Coronavirus

[AlanSubie4Life]

More people infected, more chances for virus to make someone dead. Serology tests in NYC show ~25% people in NYC have coronavirus antibodies.

But that’s exactly what I said (except I used the last figure of 20% I was aware of).

I think you lost the thread somewhere. The discussion was an extrapolation of mortality to the rest of the US based on NYC mortality. So infection rates don’t matter (unless the attack rates and incidence are stratified by age differently in other locations, as discussed).

What we were not discussing was the likelihood of achieving a certain level of incidence in a given location.

You may want to go back about 5-6 posts to see the original (silly!) contention about how this is no big deal because not that many people have died relative to all the other reasons people have died.

 

[scottf200]

Ex-Obama health care head. -- Andy Slavitt @ on Twitter
"COVID-19 may be under control in Florida but they have an unprecedented “pneumonia” crisis."

 

[davepsilon]

If you take into account incubation periods we are still "in phase 1", while phase 2 started two days ago we haven't had time for phase 2 incubations to turn in to new cases.

That's why I'm looking at SAH vs Phase 1 now. We still have 2 more days left in this week of phase one numbers. It'll be another week or two before we have any idea about phase 2.

Yes. rt.live website has been showing slight upticks in their modeled effective transmission rate as the SAH ages out. I will be following it closely due to the potential economic implications. Though I think the US public has little patience for a second shutdown so I doubt any state that starts sloping upward will act pre-emptively before hospitals are basically overflowing. But with any luck won’t get there.

Flat or slowly oscillating disease rates as the public health measures are eased are not a problem. It’s a potential return to exponential growth scenario that worries me. I think that’s a low probability. China experience thus far is encouraging. But probability of relapse to exponential transmission is uncomfortably high given the implications. Maybe 1-10% (my own wag)

 

[AlanSubie4Life]

China experience thus far is encouraging.

How applicable is China’s situation to this situation? Approximately how many untracked infections did they have in Hubei province when they finally reopened? What were their daily case counts at that point?

I too think it is possible for us to thread the needle, if we get a bit lucky, but I haven’t been looking to China for any guidance on what our chances are for success, because my impression was that they had virtually no disease circulation (community transmission) upon reopening. But maybe I’m missing something.

Interesting how different the rt.live data is from the covid19-projections.com website Rt values. I guess we have to ask whether we should trust machine learning, or Instagram.

COVID-19 Infections Tracker

About covid19-projections.com

Rt Covid-19

 

[MC3OZ]

JPR007 on Twitter

In the week since U.S. auto factories reopened after coronavirus lockdowns, workers at all three Detroit automakers have tested positive for COVID-19 but only Ford Motor Co has temporarily closed plants

Personally I don't think factories need to close if they have a good COVID-19 risk reduction strategy in place and all possibly infected workers self-isolate...

My point is, there is no moral outrage in the media or much reporting at all in the media, if a worker tests positive at Tesla it will be big news.

 

[MC3OZ]

This data isn't exact some dates are missing or possibly wrong. but gives an idea of the trend.

If we factor in increased testing, not too bad, but too early to claim a definite decline..

4th May - California - 1,320 new cases Alameda 27 new cases
5th May - California - 2,530 new cases Alameda 33 new cases
6th May - California - 1,879 new cases Alameda 54 new cases
7th May - California - 1,441 new cases Alameda 54 new cases
8th May - California - 1,533 new cases Alameda 44 new cases
9th May - California - 2,580 new cases Alameda 62 new cases
10th May - California - 1,230 new cases Alameda 41 new cases
11th May - California - 1,286 new cases Alameda 37 new cases
12th May - California - 1,735 new cases Alameda 32 new cases
13th May - California - 1,967 new cases Alameda 45 new cases
14th May - California - 1,880 new cases Alameda 56 new cases
15th May - California - 2,034 new cases Alameda 66 new cases
16th May - California - 2,866 new cases Alameda 51 new cases
17th May - California - 1,580 new cases Alameda 41 new cases
18th May - California - 2,093 new cases Alameda 65 new cases
20th May - California - 2,081 new cases Alameda 38 new cases
21th May - California - 2,316 new cases Alameda 49 new cases
22nd May - California - 2,106 new cases Alameda 99 new cases
23rd May - California - 1,947 new cases Alameda 59 new cases
24th May - California - 1,951 new cases Alameda 80 new cases
25th May - California - 2,382 new cases Alameda 27 new cases
26th May - California - 2,908 new cases Alameda 112 new cases
27th May - California - 1,779 new cases Alameda 63 new cases
28th May - California - 2,242 new cases Alameda 48 new cases

California Coronavirus: 103,797 Cases and 4,039 Deaths (COVID-19 ) - Worldometer

 

[TespaceX]

This data isn't exact some dates are missing or possibly wrong. but gives an idea of the trend.

If we factor in increased testing, not too bad, but too early to claim a definite decline..

With no information on what the number of tests conducted is, how can you make that conclusion?

 

[AlanSubie4Life]

there is no moral outrage in the media or much reporting at all in the media, if a worker tests positive at Tesla it will be big news.

This is probably true, though I would argue that Elon has kind of acted like a lightning rod, and if that unfortunate event transpires, has likely brought the attention upon himself with his demonstrably false tweets about coronavirus. Not that the media would have treated Tesla well in that event, regardless. But it's definitely going to be more of a focus now due to the recent history, if (when?) it happens.

I like how he doubled down on his end of April prediction. Lol. Guess he must be short of sleep with the new baby.

Bloomberg -
Elon Musk Is the Hero America Deserves

"As for the virus and his predictions of its imminent disappearance, Musk refuses to back down, despite clear evidence to the contrary. “I think the statistics became unreliable at the point at which they included those who weren’t actually tested for Covid but simply had Covid-like symptoms,” he says. “The statistics became bogus probably around mid-April. There’s about a hundred Covid-like symptoms. Basically anything. And then the stimulus bill gave a major incentive to have someone regarded as having Covid. The data is no longer valid. That said, I would say I was off by maybe three or four weeks.”"

 

[MC3OZ]

With no information on what the number of tests conducted is, how can you make that conclusion?

There are report of testing being ramped up... bit no official number that I know of

 

[golfcart]

I wish these studies finding a very large percentage of asymptomatic cases would clarify whether that number is at the time of testing or whether they followed up later too. Everyone will test positive before they show symptoms.

It isn't perfect but they said the following:

"The first recorded fever on board the ship was a febrile passenger on day 8. Isolation protocols were immediately commenced, with all passengers confined to cabins and surgical masks issued to all. Full personal protective equipment was used for any contact with any febrile patients, and N95 masks were worn for any contact with passengers in their cabins. The crew still performed duties, including meal services to the cabin doors three times a day, but rooms were not serviced. Expedition staff helped with crew duties at meal service."

Given that they were completely isolated from the outside world for 28 days and isolation protocols on the ship commenced on day 8, my opinion is that it is fair to assume (based on our knowledge of how the virus typically progresses) that most who tested positive would have developed symptoms by day 28 (after 20 days of isolation protocols) if they were going to develop them. Thoughts?

 

[SageBrush]

Ex-Obama health care head. -- Andy Slavitt @ on Twitter
"COVID-19 may be under control in Florida but they have an unprecedented “pneumonia” crisis."

I expect close to 100% of Covid-19 deaths to list pneumonia on the death cert, along with Covid-19 infection.

Nothing nefarious going on in the death cert reporting from what has been shown, although I suppose it leaves some wiggle-room for corrupt tabulations by politically motivated or directed counters.

 

[byeLT4]

I find the pneumonia information interesting. To me the Texas death numbers for Covid seem really low vs cases, however I am absolutely, positively in no way trying to say our red state is covering up anything....

Back to facts. 1 new case in my county yesterday. And my neighbor who I believe already lost toes was transferred back to the hospital and may possibly lose a leg now. Not sure if this is from clotting or?? She is in her 70s and was on a respirator for 11 days IIRC. Although it hasn't taken her life it has taken an incredible toll on her. She is a hell of a fighter and I wish her the best.

 

[SageBrush]

I expect close to 100% of Covid-19 deaths to list pneumonia on the death cert, along with Covid-19 infection.

Nothing nefarious going on in the death cert reporting from what has been shown, although I suppose it leaves some wiggle-room for corrupt tabulations by politically motivated or directed counters.

I just checked Florida. They report 2,364 Covid deaths to date. Since historical pneumonia cases are < 1000 for Feb-May and Pneumonia is 4259 in 2020 for the same time period, it would not be surprising to find that Covid-19 deaths are being under-reported.

Florida really should review its pneumonia cases.

 

[SageBrush]

I have written before on the completion of a death certificate, but it is topical again so here goes:

The general outline is:
1. Immediate cause of death
2. Due to
2a. Primary
2b. Secondary
3. Contributing factors

If I was completing a Covid-19 death in an elder with co-morbidities it might look like this:

1. Pneumonia
2a. Covid-19
2b. Chronic lung disease
3. Age, obesity, diabetes, htn

----
It is not uncommon for physicians to complete death certs incorrectly, but pneumonia and Covid-19 will show up on the cert *somewhere*
An incorrectly filled out death cert of the above patient might look like this:
1. Respiratory arrest
2. Pneumonia
3. Covid-19, Diabetes, HTN

 

[cusetownusa]

Can you propose some ways in which NYC would be significantly different? I believe the demographic pyramid is quite similar between NY and the rest of the US, and that seems like it would be the main determinant of IFR.

To me it seems like the main thing that would be different would be the speed of spread and Rt, but that wouldn't affect the IFR, unless the hospitals were overwhelmed (they weren't).

I've heard a lot of people claiming NYC can't possibly be representative, but I'm looking for good reasons why it wouldn't be. The most recent "reason" I have heard is how they were stuffing sick people into nursing homes, and that drove up their IFR. But those types of deaths (care facilities) only make up 1/4 of the total deaths as I recall, and the actual additional number caused by poor policies would clearly be a smaller fraction than that. So I don't see that as being a particularly significant factor.

To me it seems like quite a good dataset to extrapolate from - and it's one of the only ones we have, due to the serology testing and high incidence (which makes the serology testing more reliable).

you said it yourself. The main difference between nyc and where I live in NY is the population density and public transportation. Not sure extrapolating from NYC to the rest of the country makes sense. Not to mention, at this point there is a lot more testing, tracing and isolation, and more awareness with social distancing and hygiene.

I never said the virus is more deadly in nyc than everywhere else. Just that you can’t assume it would spread as quickly as it did in NYC. If a vaccine never happens then I would expect it to eventually spread to about 80% of the population, but that would be over many years.

also, I am not claiming I know anything special.

 

[davepsilon]

How applicable is China’s situation to this situation? Approximately how many untracked infections did they have in Hubei province when they finally reopened? What were their daily case counts at that point?

I too think it is possible for us to thread the needle, if we get a bit lucky, but I haven’t been looking to China for any guidance on what our chances are for success, because my impression was that they had virtually no disease circulation (community transmission) upon reopening. But maybe I’m missing something.

Interesting how different the rt.live data is from the covid19-projections.com website Rt values. I guess we have to ask whether we should trust machine learning, or Instagram.

COVID-19 Infections Tracker

About covid19-projections.com

Rt Covid-19

I highlight China just because it shows what is possible with very strict interventions despite a highly contagious virus. Less strict interventions may also be sufficient. The US will not have the same public health interventions as China!

What odds do you assign to the negative cases of exceeding previous peak or second shutdown in US?

 

[AlanSubie4Life]

Just that you can’t assume it would spread as quickly as it did in NYC. If a vaccine never happens then I would expect it to eventually spread to about 80% of the population, but that would be over many years.

I’m pretty sure that without mitigation the virus would spread to the point of herd immunity (which with overshoot might be as high as 80%) in places other than NYC within a few months. It spread to 20% of the NYC population in a matter of just a few weeks! I don’t think it would take years at all.

As I said, in the context of the discussion where the comment from @Norbert was made, we weren’t talking about the rate of spread anyway. The point was to illustrate the small percentage of the populace infected to date, and extrapolate the NYC mortality numbers to the nation. That does not have anything to do with the rate of spread.

What odds do you assign to the negative cases of exceeding previous peak or second shutdown in US?

I think it makes sense to think regionally. I think there is nearly zero chance we will exceed the previous (true) case peak. And I don’t think the administration would ever call for a second shutdown or pause.

I’m not making any predictions. I find it very difficult to predict what will happen going forward, now. It is much more difficult than it was when everyone was oblivious to the danger from the virus.

That being said, if I am forced to say something (which is not a prediction): I think regionally it is quite possible/likely that a few states will exceed their previous case peaks significantly due to reopening too quickly. I’d say significantly larger regional second peaks have about a 70% chance of occurring somewhere this summer. Again, this is not a prediction. I’m not making predictions; there are just too many unknowns. Just a guess.

 

[JRP3]

Chinese CDC says animal to human transition did not happen at the wet market

The Chinese CDC now says the coronavirus didn't jump to people at the Wuhan wet market — instead, it was the site of a super-spreader event

 

[SageBrush]

I have written before on the completion of a death certificate, but it is topical again so here goes:

The general outline is:
1. Immediate cause of death
2. Due to
2a. Primary
2b. Secondary
3. Contributing factors

If I was completing a Covid-19 death in an elder with co-morbidities it might look like this:

1. Pneumonia
2a. Covid-19
2b. Chronic lung disease
3. Age, obesity, diabetes, htn

----
It is not uncommon for physicians to complete death certs incorrectly, but pneumonia and Covid-19 will show up on the cert *somewhere*
An incorrectly filled out death cert of the above patient might look like this:
1. Respiratory arrest
2. Pneumonia
3. Covid-19, Diabetes, HTN

Now, lets change the above patient to someone who is admitted with a subsequently fatal MI ("heart attack") but is diagnosed Covid-19 positive during their hospitalization. That death cert might be filled out as follows:

1. Myocardial Infarction
2. ASCVD
3. Diabetes, Obesity
4. Other: Covid-19, toenail fungus

 

[SageBrush]

Now, lets change the above patient to someone who is admitted with a subsequently fatal MI ("heart attack") but is diagnosed Covid-19 positive during their hospitalization. That death cert might be filled out as follows:

1. Myocardial Infarction
2. ASCVD
3. Diabetes, Obesity
4. Other: Covid-19, toenail fungus

Let's get complicated. Say the above patient dies on a vent from hospital acquired pneumonia. Now a reasonable completion of the death cert might be

1. Nosocomial Pnemonia (specified agent if known)
2. Myocardial infarction
3. ASCVD, Diabetes, Obesity
4. (other): Covid-19, toenail fungus