a tongue-in-cheek comment
theNetSmith
How long do you want to ignore this user?
AG
noted, Cav. I treasure you as my #1 antagonist.
cavscout96
How long do you want to ignore this user?
FlyRod
How long do you want to ignore this user?
https://www.kbtx.com/2020/12/10/health-district-reports-25-covid-19-deaths-after-reporting-error/
MiMi
How long do you want to ignore this user?
S
Today's numbers: 134 new confirmed cases, 43 probable cases, 26 Brazos county residents hospitalized, 1,731 tests, 25 deaths

From KBTX:
Quote:

25 new deaths have been reported in the last 24 hours. BCHD says these 25 COVID-19 reach back to June 2020 and were reported directly to the Texas Department of State Health Services and were not reported locally to the Brazos County Health District. The health district has worked on a process to ensure that deaths are reported locally in the future.
Bunk Moreland
How long do you want to ignore this user?
death drop from months past + a large increase in positives.
cavscout96
How long do you want to ignore this user?
AG
What a goat-screw. >25% error in reporting. Completely incompetent and wholly unacceptable.
cavscout96
How long do you want to ignore this user?
AG
Thanksgiving?
Slocum on a mobile
How long do you want to ignore this user?
AG
FlyRod said:

https://www.kbtx.com/2020/12/10/health-district-reports-25-covid-19-deaths-after-reporting-error/


When I read this "Send in the Clowns" played in my head.

The very same people that are advising the local politico types are the same ones that missed a large chunk of deaths attributable to Covid. I would say I'm in shock, but I really don't expect any better with the local $h**show.

We pay taxes to support this crap. Between a press that throws softballs, to a County Health director who won't give basic info (age, location, etc - which they *do* in many counties), I can't help but feel that this whole rigamarole has done nothing to boost the public's confidence. It has marred it further.

DO BETTER, BRAZOS COUNTY. This isn't Robertson County.

Edit to add - I'm sure the local press will cry "THE SKY IS FALLING!" Not, the "County Health Department are dumb@$$es..."
FlyRod
How long do you want to ignore this user?
I agree. Troubling.
lockett93
How long do you want to ignore this user?
AG
MiMi said:

Today's numbers: 134 new confirmed cases, 43 probable cases, 26 Brazos county residents hospitalized, 1,731 tests, 25 deaths

From KBTX:
Quote:

25 new deaths have been reported in the last 24 hours. BCHD says these 25 COVID-19 reach back to June 2020 and were reported directly to the Texas Department of State Health Services and were not reported locally to the Brazos County Health District. The health district has worked on a process to ensure that deaths are reported locally in the future.




I brought up the discrepancy between John Hopkins / worldometers data and BCHD data on here a long time ago...

How could someone at BCHD not have also seen the same discrepancy?
lockett93
How long do you want to ignore this user?
AG
We also know that BCHD isnt or wasn't getting the testing number correct. Impossible with the number of tests being done on campus.
Rapier108
How long do you want to ignore this user?
It is the same song, different verse that we've been saying in this all began.

The BCHD is either completely incompetent, there is some level of malfeasance going on, and probably both.

This is the same kind crap we see with the "positivity rate" where they report 0 tests for days on end so the rate goes sky high, when in reality it is an completely artificial number. Then they'll dump several hundred to several thousand at once, then go back to reporting 0.

That said, the media and doomers will act like 25 people died yesterday because that is what they do.
lockett93
How long do you want to ignore this user?
AG
Thursday's results include:

90's - 0
80's - 1
70's - 7
19 from a cluster

64 (+4) Covid in hospital
+62 PCR cases today than this day last week
28% in the 18-24 age group (38 people)

Spreadsheet updated, chart tab updated.

LINK TO SPREADSHEET AND CHARTS
The Brazos Kid
How long do you want to ignore this user?
It shall be blamed on BCHD "budget cuts"...
Slocum on a mobile
How long do you want to ignore this user?
AG
lockett93 said:


I brought up the discrepancy between John Hopkins / worldometers data and BCHD data on here a long time ago...

How could someone at BCHD not have also seen the same discrepancy?
They are oblivious. A likely scenario I can put together is that the low level employee charged with adding together the numbers and update the PDFs, the thought never crossed that person's mind that they might be wrong or under-reporting. I'm betting that the error was spotted from someone higher up the stack, or on the other side.(like the state).

If this was Harris county, Channel 13 would be down there stuck in Sullivan's face demanding answers.

Our press: "Awww, shucks.... Sorry that happened, and y'all can't count... No more questions!"
lockett93
How long do you want to ignore this user?
AG
TxFoundry said:

It shall be blamed on BCHD "budget cuts"...

They will blame the reporting entity.

Speaking of that, I wonder who it was that wasn't reporting their deaths to the BCHD? St Joes, S&W, Physicians Center? The medical examiner? If it was one of the first 3, don't you think BCHD would say "wow, that's odd, we haven't gotten any deaths from XYZ..."

This is a chart I made from the state data. Doesn't look like June to me. It looks like October through December is where the county was off... But the deaths could have been June. These are the reported dates.
trouble
How long do you want to ignore this user?
AG
Pretty much just glossed over the missing 25 deaths. The audio is horrible so I'm not going to wait and listen to the softballs.

My guess is that they were reconciling death certificates and realized they missed a bunch. If they were actually doing tracking like they originally told us they were, those wouldn't have been missed.
trouble
How long do you want to ignore this user?
AG
That's after the person who was doing social media was laid off and we started getting all kinds of screwy numbers in their graphics.
plain_o_llama
How long do you want to ignore this user?
Not trying to accuse or defend anyone here.....

I suspect this is obvious to most but one of the problems in reporting is how to attribute locale to patient hospitalizations and deaths. I don't know if it applies with Brazos County, but some places have been more careful about reporting deaths of only their residents. So, if someone gets transferred out of the community for treatment and die there, how, when, and where is that death reported? And who is ultimately reconciling the data and reporting? The hospitals, the county, the state, the CDC, TexAgs, or twitter?

YMMV
trouble
How long do you want to ignore this user?
AG
It should be filed with their county of residence. The place of death can be listed as a different county.

It's an electronic process so I really don't understand how they got so off for so long.
LSCSN
How long do you want to ignore this user?
how hard is it to have a spreadsheet with the date, number of tests performed that day, and the positive/negative counts??
cavscout96
How long do you want to ignore this user?
AG
plain_o_llama said:

Not trying to accuse or defend anyone here.....

I suspect this is obvious to most but one of the problems in reporting is how to attribute locale to patient hospitalizations and deaths. I don't know if it applies with Brazos County, but some places have been more careful about reporting deaths of only their residents. So, if someone gets transferred out of the community for treatment and die there, how, when, and where is that death reported? And who is ultimately reconciling the data and reporting? The hospitals, the county, the state, the CDC, TexAgs, or twitter?

YMMV
Lockett93

The answer is Lockett93......

he is the only one in Brazos County compiling a detailed list of this with any sort of usable info. Even with suspect data his SS is clear cut and straight forward. Several others have provided input and analysis. NO ONE from BCHD or the local media has provided a scintilla of worthwhile information or analysis, just continued bad data and non-existent or flawed analysis.


WHAT.A.JOKE
isitjustme
How long do you want to ignore this user?
AG
In the presser today, Sullivan was asked about the spike in cases. He went on about being after Thanksgiving, gatherings, etc. blah blah blah!!

There were several days of highly publicized drive-through and on-campus testing days, and they were encouraging everyone to take a test before going anywhere for the next holidays. As a result, and according to Lockett's spreadsheet, 11,549 tests have been conducting during these first 10 days of December with 718 positive results. That's just over 6%. More test = more positives. That should have been his answer.
trouble
How long do you want to ignore this user?
AG
LSCSN said:

how hard is it to have a spreadsheet with the date, number of tests performed that day, and the positive/negative counts??


It has to be more involved than that to match tests with results. They all come back at varying intervals.
lockett93
How long do you want to ignore this user?
AG
LSCSN said:

how hard is it to have a spreadsheet with the date, number of tests performed that day, and the positive/negative counts??
Data is only as good as the accuracy of those providing it. The BCHD is at the mercy of those providing them the data to do it, and do it correctly. But, they need to be the ones making sure those providing data are doing so timely and accurately.

The state is finally getting some things better organized to get USEFUL information. They are going to begin reporting cases and adjusting data based on TEST date and not result date. Reporting the test on Monday and the result on Thursday isn't useful information.

Same thing with deaths... The date the person actually died is extremely relevant, not just the total number. For charting purposes, the BCHD should create a plot like the one I did above but that shows the actual date of death and not the total number reported as of that date.
91_Aggie
How long do you want to ignore this user?
AG
cavscout96 said:

plain_o_llama said:

Not trying to accuse or defend anyone here.....

I suspect this is obvious to most but one of the problems in reporting is how to attribute locale to patient hospitalizations and deaths. I don't know if it applies with Brazos County, but some places have been more careful about reporting deaths of only their residents. So, if someone gets transferred out of the community for treatment and die there, how, when, and where is that death reported? And who is ultimately reconciling the data and reporting? The hospitals, the county, the state, the CDC, TexAgs, or twitter?

YMMV
Lockett93

The answer is Lockett93......

he is the only one in Brazos County compiling a detailed list of this with any sort of usable info. Even with suspect data his SS is clear cut and straight forward. Several others have provided input and analysis. NO ONE from BCHD or the local media has provided a scintilla of worthwhile information or analysis, just continued bad data and non-existent or flawed analysis.


WHAT.A.JOKE


If local media employees or govt employees could do detailed and accurate statistical analysis they would have "tested" out of their current jobs.
cavscout96
How long do you want to ignore this user?
AG
trouble said:

LSCSN said:

how hard is it to have a spreadsheet with the date, number of tests performed that day, and the positive/negative counts??


It has to be more involved than that to match tests with results. They all come back at varying intervals.
yes, they must be correlated to provide useful data, but it isn't that hard.
trouble
How long do you want to ignore this user?
AG
cavscout96 said:

trouble said:

LSCSN said:

how hard is it to have a spreadsheet with the date, number of tests performed that day, and the positive/negative counts??


It has to be more involved than that to match tests with results. They all come back at varying intervals.
yes, they must be correlated to provide useful data, but it isn't that hard.


Agreed but it's more involved than the spreadsheet that was proposed.
LSCSN
How long do you want to ignore this user?
cavscout96 said:

trouble said:

LSCSN said:

how hard is it to have a spreadsheet with the date, number of tests performed that day, and the positive/negative counts??


It has to be more involved than that to match tests with results. They all come back at varying intervals.
yes, they must be correlated to provide useful data, but it isn't that hard.


What's more involving about listing how many tests are performed each day and their subsequent result?
cavscout96
How long do you want to ignore this user?
AG
True. I didn't take your response in context to the original question.
Nosmo
How long do you want to ignore this user?
AG
I was going to post this previously but didn't want to derail the local Covid number discussion.

I normally look at local, state, and then USA numbers. At one time I was doing it daily and then moved to weekly.

Back at the end of July (I think 27 th), the state started using death certificate date of death. It screwed up the state numbers at the time as it added over 600 deaths (cumulative) to July 27th, and made national headlines.

I started tracking "certificate date of death" reporting vs cumulative "to date" reporting. It turns out that the state is only reporting about 1/2 the deaths for a calendar day after about 2 weeks. In other words, if the state reported 306 death on August 6, none of those deaths occurred August 6 or August 7. By August 13, one week later, there was 45 deaths attributed to August 6. Two weeks later, the August 20 state report shows 145 deaths occurred August 6. By September 1, it was 204. The number slowly increases and by October 31, that number was 229, and remained as of Saturday December 6.

Mid-September, I averaged numbers since the end of July, and came up with the following time lag for "actual date of death" vs "date reported":

After 5 days, 5% of actual deaths reported.
After 10 days, 30% of actual deaths reported.
After 15 days, 50% of actual deaths reported.
After 20 days, 65% of actual deaths reported.
After 25 days 80% of actual deaths reported.
After 45 days 95% of actual deaths reported.

When the state says there's 300 deaths today, most of those deaths have occurred over past 6 weeks or so.

So what does this have to do with today's numbers?

I am not sure, but I am guessing it is somewhat a dilemma, as the local media want's instant information, and the state appears to be using numbers it can back-up with death certificates.

I assume that since names are "secret" due to HIPAA, that the numbers can get hard to track (for the media), especially since a person gets counted back to their residence county.

I am not justifying the confusion or defending the inaccurate reporting, but I can see there's a lot of moving parts.

But you'd think a group dedicated to this (BCHD) could figure it out. If the state is reporting by county, then why aren't they (BCHD) on the same page? And if the media is keeping the public informed, why isn't BCHD (who know all the names) doing a better job?

For something that's got the whole country in turmoil, you'd think this was kind of important.

Pretty much a "cluster" (and not the Covid kind).

trouble
How long do you want to ignore this user?
AG
LSCSN said:

cavscout96 said:

trouble said:

LSCSN said:

how hard is it to have a spreadsheet with the date, number of tests performed that day, and the positive/negative counts??


It has to be more involved than that to match tests with results. They all come back at varying intervals.
yes, they must be correlated to provide useful data, but it isn't that hard.


What's more involving about listing how many tests are performed each day and their subsequent result?


Because they don't all come back in the same time period. Positives typically come back faster than negatives. Different tests take different times to develop.

You would have to have someone checking each positive to find the testing date it matches.
jeffk
How long do you want to ignore this user?
AG
Interesting, thanks.
 
×
subscribe Verify your student status
See Subscription Benefits
Trial only available to users who have never subscribed or participated in a previous trial.