Coronavirus

[Margaritaboy]

Wow. As a physician myself, it is a bit disheartening to see logic thrown out the window in favor of broad based data-free proclamations.

While I concede there are many uncertainties and gaps with a brand new virus causing a pandemic, may I suggest a few best practices for people on this forum? Try these:

1. State the underlying premise for your arguments. If you are arguing numbers, then disclose your foundational assumptions.
2. State your argument, clearly and concisely.
3. Attack an argument, not the person behind the argument.
4. Be forgiving of others’ arguments, and be open to the fact you may be wrong yourself.
5. This isn’t the Roman colleseum. The goal isn’t to win and eviserate your enemy.
6. Try to take the opposing view point. You might learn something.

There have been some helpful links posted by members here. The following is a decent one for numbers, but also good information on the virus itself.

COVID-19 Map

Cheers and stay safe everyone!

 

[AlanSubie4Life]

You have no proof of that. That is pure speculation. You have no proof that under normal circumstances, X number of people will die. Nobody knows the answer to that because is hasn't happened, anywhere.

It's an educated guess, based on general epidemiologist estimates of R0 being about 2.5-3 for this virus. It may be 70%. 1-1/R0.

As far as the number of people who will die: The available evidence points to an IFR of about 1%. It could be as low as 0.5%. Obviously it depends on the age & health of the population in question.

 

[dfwatt]

These are the kinds of conclusions I'm drawing as well, for NYC anyway. Most evidence is pointing to spread being closer to 60% than 10%. I don't think it's a stretch to say nearly half of New Yorkers have been in contact with the virus and have developed some degree of antibodies.

That's more than a stretch it's just made up s*** again.

Show me a single responsibly collected responsibly sampled not cherry-picked antibody IGG based epidemiology showing anything like that.

 

[dfwatt]

So you are basing your estimated 20-40% of NYC infection rate on teachers! OMG!!

And I just saw you bumped up your infection rate in New York from 20 to 40% to 60%? I want to know first of all what you're smoking? It must be really good s***! Can I have some??

 

[EVNow]

I don't think so -- the lower chart is based on an assumption of an 11% "true" rate of infections, based on an earlier study of close household contacts in Shenzen. This is substantially lower than the rate of infections in the Johns Hopkins study.

Yes - so, if you think 10% of population in NYC is infected, that would be the likelihood. In other cities it would be lesser.

My city has 55 positives, 3 fatalities (was 0 a week back). With an IFR assumption of 0.5% to 3%, we would get a range of 100 to 600 infections. That would be 2x to 10x the positives. With a population of 80k, that would be 0.12% to 0.75% infection rate in the population. But all these calculations would be quite different if I had done it 2 days back (2 deaths) or 4 days back (1 death).

Anyway, if my city has an infection rate of 1% - a negative PCR test after 21 days would mean 0.7% chance of Covid.

 

[dqd88]

Already spread ...
How much ?

Then it's too late ...
For what ?
To do what ?

We don't know for sure. You're asking me for numbers; no one knows; it's just a guess.
But you can surmise that it is a substantial amount since there is high density there and a major air travel hub where travelers would have likely brought the virus.

If many are infected by mid-March, then it's too late to do a SAH. That would trigger a tsunami of cases; which it did.
I think to do a SAH if you have to do it much earlier. They should have done it in early Feb.
But even then, what do you do after SAH? Won't the problem re-arise?
I think better to wear masks and other measures. More sustainable.

 

[N5329K]

How the COVID-19 virus attacks the immune system by targeting T-Cells (almost like HIV). Really explains a lot.
Robin

 

[MXLRplus]

I witnessed a vector location in Orange County, California. Day laborers massed together with no PPE. They get dispersed to various locations each day, then return to their point of departure. Perfect design for a vector during a lockdown.

I would guess the PD either cannot deal with it, or doesn't want to deal with. They have to see it, it's at the intersection of I-5 and Redhill.

For Riverside County, we have a number to call to report PPE offenders. I have no idea if Orange County does this, I cannot find one.

I had to pick up medical device testing samples in the area.

 

[AlanSubie4Life]

That would trigger a tsunami of cases; which it did.

Think of what we're dealing with as one of those 6-inch tsunamis. It could have been so much worse.

They should have done it in early Feb.

Yes. Fortunately some states took action in late February and early March.

But even then, what do you do after SAH? Won't the problem re-arise?

That has been extensively discussed here: Test, trace, and isolate. Test, test, test. Use stuff we did not have in place the first time around. Use this time to our advantage (unlike the time we did not use to our advantage after the incomplete China travel ban - it's not actually clear to me that that ban bought us any significant time - but we did have time).

I think better to wear masks and other measures. More sustainable.

Do you have data showing that this alone will bring R0 down to below 1? I'm not convinced at all that it will. I think it will definitely help. But you'd have to have data showing that that alone is sufficient if you wanted to rely on that alone.

With this virus, you can't take any chances. You have to do overkill, and then gradually ratchet things back to figure out what works. We have to learn on the fly without allowing things to get back to the situation we had in early/mid-March.

 

[dhanson865]

French team had to bring the temperature to almost boiling point to kill virus

• French scientists had to bring the temperature to almost boiling point to kill virus
• Results have implications for the safety of lab technicians working with the virus

Professor Remi Charrel and colleagues at the Aix-Marseille University in southern France heated the virus that causes Covid-19 to 60 degrees Celsius (140 Fahrenheit) for an hour and found that some strains were still able to replicate.

The scientists had to bring the temperature to almost boiling point to kill the virus completely, according to their non-peer-reviewed paper released on bioRxiv.org on Saturday. The results have implications for the safety of lab technicians working with the virus.

The team in France infected African green monkey kidney cells, a standard host material for viral activity tests, with a strain isolated from a patient in Berlin, Germany. The cells were loaded into tubes representing two different types of environments, one “clean” and the other “dirty” with animal proteins to simulate biological contamination in real-life samples, such as an oral swab.

After the heating, the viral strains in the clean environment were thoroughly deactivated. Some strains in the dirty samples, however, survived.

The heating process resulted in a clear drop in infectivity but enough living strains remained to be able to start another round of infection, said the paper.

 

[AlanSubie4Life]

Per the governor's press conference, for California/Cascadia, the following items will be needed to re-open:

1) Ability to monitor & protect communities through testing, tracing, isolating, and supporting

2) Ability to prevent infection in those who are at severe risk of COVID-19.

3) Ability of hospitals to handle surges.

4) Ability to develop therapeutics to meet demand.

5) Ability of businesses, schools, and child care facilities to implement physical distancing.

6) Ability to reinstitute certain measures, such as stay-at-home orders.

Restaurants will reopen with fewer tables and masks will be commonplace.

If California can build this capacity, and in addition, in parallel, they see a reduction in hospitalizations & ICU admissions over the course of three weeks or so, they will relax some of the restrictions. Per Gov. Gavin Newsom. He said as far as the timeline, ask him again after all this stuff is in place, and they have seen significant reductions in case metrics for two weeks. So assuming case numbers start heading down, ask him again around May 1st what the timeline is. Seems about right. Looks like if things go well we'll be looking at mid-May.

He also said large-scale gatherings are very likely not in the cards as long as a vaccine or some revolutionary therapeutic is not available. So June, July, August will not have large-scale gatherings in California. No sports, etc.

 

[dqd88]

Think of what we're dealing with as one of those 6-inch tsunamis. It could have been so much worse.



Yes. Fortunately some states took action in late February and early March.



That has been extensively discussed here: Test, trace, and isolate. Test, test, test. Use stuff we did not have in place the first time around. Use this time to our advantage (unlike the time we did not use to our advantage after the incomplete China travel ban - it's not actually clear to me that that ban bought us any significant time - but we did have time).



Do you have data showing that this alone will bring R0 down to below 1? I'm not convinced at all that it will. I think it will definitely help. But you'd have to have data showing that that alone is sufficient if you wanted to rely on that alone.

With this virus, you can't take any chances. You have to do overkill, and then gradually ratchet things back to figure out what works. We have to learn on the fly without allowing things to get back to the situation we had in early/mid-March.

If I were in charge, I would tell everyone they need to walk/bike/etc at least a total of 1 hour a day. If people exercise and lose weight they will be less susceptible. More adipose tissue makes it harder to breathe and steals oxygen.
I would close indoor events with close contact, like nightclubs, but I would leave all outdoor events open.
I would leave open all stores and shopping malls as long as people can spread out.
I would leave indoor events open (eg NBA, NCAA) if they maintain one empty seat between people.
I would seal off elderly facilities.
Leave all schools open. Retire old teachers, use young teachers, esp for kindergarten where kids are yucky.
Invent new tools to capture and remove sneeze aerosol from air.
Make it illegal to sneeze into the open air. 1 warning, $500 fine.

 

[EinSV]

Can anyone track down the false-positive rate for C19 PCR tests?
I'm having a hard time locating a number.

My understanding is that covid-19 false positive (assuming tests are conducted properly) are effectively zero (or close to zero). If that's incorrect, hopefully one of the other forum members can chime in (ideally with a citation to a report that says otherwise).

This is significant due to the stories of persistent C19 infection after symptoms subside. If they were never SARS-CoV-2 positive to begin with they could test as positive later. Whether it's due to the other coronavirus strains, or SARS-CoV-2, or even mishandling is immaterial in that regard.

It is critical at this stage to determine if you can get SARS-CoV-2 infection twice, or if the infection is persistent. No other quarantine decisions should be made without that knowledge.

My understanding is that the question with these is not false positives but the potential for false negatives that suggest a person is "cured" when they are not. The Johns Hopkins study I linked above (and others) have concluded that the likelihood of false negatives increases as time goes on, even though the infection lingers.

To my knowledge it is a subject of debate whether these "reinfections" are in fact reinfections or the result of a false negative. Here is a discussion yesterday from Marc Lipsitch (Harvard epidemiologist):

One concern has to do with the possibility of reinfection. South Korea’s Centers for Disease Control and Prevention recently reported that 91 patients who had been infected with SARS-CoV-2 and then tested negative for the virus later tested positive again. If some of these cases were indeed reinfections, they would cast doubt on the strength of the immunity the patients had developed.​

An alternative possibility, which many scientists think is more likely, is that these patients had a false negative test in the middle of an ongoing infection, or that the infection had temporarily subsided and then re-emerged. South Korea’s C.D.C. is now working to assess the merit of all these explanations. Opinion | Who Is Immune to the Coronavirus?​

 

[EinSV]

Yes - so, if you think 10% of population in NYC is infected, that would be the likelihood. In other cities it would be lesser.

Let's stick with the New York City hospital study for a moment. That study recorded 15.4% positives.

Johns Hopkins estimated 26-100% of covid-positive people would get a negative result in a PCR test depending on timing. Taking the bottom of this range (26%), suggests that conservatively at least 20.8% were infected in the population screened in the hospital study (15.4/(1-0.26)).

Based on these studies, the true percentage of covid-positives in the population tested is more likely than not to be substantially higher than that because infections in the four days before symptoms emerge have an estimated 61-100% false negative rate (with large error bars) and 10 days after symptom onset false negatives exceed 50% again.

Also, the data collection for the hospital study ended 10 days ago. I don't know what the doubling rate has been for New York City in that time period, but 10 days is likely conservative, so the number of infected could easily have doubled since then. At a bare minimum, there is no question that the percentage of infected individuals in New York City has increased substantially in the 10 days since April 4.

So the main question in my mind is whether the population in the hospital study is representative of New York City, or whether there are other statistical quirks that led them to an incorrect result. Otherwise, especially given the passage of 10 additional days and the high false negative rate, it's not hard to see New York City in the 30-50% infected range now.

 

[EVNow]

So the main question in my mind is whether the population in the hospital study representative of New York City

Was it a random sample ? No. So, it doesn't represent the population.

 

[EVNow]

In other news ...

No evidence of clinical efficacy of hydroxychloroquine in patients hospitalized for COVID-19 infection with oxygen requirement: results of a study using routinely collected data to emulate a target trial : COVID19
​

"However, it is not a randomized trial, and in particular, the HCQ group was slightly younger, none were reported as confused at admission, but had higher co-morbidities than the non-HCQ group. IPCW is a statistically robust estimation approach to adjust for these differences, and sensitivity analyses of other modeling approaches found similar results."​

​

 

[dqd88]

It's an educated guess, based on general epidemiologist estimates of R0 being about 2.5-3 for this virus. It may be 70%. 1-1/R0.

As far as the number of people who will die: The available evidence points to an IFR of about 1%. It could be as low as 0.5%. Obviously it depends on the age & health of the population in question.

The R0 is unknown. It was based on initial data from China. Also, R0 is malleable. It partly is a function of the virus but also a function of how people respond to the virus.
You don't know though, under normal circumstance, how many will get the disease. It is only estimates, and most estimates have been horribly off.

 

[EinSV]

Was it a random sample ? No. So, it doesn't represent the population.

My understanding was that they screened every single pregnant woman who came to the hospital for delivery. So it was not random for the entire population, but it was better than most covid-19 testing which tends to focus on symptomatic people or is skewed in other ways.

I like to be data-driven. Is there a better study for New York City?

 

[alloverx]

Re estimates from this paper Physical distancing is working

 

[AlanSubie4Life]

It is only estimates, and most estimates have been horribly off.

No it's not. Also, which estimates have been horribly off? I'm not aware of any that have been way off other than conservative estimates about hospital ventilator requirements (which would obviously be biased towards the worst case).

The R0 is unknown.

No, it's not - we know the ballpark value. See below.

It was based on initial data from China.

No, it's not.

From this paper

Also, see (I've posted these before):

Covid-19: Global summary

This is something that epidemiologists can figure out...

Of course it's malleable. And of course it depends on the particular country/region/group in question. That's why I was asking whether you thought masks would bring it to 1.