Corona Virus/COVID19: Local Impact

[deuce]

Perhaps drowning in data and starving for information. If testing were conducted on a consistent basis across a reasonably valid cross-section of the population - especially since there is a reasonable chance that many, many asymptomatic carriers are in general circulation - the data would be more useful. It still may be useful if the professionals know how to massage and scrub the data for all of those variables. But those data come with a boatload of embedded variation.

So lets say that we figure out the asymptomatic fraction. I'm never opposed to more data mind you, though I may object to how it's spun. So let's say that we determine that 50% of infections are asymptomatic. That means that our mortality rate drops from 4.0% worldwide to 2.0% (or 1.4% to 0.7% in the US). Then what?

So people can claim "Hey! It's no deadlier than the flu!"?

Except that flu numbers are not predicated on calculating an asymptomatic fraction, so it would be like comparing apples to oranges. As I mentioned, if we determined the asymptomatic fraction of influenza viruses, death rates - on average- would also drop by about half ... which again would make coronavirus a much deadlier disease.

If it's used to develop a model to calculate herd immunity (though a simple antibody test won't prove immunity), that's one thing ... but the discussion was about "determining how deadly the disease is", and we already use standard metrics to make those calculations.

 

[Nate]

SMH example from Tuesday:
I used a public restroom in a large building. A baby boomer (my estimate) was already there and he practiced social distancing by selecting the farthest urinal. He was also wearing a mask. Upon completion, he walked past the sink and out the door.

The distance and mask are good but come on...

I’m sure he had a good reason. True story...about 5 years ago had to have a conversation w/ a guy that worked w/ us because he would always leave the restroom and never flush (wasn’t auto at the time). Said it was gross and didn’t want to touch the handles. Ummmm, got ya but someone has to!!!

 

[Vortex]

Agree * 1000. Social media and electric hand dryers will prove to be the downfall of our civilization.

I almost mentioned hand blow driers. Like many people, I thought it was the better solution until reviewing a study about those germ pools. PSA: wipe hands on pants before using one of those machines.

 

[tahoebum]

So the steady stream of reported rates (# tested, # confirmed cases, # deaths) coming out of every country impacted doesn't equate to plenty of data? Errr, ok.

How many have been infected in a a country like the UK? 10%, 20%?? Nobody knows the correct answer yet but its many times greater than the tested and confirmed cases. Oxford Covid 19 modeling experts are suggesting that 50% of the population in the UK have already been infected going back to mid January. That would give it so far a mortality rate much smaller than the seasonal flu. They shortly will begin testing to see if their hypothesis is correct by testing the general public for Covid 19 antibodies.

Oxford Model: Coronavirus May Have Already Infected Half of U.K. Population

If the hypothetical model is proven by antibody testing, it would mean minuscule hospitalization rates for coronavirus cases in the country.

nymag.com

 

[deuce]

How many have been infected in a a country like the UK? 10%, 20%?? Nobody knows the correct answer yet but its many times greater than the tested and confirmed cases. Oxford Covid 19 modeling experts are suggesting that 50% of the population in the UK have already been infected going back to mid January. That would give it so far a mortality rate much smaller than the seasonal flu. They shortly will begin testing to see if their hypothesis is correct by testing the general public for Covid 19 antibodies.

Oxford Model: Coronavirus May Have Already Infected Half of U.K. Population

If the hypothetical model is proven by antibody testing, it would mean minuscule hospitalization rates for coronavirus cases in the country.

nymag.com

Again, influenza also is potentially asymptomatic in a large portion of cases, which would drop the influenza mortality rate. You can't calculate the asymptomatic fraction of one disease, not do it for the other, and then say the numbers are comparable.

 

[tequila4kapp]

Rates are impacted by response. Yes, Korea and the US's rates differ because, as I said before ... Korea was on top of it, and we were not.

I’m pretty sure the numbers I saw yesterday had SK at 1.3% and the US at 1.4%. This contributes to my question about what the rate is.

On one hand we could conclude that since two countries with sophisticated health care systems which had very different response but both managed to stay (so far) below ICU capacity are at virtually the same rate so the mortality rate is about 1.3-1.4%.

On the other hand we could look at other data from other countries, large cities, controlled environments (Diamond Princess) and question sample sizes, demographics, quality of healthcare and say the mortality rate is higher or lower.

That’s why I asked what the number is. Again, I’m not looking to argue. I’m looking to understand.

 

[Reframmellator]

So lets say that we figure out the asymptomatic fraction. I'm never opposed to more data mind you, though I may object to how it's spun. So let's say that we determine that 50% of infections are asymptomatic. That means that our mortality rate drops from 4.0% worldwide to 2.0% (or 1.4% to 0.7% in the US). Then what?

So people can claim "Hey! It's no deadlier than the flu!"?

Except that flu numbers are not predicated on calculating an asymptomatic fraction, so it would be like comparing apples to oranges. As I mentioned, if we determined the asymptomatic fraction of influenza viruses, death rates - on average- would also drop by about half ... which again would make coronavirus a much deadlier disease.

I'm not an epidemiologist, so I have probably said more than I know.

I'm thinking of how we model "the curve" we are all trying to flatten. I guess I'm focusing on modeling number of cases rather than fatality rates. If I were asked to do that, I'd like to know three things: how many people have it (points of contact), how contagious it is (how easily spread), and the rate of circulation of carriers (how many opportunities to spread). I'd need two of those to determine the third, and I'd probably prefer to know the nnumber of carriers and rates of circulation to determine ease of transmission.

 

[Stemmy66]

We don’t know With any certainty how deadly or contagious Covid 19 is relative to the flu and won’t know until we randomly test the general population for the Covid antibodies.

An Oxford MD/professor is starting that random testing very soon and in a month we will have much better data.

In my opinion the results or conclusions will still be biased data. How can on make the correlation to how contagious or deadly when many of the population are self quarantined or practicing social distancing for a relatively significant time period? Wonder how they will account for this in terms of analysis or conclusions once broad testing is conducted. This is confounding even in South Korea where they tested very broadly, but also implemented fairly strict social distancing.

 

[tahoebum]

Again, influenza also is potentially asymptomatic in a large portion of cases, which would drop the influenza mortality rate. You can't calculate the asymptomatic fraction of one disease, not do it for the other, and then say the numbers are comparable.

But we have decades of good data on the seasonal flu with millions of tests and know within a range about how deadly it is. The 0.1% flu death rate is based on that. Right now we don’t have even a rough idea of the number of those infected in the U.S. or any other country. We all know the number of infected in the U.S. is certainly not the current 93,000 confirmed cases. My buddies son got it and his entire household got the classic symptoms of fever and cough but the four of them only count as one official Covid case because doctors told them it was 99% likely that they all had it based on their symptoms and that their son tested positive. They refused to give the test to the other three in his family. They are all recovering nicely with the youngest having very mild symptoms, two of them having mild flu like symptoms with fever and cough, and his son having more serious fever and cough. Millions of others are likely similarly recovered but never tested or counted in any of the numbers being reported worldwide.

The number in the U.S. infected with Covid 19 these last few months could be 500,000, 5 million, or 50 million. We will know a lot more when random testing takes place to make an educated guess at what that number is. Right now it’s an uneducated guess and you can’t calculate a death rate if you don’t know the denominator.

 

[arydolphin]

This is so awful:

 

[mpeterson]

I'm not an epidemiologist, so I have probably said more than I know.

I'm thinking of how we model "the curve" we are all trying to flatten. I guess I'm focusing on modeling number of cases rather than fatality rates. If I were asked to do that, I'd like to know three things: how many people have it (points of contact), how contagious it is (how easily spread), and the rate of circulation of carriers (how many opportunities to spread). I'd need two of those to determine the third, and I'd probably prefer to know the nnumber of carriers and rates of circulation to determine ease of transmission.

There is a massive, massive body of literature on exactly this, given incomplete datasets like we get during outbreaks.

 

[Reframmellator]

There is a massive, massive body of literature on exactly this, given incomplete datasets like we get during outbreaks.

So you're saying I'm right . . . .

I'm not an epidemiologist, so I have probably said more than I know.

 

[tahoebum]

Again, influenza also is potentially asymptomatic in a large portion of cases, which would drop the influenza mortality rate. You can't calculate the asymptomatic fraction of one disease, not do it for the other, and then say the numbers are comparable.

Asymptomatic could be just a small portion of the denominator that we are missing. It’s possible than millions of people that are symptomatic with Covid 19 have recovered but never tested or even been to the doctor. We have been very short on testing outside of a few countries like SK and Iceland so many that are symptomatic won’t show up anywhere in the numbers as they weren’t tested. I known of this personally because my whole buddies family has it but they only show up as 1 confirmed case.

What I’m saying is the death rate could be 4%, 2%, 1%, 0.1% or any other number. We don’t have the data of percentage of the general population currently with antibodies to know actual death rates within a factor of 100 with any amount of certainty.

 

[obedt]

Asymptomatic could be just a small portion of the denominator that we are missing. It’s possible than millions of people that are symptomatic with Covid 19 have recovered but never tested or even been to the doctor. We have been very short on testing outside of a few countries like SK and Iceland so many that are symptomatic won’t show up anywhere in the numbers as they weren’t tested.

And the same applies to flu as well as many other infectious diseases. That's why the ratios & comparison can stand.

 

[tahoebum]

And the same applies to flu as well as many other infectious diseases. That's why the ratios & comparison can stand.

Are you making the argument that we know as much about Covid 19 that has been around for 4 months as we do smallpox, Ebola, the seasonal flu, or malaria????

What if we know in 60 days that ten or a hundred times the number of people were infected with Covid 19 than ever tested positive? Will out social distancing change next year if it’s about as deadly as the seasonal flu and we have a vaccine and medicines for it? Will more people get the flu shot next year when we’ve never as a country been able to get more than half the population to get one even though it’s widely available and free or super cheap?

 

[th6252]

Mind bottling...

 

[tahoebum]

Mind bottling...

If we knew or saw the unsanitary conditions that our food is prepared in at many food establishments the restaurant business would me a much smaller part of the economy. A lot of us would eat at home all the time, lol.

 

[JDax]

Tahoebum said with “certainty”. Do we know that? What is it? That’s a serious question, not being argumentative, because there is pretty substantial variation in the rate for S. Korea / US vs Italy vs China vs Diamond Princess, vs NY, etc.

I’ve always thought that the real number would settle at our around the Diamond Princess’ , it’s best controlled test of the whole curve that we have... The data is clean in that case.

 

[JDax]

So lets say that we figure out the asymptomatic fraction. I'm never opposed to more data mind you, though I may object to how it's spun. So let's say that we determine that 50% of infections are asymptomatic. That means that our mortality rate drops from 4.0% worldwide to 2.0% (or 1.4% to 0.7% in the US). Then what?

So people can claim "Hey! It's no deadlier than the flu!"?

Except that flu numbers are not predicated on calculating an asymptomatic fraction, so it would be like comparing apples to oranges. As I mentioned, if we determined the asymptomatic fraction of influenza viruses, death rates - on average- would also drop by about half ... which again would make coronavirus a much deadlier disease.

If it's used to develop a model to calculate herd immunity (though a simple antibody test won't prove immunity), that's one thing ... but the discussion was about "determining how deadly the disease is", and we already use standard metrics to make those calculations.

The data is being used in models that are driving policy decisions, that’s why the accuracy of the data is so important.

Can we all at least agree that we aren’t Italy, at this point?

 

[obedt]

Are you making the argument that we know as much about Covid 19 that has been around for 4 months as we do smallpox, Ebola, the seasonal flu, or malaria????

What if we know in 60 days that ten or a hundred times the number of people were infected with Covid 19 than ever tested positive? Will out social distancing change next year if it’s about as deadly as the seasonal flu and we have a vaccine and medicines for it? Will more people get the flu shot next year when we’ve never as a country been able to get more than half the population to get one even though it’s widely available and free or super cheap?

No, that's not the case I'm making. I'm saying the way the metrics work for them is that we go off diagnosed cases. There's lots of people that don't get tested or properly diagnosed regardless of the disease.

 

[obedt]

The data is being used in models that are driving policy decisions, that’s why the accuracy of the data is so important.

Can we all at least agree that we aren’t Italy, at this point?

I'd think that's a given. Italy and Spain both have really suffered incredibly.

 

[tahoebum]

Iceland has done by far the most per capita testing for Covid 19(way more than South Korea). More importantly they are the only country that has tested a decent size sample of the general population that doesn’t show symptoms. So far they have found that 1% of their population is asymptomatic carriers which would mean we would have about 3,300,000 million of those people in the U.S. or will have shortly. How do we identify and isolate those people? How contagious are they? Is it already too late and in some countries is half the population already infected?

There is obviously lots of questions and lots to learn about this virus and hopefully in a few months we will have a great deal more clarity.

Detectives aid COVID-19 tracking in Iceland's fight to contain spread

"We are using science and the best evidence there is," Iceland’s health director, Dr. Alma Moeller, says.

www.nbcnews.com

 

[mpeterson]

I'd think that's a given. Italy and Spain both have really suffered incredibly.

One thing I'd be careful about declaring "victory" in that regard there is to not forget that those places have a "head start" in terms of outbreak. Maybe "we" as a country aren't going to see what Italy is seeing, but there is a chance that places within the country absolutely do.

Take a look at this chart, that compares NYC with a number of places, including Spanish and Italian cities at the same times in the outbreak.

 

[obedt]

One thing I'd be careful about declaring "victory" in that regard there is to not forget that those places have a "head start" in terms of outbreak. Maybe "we" as a country aren't going to see what Italy is seeing, but there is a chance that places within the country absolutely do.

Take a look at this chart, that compares NYC with a number of places, including Spanish and Italian cities at the same times in the outbreak.

Oh I'm not declaring victory, or anything close to it. I think the mortality rate in US is going up, not down, over the next couple months, and think the government was slower to prepare itself than I would have liked. But I think taking the very worst spot isn't representative of the entire nation either. We'll see how it plays out

 

[Sean]

People coming into Massachusetts now have to self-quarantine for 14 days. I don't know how they will enforce it, but I wonder if other states will follow suit.