Many new cases share a surprising common risk factor - A previous positive test for COVID-19

a reply to: MotherMayEye

You are dead wrong. I have followed Florida numbers every day. On the graphic you posted it says right below the headings that only a first positive is counted. The number of new cases reported has always had second positives backed out. I.E. People testing positive a second time are not included in the total count.

What is happening instead is that the % of people testing positive the first time is extremely understated. While a second positive test doesn't count in those calculations, second negatives do indeed count when calculating that number.

For example. Lets look at 4/30 for palm beach. 143 positives and 63 new positives. Only the 63 new positives are added to the case numbers.

originally posted by: Gryphon66

originally posted by: OccamsRazor04
a reply to: hombero

Because it's not a reinfection. If you have 3 tests for the SAME infection, that is counted as 3 new cases. It's not, it's one new case.

But not universally; the retests have been excluded in some cases.

Which makes the data more dirty, but doesn't invalidate it.

The data is fine, it's their use of the data that needs to be changed, it's very misleading.

originally posted by: OccamsRazor04

originally posted by: Gryphon66

originally posted by: OccamsRazor04
a reply to: hombero

Because it's not a reinfection. If you have 3 tests for the SAME infection, that is counted as 3 new cases. It's not, it's one new case.

But not universally; the retests have been excluded in some cases.

Which makes the data more dirty, but doesn't invalidate it.

The data is fine, it's their use of the data that needs to be changed, it's very misleading.

Well, I disagree at least with the first part. The data is not "fine." The data has to be carefully "cleaned' before it should be used in any sort of scientific or medical publication (which means, if you don't know, that gross errors are removed or corrected in the given data set based on standard procedures).

As far as misusing data? Hell yes. On all sides.

originally posted by: Gryphon66

originally posted by: OccamsRazor04

originally posted by: Gryphon66

originally posted by: OccamsRazor04
a reply to: hombero

Because it's not a reinfection. If you have 3 tests for the SAME infection, that is counted as 3 new cases. It's not, it's one new case.

But not universally; the retests have been excluded in some cases.

Which makes the data more dirty, but doesn't invalidate it.

The data is fine, it's their use of the data that needs to be changed, it's very misleading.

Well, I disagree at least with the first part. The data is not "fine." The data has to be carefully "cleaned' before it should be used in any sort of scientific or medical publication (which means, if you don't know, that gross errors are removed or corrected in the given data set based on standard procedures).

As far as misusing data? Hell yes. On all sides.

By the data is fine I mean these states actually have the data on which tests are what. They are simply lumping them all together. I agree with what you are saying though the way they are using it is dirty.

originally posted by: sligtlyskeptical
For example. Lets look at 4/30 for palm beach. 143 positives and 63 new positives. Only the 63 new positives are added to the case numbers.

No, you are dead wrong. You can only find that data at the county level and ONLY since May 3. I quoted DeSantis and sourced to the county level reports.

The state dashboard still includes retests within their reported number of people tested and they do not indicate that retests are included. I have sourced this in my OP.

Please do post your source from the state dashboard for your claim that retests are excluded. You won't find one so I won't hold my breath.

originally posted by: Boadicea
a reply to: MotherMayEye

Great thread, Mom... good research and very well explained!!!

So there is very obviously an effort to inflate the numbers. I can't believe anything less than this is deliberate and calculated.

And interestingly enough, there seems to be an effort to discourage testing for antibodies. There was a recent ATS thread by a member who told of his experience requesting an antibody test, and being told they weren't accurate enough "yet" and not to waste his/her time.

My husband told me yesterday that his work was looking at having everyone who works in the field tested for antibodies, and they were told it's "too soon," because the virus has not run its course. This made no sense to me (or them), but it was through Concentra, which I think is mostly for industrial injuries and physicals and work related things. They may go through Quest Diagnostics instead. They think the virus rampaged through their shop in February, and just want to be able to tell their customers that their guys are not infectious when they go out on jobs.

Whether intended or unintended, the end result is the same: Ever increasing positive tests (new patients), but little to no antibody tests (past patients now immune).

Thanks Boadicea!

My impression, I think the antibody tests are more of a looming problem as far as new case data goes. Although many states are rolling those numbers into their case totals, those tests aren't as widespread...yet. But there are problems with some kinds of antibody tests -- namely they can't distinguish between four types of coronaviruses (IIRC). But as that kind of testing picks up, positive antibody tests will continue to be rolled in by many states and the CDC, too.

originally posted by: Gryphon66
To say it's "less credible" is very vague ... are you suggesting anything specific?

Yes. In the beginning, including retest results with the total cases was likely not all that statistically significant. But as time goes on, positive retests make up an increasing number of new/total cases.

Given that the number of 'new cases' are a core part of the rubric for reopening, then any state including retests with their daily new case numbers (and the CDC) are effectively making the reopening goals more difficult to accomplish.

In the alternative, there are reopening rubrics regarding tests done on a goal number of people. Including retests (negative and positive results) gives the false impression that more people are being tested than there actually are.

***

Full disclosure: Personally, I am not compelled by the reopening rules regarding the number of people tested. I find that to be a somewhat arbitrary 'rule' for reopening. As long as tests are available for anyone/employer that needs/wants them, then I am satisfied.

Can we get a compilation thread for all the ways data analysis and testing experts have bolloxed this whole pandemic response situation? Bonus points if you can dig up all the memos and notices that were ignored last year. Let's make formal appeal out of this mess.

a reply to: MichiganSwampBuck

Ah, that's ok...I was just concerned the thread might be derailed with talk of reinfections. So far as I know, reinfection hasn't been clearly established yet and/or it's thought that once people have the antibodies, they are protected for a period of time where reinfections would not be statistically significant.

originally posted by: Gryphon66

originally posted by: ketsuko
So this whole thing has royally exposed the Centers for Disease Control hasn't it?

They basically have one function, and they can't handle it. So what good are they? I am pretty sure the postmortem of this will see them crying in congress making claims that they'd have done better if only they had more money, but I'm also pretty sure that time and again their rules and regs and red tape just plain got in the way of timely response.

They also seem to have inadequate means of record keeping too and lack of some kind of infrastructure and system for reliably keeping and organizing that data.

Where have they been spending all the money they've been getting from us thus far? Perhaps an audit is in order.

Royally exposed the CDC? That's one take on it, surely. Vastly oversimplified of course.

The Trump Administration has been targetting the CDC for budget cuts, has meddled with their procedures, and has disregarded their warnings. Also, the methods of reporting have been controlled by the White House as well, to keep them in line with the President's "message."

That's another take on it. If you're interested in the budget, it's available online. I doubt that though; you're merely trying to move the responsibility from the Trump regime to the CDC.

That's okay; it's an opinion.

The CDC's budget allocation(750,000 range) for respiratory immunizations has been basically the same since 2015, they could have got every dollar they wanted and the allocation for this situation would have been about the same.

2015 budget for RI $748,066
2020 budget for RI $730,231

in comparison

$1,318,056 for HIV and AIDS prevention

$628,839 for injury prevention and control

$825,000 for Public health and preparedness response

The largest cut was 167,000 on immunization programs. Obviously some of this may or may or may not made a difference in their COVID response and effectiveness. Point being they have had about the same percentages of the total budget allocated to respiratory diseases since 2013.

originally posted by: Asktheanimals
Pandemic? $2.3 trillion "lost"?
Simple accounting errors.
Oh gosh darn it, we'll fix it next time.

NYC is presently remaining shut and suffering catastrophic economic loss because they need 3%-4% more hospitaal beds and ICU beds and local government needs to hire 1000-2250 contact tracers to make phone calls.

That's it. All the other goals have been met. It's not the virus keeping NYC closed.

Fulfilling those goals is a drop in the bucket compared to broad economic losses the city is facing. Are all hands on deck to get those last hospital and ICU beds in place and hire those contact tracers? People can make phone calls from home to contact trace, even. Offer triple the salary and NYC would still be effectively saving money. Residents are leaving en masse, not paying rent, office rent is not being paid, tax revenue loss, etc...

originally posted by: ketsuko
So this whole thing has royally exposed the Centers for Disease Control hasn't it?

There have been some inexplicable actions and decisions coming from them considering the center's Mission and Pledge to the American People:

Mission

CDC works 24/7 to protect America from health, safety and security threats, both foreign and in the U.S. Whether diseases start at home or abroad, are chronic or acute, curable or preventable, human error or deliberate attack, CDC fights disease and supports communities and citizens to do the same.

CDC increases the health security of our nation. As the nation’s health protection agency, CDC saves lives and protects people from health threats. To accomplish our mission, CDC conducts critical science and provides health information that protects our nation against expensive and dangerous health threats, and responds when these arise.

Pledge to the American People

• Be a diligent steward of the funds entrusted to our agency
• Provide an environment for intellectual and personal growth and integrity
• Base all public health decisions on the highest quality scientific data that is derived openly and objectively
• Place the benefits to society above the benefits to our institution
• Treat all persons with dignity, honesty, and respect

Link

a reply to: Gryphon66

That money lapsed before the renewal bill was passed. And often, the "cuts" are reductions in the annual assumed rates of budget increase that every department assumes it will get. That isn't actually a "cut" just less of a raise.

Is it true or not that part of the reason why we didn't go with the test kits the rest of the world was using was because the CDC wanted to go with their own, more elaborate ones? Those later proved to be less reliable. Then one of the reagents wound up being botched. All of that slowed down testing.

It was regulations that prevented many places from being able to become testing centers. There was one place in Washington that was doing influenza research. They had dozens of nasal swab samples already on hand, and they were capable and able to begin testing those samples for COVID-19, but the CDC and FDA shut them down because they weren't an FDA approved site for that kind of testing.

The CDC insisted on a centralized, highly regulated and regimented process that ended up being clunky and wholly inadequate for tracing and controlling the rapid spread of this disease.

Usually the people who have these high ranking government jobs and high education do not have much common sense. Memorization of things is different than IQ. They believe they are smart though, but are often dumb as a box of rocks. Same with the people working under our governor, they cannot see that the governor is destroying our economy and destroying people's lives, they can only see facts, figures, graphs, and cannot see the whole picture. Finally they started to see that more people are hurt by shutting down hospitals for people who need surgeries than are getting hurt by the virus.

Also, people got a second chance to try to adapt and do not feel they actually need the surgeries anymore and are getting better...no need for the knee replacement anymore...and they are telling their friends...so hospitals will be hurting for a long time. Same with therapies, people healed fine without the therapies, they tell their friends.

Gee, it is dangerous when over a quarter of the countries jobs are in healthcare and people discover that it is not needed nearly as much as they thought. They did research and found other ways to fix their problems....because they were stuck at home and did not have to work...handy, the medical industry will not be recovering very well.

a reply to: MotherMayEye

See, if you count the same people multiple times then your Case Fatality Rate is greatly decreased.

a reply to: infolurker

Yes, it would seem to imply that. However, there are issues with reporting COVID-19 deaths, as well. Other threads cover some of those issues.

originally posted by: MotherMayEye
a reply to: MichiganSwampBuck

Ah, that's ok...I was just concerned the thread might be derailed with talk of reinfections. So far as I know, reinfection hasn't been clearly established yet and/or it's thought that once people have the antibodies, they are protected for a period of time where reinfections would not be statistically significant.

I don't have a problem being corrected if I'm wrong or misinterpreting something, I'm glad neither of us is copping an attitude, that is very cool. Also, I'm glad you are making this important distinction, it makes things clearer. Have a good one.

I bet they are including at least some if not all retests.

For them to be excluded, they have to be able to track who is taking the test. The only way to do that is give each person a unique number that they use for retests or go by the social security number.

I took two relatives for testing. One at the National Guard testing site and the other at a private doctor. The National Guard only asked for phone #s. So yes, they could very well double count.

Maybe they have a way to exclude hospitalized patients or ones in nursing homes.

See, if you count the same people multiple times then your Case Fatality Rate is greatly decreased.

a reply to: infolurker
Absolutely correct . Lets say half the confirmed cases were retests the US mortality rate jumps to 11.76 percent . But its just the flu right .

a reply to: hutch622

Current outcomes:

www.worldometers.info...

Deaths:
100,572

Recovered:
479,969