Lots of good news today

[ERD50]

Originally Posted by ERD50 View Post
..... Bad news is, the local media and politicians are focused (like many in this thread) on "cases! cases! cases!!! People, cases are a function of testing. "Cases" w/o context is misleading, very misleading. More tests can mean more "cases", even if things are improving. ...

But aren't cases indicative of people who have been tested and confirmed to be infected with the virus... and were contagious for some period before they tested positive and will be contagious for some period after they test positive?

If so, wouldn't more cases (especially recent cases) possibly be a future indicator of spread?

Now OTOH, of those who do get infected less hospitalizations and deaths is obviously good news.

Not necessarily, it is simple arithmetic. [NOTE, the following is just hypothetical to illustrate the point] If one week say you test 10,000 people and come up with 4,000 cases, there's your number. What if the next week, you test 20,000 people, and come up with 5,000 cases? That's more cases, more recent cases. But are things getting worse? All else being equal, we'd expect 8,000 cases (twice the number of cases from twice the number of tests). So we had 3,000 fewer than expected.

But if the media reports "cases are up by 25%!!!!!", and provides no other context, it sounds bad, while it appears we are doing much better ( a 25% infection rate among those recently tested, versus a 40% infection rate the week before).

In my career, if I provided action items based on the "number", rather than percentages, well, I wouldn't have been able to retire, I would have been let go and would have had to find a minimum wage job. This isn't rocket science, it's 5th grade arithmetic.

I know that Hanlon's Razor states: Never attribute to malice that which is adequately explained by stupidity. But I really hope the people reporting these numbers are not that stupid. I suspect a motivation behind it. Just like plane crashes are news, but 100,000 safe landings are not. But I'll leave it at that.

-ERD50

 

[harllee]

Not necessarily, it is simple arithmetic. [NOTE, the following is just hypothetical to illustrate the point] If one week say you test 10,000 people and come up with 4,000 cases, there's your number. What if the next week, you test 20,000 people, and come up with 5,000 cases? That's more cases, more recent cases. But are things getting worse? All else being equal, we'd expect 8,000 cases (twice the number of cases from twice the number of tests). So we had 3,000 fewer than expected.

But if the media reports "cases are up by 25%!!!!!", and provides no other context, it sounds bad, while it appears we are doing much better ( a 25% infection rate among those recently tested, versus a 40% infection rate the week before).

In my career, if I provided action items based on the "number", rather than percentages, well, I wouldn't have been able to retire, I would have been let go and would have had to find a minimum wage job. This isn't rocket science, it's 5th grade arithmetic.

I know that Hanlon's Razor states: Never attribute to malice that which is adequately explained by stupidity. But I really hope the people reporting these numbers are not that stupid. I suspect a motivation behind it. Just like plane crashes are news, but 100,000 safe landings are not. But I'll leave it at that.

-ERD50

In my state we saw an increase of positive cases back in June but we had more tests also so that did not tell us much. What did tell us something was a few weeks later we saw an increase in hospitalizations and then a few weeks after that we saw an increase in deaths--that did tell us that things were getting worse. Fortunately when the cases started increasing our Governor imposed a statewide mask requirement. Now our positive cases have started to drop and I am hopeful that our hospitalizations and deaths will start to drop soon.

There is a lag time in Covid. In most places cases start increasing and then a few weeks later hospitalizations start to increase and then a few weeks later deaths start to increase.

 

[pb4uski]

Not necessarily, it is simple arithmetic. [NOTE, the following is just hypothetical to illustrate the point] If one week say you test 10,000 people and come up with 4,000 cases, there's your number. What if the next week, you test 20,000 people, and come up with 5,000 cases? That's more cases, more recent cases. But are things getting worse? All else being equal, we'd expect 8,000 cases (twice the number of cases from twice the number of tests). So we had 3,000 fewer than expected.

But if the media reports "cases are up by 25%!!!!!", and provides no other context, it sounds bad, while it appears we are doing much better ( a 25% infection rate among those recently tested, versus a 40% infection rate the week before).

In my career, if I provided action items based on the "number", rather than percentages, well, I wouldn't have been able to retire, I would have been let go and would have had to find a minimum wage job. This isn't rocket science, it's 5th grade arithmetic.

I know that Hanlon's Razor states: Never attribute to malice that which is adequately explained by stupidity. But I really hope the people reporting these numbers are not that stupid. I suspect a motivation behind it. Just like plane crashes are news, but 100,000 safe landings are not. But I'll leave it at that.

-ERD50

I dunno, I guess if you assume that if you had tested 20,000 in that first week that you would have had 8,000 cases then perhaps... and we all know about assuming... you'll never really know since you didn't test 20,000 cases that first week.

It seems to me that knowing that there are 5,000 people that have been out and about infected is worse than knowing that there are 4,000 people that have been out and about infected.

I think you're too sensitive to the reporting on this. BTW, many reporters are blooming idiots so it might well be stupidity rather than some ulterior motive.

Your plane crashes are news but safe landings are not is very strange... you don't seriously expect the press to cover safe landings do you?

 

[Gumby]

The virus has its own natural course and the number of actual infections will be what it is, regardless of what is reported. I mean, we're not talking Schrodinger's cat here - where the cat is both infected and not infected until the moment we test and establish the quantum state.

If you look at rt.live, they adjust their results for testing volume and use running averages to smooth the daily fluctuations. So the information is out there if you are interested.

As far as the press goes, they are most certainly not experts in either epidemiology or statistics. And shocking headline numbers sell newspapers. Nuance doesn't.

 

[W2R]

You want lots of good news today?
This article says Louisiana's second COVID-2 peak is in clear decline at the moment:

Louisiana is in clear decline from its second coronavirus peak, with daily numbers of new cases and other important indicators falling in every part of the state.

Some examples given were that

Fewer than 8,000 new cases were reported across the state during the past week, a drop of more than 25% from the week before and a 46% decrease from the 15,025 cases diagnosed in the week ending July 27. So far, that week saw the highest number of cases reported during the outbreak.

The percentage of tests coming back positive has also been dropping, falling to 6% this week.

Also,

Hospitalizations have also been on a steady, if not unbroken, decline from a peak of 1,600 near the end of July.

The article suggests these encouraging changes may be due to behavioral factors such as improved compliance with masking, social distancing, and similar restrictions, but from what I have read (and seen locally) this seems to be the opposite of what is happening.

 

[Badger]

...The article suggests these encouraging changes may be due to behavioral factors such as improved compliance with masking, social distancing, and similar restrictions, but from what I have read (and seen locally) this seems to be the opposite of what is happening.

So many variations of how people are responding in this country as well as world wide. It increases the difficulty with how to manage the pandemic when we don't have a control group or standard to compare to every isolated behavior variable. This virus is sneaky and we are still somewhat in the dark as to how it behaves regarding transmission, rate of mutation, its "Achille' heel", host age susceptibility, etc. At this point in time we are still making hypotheses.


Cheers!

 

[explanade]

What did LA do to try to lower the infections?

 

[timo2]

New Mexico had a quarantine placed on the state back in April. Anyone entering had to quarantine for 14 days; we changed our travel plans in June because of it. It is still in effect.

Hawaii also had a 14 day quarantine period for arriving residents and visitors. There were arrests made for folks who violated the quarantine, specifically those who stopped for groceries on their way home. It is still in effect.

New Zealand has also been in complete lockdown for months, their rate was extremely low for cases and deaths. Their lockdown was still in place, and just got stricter.

All three now have growing amount of cases and deaths, despite their efforts.

Regarding NM, things have improved enough that New York just removed NM off of their travel advisory list. "Governor Andrew Cuomo announced Tuesday that five states have been removed from New York’s 14-day quarantine travel list.

The states that have been removed from the 14-day mandatory quarantine travel list are:

Alaska
New Mexico
Ohio
Rhode Island
Washington"

https://www.wgrz.com/article/news/l...-list/71-b98acc74-29d1-40ae-9eae-0023a2d360a8

But Hawaii, South Dakota, & Virgin Islands were added on to the list.

 

[ut2sua]

Good news/bad news in IL.

Good news is daily deaths are way down, about 4x fewer than in May/June , and about flat at that lower level for over a month

More good news is that hospitalizations are also down to about 3x fewer than in May, and have been flat at this lower level for about a month and a half.

Bad news is, the local media and politicians are focused (like many in this thread) on "cases! cases! cases!!! People, cases are a function of testing. "Cases" w/o context is misleading, very misleading. More tests can mean more "cases", even if things are improving. The case numbers are getting people worked up for all the wrong reasons. Not only the number of tests, but the profile of the group being tested could be another variable (it hasn't been a random cross section at anytime, as far as I know).

Even the hospitalizations and deaths are somewhat questionable, as we can't be sure the same methodology has been followed for the reporting, but it is far more meaningful than "cases".

edit/add: And he news this morning (I generally tune out, I'm so frustrated at the bad reporting), keeps reporting single day numbers. Look at those graphs, the daily numbers are all over the place. Look at 7 day averages (I wish they had 30 day as well). I'm used to working with "noisy" data from my career, you need to apply some analysis to it or you come to very bad conclusions.

https://covidtracking.com/data/state/illinois#historical

-ERD50

Interestingly, from another angle, the "bad news" may help to bring about the "good news". As when folks realize that there are xxxxx number of infected folks out there, they will follow safe practices more strictly, thus effectively reducing death numbers. We need to keep in mind the improved death numbers also got help as our treatment improved (reducing the death rate). The increasing-in-cases reporting remind us to be more careful to avoid getting infected. Would you agree that is a good side of the "bad news"? By not looking at things from a political angle, it may help you to get less frustrated (?)

 

[Car-Guy]

Regarding NM, things have improved enough that New York just removed NM off of their travel advisory list. "Governor Andrew Cuomo announced Tuesday that five states have been removed from New York’s 14-day quarantine travel list.

The states that have been removed from the 14-day mandatory quarantine travel list are:

Alaska
New Mexico
Ohio
Rhode Island
Washington"

https://www.wgrz.com/article/news/l...-list/71-b98acc74-29d1-40ae-9eae-0023a2d360a8

But Hawaii, South Dakota, & Virgin Islands were added on to the list.

At least Texas is still on their list.... From my POV they can keep Texas on the restricted list long after the virus is gone.

 

[eytonxav]

At least Texas is still on their list.... From my POV they can keep Texas on the restricted list long after the virus is gone.

+1, also hope Texas does that to the West coast states.

 

[Aerides]

mod note:

Turning this thread into yet another political state-bashing "mine's better, stay in yours" diatribe, is a great way to get this thread closed.

 

[ERD50]

catching up from a few days away...

I dunno, I guess if you assume that if you had tested 20,000 in that first week that you would have had 8,000 cases then perhaps... and we all know about assuming... you'll never really know since you didn't test 20,000 cases that first week.

It seems to me that knowing that there are 5,000 people that have been out and about infected is worse than knowing that there are 4,000 people that have been out and about infected. ...

As Gumby said (bold mine),

The virus has its own natural course and the number of actual infections will be what it is, regardless of what is reported. I mean, we're not talking Schrodinger's cat here - where the cat is both infected and not infected until the moment we test and establish the quantum state.

And he did it with style, with the "Schrodinger's cat" reference

The safe airplane landings was just my way of saying, the "news" goes for sensationalism.

So in one way, more information can be good. But as I learned in my career, only if you have a plan for how to use that data. Otherwise, it just swamps you, and sometimes you get pushed into analysis paralysis, or just become numb to the numbers.

So unless they are doing something specific with those extra found cases, like serious contact tracing (which I don't think is really happening much in any meaningful way, just too many cases to track), then no, it does no good to "know" (through testing) that there are 5,000 versus 4,000. Sampling theory could have easily told you what the number was with a great degree of confidence (if you test a random sample and get X% positive, you can assume X% applies to a larger population).

Interestingly, from another angle, the "bad news" may help to bring about the "good news". As when folks realize that there are xxxxx number of infected folks out there, they will follow safe practices more strictly, thus effectively reducing death numbers. We need to keep in mind the improved death numbers also got help as our treatment improved (reducing the death rate). The increasing-in-cases reporting remind us to be more careful to avoid getting infected. Would you agree that is a good side of the "bad news"? By not looking at things from a political angle, it may help you to get less frustrated (?)

No. Here's the problem I have with that - numbers tell a story, numbers don't have an opinion, bias, or an agenda. They are what they are. If "someone" decides that sensationalizing/misrepresenting the numbers is "good", then the public is subject to that "someone's" opinion, bias, or agenda. What about some other "someone" who believes just as fervently in the opposite view? Do they get to decide what is "good" for us? Then who decides who gets to decide?

No, we should be told the cold, hard facts, w/o spin. Then we can decide. The "positive" spin can be counter-productive as well - if you tell someone "X" is great, then they learn that "X" is not so great, and you were misrepresenting it, then how about when the truly great "Y" comes along? The public might not take you seriously. Best to be straight about it, in all cases.

So no, the end does not justify the means, regardless of (or maybe especially because of) "good" intentions.

-ERD50

 

[ut2sua]

Bad news is, the local media and politicians are focused (like many in this thread) on "cases! cases! cases!!! People, cases are a function of testing. "Cases" w/o context is misleading, very misleading. More tests can mean more "cases", even if things are improving. The case numbers are getting people worked up for all the wrong reasons. Not only the number of tests, but the profile of the group being tested could be another variable (it hasn't been a random cross section at anytime, as far as I know).

Even the hospitalizations and deaths are somewhat questionable, as we can't be sure the same methodology has been followed for the reporting, but it is far more meaningful than "cases".

-ERD50

No. Here's the problem I have with that - numbers tell a story, numbers don't have an opinion, bias, or an agenda. They are what they are. If "someone" decides that sensationalizing/misrepresenting the numbers is "good", then the public is subject to that "someone's" opinion, bias, or agenda. What about some other "someone" who believes just as fervently in the opposite view? Do they get to decide what is "good" for us? Then who decides who gets to decide?

No, we should be told the cold, hard facts, w/o spin. Then we can decide. The "positive" spin can be counter-productive as well - if you tell someone "X" is great, then they learn that "X" is not so great, and you were misrepresenting it, then how about when the truly great "Y" comes along? The public might not take you seriously. Best to be straight about it, in all cases.

So no, the end does not justify the means, regardless of (or maybe especially because of) "good" intentions.

-ERD50

My bold. I thought your initial idea was the opposite from what you have just mentioned here. That you didn't want "cold, hard" numbers (ie. case number) without the "correct" context. I never suggested that we should make up case number to get people's attention.
I do think that people should be told the real numbers (both cases and deaths). By ignoring the case number (as you have implied), folks will be left with half the story. So, by ignoreing the case numbers, your thinking is tilted just like that of the "bad" media and politicians. You also questioned the hospitalization and death numbers, so I guess "cold, hard" facts need to fit into your own context?
Frankly, I am not trying to change your thinking. Just want to point out the double sided rationale in your argument.

 

[ERD50]

My bold. I thought your initial idea was the opposite from what you have just mentioned here. That you didn't want "cold, hard" numbers (ie. case number) without the "correct" context. I never suggested that we should make up case number to get people's attention.
I do think that people should be told the real numbers (both cases and deaths). By ignoring the case number (as you have implied), folks will be left with half the story. So, by ignoreing the case numbers, your thinking is tilted just like that of the "bad" media and politicians. You also questioned the hospitalization and death numbers, so I guess "cold, hard" facts need to fit into your own context?
Frankly, I am not trying to change your thinking. Just want to point out the double sided rationale in your argument.

No, no, no. I did not imply we ignore the case numbers, I'm saying they don't provide useful information w/o context, and the media/politicians are shouting "cases!" without the meaningful context. That doesn't help anyone.

It's not double sided at all, context is everything.

It's like saying our factory made 10,000 widgets on yesterday's shift, but only 5,000 today. OMG! Terrible, production is down 50%!!!!

Oh wait, some context - we shut down to install some new equipment (and had a training session for people, so their time wasn't wasted), got the equipment running and checked out, and those 5,000 were made in the last hour. So tomorrow, we expect to do 40,000 on the shift.

So what sounds like bad news can be very good news with some context. It happens all the time.

And to your last point, sure, all the data is questionable to some degree - there are subjective measures to it, and apparently some changes in protocol/procedures. That doesn't mean the data isn't useful, but you need to keep in mind this isn't something as straightforward as weighing something with a calibrated scale, where you can have assurance of 1% accuracy. You need to consider all these effects before declaring anything better/worse/indifferent.

Here you go:

https://covidtracking.com/data/state/illinois#historical

The 7-day average of new cases in IL have about tripled since mid-June. But deaths are about 1/3rd (a 2/3rds reduction) over that time, a 6:1 disconnect. Hospitalizations are about flat, so a 3:1 disconnect. Certainly some context is needed here, the numbers are vastly different.

-ERD50

 

[RobbieB]

It'll take a few weeks for the new cases to get sick enough for the hospital.

 

[ERD50]

It'll take a few weeks for the new cases to get sick enough for the hospital.

Yes, hospitalizations and deaths are a lagging indicator.

But as mentioned earlier by a few of us, which "new cases" are we talking about? All new cases will have some percentage that will result in hospitalizations and deaths. That is true whether the "new case" was found by testing or not found by testing.

Heck, if it was that simple, we should stop testing altogether! Fewer new cases!

That is what is misleading about focusing on "new cases", w/o also providing some context regarding the amount of testing that was done. There was a solid increase in new cases in IL going back about 2 months, but those haven't shown up (at least not yet), in hospitalizations and deaths.

I'm sure part of the lower death rates are improved treatment knowledge. I also wonder if many of the weakest have already succumbed. It would be interesting to see if the age distribution of deaths in IL has changed over the past 3 months or so.

-ERD50

 

[Montecfo]

Good news on vaccines

Pfizer and BioNtech announced that their 2nd vaccine candidate in phase 1 trial showed effectiveness and may be safer than its other candidate.

https://finance.yahoo.com/news/pfizer-biontech-second-coronavirus-vaccine-022707876.html

Their earlier candidate could be submitted for approval as early as October.

https://www.bloomberg.com/news/articles/2020-08-21/pfizer-virus-vaccine-on-track-for-regulatory-review-in-october

 

[JoeWras]

Pfizer and BioNtech announced that their 2nd vaccine candidate in phase 1 trial showed effectiveness and may be safer than its other candidate.

https://finance.yahoo.com/news/pfizer-biontech-second-coronavirus-vaccine-022707876.html

Their earlier candidate could be submitted for approval as early as October.

https://www.bloomberg.com/news/arti...ine-on-track-for-regulatory-review-in-october

My friend had a version of the Pfizer vaccine in a trial. Her major side effect was significant arm pain, as in needing to ice it and hold it down for 2 days. Maybe it was B1. She's pretty sure she didn't get the placebo. BTW, she's fine now.

Turns out a bunch of our friends have been enrolled in various trials. We didn't get the memo because we aren't on their facebook or something.

Lots of activity for trials here in central NC.
https://www.newsobserver.com/news/coronavirus/article245077375.html

 

[Aerides]

Yes, hospitalizations and deaths are a lagging indicator.

But as mentioned earlier by a few of us, which "new cases" are we talking about? All new cases will have some percentage that will result in hospitalizations and deaths. That is true whether the "new case" was found by testing or not found by testing.
-ERD50

I don't care to parse all the details, and I'm baffled by the "testing" arguments (don't try to explain it to me again, doesn't matter).

All that matters to me:
In FL for about 4 weeks, we were getting 10k cases per day. Lagging that by about 4 weeks, we had about 100-200 deaths per day. The cases have leveled off, the deaths still too soon to call.

It won't make me feel better if there were really 100k cases per day, because, me personally, I'm trying to make sure that my family and I stay in the zero-cases-per-ever category.

 

[Winemaker]

It won't make me feel better if there were really 100k cases per day, because, me personally, I'm trying to make sure that my family and I stay in the zero-cases-per-ever category.

I sincerely doubt that will ever happen. We are dealing with an extremely contagious virus that affects everybody in various ways. Isolated areas with strict lockdowns such as New Zealand and Hawaii now have increasing cases, but not alarming numbers. Prisons and nursing homes in the US are still the leading primary cases, and they are also under lockdown. The virus or its mutation, is out there, and it is going to get us, one way or another.

 

[Montecfo]

Suggestion: there are.many threads where people debate numbers. Can we keep this thread to the selected topic?

 

[Gumby]

Happens all the time.

 

[Montecfo]

Home sales increase almost 25 percent.
Median price up 8.5 percent.

https://www.cnbc.com/2020/08/21/july-home-sales-spike-a-record-24point7percent-as-prices-set-a-new-high.html

 

[Chuckanut]

A fishing vessel with a crew of 122had the coronavirus pop up and it infected 100 members of the crew while at sea. Three crewmen had already had CV. None of them got it again. Apparently, they did have some useful level of immunity.