Author Topic: How long can we wait while flattening the curve?  (Read 253661 times)

AnnaGrowsAMustache

  • Handlebar Stache
  • *****
  • Posts: 1919
  • Location: Noo Zilind
Re: How long can we wait while flattening the curve?
« Reply #600 on: April 14, 2020, 11:30:11 PM »
Herd (community) immunity doesn't have a specific timeline. It depends on how long it takes for a critical percentage of a population to be immunized (either through being infected or vaccination). For example, the polio vaccines provided herd immunity within a period of a few years due to their efficacy and the long-term immunity they provided. (https://www.cdc.gov/vaccines/pubs/pinkbook/polio.html - Section on Secular Trends in the United States).

Similarly, the measles vaccines when they first came out did the same thing. https://www.cdc.gov/vaccines/pubs/pinkbook/meas.html (section on Secular Trends in the United States)

Smallpox, in contrast, took longer to eradicate as it was (when still endemic) the most contagious of the human viruses. Now there's actually no herd immunity as the smallpox vaccine is no longer administered.

HPV, which is probably the most common chronic viral infection in adult humans, can be suppressed with only 20% vaccination. Community immunity in some populations has been identified in 4 years. This is due to its lower risk of transmission (as it is not transmitted by droplets). Lancet Public Health. 2016 Nov;1(1):e8-e17 - Figure 6.

If a vaccine is developed and sufficient number of people are immunized with it, then herd (community) immunity can be achieved in a relatively short period of time. There is no reason to suspect it can't be done. Also, if an isolated population is exposed, then herd immunity can be developed the hard way (through large % being infected and high number of deaths). Both of these assume that there are no new people entering into the population. This is obviously not ideal in any scenario, and probably not feasible in a big city like New York. Regarding whether it is happening, I think that is unlikely. Assuming the "true" case fatality rate of this virus is ~1% (based on S Korean and German data), and with a current 4% case fatality rate in New York, we can assume 4x as many people are infected as have been identified. That means 200k x4 = 800k people have it in New York, but there's about 14m people in that metropolitan area. We'd need 14x more people infected (14m * 0.8 = 11.2m / 800k = 14). That'd amount to 140,000 deaths in NY alone. These are clearly very rough calculations but demonstrate that we can't allow this virus to run through our population.

Vaccines work by speeding up the process massively. That's the point of vaccines. Without vaccines, herd immunity takes a very, very long time, and a LOT of people without the right immune response being killed off.


Abe

  • Handlebar Stache
  • *****
  • Posts: 1658
Re: How long can we wait while flattening the curve?
« Reply #601 on: April 14, 2020, 11:44:44 PM »
Also, to clarify we don't have herd immunity to most coronaviruses. The reason for this is unclear, but non-vaccine immunity (there are no commercially available vaccines for coronavirii) seems to last less than a year (~40 weeks). This is also true for influenza (thus the intermittent recurrence of various strains every few years). So in this context, herd immunity would only apply for a finite period of time, and would not be the long-term immunity seen with the diseases I mentioned before. It's a bummer.

former player

  • Walrus Stache
  • *******
  • Posts: 5813
  • Location: Avalon
Re: How long can we wait while flattening the curve?
« Reply #602 on: April 15, 2020, 02:01:33 AM »
Also, to clarify we don't have herd immunity to most coronaviruses. The reason for this is unclear, but non-vaccine immunity (there are no commercially available vaccines for coronavirii) seems to last less than a year (~40 weeks). This is also true for influenza (thus the intermittent recurrence of various strains every few years). So in this context, herd immunity would only apply for a finite period of time, and would not be the long-term immunity seen with the diseases I mentioned before. It's a bummer.
So, the end of life as we have known it so far?  That would be worrying, given that we are currently finding out how much of western society is based on activities that may never be "safe" again.

mwulff

  • Bristles
  • ***
  • Posts: 290
Re: How long can we wait while flattening the curve?
« Reply #603 on: April 15, 2020, 03:29:50 AM »
Also, to clarify we don't have herd immunity to most coronaviruses. The reason for this is unclear, but non-vaccine immunity (there are no commercially available vaccines for coronavirii) seems to last less than a year (~40 weeks). This is also true for influenza (thus the intermittent recurrence of various strains every few years). So in this context, herd immunity would only apply for a finite period of time, and would not be the long-term immunity seen with the diseases I mentioned before. It's a bummer.
So, the end of life as we have known it so far?  That would be worrying, given that we are currently finding out how much of western society is based on activities that may never be "safe" again.

Give it enough time and Sars-covid-2 will hopefully mutate into something a little less deadly. This is something that has happened with many viruses given enough spread and time. Simply due to the fact that a living host is "worth" more to the virus than a dead host. But the timescale for this to occur can vary and there are no guarantees.

See this for more information https://phys.org/news/2013-05-diseases-rapidly-evolve-moreor-lessvirulent.html

I am not an expert but I sincerely hope we find a vaccine.

But anyways this is in my opinion the end of life as we knew it up until 2020. Because we have to face the fact that at some point we will encounter sars-covid-3 and 4 and 5. Humanity needs better defense against the germs, bugs and viruses that are trying to kill us. This is not the end of civilized society, it will just have to change.

And we can't have a globalized society where a virus outbreak in a major country can cause a global pandemic inside of a week. So travel will have to change going forward.

Buffaloski Boris

  • Handlebar Stache
  • *****
  • Posts: 2129
Re: How long can we wait while flattening the curve?
« Reply #604 on: April 15, 2020, 08:27:45 AM »
Some interesting theoretical models regarding social distancing and its efficacy from Harvard's Public Health department (https://science.sciencemag.org/content/early/2020/04/14/science.abb5793)

They created a model of transmission using two known coronavirii, fitted this model to prior records, then added the new coronavirus using estimated parameters from current data (the major limitation of this study).

The basic summary is: social distancing duration (4, 8, 20 weeks) only affects the onset of a second wave, but not its ability to overwhelm current hospital capacity. If the social distancing is very effective (60% reduction in transmission rate from a baseline 2.2 estimated from other coronavirii), then the second peak is nearly the same as without distancing measures. (Figures 4 - assuming no seasonal variation and 5 assuming seasonal variation, left panels). In nearly all scenarios, current critical care bed capacity will be overwhelmed at some point regardless of measures taken, other than indefinite social distancing until a vaccine is developed. In this case, a 40% or 60% reduction would stay below this threshold. The obvious advantages of longer social distancing time are the ability to increase critical care capacity and longer time to develop a vaccine. Obvious disadvantage is the economic fall-out.

The second prediction is that having cut-offs for opening up and shutting down society will allow intermittent exposure and ultimately herd immunity without overwhelming critical care capacity (Figure 6), but this would take to 2025 with current capacity, or until 2022 with doubling of capacity, with >75% of the time spent shut down in the former scenario. If we get a break in the summer, then significantly less time is spent shut down (especially if critical care capacity increases).

So it seems the best option is moderate social distancing to spread out the pain, and ramping up both ventilator capacity and re-tooling the majority of inpatient healthcare for the next 1-2 years towards managing COVID. Unfortunately there really is no good scenario in any of this.

Two trials on chloroquine/azithromycin were stopped due to high risk of fatal arrhythmias and lack of efficacy in reducing viral load. Other trials are ongoing.

In more positive news, there is promising signs on vaccine development. Both the NHS and Johnson & Johnson estimate clinical randomized trials to start by September, with preliminary results by December and potential manufacturing to start in January. Everyone hang in there, it's going to be a bumpy ride.

Thanks @Abe for that link.  I read the paper, but came to a little different conclusion. Of course I'm not a physician and I don't play one on TV, but what I got out of it is that the huge unknown is what level of long term immunity there will be (by coming down with it or eventually a from a vaccine - assuming there will be one).  If it's a shorter time frame, say 40 weeks, then we're glued, screwed, and tattooed for the next several years.  If it's longer term immunity, we're still well-glued.  With the bonus being that we won't know what that level of immunity is until we do a lot more serological testing, which we aren't really doing yet. With yet another added bonus being that widespread shutdowns and social distancing, if very effective, really just push the infection peak to the fall.  That's not nothing as it gives time to figure out better treatments, develop drugs that might work, possibly develop a vaccine, and boost critical care capacity. Still given that round 2 of this will likely come about this fall or winter, what we seem to be doing is mostly delaying the day of reckoning rather than avoiding it.

Am I missing something?   

BTW, I really like the understatement.  Looks like a bumpy ride to me as well. 

fattest_foot

  • Pencil Stache
  • ****
  • Posts: 844
Re: How long can we wait while flattening the curve?
« Reply #605 on: April 15, 2020, 09:10:07 AM »
You're talking about herd immunity like you know what it means, and you clearly don't. We have herd immunity to the common cold - that's taken a thousand years and we still have seasonal outbreaks. Covid 19 is showing signs of being similar in the kind of immunity you might get from an exposure - ie not total and not long term. If that's the case, even with a vaccine, even with herd immunity, you will still get seasonal outbreaks that may kill thousands. That could go on for a thousand years, and THEN it'll be no more than a common cold to us. Do you understand? Herd immunity is a concept that applies to a species over a long period of time. It is NOT the result of one outbreak. There's no "herd immunity" that can protect the human species from covid-19 now, or this year, or in the next ten years..... unless we develop a vaccine and people are regularly vaccinated.

Isn't the common cold just a series of many different types of viruses? You're comparing a basket of apples to a single apple.

Ironically, several of them are coronavirus. We should be adding this to the basket, not implying that it's an entirely separate basket.

Abe

  • Handlebar Stache
  • *****
  • Posts: 1658
Re: How long can we wait while flattening the curve?
« Reply #606 on: April 15, 2020, 06:20:14 PM »
Some interesting theoretical models regarding social distancing and its efficacy from Harvard's Public Health department (https://science.sciencemag.org/content/early/2020/04/14/science.abb5793)

They created a model of transmission using two known coronavirii, fitted this model to prior records, then added the new coronavirus using estimated parameters from current data (the major limitation of this study).

The basic summary is: social distancing duration (4, 8, 20 weeks) only affects the onset of a second wave, but not its ability to overwhelm current hospital capacity. If the social distancing is very effective (60% reduction in transmission rate from a baseline 2.2 estimated from other coronavirii), then the second peak is nearly the same as without distancing measures. (Figures 4 - assuming no seasonal variation and 5 assuming seasonal variation, left panels). In nearly all scenarios, current critical care bed capacity will be overwhelmed at some point regardless of measures taken, other than indefinite social distancing until a vaccine is developed. In this case, a 40% or 60% reduction would stay below this threshold. The obvious advantages of longer social distancing time are the ability to increase critical care capacity and longer time to develop a vaccine. Obvious disadvantage is the economic fall-out.

The second prediction is that having cut-offs for opening up and shutting down society will allow intermittent exposure and ultimately herd immunity without overwhelming critical care capacity (Figure 6), but this would take to 2025 with current capacity, or until 2022 with doubling of capacity, with >75% of the time spent shut down in the former scenario. If we get a break in the summer, then significantly less time is spent shut down (especially if critical care capacity increases).

So it seems the best option is moderate social distancing to spread out the pain, and ramping up both ventilator capacity and re-tooling the majority of inpatient healthcare for the next 1-2 years towards managing COVID. Unfortunately there really is no good scenario in any of this.

Two trials on chloroquine/azithromycin were stopped due to high risk of fatal arrhythmias and lack of efficacy in reducing viral load. Other trials are ongoing.

In more positive news, there is promising signs on vaccine development. Both the NHS and Johnson & Johnson estimate clinical randomized trials to start by September, with preliminary results by December and potential manufacturing to start in January. Everyone hang in there, it's going to be a bumpy ride.

Thanks @Abe for that link.  I read the paper, but came to a little different conclusion. Of course I'm not a physician and I don't play one on TV, but what I got out of it is that the huge unknown is what level of long term immunity there will be (by coming down with it or eventually a from a vaccine - assuming there will be one).  If it's a shorter time frame, say 40 weeks, then we're glued, screwed, and tattooed for the next several years.  If it's longer term immunity, we're still well-glued.  With the bonus being that we won't know what that level of immunity is until we do a lot more serological testing, which we aren't really doing yet. With yet another added bonus being that widespread shutdowns and social distancing, if very effective, really just push the infection peak to the fall.  That's not nothing as it gives time to figure out better treatments, develop drugs that might work, possibly develop a vaccine, and boost critical care capacity. Still given that round 2 of this will likely come about this fall or winter, what we seem to be doing is mostly delaying the day of reckoning rather than avoiding it.

Am I missing something?   

BTW, I really like the understatement.  Looks like a bumpy ride to me as well.

Youíre right about length of immunity being a big question mark. I think long term if we have a vaccine itíll be less of an issue other than it may increase the frequency of immunizations to every year or even every 6 months potentially. If we dont have a vaccine, then thatís a major problem like you mentioned. How good a prediction the 40 weeks is based on other viruses is outside of my field so I canít comment on that being a likely scenario or not. Iíll ask one of my friends in infectious diseases.

AnnaGrowsAMustache

  • Handlebar Stache
  • *****
  • Posts: 1919
  • Location: Noo Zilind
Re: How long can we wait while flattening the curve?
« Reply #607 on: April 16, 2020, 06:12:37 AM »
You're talking about herd immunity like you know what it means, and you clearly don't. We have herd immunity to the common cold - that's taken a thousand years and we still have seasonal outbreaks. Covid 19 is showing signs of being similar in the kind of immunity you might get from an exposure - ie not total and not long term. If that's the case, even with a vaccine, even with herd immunity, you will still get seasonal outbreaks that may kill thousands. That could go on for a thousand years, and THEN it'll be no more than a common cold to us. Do you understand? Herd immunity is a concept that applies to a species over a long period of time. It is NOT the result of one outbreak. There's no "herd immunity" that can protect the human species from covid-19 now, or this year, or in the next ten years..... unless we develop a vaccine and people are regularly vaccinated.

Isn't the common cold just a series of many different types of viruses? You're comparing a basket of apples to a single apple.

Ironically, several of them are coronavirus. We should be adding this to the basket, not implying that it's an entirely separate basket.

You've missed the point. Take just one of the common cold corona viruses. It kills if you have no immunity. We see that every time a population that has not been exposed to it before, is. Think colonisation examples. It doesn't kill the herd of the coloniser because they've developed an immunity over multiple generations. They still get the illness but it's vastly reduced in severity. That is what herd immunity is. There are two ways to get herd immunity - A) multiple generations of exposure to the virus with many people severely ill and dying, kind of a natural selection, or B) a vaccine.

We have herd immunity to all of the common cold viruses, to the vast degree. We also have it to common illnesses like chicken pox and measles. Yes, you will get sick. Yes, some people die. However, we get nowhere near the severity of illness that a population with no history of exposure gets. Again, look to colonisation examples. We now have herd immunity to diseases like polio and smallpox, because we have vaccines. Those diseases will no longer kill off our herd.

We have no immunity to Covid-19. Because it's new. Even if you exposed every single person in the world to it, the survivors would not have herd immunity. That's because our offspring would not necessarily have an easier time with the illness. That's because we might be able to get it more than once, and we're not sure that the second time around would be easier or less fatal. It would take multiple generations to say that, yes, we have developed some kind of herd immunity to this. You would need to show that those generations have a far less severe reaction to covid-19 than our current generation is having. OR.... a vaccine.

nereo

  • Senior Mustachian
  • ********
  • Posts: 13562
  • Location: Just south of Canada
    • Here's how you can support science today:
Re: How long can we wait while flattening the curve?
« Reply #608 on: April 16, 2020, 07:04:54 AM »
No... most viruses do not kill the host even if it has no immunity.  This is particularly true with other corona viruses, which typically present with mild symptoms in healthy individuals.

Pathogens that kill their host are outliers.  We notice them more because of this lethality, but you are confusing immunity, a normal immunosuppression response and the diseases which, through some quirk, are not effectively recognized by our immune system.

T-Money$

  • Stubble
  • **
  • Posts: 218
  • Location: New York
Re: How long can we wait while flattening the curve?
« Reply #609 on: April 16, 2020, 07:58:28 AM »
No... most viruses do not kill the host even if it has no immunity.  This is particularly true with other corona viruses, which typically present with mild symptoms in healthy individuals.

Pathogens that kill their host are outliers.  We notice them more because of this lethality, but you are confusing immunity, a normal immunosuppression response and the diseases which, through some quirk, are not effectively recognized by our immune system.

This is very accurate.

Lymphocytes are triggered to make antibodies.  This is called adaptive immunity.  This occurs with the overwhelming majority of viruses, and in the case of COVID-19 occurs with the overwhelming majority of those infected.

Per the Iceland data, about 1/2 of those infected with COVID-19 do not have a significant immunological response (they remain asymptomatic for the duration of their infection).  Perhaps immunity is already established from other coronaviruses that maybe similar.  Who knows?

The good news is that through time this virus is not living up to the dire predictions of the most publicized scientific studies.  As the economy opens up and people begin to socialize hopefully it will continue to spread without the predicted impacts, and hopefully people will adjust their psychology to something a bit more rational.  It will perhaps be something we will have to live with indefinitely, but as it spreads, mutates and immunity develops it will hopefully fade into something that is not so troublesome to people. 

Jon Bon

  • Handlebar Stache
  • *****
  • Posts: 1167
  • Location: Midwest
Re: How long can we wait while flattening the curve?
« Reply #610 on: April 16, 2020, 08:06:26 AM »
We have 22 million newly unemployed people. Approximately 18% unemployment rate.

So my answer to the original question is: Not long, not long at all.

https://www.washingtonpost.com/business/2020/04/16/unemployment-claims-coronavirus/

T-Money$

  • Stubble
  • **
  • Posts: 218
  • Location: New York
Re: How long can we wait while flattening the curve?
« Reply #611 on: April 16, 2020, 08:08:56 AM »
We have 22 million newly unemployed people. Approximately 18% unemployment rate.

So my answer to the original question is: Not long, not long at all.

https://www.washingtonpost.com/business/2020/04/16/unemployment-claims-coronavirus/

Bingo.  People will be willing to risk COVID-19 so they can continue to feed their family, then they'll realize the sky isn't falling after all.

nereo

  • Senior Mustachian
  • ********
  • Posts: 13562
  • Location: Just south of Canada
    • Here's how you can support science today:
Re: How long can we wait while flattening the curve?
« Reply #612 on: April 16, 2020, 08:16:36 AM »
We have 22 million newly unemployed people. Approximately 18% unemployment rate.

So my answer to the original question is: Not long, not long at all.

https://www.washingtonpost.com/business/2020/04/16/unemployment-claims-coronavirus/
I view these numbers a bit differently. I see that, despite the drastic measures, we still have ~80% of the workforce, and a government that presently is committed to providing social assistance to almost all who arenít working.

Seems to me we can continue this for quite a bit longer, painful as it may be to many.
Big question in my mind is if the small business grant program will get more funding and a better rollout.

Davnasty

  • Magnum Stache
  • ******
  • Posts: 2582
Re: How long can we wait while flattening the curve?
« Reply #613 on: April 16, 2020, 08:26:19 AM »
No... most viruses do not kill the host even if it has no immunity.  This is particularly true with other corona viruses, which typically present with mild symptoms in healthy individuals.

Pathogens that kill their host are outliers.  We notice them more because of this lethality, but you are confusing immunity, a normal immunosuppression response and the diseases which, through some quirk, are not effectively recognized by our immune system.

This is very accurate.

Lymphocytes are triggered to make antibodies.  This is called adaptive immunity.  This occurs with the overwhelming majority of viruses, and in the case of COVID-19 occurs with the overwhelming majority of those infected.

Per the Iceland data, about 1/2 of those infected with COVID-19 do not have a significant immunological response (they remain asymptomatic for the duration of their infection).  Perhaps immunity is already established from other coronaviruses that maybe similar.  Who knows?

The good news is that through time this virus is not living up to the dire predictions of the most publicized scientific studies.  As the economy opens up and people begin to socialize hopefully it will continue to spread without the predicted impacts, and hopefully people will adjust their psychology to something a bit more rational.  It will perhaps be something we will have to live with indefinitely, but as it spreads, mutates and immunity develops it will hopefully fade into something that is not so troublesome to people.

Do you have a citation for this. I've been following Iceland's situation with interest but everything I've read suggests that ~50% are asymptomatic at the time of testing. I haven't seen any kind of follow up that suggests this remains true throughout the duration of infection.

https://www.barrons.com/news/loans-wont-cut-it-for-flailing-small-businesses-51586982045

Quote
43% of people who tested positive either had no symptoms, or at least they did not at the time of their testing. This provides further evidence that many carriers of the virus are asymptomatic or presymptomatic.

Also worth noting, that's 43% of 1,720 confirmed cases in Iceland so it's still a relatively small sample size

And while it may end up being the case that 50% or even more of the population never experiences any symptoms, the fatality rate remains a critical part of the decision making formula. 50% asymptomatic and 1% fatality would still be a very bad outcome.

waltworks

  • Magnum Stache
  • ******
  • Posts: 4277
Re: How long can we wait while flattening the curve?
« Reply #614 on: April 16, 2020, 08:28:33 AM »
We have 22 million newly unemployed people. Approximately 18% unemployment rate.

So my answer to the original question is: Not long, not long at all.

https://www.washingtonpost.com/business/2020/04/16/unemployment-claims-coronavirus/

Yeah, the economic damage is getting too extreme, I think. I'm guessing most states will reopen things pretty soon.

We blew it, repeatedly...sigh.

-W

fattest_foot

  • Pencil Stache
  • ****
  • Posts: 844
Re: How long can we wait while flattening the curve?
« Reply #615 on: April 16, 2020, 08:50:35 AM »
When did the goal shift from "stay home to flatten the curve" to "we all need to stay home to prevent a single transmission?"

We were trying to prevent hospitals from being overwhelmed. Preventing young, healthy people from leaving their home and never getting the virus is counter-productive. We need people to get the virus. Restricting everything and preventing transmission completely is the opposite of what we should be doing.

But we have tons of people even here that are so panic stricken that they believe we need to stay locked up for at least another 2 or 3 months for...what exactly?

mathlete

  • Handlebar Stache
  • *****
  • Posts: 1647
Re: How long can we wait while flattening the curve?
« Reply #616 on: April 16, 2020, 08:55:33 AM »
COVID-19 is the number one daily killer in the US right now. Seems strange to get cold feet on the lockdown now of all times.

We have financial magic tricks we can do to ease people's suffering while we come up with the medical magic trick. Let's do that.

If we find safe ways to re-open (i.e. public gatherings or dining in on a limited basis) then sure, let's do that.

Laserjet3051

  • Pencil Stache
  • ****
  • Posts: 845
  • Age: 92
Re: How long can we wait while flattening the curve?
« Reply #617 on: April 16, 2020, 08:58:13 AM »
When did the goal shift from "stay home to flatten the curve" to "we all need to stay home to prevent a single transmission?"

.

We were trying to prevent hospitals from being overwhelmed. Preventing young, healthy people from leaving their home and never getting the virus is counter-productive. We need people to get the virus. Restricting everything and preventing transmission completely is the opposite of what we should be doing.

But we have tons of people even here that are so panic stricken that they believe we need to stay locked up for at least another 2 or 3 months for...what exactly?

Shhhhhh....you were not to supposed to notice moving the goal posts. Ive even heard some "epidemiologists" urging that current social isolation strategies be kept in place until 2022. Fear is a very effective tool for control. Take a look around, even here on MMM

Davnasty

  • Magnum Stache
  • ******
  • Posts: 2582
Re: How long can we wait while flattening the curve?
« Reply #618 on: April 16, 2020, 09:03:28 AM »
When did the goal shift from "stay home to flatten the curve" to "we all need to stay home to prevent a single transmission?"

It hasn't. Where did you get that idea?

Quote
We were trying to prevent hospitals from being overwhelmed. Preventing young, healthy people from leaving their home and never getting the virus is counter-productive. We need people to get the virus. Restricting everything and preventing transmission completely is the opposite of what we should be doing.

But we have tons of people even here that are so panic stricken that they believe we need to stay locked up for at least another 2 or 3 months for...what exactly?

If preventing all transmission was possible, that would be an amazing outcome, not counterproductive. Unfortunately I don't believe that's a possibility right now.

boy_bye

  • Handlebar Stache
  • *****
  • Posts: 2472
Re: How long can we wait while flattening the curve?
« Reply #619 on: April 16, 2020, 09:04:10 AM »
When did the goal shift from "stay home to flatten the curve" to "we all need to stay home to prevent a single transmission?"

We were trying to prevent hospitals from being overwhelmed. Preventing young, healthy people from leaving their home and never getting the virus is counter-productive. We need people to get the virus. Restricting everything and preventing transmission completely is the opposite of what we should be doing.

But we have tons of people even here that are so panic stricken that they believe we need to stay locked up for at least another 2 or 3 months for...what exactly?

Because we don't yet have the mass testing ability to know when people are carriers and when they are not.

Because we don't have any understanding yet of whether someone can be re-infected (seeing several reports out of China and South Korea about folks who tested positive, then negative, then positive again -- could be testing errors but we don't know).

Because we don't know the long-term damage this virus can do to even healthy people who get a massive viral load -- death is not the only shitty outcome -- I am also seeing reports of lasting lung damage, kidney damage, and neurological problems.

Basically because of the bungled response to this in the USA, and the fact that there's not an excellent and transparent process for scientists to quickly share and evaluate and learn from new information, we don't know shit. So we have to either stay home until we DO know shit, or risk hundreds of thousands of deaths / newly disabled people in this country alone.

We are staying home to buy ourselves time to figure out wtf to do. Literally no one knows at this point.
« Last Edit: April 16, 2020, 09:05:57 AM by madgeylou »

fattest_foot

  • Pencil Stache
  • ****
  • Posts: 844
Re: How long can we wait while flattening the curve?
« Reply #620 on: April 16, 2020, 09:09:58 AM »
When did the goal shift from "stay home to flatten the curve" to "we all need to stay home to prevent a single transmission?"

It hasn't. Where did you get that idea?

Quote
We were trying to prevent hospitals from being overwhelmed. Preventing young, healthy people from leaving their home and never getting the virus is counter-productive. We need people to get the virus. Restricting everything and preventing transmission completely is the opposite of what we should be doing.

But we have tons of people even here that are so panic stricken that they believe we need to stay locked up for at least another 2 or 3 months for...what exactly?

If preventing all transmission was possible, that would be an amazing outcome, not counterproductive. Unfortunately I don't believe that's a possibility right now.

What do you mean it hasn't? 49 states not named New York are closed right now. Why?

Even with New York, not a single hospital is overwhelmed. Open everything back up and if we see systems trending back to becoming overwhelmed, you re-implement quarantines.

Look at the post above mine. Just complete fear. "Well, it might kill hundreds of thousands. No choice but to not let anyone leave the house!" This is an insane response. I hope you realize that; maybe in retrospect.

OtherJen

  • Magnum Stache
  • ******
  • Posts: 3621
  • Location: Metro Detroit
Re: How long can we wait while flattening the curve?
« Reply #621 on: April 16, 2020, 09:16:06 AM »
When did the goal shift from "stay home to flatten the curve" to "we all need to stay home to prevent a single transmission?"

It hasn't. Where did you get that idea?

Quote
We were trying to prevent hospitals from being overwhelmed. Preventing young, healthy people from leaving their home and never getting the virus is counter-productive. We need people to get the virus. Restricting everything and preventing transmission completely is the opposite of what we should be doing.

But we have tons of people even here that are so panic stricken that they believe we need to stay locked up for at least another 2 or 3 months for...what exactly?

If preventing all transmission was possible, that would be an amazing outcome, not counterproductive. Unfortunately I don't believe that's a possibility right now.

What do you mean it hasn't? 49 states not named New York are closed right now. Why?

Even with New York, not a single hospital is overwhelmed. Open everything back up and if we see systems trending back to becoming overwhelmed, you re-implement quarantines.

Look at the post above mine. Just complete fear. "Well, it might kill hundreds of thousands. No choice but to not let anyone leave the house!" This is an insane response. I hope you realize that; maybe in retrospect.

You do realize that the hospitals are not overwhelmed *because* of the strict shutdown measures and not in spite of them, right?

Hokkaido prefecture in Japan re-issued a state of emergency and lockdown after lifting the previous one apparently too soon. From my colleagues in Hong Kong, they had a similar experience once they relaxed the first round of restrictions. We should be watching places like Austria now as they begin to ease up, and hopefully learn from these experiences.

I don't think rational people are advocating an indefinite lockdown, but they also aren't advocating a no-holds-barred, open everything and leave it to chance approach. There's a middle ground, and those of us who support that middle ground are not fear-mongering.

mathlete

  • Handlebar Stache
  • *****
  • Posts: 1647
Re: How long can we wait while flattening the curve?
« Reply #622 on: April 16, 2020, 09:16:57 AM »
What do you mean it hasn't? 49 states not named New York are closed right now. Why?

Even with New York, not a single hospital is overwhelmed. Open everything back up and if we see systems trending back to becoming overwhelmed, you re-implement quarantines.

Look at the post above mine. Just complete fear. "Well, it might kill hundreds of thousands. No choice but to not let anyone leave the house!" This is an insane response. I hope you realize that; maybe in retrospect.

There's almost too much inaccuracy and misunderstanding here to correct it all.

Hospitals aren't overwhelmed because we're doing what we're doing.

People can leave the house. You can go grocery shopping. You can go get drive through food. You can take a walk in your neighborhood or ride your bike or take a pleasure drive if that's available to you.

To reiterate; COVID19 is the number one daily killer in the United States right now. And that's after a month of pretty aggressive mitigation measures to limit the spread. This is a rational response to a novel mortality risk that has shot up to being the number one daily killer. Calling the response "complete fear" is uneducated. It reads as a lazy attempt to feel superior to others. "Look, you're afraid and I'm not!!"

This is rationality, not fear.

Davnasty

  • Magnum Stache
  • ******
  • Posts: 2582
Re: How long can we wait while flattening the curve?
« Reply #623 on: April 16, 2020, 09:20:08 AM »
When did the goal shift from "stay home to flatten the curve" to "we all need to stay home to prevent a single transmission?"

It hasn't. Where did you get that idea?

Quote
We were trying to prevent hospitals from being overwhelmed. Preventing young, healthy people from leaving their home and never getting the virus is counter-productive. We need people to get the virus. Restricting everything and preventing transmission completely is the opposite of what we should be doing.

But we have tons of people even here that are so panic stricken that they believe we need to stay locked up for at least another 2 or 3 months for...what exactly?

If preventing all transmission was possible, that would be an amazing outcome, not counterproductive. Unfortunately I don't believe that's a possibility right now.

What do you mean it hasn't? 49 states not named New York are closed right now. Why?

Even with New York, not a single hospital is overwhelmed. Open everything back up and if we see systems trending back to becoming overwhelmed, you re-implement quarantines.

Look at the post above mine. Just complete fear. "Well, it might kill hundreds of thousands. No choice but to not let anyone leave the house!" This is an insane response. I hope you realize that; maybe in retrospect.

"we all need to stay home to prevent a single transmission?" ≠ Closing down non-essential businesses.

Not sure what you're trying to say

This statement right here...

Quote
Open everything back up and if we see systems trending back to becoming overwhelmed, you re-implement quarantines

Tells me you have no clue how this virus works. I would recommend doing a bit of reading if you insist on having strong opinions about our strategy moving forward.


T-Money$

  • Stubble
  • **
  • Posts: 218
  • Location: New York
Re: How long can we wait while flattening the curve?
« Reply #624 on: April 16, 2020, 09:21:56 AM »
When did the goal shift from "stay home to flatten the curve" to "we all need to stay home to prevent a single transmission?"

It hasn't. Where did you get that idea?

Quote
We were trying to prevent hospitals from being overwhelmed. Preventing young, healthy people from leaving their home and never getting the virus is counter-productive. We need people to get the virus. Restricting everything and preventing transmission completely is the opposite of what we should be doing.

But we have tons of people even here that are so panic stricken that they believe we need to stay locked up for at least another 2 or 3 months for...what exactly?

If preventing all transmission was possible, that would be an amazing outcome, not counterproductive. Unfortunately I don't believe that's a possibility right now.

What do you mean it hasn't? 49 states not named New York are closed right now. Why?

Even with New York, not a single hospital is overwhelmed. Open everything back up and if we see systems trending back to becoming overwhelmed, you re-implement quarantines.

Look at the post above mine. Just complete fear. "Well, it might kill hundreds of thousands. No choice but to not let anyone leave the house!" This is an insane response. I hope you realize that; maybe in retrospect.

You do realize that the hospitals are not overwhelmed *because* of the strict shutdown measures and not in spite of them, right?

Hokkaido prefecture in Japan re-issued a state of emergency and lockdown after lifting the previous one apparently too soon. From my colleagues in Hong Kong, they had a similar experience once they relaxed the first round of restrictions. We should be watching places like Austria now as they begin to ease up, and hopefully learn from these experiences.

I don't think rational people are advocating an indefinite lockdown, but they also aren't advocating a no-holds-barred, open everything and leave it to chance approach. There's a middle ground, and those of us who support that middle ground are not fear-mongering.

I think that your argument is confirmation bias.  The initial fatality projections of COVID-19 were off my multiple factors of magnitude. 
Sweden, while practicing social distancing has not had lockdowns and their hospitals are well below capacity (Sweden does have more capacity than the Americas).  Taiwan is similar. 

During the peak of the virus New York City was only using 18% of the hospital beds it had predicted. 

Social distancing does reduce the spread temporarily, but the worst case scenarios also did not come to fruition. 

The virus will spread and there will be more cases as lockdowns are eased.  It is inevitable and should be expected.

« Last Edit: April 16, 2020, 09:23:47 AM by egillespie »

mathlete

  • Handlebar Stache
  • *****
  • Posts: 1647
Re: How long can we wait while flattening the curve?
« Reply #625 on: April 16, 2020, 09:24:02 AM »
Tells me you have no clue how this virus works.

A lot of that post does that. Particularly, the "49 states not named New York." part.

Turn back the clocks four months at it becomes "X cities not named Wuhan". Turn back three months and it's, "Why should Italy and Spain lock down for a disease in China?"

« Last Edit: April 16, 2020, 09:28:54 AM by mathlete »

dandarc

  • Magnum Stache
  • ******
  • Posts: 4034
  • Age: 37
Re: How long can we wait while flattening the curve?
« Reply #626 on: April 16, 2020, 09:25:53 AM »
Should be required reading for anyone posting in this thread:

https://medium.com/@tomaspueyo/coronavirus-the-hammer-and-the-dance-be9337092b56

And in the US we're not even really approaching the measures taken in Italy, let alone what they were able to do in other places. Saying our "lockdowns" are designed to prevent a single transmission is stunning in its ignorance.

We're on a forum where we're aiming to take advantage of compound interest and people still don't seem to grasp how quickly "one day delay in implementing distancing measures leads to a 40% higher peak" can escalate.

mathlete

  • Handlebar Stache
  • *****
  • Posts: 1647
Re: How long can we wait while flattening the curve?
« Reply #627 on: April 16, 2020, 09:28:26 AM »
I think that your argument is confirmation bias.  The initial fatality projections of COVID-19 were off my multiple factors of magnitude.
Sweden, while practicing social distancing has not had lockdowns and their hospitals are well below capacity (Sweden does have more capacity than the Americas).  Taiwan is similar. 

During the peak of the virus New York City was only using 18% of the hospital beds it had predicted. 

Social distancing does reduce the spread temporarily, but the worst case scenarios also did not come to fruition. 

The virus will spread and there will be more cases as lockdowns are eased.  It is inevitable and should be expected.

The initial estimates were without social distancing assumed. Apples and oranges. And they were on the order of a million deaths. Revised estimates, which account for distancing, have 100K deaths still in the fat part of the probability distribution. So even if you wanted to make an apples to apples comparison, that's off by one order of magnitude, not multiple.


T-Money$

  • Stubble
  • **
  • Posts: 218
  • Location: New York
Re: How long can we wait while flattening the curve?
« Reply #628 on: April 16, 2020, 09:49:45 AM »
I think that your argument is confirmation bias.  The initial fatality projections of COVID-19 were off my multiple factors of magnitude.
Sweden, while practicing social distancing has not had lockdowns and their hospitals are well below capacity (Sweden does have more capacity than the Americas).  Taiwan is similar. 

During the peak of the virus New York City was only using 18% of the hospital beds it had predicted. 

Social distancing does reduce the spread temporarily, but the worst case scenarios also did not come to fruition. 

The virus will spread and there will be more cases as lockdowns are eased.  It is inevitable and should be expected.

The initial estimates were without social distancing assumed. Apples and oranges. And they were on the order of a million deaths. Revised estimates, which account for distancing, have 100K deaths still in the fat part of the probability distribution. So even if you wanted to make an apples to apples comparison, that's off by one order of magnitude, not multiple.

I strongly disagree.  The below graphic shows various models, some of which included social distancing and isolation.  Reality is well below all of the predictions.  The main reason for the lack of death and significant illness is because the models were wrong. 

Jon Bon

  • Handlebar Stache
  • *****
  • Posts: 1167
  • Location: Midwest
Re: How long can we wait while flattening the curve?
« Reply #629 on: April 16, 2020, 10:09:37 AM »
I saw this somewhere so I can't take credit for it.

1. The virus is dangerous
2. High unemployment and a wrecked economy is dangerous
3. Authoritative government policies are dangerous

None of these are mutually exclusive.

.






mathlete

  • Handlebar Stache
  • *****
  • Posts: 1647
Re: How long can we wait while flattening the curve?
« Reply #630 on: April 16, 2020, 10:20:16 AM »
I strongly disagree.  The below graphic shows various models, some of which included social distancing and isolation.  Reality is well below all of the predictions.  The main reason for the lack of death and significant illness is because the models were wrong.

I do statistical modeling for a living so I just want to get that out of the way before I go to bat for models in general. My biases have been declared.

I'm very familiar with the IHME model and I know that as you mentioned, it does include the impact of social distancing and isolation. So we can agree there. The graphic shows projection of hospital beds needed. Of all the metrics modeled, this is probably the most difficult one to get right. Modelers are still asked to model it though because it's something that we can potentially take action on.

Since the IHME model was released, daily deaths have generally fit within their cone of uncertainty.

"The models were wrong" is a truism. There is a saying in modeling, "All models are wrong but some are useful." The IHME model in particular has been very useful. I say this as someone who is being paid a lot of money to model this stuff right now. Actual results almost always come in above or below the mean modeled outcome. Sometimes, they come in above or below the 95% confidence intervals too. This is especially the case when we're trying to model something that there is very little information on, such as a novel virus. Pointing at model results, identifying them as "wrong", and using that as a platform to reject empirically based decision making is incredibly stupid though.

The main reason for the lack of death

And for a third time now, COVID19 is the leading daily cause of death in the United States. There is no lack of death.

mathlete

  • Handlebar Stache
  • *****
  • Posts: 1647
Re: How long can we wait while flattening the curve?
« Reply #631 on: April 16, 2020, 10:27:26 AM »
Imagine dropping someone with no prior knowledge of geography in the middle of Kansas and telling them to get to New York City.

Then you give them a model. Wait for the sun to rise every day, and when it does, head in the direction of the sunrise. They'll probably end up several states away from New York, so the model was wrong. But they won't end up in Mexico or California, so the model was probably useful.

Abe

  • Handlebar Stache
  • *****
  • Posts: 1658
Re: How long can we wait while flattening the curve?
« Reply #632 on: April 16, 2020, 10:34:15 AM »
@mathlete - thanks for trying to explain this. Another thing Iíd add is obviously initial models will be incorrect if we do social distancing and reduce transmission, which is basically what all the models showed. You canít win with these guys because the only scenario where they will admit they are wrong is if we didnít do social distancing and hundreds of thousands died. Then we can do the ďtold-you-soĒ dance on the mountains of dead, but us rational people decided it was a bad idea just to prove the contrarians wrong. Thereís no way to prove anything to their satisfaction, and itís no use applying facts or projections to their opinions, which are fixed and not changeable. Iíd say itís not worth getting into a debate. Let them have their opinions, out of context data, and the rest of the us community have our facts. I know youíre trying to prevent misinformation from spreading but at some point itís not worth it.
There are several false statements made above but Iím not got the time to correct them. Suffice to say hospitals in NYC are overwhelmed, so go volunteer if interested in proving that wrong. My cousin and friends would like a break. I canít because Iíve got my patients here to look after.

In other news, NJ had a big spike in deaths reported two days ago and NYC yesterday - these were retrospective reporting of people who had died previously but were not tested. So those spikes arenít signs of a significant change in the overall course. Unfortunately still about 800-1000 deaths per day in that area.
« Last Edit: April 16, 2020, 10:40:00 AM by Abe »

T-Money$

  • Stubble
  • **
  • Posts: 218
  • Location: New York
Re: How long can we wait while flattening the curve?
« Reply #633 on: April 16, 2020, 10:37:17 AM »
I'm involved with numerical modelling too, but in a physical science field.  The models I use are much more complicated than the basic statistical models used in epidemiological modelling.  The track record of disease modelling, the inaccuracy, would be inexcusable in hard sciences. 

As Dr. Fauci has said:

Fauci said he personally remains skeptical about models because theyíre ďonly as good as the assumptions you put into the model.Ē

Whatís more important, he said, is accumulating data that eventually ďtrumps any modelĒ and allows people to modify it.

ďI have no problem with people who are critical of modeling because modeling is inherently an imperfect science. So I donít really have any quibbling with that,Ē he said.


Those quotes are 8 days old, by the way.  The inaccuracies in disease modeling have been known for over a decade now, and those that study how inaccurate they are have recommended in medical journals they are to be met with skepticism when forming public health policy.

The current data shows that the models (including the ones you work with) are not providing accurate disease predictions.  Dr. Fauci acknowledges this.  They are the excuse to promote hysteria and fear, and create public policy which is ultimately destructive. 

Perhaps disease modelling will be more accurate, but only when medicine has international standards in diagnosis and the epidemiological models can use ensemble spreads and advanced methods in output confidence can be established.  Until then, relying on these models, such as the ones you work with, well they wouldn't make it past a 1914 course in Geology, quite frankly.

nereo

  • Senior Mustachian
  • ********
  • Posts: 13562
  • Location: Just south of Canada
    • Here's how you can support science today:
Re: How long can we wait while flattening the curve?
« Reply #634 on: April 16, 2020, 10:39:13 AM »
FWIW, a key component of current efforts is to predict the caseloads where reopening some currently closed sectors will not result in localize outbreaks.  Itís not zero infected patients, but it is most likely fewer than thousands of patients in a given region.

Thereís also good economic and historical data showing that sticking to isolation protocols and business closures for a longer period of time results in LESS economic pain, not more.  Because sick people and fear of becoming sick has an enormous financial cost.

Finally, the majority of the economy remains operational, even its drastically different and front-facing businesses remain closed. This isnít an all-or-nothing proposition here, and we are finding ways of getting more of those operational in the weeks to come.

Unlike some, I do not view this as a choice between saving some people and saving the economy.  Once there was widespread community transmission the economic costs were going to be high regardless. Our current and continued strategy may likely save a substantial number of lives while also minimizing economic impacts.  Thatís the idea, anyway... uncharted waters and all.

mathlete

  • Handlebar Stache
  • *****
  • Posts: 1647
Re: How long can we wait while flattening the curve?
« Reply #635 on: April 16, 2020, 10:41:16 AM »
@mathlete - thanks for trying to explain this. Another thing I’d add is obviously initial models will be incorrect if we do social distancing and reduce transmission, which is basically what all the models showed. You can’t win with these guys because the only scenario where they will admit they are wrong is if we didn’t do social distancing and hundreds of thousands died. Then we can do the “told-you-so” dance on the mountains of dead, but us rational people decided it was a bad idea just to prove the contrarians wrong. There’s no way to prove anything to their satisfaction, and it’s no use applying facts or projections to their opinions, which are fixed and not changeable. I’d say it’s not worth getting into a debate. Let them have their opinions and the rest of the us community have our facts. I know you’re trying to prevent misinformation from spreading but at some point it’s not worth it.

Lol thanks.

To be a statistical modeler, you have to have thick skin. In 18 months, it could turn out that a whole bunch of people had COVID19 months ago and thus our denominators were all messed up. Maybe it's no deadlier than the flu. I don't think that outcome is likely based on what we've seen so far, but it's possible.

If that happens though, I'll still be here defending the lockdown. Because you have to defend the process, not the outcome. That's what will help you next time. You evaluate whether or not we made the best decisions based on the information we had at the time. If the answer is yes, then the next time a pandemic happens and the mortality rate really is 10X or 20X the flu, we'll take all the same actions and save the lives then.

If we only consider outcomes though, we're going to do absolutely nothing in the face of that 10X or 20X pandemic next time. And hundreds of millions of people will die.
« Last Edit: April 16, 2020, 10:43:59 AM by mathlete »

js82

  • Bristles
  • ***
  • Posts: 469
Re: How long can we wait while flattening the curve?
« Reply #636 on: April 16, 2020, 10:45:50 AM »
We have 22 million newly unemployed people. Approximately 18% unemployment rate.

So my answer to the original question is: Not long, not long at all.

https://www.washingtonpost.com/business/2020/04/16/unemployment-claims-coronavirus/

Bingo.  People will be willing to risk COVID-19 so they can continue to feed their family, then they'll realize the sky isn't falling after all.

I'm beginning to come an unpleasant contrarian view: We're stuck between 2 ugly, ugly alternatives - and it's becoming apparent to me that we may no longer have space to "thread the needle".  While I don't disagree with the notion that keeping up this degree of social distancing is unsustainable, I think you're underestimating the ugliness(from a public health perspective) if we go back to "business as usual" (or "business as usual" plus homemade face masks).

Think New York is bad?  1% of the population has tested positive.  From what we know of estimated mortality rates, 5-10% of the population may have actually had the virus, but been asymptomatic or had a case sufficiently mild as to not merit testing(and thus have some resistance).  Based on estimated spread rates, you probably need somewhere in the range of 60-75% of the population to be infected (depending on the precise degree of contagiousness) in a no-social-distancing scenario.  So you're still easily looking at increasing casualties by a factor of 6-10x.  Maybe more than that if we open the floodgates and overwhelm hospital capacity.

And the numbers so far are dominated by a few large cities: New York, Detroit and New Orleans most notably.  Casualty rates in medium and large cities across the country would get seriously ugly under a no-social-distancing scenario.

On the other hand, another 9(or even 6) months of this level of social distancing will mean literal economic collapse - and that comes with its own massive human suffering, even if it's not captured in deaths and case counts


Reopening things rapidly will kill a lot of people, but the collective damage (maybe not in terms of death, but in terms of aggregate suffering) from trying to keep everything closed for much more than another month will also be tremendously, tremendously ugly.  I hope I'm wrong, but even the best outcomes look extremely ugly from where I'm standing - it's just a question of what form that damage takes.
« Last Edit: April 16, 2020, 10:55:33 AM by js82 »

Abe

  • Handlebar Stache
  • *****
  • Posts: 1658
Re: How long can we wait while flattening the curve?
« Reply #637 on: April 16, 2020, 10:54:25 AM »
Thatís true, and why thereís no plan for 6-9 months of social distancing at this point. Our economy is too weak to handle that. Reinfection in East Asia suggests that even with maximal efforts, there will be a second peak. We can see how Germany fares with planned reopening. Weíre going to have to bite the bullet eventually. Hopefully manufacturers can scale up critical care support equipment in time for the eventual second spike. Also, hopefully enough healthcare workers will have been exposed and recovered that it wonít overwhelm the hospitals. Weíll see what happens!
« Last Edit: April 16, 2020, 10:57:14 AM by Abe »

mathlete

  • Handlebar Stache
  • *****
  • Posts: 1647
Re: How long can we wait while flattening the curve?
« Reply #638 on: April 16, 2020, 11:00:30 AM »
Another truism about modeling is that a model is only as good as its assumptions.

Problem is, people are making those assumptions whether or not the model exists anyway. Governments ask their scientists how bad COVID19 is and the scientists say that it has an R0 of 1.4-2.7 and mortality of 1%-2%. And then those are the assumptions. If the consensus of the models is wrong, then it's likely that one of the assumptions is wrong. Then you revisit the assumption setting process. Did we have any data that said that either R0 or mortality would be much lower? If yes, then cool, we found a systemic flaw in epidemiology and we can work to correct it. If no, then that's life. The trajectory of novel viruses is extremely difficult to predict, even with some of the world's smartest people working on it.

GardenerB

  • 5 O'Clock Shadow
  • *
  • Posts: 61
Re: How long can we wait while flattening the curve?
« Reply #639 on: April 16, 2020, 11:04:04 AM »
Agreed you need thick skin for modelling.  I think we are somewhere between these 2 models/scenarios.  Both of these models track deaths/day and/or deaths/day/million people.

Both show/assume a very high infection rate and of course very low detection rate, but, both are on track fairly accurately to track the peak of the curve (for deaths).  At any time close to peak deaths, infection rates are 2 to 10% (one estimated Italy and Spain hit 20% at peak).

https://imperialcollegelondon.github.io/covid19estimates/#/details/United_Kingdom

https://medium.com/@ali_razavian/covid-19-from-a-data-scientists-perspective-95bd4e84843b

The second model assumes the virus already spread through most of all populations, which may be wrong, but the estimated deaths/day/million citizens is still tracking (log scale).  Deaths come from infections 3-4 weeks ago (5 days incubation, 2-3 weeks symptomatic).  So I assume the countries that had better lockdowns (more obedient citizens), will show a sharper decrease in deaths/day 4 weeks after lockdown compared to other countries.  E.g. compare Germany versus Spain in the second model.

Either way the peaks have to be under hospital capacity - the main goal all along.

I don't agree with the second model stating it's all over 75 days after starting.  There will be multiple waves after restrictions ease.  Then it will be similar to seasonal flu - may have vaccine in 18 months, but just like flu, doesn't work every year (new strains), and this will be yet another bad COVID virus on top of colds, flu, etc. that we deal with, but way more contagious.

GB



js82

  • Bristles
  • ***
  • Posts: 469
Re: How long can we wait while flattening the curve?
« Reply #640 on: April 16, 2020, 11:24:08 AM »
Agreed you need thick skin for modelling.  I think we are somewhere between these 2 models/scenarios.  Both of these models track deaths/day and/or deaths/day/million people.

Both show/assume a very high infection rate and of course very low detection rate, but, both are on track fairly accurately to track the peak of the curve (for deaths).  At any time close to peak deaths, infection rates are 2 to 10% (one estimated Italy and Spain hit 20% at peak).

https://imperialcollegelondon.github.io/covid19estimates/#/details/United_Kingdom

https://medium.com/@ali_razavian/covid-19-from-a-data-scientists-perspective-95bd4e84843b

The second model assumes the virus already spread through most of all populations, which may be wrong, but the estimated deaths/day/million citizens is still tracking (log scale).  Deaths come from infections 3-4 weeks ago (5 days incubation, 2-3 weeks symptomatic).  So I assume the countries that had better lockdowns (more obedient citizens), will show a sharper decrease in deaths/day 4 weeks after lockdown compared to other countries.  E.g. compare Germany versus Spain in the second model.

Either way the peaks have to be under hospital capacity - the main goal all along.

I don't agree with the second model stating it's all over 75 days after starting.  There will be multiple waves after restrictions ease.  Then it will be similar to seasonal flu - may have vaccine in 18 months, but just like flu, doesn't work every year (new strains), and this will be yet another bad COVID virus on top of colds, flu, etc. that we deal with, but way more contagious.

GB

The nice thing about science is that you can easily, instantly refute hypotheses with evidence that that clearly demonstrates that they are incorrect.  This is(unfortunately) the case for the death rate used in the 2nd article.

Consider:
1) The current death rate in New York State is already at 0.08% of the population.  That's POPULATION, not infections.
2) Those deaths are overwhelmingly concentrated in the NYC area, which has ~60% of the state's population if you include Long Island
3) We're not done yet - deaths are going down in NYC, but they're nowhere near over, and it's quite likely that plenty of people still haven't been infected because they've been hunkering down.
4) Upstate NY(~40% of the state's population) has barely been hit(thusfar) in comparison with NYC.

All this points to the real mortality rate being at least several times higher than the 0.045%/0.06% figure the author of the 2nd article is basing his analysis on.  I'd love for him to be right and have the real death rate be 0.06% - but the data doesn't back it up.  A study from Germany yielded an estimate closer to 0.4%, which is in the realm of believabilty when you look at actual data, subject to adjustments from the quality of a country's medical system.
« Last Edit: April 16, 2020, 11:27:45 AM by js82 »

GardenerB

  • 5 O'Clock Shadow
  • *
  • Posts: 61
Re: How long can we wait while flattening the curve?
« Reply #641 on: April 16, 2020, 11:46:44 AM »
Models are averaging out across entire countries, not down to the city level of course, so there can easily be 10-fold differences in rates based on population densities, co-morbidity effects etc.  Similar to how data for the town of Vo in northern Italy had it's own unique IFR and deaths/day (more old people, pre-existing conditions etc.)

Agreed it all comes down to is it/was it much more contagious than first thought and lower IFR (infection fatality rate), or less contagious and more deadly.  Still, interesting to track/see since both models are tracking with good accuracy.

GB

Jon Bon

  • Handlebar Stache
  • *****
  • Posts: 1167
  • Location: Midwest
Re: How long can we wait while flattening the curve?
« Reply #642 on: April 16, 2020, 11:49:34 AM »
Honest question:

Does anyone have access to raw data concerning age/race/gender when it comes to cases and deaths?

I know in my state currently the median age of death is 78 years old. I I am curious if the data is similar nationwide?

T-Money$

  • Stubble
  • **
  • Posts: 218
  • Location: New York

T-Money$

  • Stubble
  • **
  • Posts: 218
  • Location: New York
Re: How long can we wait while flattening the curve?
« Reply #644 on: April 16, 2020, 12:52:52 PM »
Honest question:

Does anyone have access to raw data concerning age/race/gender when it comes to cases and deaths?

I know in my state currently the median age of death is 78 years old. I I am curious if the data is similar nationwide?

Iíd love to see this too.   Anyone?

I know it exists for other countries.   Itís very difficult to find for the US.  I assume there is a politically motivated reason for this.

I saw a reference in a New Jersey newspaper a few days ago stating 1/2 of NJ state deaths were from residents of nursing homes and hospice. 

mathlete

  • Handlebar Stache
  • *****
  • Posts: 1647
Re: How long can we wait while flattening the curve?
« Reply #645 on: April 16, 2020, 01:16:59 PM »
Honest question:

Does anyone have access to raw data concerning age/race/gender when it comes to cases and deaths?

I know in my state currently the median age of death is 78 years old. I I am curious if the data is similar nationwide?

NYC reports cases and deaths by age bracket and sex on a daily basis:

https://www1.nyc.gov/site/doh/covid/covid-19-data.page#download

mathlete

  • Handlebar Stache
  • *****
  • Posts: 1647
Re: How long can we wait while flattening the curve?
« Reply #646 on: April 16, 2020, 02:02:07 PM »
NYC reports cases and deaths by age bracket and sex on a daily basis:

https://www1.nyc.gov/site/doh/covid/covid-19-data.page#download

Using yesterday's data (because today's is a little odd since they started reporting deaths with probable diagnosis for the first time)

Code: [Select]
Age Cases Deaths CFR Pct of Deaths
0-17 2150 3 0.1% 0.0%
18-44 42127 309 0.7% 4.5%
45-64 40559 1,581 3.9% 23.1%
65-74 14025 1,683 12.0% 24.6%
75+ 12331 3,263 26.5% 47.7%

CFR is "case fatality rate" or just simply confirmed deaths over confirmed cases. Pct. of Deaths is the percentage of total deaths attributable to that age bracket.

secondcor521

  • Magnum Stache
  • ******
  • Posts: 3137
  • Age: 51
  • Location: Boise, Idaho
  • Big cattle, no hat.
    • Age of Eon - Overwatch player videos
Re: How long can we wait while flattening the curve?
« Reply #647 on: April 16, 2020, 02:09:43 PM »
Honest question:

Does anyone have access to raw data concerning age/race/gender when it comes to cases and deaths?

I know in my state currently the median age of death is 78 years old. I I am curious if the data is similar nationwide?

Idaho reports cases by age and gender, and deaths by race and ethnicity:

https://public.tableau.com/profile/idaho.division.of.public.health#!/vizhome/DPHIdahoCOVID-19Dashboard_V2/Story1

Unfortunately, to assess relative risk I think you'd have to normalize it by the relative sizes of those populations.  So, for example, it's not really surprising that most deaths in Idaho are among white non-Hispanics, since Idaho is a very white, non-Hispanic state.

Anecdotally, and similar to most other places I think, Idaho's deaths are more frequent among the elderly and those with known health risk factors.

GardenerB

  • 5 O'Clock Shadow
  • *
  • Posts: 61
Re: How long can we wait while flattening the curve?
« Reply #648 on: April 16, 2020, 02:34:08 PM »
Wondering for NYC and NY State what true infection rate will end up being.  Most reports show 500,000 tests done, 41% positive.  That's only 2.6% of the state tested (or 6% if assumed most are in NYC alone).

Doesn't account for the majority of infected who never get symptoms but still can spread.  From other countries' data this is as high as 80% of all infected people (asymptomatic).  Only Iceland tested all people randomly, unlike most countries who only did limited testing on people who were already sick.  Their data/histogram showed equal distribution across whole population, whereas countries testing only sick showed histograms weighted to older populations.

It could still point to (especially for NYC like other dense cities), that the infection rate is closer to 40 to 50% for the whole city.  That would make the IFR much lower than CFR and would make more sense.  And makes the assumed IFR closer to other dense countries as shown in the Imperial College models (IFR around 0.4%).

So it seems the same trade off is there - either way more people are infected than estimated or IFR is higher than estimated.

the_gastropod

  • Stubble
  • **
  • Posts: 237
  • Age: 34
  • Location: Brooklyn, NY
Re: How long can we wait while flattening the curve?
« Reply #649 on: April 16, 2020, 02:49:39 PM »

Even with New York, not a single hospital is overwhelmed. Open everything back up and if we see systems trending back to becoming overwhelmed, you re-implement quarantines.

Look at the post above mine. Just complete fear. "Well, it might kill hundreds of thousands. No choice but to not let anyone leave the house!" This is an insane response. I hope you realize that; maybe in retrospect.

Hi. New Yorker here. I went on a be-masked walk yesterday and saw (and smelled) the mobile morgues (refrigerated trailers) setup next to the Brooklyn Hospital. Please stop downplaying this. Itís dangerous, insensitive, and plainly ignorant.