Jump to content

Menu

Recommended Posts

3 hours ago, lewelma said:

I'm still catching up on the thread, but I can answer questions about modelling.  My PhD is in mathematical/statistical modelling of population dynamics.. 🙂 

Time series models make notoriously poor forecasts. Because the infections today are based on the infections yesterday which are based on the infections the day before, it is like the butterfly effect.  They are *very* sensitive to initial conditions.  Basically, they compound the error in the model for every day out you try to predict.  So if you are building your model out of daily data (rather than hourly or monthly), you can make a decent prediction with large error bars ONE day out. By 2 days out, all the errors compound. 

In addition, the models can be built from both a data-fitting approach or from a biology-driven approach.  The data fitting approach is only as good as the data we have, which is obviously poor given the lack of testing, and poor given the uncertainty of the compliance with the lockdown.  The biology-driven approach is only as good as the knowledge about the virus, and that is poor.  They are not clear as to death rate or infection rate or spread conditions.  So neither approach will be great, so I'm sure they are trying both, and interweaving them.  

The assumptions required to be made for these types of models are just huge, and each modeller is trying different things.  I'm sure that they are also holding back current data, to then test the robustness of their model, and then if the forecasts are good, they will use those parameters with the full data set.  But clearly, they also have a problem with the variability in the data.  If all things were known, the growth of the virus would be completely smooth, but even the data coming out of NZ which has good testing, is quite ragged. So their models might only explain 70% of the variability in past data, and all their forecasts beyond one day, predict based on the forecasts from the previous day, etc. So compounding the uncertainty to make predictions useless.

There is also a massive problem with the spatial component. Every region in the US has a different type of lockdown and different compliance.  It is really fractal, with an equal amount of variability at each spatial scale measured.  This is Huge, and close to impossible to deal with.  Each model really needs to be for a very small physical area, and even then there would be variability in both the data for the dependent and independent variables. That is a crazy large number of models for a place as big as the US. Each of the models we see being published are clearly using a subset of the data in a constantly moving landscape.

I think the models were more consistent in the beginning because there was really only one approach for the independent variables and that was 'do nothing.' So a pure spread.  Now that there are all sorts of lockdown rules, that are being followed or not, that really shakes up the models and makes them even less accurate because the data over the time scale you are working on has the independent variables changing at different points along the x axis.  This makes for a messy situation, because when you add in more and more variables to explain everything, you can actually make your model less and less accurate because you have overfitted it. 

Finally, I spent 4 years of my life trying to make a reasonable model. Sure I was a newbie to modelling and the biology of my animal, but doing these models in just weeks is a BIG ask. The understanding of the biology keeps changing -- so the time based relationship between the data in the dependent variable. And the independent variables keep changing  -- lock down, masks for the public, etc. You can say that there are just a crazy number of modellers out there all working together, but 9 women cannot make a baby in a month.  

Point being. The models are no good, and I have been ignoring them.

I also notice that in Australia they are claiming to be flattening the curve.  I feel like that’s not really accurate because growth in cases till now has been mostly imported.  International Travel ban might flatten that curve (or stop those seeds being sown) but community spread is a separate curve that’s still growing.  Would that be correct?

  • Like 3
Link to post
Share on other sites
  • Replies 16.7k
  • Created
  • Last Reply

Top Posters In This Topic

  • Ausmumof3

    3093

  • Pen

    2366

  • Arcadia

    1337

  • prairiewindmomma

    305

Top Posters In This Topic

Popular Posts

DS got home 3 hours ago!  ❤️❤️

Update-  my youngest is not only short of breath, coughing, dizzy, nausaues, and with headache-  she is also confused.  I called our doctor and talked with him and she is going to be going to the ER.

That's not a blanket right.  If my religion required human sacrifice, I can't practice it.  If my religion required sexual assault, I can't practice it. Freedom of religion isn't a blanket right

Posted Images

1 minute ago, Ausmumof3 said:

I also notice that in Australia they are claiming to be flattening the curve.  I feel like that’s not really accurate because growth in cases till now has been mostly imported.  International Travel ban might flatten that curve (or stop those seeds being sown) but community spread is a separate curve that’s still growing.  Would that be correct?

They would not use total numbers in their models, because it would muck up the infectious rate they are using.  When you build a model, the population has dependent factors that link the population size to the previous day based on the biology of the animal. The imported cases won't work with that biology.  So they are likely to be adding them in as a separate variable, that then in the following day integrating them into the biological relationship.  However, here in NZ the infection rate will be different for the imported cases who are under quarantine and the community spread cases.  So they may be getting their dependent variable as a sum of different sub-variables that are linked to each other in time using different time-linked relationships. 

  • Like 2
Link to post
Share on other sites
11 minutes ago, Ausmumof3 said:

I also notice that in Australia they are claiming to be flattening the curve.  I feel like that’s not really accurate because growth in cases till now has been mostly imported.  International Travel ban might flatten that curve (or stop those seeds being sown) but community spread is a separate curve that’s still growing.  Would that be correct?

Oh, I don't think I answered your question.  I agree with you.  If the growing case load in Australia was originally based on imported cases, and now that is stopped so the growth rate has declined, that says nothing about community spread.  Flattening the curve is all about community spread.  Have you guys been testing for this?  

  • Like 2
Link to post
Share on other sites
5 minutes ago, lewelma said:

Oh, I don't think I answered your question.  I agree with you.  If the growing case load in Australia was originally based on imported cases, and now that is stopped so the growth rate has declined, that says nothing about community spread.  Flattening the curve is all about community spread.  Have you guys been testing for this?  

Only in cases of significant unexplained pneumonia.  Two states were expanding testing for any unexplained respiratory illness from yesterday I think and ACT are starting to do sentinel testing.  I’ve forgotten the finer details but basically up until Sunday not really.  They have detected some community spread even with those criteria.  I know at least two people whose GPS wanted testing but were refused because they didn’t mean the international travel criteria.

  • Like 2
Link to post
Share on other sites

Abc: Ms Berejiklian (NSW premier) was blunt when talking about how long we're in this for:

  

"Because the reality is that until we find a vaccine, we all have to live with this virus. And no matter what restrictions there are in the future, no matter what restrictions are potentially eased in the future, until a vaccine is found, social distancing is a way of life now. That is the new normal. Not having physical contact, socially distancing is, at the very minimum, a way of life for us. Because we have to be honest with ourselves — until a vaccine is found, there is no simple way to deal with this issue," she said.

  • Like 1
  • Sad 4
Link to post
Share on other sites

You also have the problem with international travellers that they are a different subset of the population (often younger) so the death rate will also be different than for the general population.  Very tricky to integrate all these different variables into the data.  You actually could run into a degrees of freedom problem with too many variables and not enough data.  The data points are by DAY (not by hour) as far as I can tell. So you have very limited data to build a complex model. 

  • Like 1
  • Thanks 1
Link to post
Share on other sites
Just now, lewelma said:

You also have the problem with international travellers that they are a different subset of the population (often younger) so the death rate will also be different than for the general population.  Very tricky to integrate all these different variables into the data.  You actually could run into a degrees of freedom problem with too many variables and not enough data.  The data points are by DAY (not by hour) as far as I can tell. So you have very limited data to build a complex model. 

And on the flip side we have significant case load from cruise ships which skew to and older population as well.

  • Like 2
Link to post
Share on other sites

For example, I had 1 dependent variable and 3 independent variables an 200 data points (annual data for 22 years but with some missing data).

Link to post
Share on other sites
1 minute ago, Ausmumof3 said:

And on the flip side we have significant case load from cruise ships which skew to and older population as well.

Yup.  Plus, the spread variable will be mucked up because those people got on planes and flew home, so the isolated communities are not isolated like the rest of the model would be assuming. 

Link to post
Share on other sites

Zinc:

  • COVID-19 UPDATE: Zinc lozenges are being promoted to help prevent or treat COVID-19, the infection caused by the SARS-CoV-2 virus. There is no direct evidence proving that zinc lozenges can do this, although supplementing with appropriate amounts of zinc may benefit people deficient in zinc. See the  COVID-19section for details.

https://www.consumerlab.com/m/reviews/Zinc-Supplements-Lozenges-Review/zinc/

 

NB: this is irrespective of whether it is a “homeopathic “ or a conventional zinc lozenge

 

and this is another reason I am not supplementing with extra zinc (beyond whatever I may get in multivitamin)—unlike with vitamin D, I have no particular reason to believe myself deficient.

 

Edited by Pen
  • Like 2
Link to post
Share on other sites

Potassium and magnesium 

while on zinc I thought potassium (and also magnesium) might be worth mentioning.  One must be extremely careful with potassium because fatalities can occur from either too much or too little. But I think emphasizing potassium rich food especially liked raw vegetables and fruits, as well as cooked potatoes, beans etc is a safe way to get potassium.   And all that fresh food and even milk now going to waste 😢 seems so dreadful when it’s so much better than the processed stuff people seem to be buying so much... ach, I mustn’t get started on a vent...

 

Anyway, potassium is extremely important for healthy lung functions and modern commercial diets may mean a lot of people don’t get enough. (And also magnesium is important)

Low magnesium and potassium intakes were associated with lower lung function. Girls with low magnesium intake had lower forced expiratory flow at 75% of the ...
 
Nov 7, 2018 · Potassium is vital to lung function, so a potassium deficiency can cause breathing issues.
 
 
Edited by Pen
  • Like 3
Link to post
Share on other sites
2 minutes ago, Pen said:

Potassium

while on zinc I thought potassium might be worth mentioning.  One must be extremely careful with potassium because fatalities can occur from either too much or too little. But I think emphasizing potassium rich food especially like raw vegetables and fruits, as well as cooked potatoes, beans etc is a safe way to get potassium.   And all that fresh food and even milk now going to waste 😢 seems so dreadful when it’s so much better than the processed stuff people seem to be buying so much... ach, I mustn’t get started on a vent...

 

Anyway, potassium is extremely important for healthy lung functions and modern commercial diets may mean a lot of people don’t get enough.

Low magnesium and potassium intakes were associated with lower lung function. Girls with low magnesium intake had lower forced expiratory flow at 75% of the ...
 
Nov 7, 2018 · Potassium is vital to lung function, so a potassium deficiency can cause breathing issues.
 
 

 

I’m starting to think the recommended intake of daily fruits and vegetables plus a good multivitamin will cover most of this. 

  • Like 2
Link to post
Share on other sites
1 hour ago, square_25 said:

 

What I want to know is where NYC's modelers were in February... (Yes, we didn't have data, but I think it was something people should have been thinking about.) 

Denial.  At that point in time the politics were such that nobody would admit it was an issue.

Link to post
Share on other sites
2 hours ago, dmmetler said:

I really wonder if that is one reason why Memphis is acting much more aggressively than most of the region, and then the big modeling sites seem to indicate. I know the public health folks at the Med school are actively involved and maybe they are able to predict more closely because they know the local variables better. Reading local news is apt to give you mental whiplash because the newspapers are reporting much more positive models, which show the state generally being in good shape for the peak to come, but the local folks are preparing for a NYC level surge. 

I have been wondering about Memphis, too, because that's where my family is, and it's such an outlier for the state, policy wise.  I also wonder if doctors are seeing things that are not covered by official numbers, such as large numbers of suspected cases or a surge in unexplained deaths.  

  • Like 1
Link to post
Share on other sites

Has this been posted yet?

https://www.cnn.com/2020/04/06/us/new-york-park-burial-morgue-capacity/index.html

Apparently NY is considering temporarily burying bodies in city parks.  It looks like they've done it before, too.  I'm sure it's a last resort, but it's so crazy to even think about. 

  • Sad 2
Link to post
Share on other sites
32 minutes ago, Cnew02 said:

Has this been posted yet?

https://www.cnn.com/2020/04/06/us/new-york-park-burial-morgue-capacity/index.html

Apparently NY is considering temporarily burying bodies in city parks.  It looks like they've done it before, too.  I'm sure it's a last resort, but it's so crazy to even think about. 

I saw a "conspiracy" video with them loading bodies in bags in the back of a semi. It's not really a conspiracy, it's kind of all over the news. 

  • Like 1
Link to post
Share on other sites

Viral load

Has this been discussed yet?  Some think that people who get a higher viral load or more of the virus will have a worse case of the illness.  Apparently, this has not been proven.  Anyone know about this?

Link to post
Share on other sites

The refrigerator trucks I've seen, the temporary morgues.  And those made perfect sense to me, morgues aren't meant to deal with thousands dying a day, they have to go somewhere.  They've been used in other disasters too, they aren't a new thing.    I just wasn't thinking about a scenario where even those weren't enough.  It looks like Cuomo is denying the whole idea, but it was tweeted out by a city council person as a potential worse case scenario option, I guess in case cemeteries can't keep up?  It was just surprising. 

Edited by Cnew02
word choice
  • Like 2
Link to post
Share on other sites
7 minutes ago, Teaching3bears said:

Viral load

Has this been discussed yet?  Some think that people who get a higher viral load or more of the virus will have a worse case of the illness.  Apparently, this has not been proven.  Anyone know about this?

I don't know if it's true with the coronavirus, but our pediatrician told me that it would happen with chicken pox.  She said with so many kids in the house, if they got it they would probably have severe cases. 

Link to post
Share on other sites

For some reason, I cannot quote people without getting logged out.

Regarding potassium, most people in the US do not get enough. Only about 2% in the US get enough potassium. The minimum daily requirement is 4700 mg. It is best to get this from food sources. Supplements should be taken while being supervised by someone knowledgeable about them. Like Pen said, it's not good to take too much.

https://nutritionfacts.org/2013/05/23/98-of-american-diets-potassium-deficient/

A lot of people in the US also do not get enough magnesium which is needed to metabolize Vitamin D among many other things.

https://www.sciencedaily.com/releases/2018/02/180226122548.htm

Vitamin D can't be metabolized without sufficient magnesium levels, meaning Vitamin D remains stored and inactive for as many as 50 percent of Americans. In addition, Vitamin D supplements can increase a person's calcium and phosphate levels even while they remain Vitamin D deficient. People may suffer from vascular calcification if their magnesium levels aren't high enough to prevent the complication.
  • Like 2
Link to post
Share on other sites
2 hours ago, Seasider too said:

 

I’m starting to think the recommended intake of daily fruits and vegetables plus a good multivitamin will cover most of this. 

 

I think it would!!!

But last time I was at store people were buying chips and cookies or even Mac and cheese and rice  and so forth from the middle of store while gorgeous fruits and vegetables were going to waste it appeared. 

  • Like 1
Link to post
Share on other sites
32 minutes ago, BeachGal said:

For some reason, I cannot quote people without getting logged out.

Regarding potassium, most people in the US do not get enough. Only about 2% in the US get enough potassium. The minimum daily requirement is 4700 mg. It is best to get this from food sources. Supplements should be taken while being supervised by someone knowledgeable about them. Like Pen said, it's not good to take too much.

https://nutritionfacts.org/2013/05/23/98-of-american-diets-potassium-deficient/

A lot of people in the US also do not get enough magnesium which is needed to metabolize Vitamin D among many other things.

https://www.sciencedaily.com/releases/2018/02/180226122548.htm

Vitamin D can't be metabolized without sufficient magnesium levels, meaning Vitamin D remains stored and inactive for as many as 50 percent of Americans. In addition, Vitamin D supplements can increase a person's calcium and phosphate levels even while they remain Vitamin D deficient. People may suffer from vascular calcification if their magnesium levels aren't high enough to prevent the complication.

 

Yes. Thanks — I had tried to edit to add magnesium to my post, but then post would not go and I had to deal with irl 

I think magnesium is safer to supplement than potassium, but vegetables and fruits are especially excellent for potassium which should not be supplemented IMO without medical professionals oversight. And at same time things like green leafy vegetables are also good for magnesium.

btw dark chocolate is also rich in magnesium and some people think chocolate craving could be a sign of needing more magnesium.  I have actually been craving cooked greens! 

  • Like 1
Link to post
Share on other sites

I feel like there's a gulf between Dr. Birx advising us not to go to the store because we're about to surge upward massively and Mayor Bowser saying we won't peak until July. Like, Dr. Birx is implying that we're on the same curve as NYC and a week or so behind. But if that's the case, then a July peak seems wild. No one is talking about NYC peaking in June - every single thing I've read suggests a sooner peak.

I do get that the models are only as good as the data, etc. And that the assumptions in the models are different and so on and so forth. WaPo had an article and I read that. But it still seems like a gulf there.

  • Like 1
Link to post
Share on other sites
2 minutes ago, square_25 said:

Oh, random math factoid that may turn out to be relevant... if a test is 90% accurate (say that it means that it gives both a positive and a negative result correctly 90% of the time), then what percentage of the positive results are actually false positives? 

Everything I've read suggests that the actual CV tests don't produce false positives often if at all, but produce false negatives far too often - possibly as much as a quarter of the time.

  • Like 4
  • Thanks 1
Link to post
Share on other sites

I'm a few pages behind, but with regard to the discussion of how can we tell what the trajectory would look like without mitigating measures vs what it is looking like with those measures, comparing the outcomes in Sweden and Norway may prove useful. They are culturally and demographically quite similar, but have taken totally opposite approaches to the pandemic, with Norway going into full lock-down on March 12th, while Sweden is still allowing schools, restaurants, and businesses to remain open. The PM of Sweden has said that their approach is basically to allow the virus to run its course and create herd immunity, while protecting those over 70 as much as possible. 

Initially the numbers in Norway and Sweden were quite similar, but the number of deaths, and rate of increase in deaths, is now much higher in Sweden. Sweden' has roughly twice the population of Norway, but six times as many deaths: 477 vs 76. Comparing deaths per million, Norway has 14 vs Sweden's 47. Norway is also testing a much higher percentage of the population, >20K per million, versus Sweden's <4K per million. There are increasing calls from Swedish scientists and medical personnel for additional restrictions, at least for Stockholm if not the rest of the country, as the number of deaths continues to rise. Yesterday Norway posted 5 new deaths, while Sweden added 76. One of the most worrying stats is that one-third of the nursing homes in Stockholm have already reported infections. ☹️

  • Thanks 2
  • Sad 16
Link to post
Share on other sites

🇮🇹 Italy 
https://www.channelnewsasia.com/news/world/italy-covid-19-cases-immunity-regions-12615600

“CORONAVIRUS IMMUNITY

Authorities in northern Italy have begun testing health workers for antibodies that may help identify individuals with immunity to the coronavirus as they look for ways to ease the lockdown imposed a month ago to contain the epidemic. 

On Monday, the northeastern region of Veneto, which saw one of Italy's earliest outbreaks of the disease in February, started blood tests on health workers, following neighbouring Emilia Romagna, which began testing last week.

Following an initial phase of testing on 2,000-3,000 health workers, tests are to be extended to staff and residents in nursing homes and workers in contact with the public.

Regional governor Luca Zaia said the aim would ultimately be to allow authorities to issue "licences" for individuals with proven immunity to the virus to return to work.”

  • Like 8
Link to post
Share on other sites
5 hours ago, Arcadia said:

😞 https://abc11.com/business/no-ones-buying-it-wisconsin-dairy-farmer-forced-to-dump-56k-pounds-of-milk/6078309/

“GREANLEAF, Wisc. -- A Wisconsin dairy farmer said he was forced to dump 56,000 pounds of milk on Friday because no one is buying it.

Mark Mueller owns Mueller Dairy Farm in Greenleaf.

The milk harvested from the nearly 1,000 cows will rot, Mueller told  WFRV.

Mueller said he's worried the COVID-19 pandemic is preventing dairy farmers from getting their milk to the market.

He also says the restaurant closures severely hurt his business because there was no place for him to haul his product.

"It's really depressing," said Mueller. "It's like all your hard work just running down the drain."

Dairy groups in Wisconsin are now calling on the U.S. Department of Agriculture to provide help through the Coronavirus Aid, Relief, and Economic Security Act or CARES Act.”

 

Sigh. There was very little milk at Whole Foods today. More at Heinen's, our regular local grocery store.

Painful to hear of such a great disjoint between supply and storefront demand. WI is not so far from me in Ohio. 

  • Like 2
  • Thanks 2
  • Sad 2
Link to post
Share on other sites
1 hour ago, Farrar said:

I feel like there's a gulf between Dr. Birx advising us not to go to the store because we're about to surge upward massively and Mayor Bowser saying we won't peak until July. Like, Dr. Birx is implying that we're on the same curve as NYC and a week or so behind. But if that's the case, then a July peak seems wild. No one is talking about NYC peaking in June - every single thing I've read suggests a sooner peak.

I do get that the models are only as good as the data, etc. And that the assumptions in the models are different and so on and so forth. WaPo had an article and I read that. But it still seems like a gulf there.

 

Is the Mayor a doctor?

 

  • Like 1
Link to post
Share on other sites
1 hour ago, Corraleno said:

I'm a few pages behind, but with regard to the discussion of how can we tell what the trajectory would look like without mitigating measures vs what it is looking like with those measures, comparing the outcomes in Sweden and Norway may prove useful. They are culturally and demographically quite similar, but have taken totally opposite approaches to the pandemic, with Norway going into full lock-down on March 12th, while Sweden is still allowing schools, restaurants, and businesses to remain open. The PM of Sweden has said that their approach is basically to allow the virus to run its course and create herd immunity, while protecting those over 70 as much as possible. 

Initially the numbers in Norway and Sweden were quite similar, but the number of deaths, and rate of increase in deaths, is now much higher in Sweden. Sweden' has roughly twice the population of Norway, but six times as many deaths: 477 vs 76. Comparing deaths per million, Norway has 14 vs Sweden's 47. Norway is also testing a much higher percentage of the population, >20K per million, versus Sweden's <4K per million. There are increasing calls from Swedish scientists and medical personnel for additional restrictions, at least for Stockholm if not the rest of the country, as the number of deaths continues to rise. Yesterday Norway posted 5 new deaths, while Sweden added 76. One of the most worrying stats is that one-third of the nursing homes in Stockholm have already reported infections. ☹️

I am interested to see the outcomes of each country *at the very end of it all* if they do stay on two separate paths. Are the death rates roughly the same?  Did they have higher death rates from last year for other causes due to hospitals being inundated? What other factors had different outcomes?  
 

Are there other countries that haven't shut down entirely?  Right now we have models based on incomplete data. I think there will be tons to dig through retrospectively that will be able to help us in future decision making. 

Edited by hopeallgoeswell
  • Like 6
Link to post
Share on other sites
4 hours ago, Pen said:

 

13.3mg  according to several online sources. 

And I already posted that somewhere above. 

😊

 

 

“Not all lozenges are created equal

In order for a lozenge to provide enough zinc to be effective, it needs to contain between 13 and 23 milligrams, Dr. Cooperman says. Yet only two of the four lozenges Consumer Lab tested—Cold-Eeze Homeopathic Cold Remedy and Nature’s Way Zinc—provided enough. (There are other brands of zinc they didn't test.)”

The relevant issue here imo  is whether the particular lozenge has an effective amount of zinc,  not whether it is called “homeopathic” or not.

For example,  Zand brand is conventional but iirc only has 5mg zinc /lozenge.  So not “homeopathic” but also apparently not an effective amount. 

That you don’t know how much zinc there is does not mean it is unknowable.  Sort of like if something were expressed in any other units you don’t understand.  It doesn’t mean it is unknowable.  Maybe that is true, maybe not. 

 

Zincum Gluconicum 2x
(13.3mg)  Zinc Gluconate 
 
 

 

 

I am not against people using supplements where the amount of said supplement is clearly labeled on the packaging. 2x is not just another unit of measurement that I don't understand and is not equivalent to 13mg in all cases. If the package says 13mg, that is something entirely different. If I have to figure out how much (if any) zinc is actually in the thing by getting it tested in a lab, no thanks.

Again, homeopathy is a specific thing. Homeopathy promises that like cures like and diluting a substance precisely so that it is undetectable in a solution makes said substance more potent. If we're talking about taking 13mg of zinc per day, that is not homeopathy.

  • Like 1
Link to post
Share on other sites
2 minutes ago, EmseB said:

I am not against people using supplements where the amount of said supplement is clearly labeled on the packaging. 2x is not just another unit of measurement that I don't understand and is not equivalent to 13mg in all cases. If the package says 13mg, that is something entirely different. If I have to figure out how much (if any) zinc is actually in the thing by getting it tested in a lab, no thanks.

Again, homeopathy is a specific thing. Homeopathy promises that like cures like and diluting a substance precisely so that it is undetectable in a solution makes said substance more potent. If we're talking about taking 13mg of zinc per day, that is not homeopathy.

 

You are quite correct that 2X is not 13.3 mg in all cases. In fact, it would not be in most cases. 

2X is 1/10 of the total important substances involved. 

 

Whether 13.3 mg/lozenge is or isn’t homeopathy is not discussion  I want to get into. 

 

 

Link to post
Share on other sites
21 minutes ago, Acadie said:

 

Sigh. There was very little milk at Whole Foods today. More at Heinen's, our regular local grocery store.

Painful to hear of such a great disjoint between supply and storefront demand. WI is not so far from me in Ohio. 

 

This was so sad to hear about. All I could think of was George Mueller’s orphans. I find it a sad irony that this farmer’s name is also Mueller. 

Edited by Seasider too
  • Sad 3
Link to post
Share on other sites

@mathnerd

Santa Clara County https://abc7news.com/health/coronavirus-expert-predicts-very-very-long-time-until-life-returns-to-normal/6081905/

“Now, as the Bay Area now enters it's fourth week sheltering in place, Dr. Cody has a grim prediction: That a return to normal life might be farther away than any of us hoped or expected.
"We're going to be at this for a very, very long time," she told ABC7 News anchor Liz Kreutz in a one-on-one interview Monday. "I just have to keep reminding myself that this is a marathon and we have to keep ourselves nourished somehow, and we have to keep our energy good."
Dr. Cody was part of the team of public health officials to enact the Bay Area's shelter-in-place order; the first region in the country to do so.
"Taking action early gives us an advantage, gave us an advantage in slowing things down," she said. "It has, of course, caused enormous social and economic disruption. But had we taken action later, we would still be sheltering in place, we would still be experiencing incredible economic and social disruption and we'd be having an incredible strain on our healthcare system as well."
When asked why officials haven't released which areas are hardest hit, she says the virus is widespread in the county. It's everywhere.
Dr. Cody believes taking early action has made a difference.
"While we're not out of the woods, I am cautiously optimistic that we've put our hospitals and healthcare partners in a better position to manage the infections that we expect will be continuing to be coming," she said. "What I will say, is taking action early gives us an advantage, gave us an advantage, in slowing things down."
Still, Dr. Cody says a surge in cases is coming, and that since a vaccine is still a long ways off, at some point everyone is likely to be infected with COVID-19.
"Yes, probably at some point," Dr. Cody replied when asked if she believes everyone will be infected. "What our shelter-in-place order does, though, is slow things down, so we spread the cases out over a long period of time, and so that we spread the number of people who are severely ill and require hospitalization over a long period of time as well."
Dr. Cody declined to give a timeline for when she anticipates life returning to normal, saying this is the "new normal" for the foreseeable future. She stopped short of saying school won't return in the Fall.”

  • Thanks 1
  • Sad 5
Link to post
Share on other sites
1 hour ago, Farrar said:

I do get that the models are only as good as the data, etc. And that the assumptions in the models are different and so on and so forth. 

No, it is way more serious than that.  There is not enough data to make a model that takes into account all the factors affecting the virus population size.  The models are no longer any good, therefore they conflict.

  • Like 5
Link to post
Share on other sites
22 minutes ago, Pen said:

 

Is the Mayor a doctor?

 

She quoted models created by epidemiologists. So, no. But my point is simply that the federal and local governments are relying on massively different models - both created by experts, but they seem to be miles apart.

Honestly, I don't find Dr. Birx to be a very good communicator. Some of the things she has said seem to stoke alarm in the wrong way and yet aren't super clear - like, "don't go to the grocery store" doesn't seem like great blanket advice to me, honestly. I'm sure she's very qualified as a scientist, but I don't think the messenger has to be the scientist  for the information to reflect the best model. I have no idea which model is right. I don't think anyone knows. Isn't that the point?

  • Like 10
Link to post
Share on other sites
9 minutes ago, lewelma said:

No, it is way more serious than that.  There is not enough data to make a model that takes into account all the factors affecting the virus population size.  The models are no longer any good, therefore they conflict.

I guess that was sort of my point though. But when I said a few pages ago that I didn't think the models made any real sense anymore, especially on these websites that claim to predict the peak in a huge array of places, I was chastised and told the models are good, the data is not but it is some places. Okay. Whatever. I guess I'm saying it wrong. But from a total non-science, non-mathy lay perspective, the information is all over the place at this point and giving these specific dates like gospel seems deeply misleading.

I think it would make a lot more sense to report it a bit more like when the hurricane is way out in the Atlantic. Of course, we have a lot more control over the spread of the virus than we do over the hurricane's path, but at least they've gotten reasonably adept at talking about the uncertainty of a hurricane's path and the predictions and models involved in trying to know.

  • Like 5
Link to post
Share on other sites
2 minutes ago, Farrar said:

She quoted models created by epidemiologists. So, no. But my point is simply that the federal and local governments are relying on massively different models - both created by experts, but they seem to be miles apart.

Honestly, I don't find Dr. Birx to be a very good communicator. Some of the things she has said seem to stoke alarm in the wrong way and yet aren't super clear - like, "don't go to the grocery store" doesn't seem like great blanket advice to me, honestly. I'm sure she's very qualified as a scientist, but I don't think the messenger has to be the scientist  for the information to reflect the best model. I have no idea which model is right. I don't think anyone knows. Isn't that the point?

 

I don’t think when an individual area is going to peak is all that relevant.  

I think taking a lesson from places that didn’t Stay Home as much as possible earlier rather than later is more relevant.

The country as a whole is not doing well in terms of case numbers and deaths. We don’t want to wait till our community is trying to figure out where to bury the dead to Stay Home 

and some places being able to send equipment to places that need it sooner is probably super helpful—rather than engaging in behavior likely to bring on more local need more quickly. 

 

my 2 cents

Link to post
Share on other sites
2 minutes ago, Pen said:

 

I don’t think when an individual area is going to peak is all that relevant.  

I think taking a lesson from places that didn’t Stay Home as much as possible earlier rather than later is more relevant.

The country as a whole is not doing well in terms of case numbers and deaths. We don’t want to wait till our community is trying to figure out where to bury the dead to Stay Home 

and some places being able to send equipment to places that need it sooner is probably super helpful—rather than engaging in behavior likely to bring on more local need more quickly. 

 

my 2 cents

I don't disagree with any of that advice per se. I mean, we're not taking walks. We're going to the grocery store once every two weeks. We're literally pacing the backyard, which is the size of the dining room. So it's not like I am not doing "the right things" or advocating otherwise. But I think it's natural to speculate about the dates and the peaks and the models. People in this thread were literally doing that. So I commented. And it would, indeed, be nice to have a sense of our timetable for various things. I think that's pretty relevant, even if, in the end, it turns out to be unknowable. And the gist of my comment was that the models seem like they're all over the place. That's really it.  

 

  • Like 9
Link to post
Share on other sites
21 minutes ago, Farrar said:

I guess that was sort of my point though. But when I said a few pages ago that I didn't think the models made any real sense anymore, especially on these websites that claim to predict the peak in a huge array of places, I was chastised and told the models are good, the data is not but it is some places. Okay. Whatever. I guess I'm saying it wrong. But from a total non-science, non-mathy lay perspective, the information is all over the place at this point and giving these specific dates like gospel seems deeply misleading.

I completely agree.  I am a mathy person who has worked in the field of population dynamics.  The models are not only no good, they are definitely misleading.  They are confusing to the public and cause people to doubt government decisions to shelter in place.  Numbers sound so strong, mathematicians and modellers have some sort of god-like status.  Like I wrote above, qualitative opinion would be more accurate in this situation than mathematical models. 

The models are all over the place because they are all wrong.  I just heard on the radio a data scientist saying that the data was contaminated but they were doing the best they could.  This is true, but also misleading.  Honestly, they should just come out and say that at this point models are not possible and they will need to rely on the opinions of people who have good hunches.  But I don't think that would sell with the public!

  • Like 10
Link to post
Share on other sites

Yikes — I did not realize until reading this article that the original IHME model was based entirely on the curve from Wuhan, which was matched/fitted to US data. The reason the curves look so different in the most recent update is because they have now added data from seven areas in Italy and Spain that have already peaked. It seems like the Wuhan data would be almost completely useless for US modeling purposes considering that (1) they imposed lock down measures that were FAR more severe than anything done in the US and (2) they were clearly under-reporting the number of cases and deaths. So basically garbage in, garbage out. They have also been assuming that the social distancing measures currently in place could literally reduce the number of new cases to zero, which makes no sense — I was wondering how in the world they were getting results showing zero new cases and zero new deaths within just a few weeks.

I would really like to know what assumptions are being made, and what data is being used, for the various other models whose predictions are so totally different from the IHME model. It's especially disturbing if, as reported in the WaPo article, the Feds are basing decisions about which states get what equipment and supplies on the IHME model, which could be way off. ☹️

Edited by Corraleno
  • Like 3
  • Thanks 1
Link to post
Share on other sites
8 minutes ago, Farrar said:

I don't disagree with any of that advice per se. I mean, we're not taking walks. We're going to the grocery store once every two weeks. We're literally pacing the backyard, which is the size of the dining room. So it's not like I am not doing "the right things" or advocating otherwise. But I think it's natural to speculate about the dates and the peaks and the models. People in this thread were literally doing that. So I commented. And it would, indeed, be nice to have a sense of our timetable for various things. I think that's pretty relevant, even if, in the end, it turns out to be unknowable. And the gist of my comment was that the models seem like they're all over the place. That's really it.  

 

 

Sure. I can understand that.

people in war situations or economic depressions or droughts or famine want to know peaks and end dates too, I expect

But when we do x of y or don’t do q or r, we change the results, change the peak.  Not by our individual self , but en masse 

its like with the TP shortages, no one person, not even the dudes buying truck loads to price gouge with, makes for the shortage, but as a group many people buying up tp does what we have seen

I have just told myself “it is going to be a very very long haul situation.”   And, “The more I and others Stay Home, the shorter and less devastating it will be.”  

 

 

 

  • Like 1
Link to post
Share on other sites

Also, if we are going to base decisions on modeling, shouldn’t we be doing what we do when trying to predict hurricane landfalls - tracking multiple models, seeing where the consensus is, and perhaps downgrading the outliers? All of the models could still end up being off but wouldn’t that be better than relying on a single model?

  • Like 4
Link to post
Share on other sites
8 minutes ago, Corraleno said:

Yikes — I did not realize until reading this article that the original IHME model was based entirely on the curve from Wuhan, which was matched/fitted to US data. The reason the curves look so different in the most recent update is because they have now added data from seven areas in Italy and Spain that have already peaked. It seems like the Wuhan data would be almost completely useless for US modeling purposes considering that (1) they imposed lock down measures that were FAR more severe than anything done in the US and (2) they were clearly under-reporting the number of cases and deaths. So basically garbage in, garbage out. They have also been assuming that the social distancing measures currently in place could literally reduce the number of new cases to zero, which makes no sense — I was wondering how in the world they were getting results showing zero new cases and zero new deaths within just a few weeks.

I would really like to know what assumptions are being made, and what data is being used, for the various other models whose predictions are so totally different from the IHME model. It's especially disturbing if, as reported in the WaPo article, the Feds are basing decisions about which states get what equipment and supplies based solely on the IHME model, which could be way off. ☹️

As I understand it, the IHME model is not the foundation for determining who gets what from the federal government. I've read several times that no one knows what the actual criteria the administration is using is and some states have gotten everything they ask for while others get very little.

The junkiness of the Chinese data gets more and more infuriating by the day.

  • Sad 1
Link to post
Share on other sites
9 minutes ago, Pen said:

 

Sure. I can understand that.

people in war situations or economic depressions or droughts or famine want to know peaks and end dates too, I expect

But when we do x of y or don’t do q or r, we change the results, change the peak.  Not by our individual self , but en masse 

its like with the TP shortages, no one person, not even the dudes buying truck loads to price gouge with, makes for the shortage, but as a group many people buying up tp does what we have seen

I have just told myself “it is going to be a very very long haul situation.”   And, “The more I and others Stay Home, the shorter and less devastating it will be.”  

 

 

 

Are you arguing against something I said? I'm genuinely confused. I mean, I understand how stay home orders are supposed to work.

If you're able to say, I'm okay to just wait and not know and ignore the predictions, then that's great for you. But again I'll just say that I don't think discussing the quality or public use of the models is off limits for this thread or anything.

Link to post
Share on other sites
44 minutes ago, lewelma said:

I completely agree.  I am a mathy person who has worked in the field of population dynamics.  The models are not only no good, they are definitely misleading.  They are confusing to the public and cause people to doubt government decisions to shelter in place.  Numbers sound so strong, mathematicians and modellers have some sort of god-like status.  Like I wrote above, qualitative opinion would be more accurate in this situation than mathematical models. 

The models are all over the place because they are all wrong.  I just heard on the radio a data scientist saying that the data was contaminated but they were doing the best they could.  This is true, but also misleading.  Honestly, they should just come out and say that at this point models are not possible and they will need to rely on the opinions of people who have good hunches.  But I don't think that would sell with the public!

The Louisiana model I was wondering about is completely out of line, though.  I can't see how they could come up with those numbers for that state and completely different numbers for a state that had similar numbers at the time the predictions were made.  Well, I could understand it of Louisiana had better stats and was taking stricter measures than the other state, but the complete reverse is true.  I was comparing Louisiana (LA - referring to state not city here) to my state (MA) because for a while we had kind of similar numbers of cases and were next to each other in the listings, except LA had more than 2x the death rate and seemed to have cases rising faster.  So I was curious what the predictions were going out.  According to the predictions, MA will be dire and LA is all done already.

So, for example, here's MA's stats from 3/31 through today:

image.png.2693bc74082abad65c6e60743329241b.png

and LA's stats:

image.png.1af43bfa281340fe568ecabf72907ed1.png

So, they started out a week ago with a bit less case than we did, now they have a bit more (so, cases going up faster), and the death toll per case is much higher.  As of 4/5, LA had 280 cases per 100K population, and MA 181 cases per 100K.

So, this model took those number and predicted these outcomes:

MA:

image.thumb.png.9591360456120d029497a00628391100.png

LA: 

image.thumb.png.19823fd91a541f9448203d92496fa7c1.png

 

Can someone tell me how the *same* model took those numbers and trends and decided that we'd end up needing 10x more beds and ventilators than Louisiana?  And that LA was already done with the peak and things were already getting better?   To look ahead to 4/6 (only, what 2-3 days in the future from when they made these predictions?)  - it predicts that MA would need 5091 beds (actual 1145) and 440 deaths (actual 260), and LA would need 1128 beds (actual 1809) and 499 deaths (actual 512) - the LA numbers are closer than the MA ones, but the thing is that the model shows them already on a downward slope, so that they'll never run out of ICU beds or ventilators - the LA governor said today that without new shipments they'll run out by Thursday/Friday.  Because it's not over there.  Why is the *same* model giving such rosy numbers for one of the worst affected states, and such scary ones for a state that's had a lower growth rate, faster shut-down, and lower death rate all along?  I'm still very worried that we'll be overwhelmed here.  But the disparities in the models for the two states just make no sense at.all. with the previous trends nor other human factors.

Also, wait, is it really true that Louisiana has, going into this, almost twice as many ICU beds as Massachusetts and also way more hospital beds generally?  Especially since there are 2.3 million more people in MA?  That seems... off... 

In fact, both of these seem so far off to me in wildly different ways that I'm not sure I can look at any of the other states and not feel like they just made stuff up, or perhaps used darts instead of math to figure where to aim the numbers...  it's one thing if a model makes predictions that are low or high across the board, but this is like some kind of random number generator type outcome...

Edited by Matryoshka
  • Like 3
Link to post
Share on other sites
5 hours ago, Pen said:

Potassium and magnesium 

while on zinc I thought potassium (and also magnesium) might be worth mentioning.  One must be extremely careful with potassium because fatalities can occur from either too much or too little. But I think emphasizing potassium rich food especially liked raw vegetables and fruits, as well as cooked potatoes, beans etc is a safe way to get potassium.   And all that fresh food and even milk now going to waste 😢 seems so dreadful when it’s so much better than the processed stuff people seem to be buying so much... ach, I mustn’t get started on a vent...

 

Anyway, potassium is extremely important for healthy lung functions and modern commercial diets may mean a lot of people don’t get enough. (And also magnesium is important)

Low magnesium and potassium intakes were associated with lower lung function. Girls with low magnesium intake had lower forced expiratory flow at 75% of the ...
 
Nov 7, 2018 · Potassium is vital to lung function, so a potassium deficiency can cause breathing issues.
 
 

Often those food stories relate to supply chain and people trying to manage prices.  Avocado farmers were dumping avocados and pleading with people to buy more last week due to the lack of cafe purchases.  Guess what price Avocados are in the supermarket.  $2.80 each.  They are usually $2 each max.  

  • Like 1
Link to post
Share on other sites
10 minutes ago, Farrar said:

As I understand it, the IHME model is not the foundation for determining who gets what from the federal government. I've read several times that no one knows what the actual criteria the administration is using is and some states have gotten everything they ask for while others get very little.

In the WaPo article I linked, a state official said that IHME figures had been explicitly cited by the Feds in rejecting requests for vents and equipment, on the grounds that the IHME model showed they would need far less equipment than the state's own models projected.

  • Like 1
Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×
×
  • Create New...