Cont: 2019-nCoV / Corona virus Pt 3

[lomiller]

Yes, it did. What is your point? You assume it will take another 12 days to go to 85,000? That's not how this works, it hasn't happened that way anywhere else, and in a few weeks all of you will see that.

in the US it took:
13 days for deaths to go from 10 to 100,
9 days to go from 100 to 1000
12 days to go from 1000 to 10 000


This regularity is a function of how infections the virus is and how often an infected person comes into contact with a susceptible person. The first won't change soon, the second should be reduced with social distancing. Without social isolation the pattern would continue until a large percentage of the population has had the disease and is no longer susceptible. In the US it would take 2 more iterations before ~50% of the population has had the virus and is no longer susceptible, assuming an IFR of 0.6%.

We will know in a couple weeks just how well social distancing is working, but I think the effect would need to be dramatic to do more than stretch that period to 20 days. Even if the social distancing and lock-downs work, the virus will still be around and case numbers will start the same pattern all over again until most people have been infected or there is a vaccine.

 

[Elagabalus]

I already had coronavirus. Confirmed by a test purchased through my son's pediatrician. So, unlikely. It was like a bad flu, excruciating headache, fever and dry cough but no breathing difficulties.

So that's why you're bitching about the precautions now?

Interesting how pneumonia deaths have magically plummeted now that everything is Covid-19. Apparently Covid-19 reduces the number of pneumonia cases. Who would've guessed?

Uh, no. The pneumonia cases went down because of the social distancing which you said is ineffective.

 

[Doghouse Reilly]

So that's why you're bitching about the precautions now?




Uh, no. The pneumonia cases went down because of the social distancing which you said is ineffective.

You can't prove that that is why the pneumonia cases have gone down any more than I can prove that patients who would have died of pneumonia are now dying WITH (not from) coronavirus.

And no, I thought the precautions were dumb from the beginning, I didn't care if I had it or not but my wife did care. I did the proper quarantine and stick to the measures that make people comfortable.

 

[Athyrio]

They look pretty on-the-nose right now.

Yes, the numbers shouldn't go as high as crescent's forecast if the US implements mitigation strategies.

But according to DR, we're not supposed to include those mitigating factors even though crescent's spreadsheet preceded them..

 

[Doghouse Reilly]

But according to DR, we're not supposed to include those mitigating factors even though crescent's spreadsheet preceded them..

No, you can go ahead and include them. Even with all the mitigation strategies, nobody predicts death rates as low as I am.

 

[marting]

We will know in a couple weeks just how well social distancing is working, but I think the effect would need to be dramatic to do more than stretch that period to 20 days. Even if the social distancing and lock-downs work, the virus will still be around and case numbers will start the same pattern all over again until most people have been infected or there is a vaccine.

I'm more optimistic. I think the social distancing that has been imposed for a few weeks now will get R down to well under 1. My guess is around .6 which is still double Wuhan's R after their stronger, full lockdown, was imposed.

However, it takes about 3 weeks for a change in R to be reflected in death rates and there is the complicating factor that households stretch things out when one member gets it there is a high probability that others will too, delayed by 5 days or so.

So, I expect R to stabilize and be largely reflected in death rates in another 3 weeks when we should see death rates decline by a factor of .6 every 5 days or so.

R0 might be lower or higher than .6. I don't expect it to get down to .3 like in Wuhan but if it can be held at close to 1.0 then the deaths per day will be roughly constant. That would probably be acceptable for the next 12 months until a vaccine if the daily rate was, say, under 100.

I expect the peak death rate to occur in the next 7 days and this will mean that R dropped below 1.0 weeks prior to that.

There is also the herd immunity aspect. Over time as more of the population gets infected and recovers R decreases relative to R0. But I don't see the difference as significant unless we get far more deaths.

 

[marting]

No, you can go ahead and include them. Even with all the mitigation strategies, nobody predicts death rates as low as I am.

Bill Gates indicated he thought the numbers of deaths could well be below the range the White House recently projected that had a low end of 100k. So he may well be in line with your predictions.

OTOH, he said that conditional on strong social distancing / shutdowns in effect and I believe your position is they make no difference.

 

[Athyrio]

No, you can go ahead and include them. Even with all the mitigation strategies, nobody predicts death rates as low as I am.

Obviously.

 

[Doghouse Reilly]

Thanks.


It may have been discussed already, but I must have missed it (my MO is to post links to previous discussions, if I know they exist).

My bad.

Back to Doghouse Reilly's claims/arguments/whatever: if covid-19 has a seasonality similar to that of the flu, then places like Indonesia can look to no respite at all.

JeanTate, with regard to answering the questions you had for me earlier, I have simply not had time to go back and answer them, nor have I had time to compile the questions I'd asked of you which you overlooked because of the rate the thread was moving. At this point I don't know if I'll get to them, but will try. Honestly I may simply not have the expertise to do so to a level that would be satisfactory, as I am not a scientist, nor even a particularly good debater or presenter--of-cases. I'm just a guy with what I think are well founded opinions about this. Time will make my case for me.

You definitely raised some genuine issues with my posts and certain inconsistencies, and taught me a few things. I think as the course of the thread has played, out, I've clarified a lot of what I believe, so I hope that will suffice for now, and you've definitely given me cause to treat the issue more carefully and not assume everything I read can be immediately interpreted in a way that supports my arguments.

 

[Doghouse Reilly]

Bill Gates indicated he thought the numbers of deaths could well be below the range the White House recently projected that had a low end of 100k. So he may well be in line with your predictions.

OTOH, he said that conditional on strong social distancing / shutdowns in effect and I believe your position is they make no difference.

Yes, thanks for mentioning this, and I'm surprised that more people in this thread don't at least share his view. But I know the social distancing thing is a nonfalsifiable claim at this point, on both sides which is why the bet isn't even <100k US coronavirus deaths for 2020 but is instead <85k US flu and coronavirus combined for 2020, which is far, far below even the most optimistic projections being discussed, even taking all mitigation strategies into account.

 

[marting]

Yes. But I know it's a nonfalsifiable claim at this point, on both sides which is why the bet isn't even <100k US coronavirus deaths for 2020 but is instead <85k US flu and coronavirus combined for 2020, which is far, far below even the most optimistic projections being discussed, even taking all mitigation strategies into account.

Not really.

IHME has the expected deaths totaling a bit under 82k with a range of 50k to 137k and quite flat by Aug 1. Of course their models assume continuing social distancing and discount secondary peaks if/when social distancing is relaxed.

https://covid19.healthdata.org/projections

 

[Doghouse Reilly]

Not really.

IHME has the expected deaths totaling a bit under 82k with a range of 50k to 137k and quite flat by Aug 1. Of course their models assume continuing social distancing and discount secondary peaks if/when social distancing is relaxed.

https://covid19.healthdata.org/projections

Am I missing what you're seeing? It looks to me like they're projecting 93,531 COVID-19 deaths by August 4, 2020. I'm predicting <85k COMBINED flu and COVID-19 by the end of 2020.

If their projections were in line with mine why would everyone be acting like I'm stupid with my head in the sand?

JeanTate thinks I'm a fool for not realizing we'll hit 85k coronavirus deaths in 2-4 weeks.

 

[EHocking]

You can't prove that that is why the pneumonia cases have gone down any more than I can prove that patients who would have died of pneumonia are now dying WITH (not from) coronavirus.

And no, I thought the precautions were dumb from the beginning, I didn't care if I had it or not but my wife did care. I did the proper quarantine and stick to the measures that make people comfortable.

I suggest that you continue to take precautions- as there is good evidence that you can catch COVID19-19 even after you believe you’ve recovered. Added to that is the fact that testing can also return false-negatives.

 

[marting]

Am I missing what you're seeing? It looks to me like
they're projecting 93,531 COVID-19 deaths by August 4, 2020.

This was directly copied from their website and was updated 4/4/2020

81,766COVID-19 deaths
projected by August 4, 2020

 

[lomiller]

No, you can go ahead and include them. Even with all the mitigation strategies, nobody predicts death rates as low as I am.

Not even the scientists who've studied it...

BTW CFR and IFR are (mostly) independent of mitigation strategies, since mitigation prevents people from getting the virus while death rates wrt people who actually get the virus.

 

[Doghouse Reilly]

This was directly copied from their website and was updated 4/4/2020

81,766COVID-19 deaths
projected by August 4, 2020

81,766 by August 4 is vastly more deaths than I'm predicting. As I said, I'm predicting in the neighborhood of that number by the end of 2020, FLU AND COVID-19 COMBINED (not shouting, just emphasizing).

 

[Doghouse Reilly]

I suggest that you continue to take precautions- as there is good evidence that you can catch COVID19-19 even after you believe you’ve recovered. Added to that is the fact that testing can also return false-negatives.

Thank you, I do genuinely appreciate the advice and concern. I feel great but will definitely take precautions.

 

[marting]

81,766 by August 4 is vastly more deaths than I'm predicting. As I said, I'm predicting in the neighborhood of that number by the end of 2020, FLU AND COVID-19 COMBINED (not shouting, just emphasizing).

No, because the graph they show levels out well before Aug. 4. They actually are projecting 0 deaths after June 22 and if mitigation was extended to the end of the year that would presumably continue to be the case.

This doesn't actually occur of course. There are always stragglers not to mention that the model clearly assumes that all critical patients have either recovered or died by then and that's not going to be true.

The flu is spread similarly to COVID. The mitigation their model uses would be expected to reduce the flu R even more since it already starts of much lower than COVID's. The other problem is that flu death are estimates. I don't believe flu deaths are required to be reported so they do surveillance estimates.

 

[Doghouse Reilly]

No, because the graph they show levels out well before Aug. 4. They actually are projecting 0 deaths after June 22 and if mitigation was extended to the end of the year that would presumably continue to be the case.

This doesn't actually occur of course. There are always stragglers not to mention that the model clearly assumes that all critical patients have either recovered or died by then and that's not going to be true.

Okay, very interesting. Is this model not considered reputable here? Nobody in this thread seems to think those figures are even remotely reasonable. Regardless, those totals are still higher than what I'm saying, because there have already been thousands of flu deaths this year that would put us over 85k.

 

[lomiller]

Okay, very interesting. Is this model not considered reputable here? Nobody in this thread seems to think those figures are even remotely reasonable.

Don't confuse the model with the scenario.