re: Alabama Board Coronavirus Thread

quote:

---

I don't even pay attention to *just* the cases. Main one I look for is hospitalizations. Thats a metric that can't be politicized and skewed. Either hospital capacity is overloaded or its not, and locally its easier to tell

---

.

Spot on. Exactly my opinion too.

As far as case number reporting goes, it is silly tin foil hat stuff to think there will be a sudden change either way. Each state is still going to report their individual numbers through their own public information portals. The states are the sources for the CDC data anyway. Every news outlet and millions of citizens will do what numerous sites and people already do and compile their own totals off state sites.

The fact that this is story is more than a side note shows the politization of every single aspect of this thing.

Been to their site? I was looking a couple of days ago and one part of it showed no covid deaths this month.

It is a hot mess as shown by the story linked earlier.

Do I think reporting will be better going through the White House first? No, I do not. The data the CDC gets from some states has been shown to be a hot mess too.

Why change? Ask the administration. They decided they wanted it first and that is their prerogative. The point is a spreadsheet and 50 mouse clicks (or just one and let Hopkins do the rest for you) gives anyone the ability to cross check - and you better believe every national media outlet will do it and run stories if there are any significant differences.

It varies by state and sometimes week by week. Alabama's has a good bit of data but it can be hard to find. The links are a bunch of buttons - all labeled with a number instead of description telling what is behind each one. There also doesn't seem to be logic behind some of the ways the data is presented.

We're about to get ready to start the school year. Our county is giving a virtual or in person option. So, I wonder how common Rona is among school age kids in Alabama. Reasonable question, right? No can have. ADPH gives you numbers for 0-5 and then 5-24.

There's also an overriding question of reliability. I now know two asymptomatic people that had positive tests and then within a couple of days had back to back negative tests (NASCAR's Jimmy Johnson did the same thing). I can only assume they counted as positives even though they didn't actually have it. On the flip side, I know 3 sisters that went to a family funeral in Florida. All 3 developed symptoms - 2 tested positive and 1 tested negative. The one that tested negative self quarantined for two weeks because she knows damned good and well she had it too.

As stomp said, the 'total cases' number doesn't mean crap. They're poorly collected, poorly reported, and unreliable. The number that tells the truth is hospitalizations. They tell the story in a way that can't be spun or misreported.

quote:

---

As stomp said, the 'total cases' number doesn't mean crap. They're poorly collected, poorly reported, and unreliable. The number that tells the truth is hospitalizations. They tell the story in a way that can't be spun or misreported.

---

I wouldn't say they don't mean crap... just less accurate day to day than hospitalizations. Total cases can be a leading indicator of hospital capacity...a warning shot of sorts. The 7 day average trend line of total cases is a better way to look at them, rather than "OMG 65k new cases!"

Not necessarily to you and I might be late to the party but, how do they figure the "Probable" cases per day? Is it people showing symptoms or is it contact tracers or what? I have no idea how that works.

quote:

---

Not necessarily to you and I might be late to the party but, how do they figure the "Probable" cases per day? Is it people showing symptoms or is it contact tracers or what? I have no idea how that works.

---

It's basically that the person meets a couple of criteria like being sick with Covid symptoms + exposure to someone tested positive but they have not had a laboratory confirmed positive test.

It's one of the areas where states are counting and reporting differently.

It's criticized by voices that want case numbers to appear as low as possible but encouraged by scientists who think it provides much more accurate data for research and case tracing.

quote:

---

the 'total cases' number doesn't mean crap.

---

I hope you’re just being hyperbolic here. Those numbers do have tremendous worth. I think most people that have been following this closely recognize there are some shortcomings with the reported numbers. There is however a strong belief among many experts that the reported cases are actually under reported. Who knows if the 2 balanace each other out, but it’s fair to say there is an unknown margin of error in those numbers. That doesn’t make the numbers useless. They can show where but spots currently are, can be used to predict where hospitalization numbers may increase, if they haven’t already, etc. they are definitely not the end all be all number though.

Another study about another old medicine that seems to neutralize Covid.

LINK

The important thing about tested cases is exactly identifying trends and hotspots and that is the data most useful to regular people.

When I can see that only 15 of 2.5 million people a day are now testing positive in Brooklyn, it means I'm now hiring back a couple of more people for my shop and extending hours back closer to normal. I'm sure there are millions of people doing that on a much larger scale than me.

When I see that 30 a day are now testing positive in my Mom's small town in Alabama, I'm doubling down on advice to her to keep staying at home aside from absolute necessities.

I agree with you about the importance of hospital rate and death rate in other decisions but I'm not basing either of my decisions above on those things. I'm basing it on how rampant the spread seems to be in a specific place.


That the numbers may not be perfect is something everyone realizes. I mean we know for a fact that it's impossible to test every single person so large amounts of people have it and are untested.

quote:

---

how hard it is to get a semi-accurate count for something this important?

---

I think the data is fairly accurate. There are mistakes made, and there is certainly some inflation of positive cases from people getting tested more than once, but I also think both of those are fairly minimal and not the norm. I also think looking at 7 or 14 day trends negates any discrepancies in daily numbers from state to state due to different reporting guidelines. I've always paid more attention to those trend lines than daily numbers anyway.