Coronavirus in the United States - news and thoughts

2238 deaths today with a few states plus afternoon updates still pending

United States Coronavirus: 811,478 Cases and 44,752 Deaths - Worldometer

 

That was implemented a week ago, so the numbers we have seen for the past week already have that included.

 

No, no, no, no, no, no, I think a little stats lesson is due.

1) They don't cancel each other out except for very specific circumstances. For example, let's say a disease has 10% prevalence, and you run a test with 90% sensitivity and specificity. You'll have 9% (10% of 90%) of the population with false positives, and 1% (10% of 10%) with false negatives. 9% does not cancel out 1%, agreed?

2) They depend on the accuracy of the test AND the prevalence of the disease. For example, if you have a disease with 0% prevalence, then any positive result is a false positive, resulting in a false positive rate of 100%. You run the same test on a population with 100% prevalence, then any positive result would be true, resulting in a false positive rate of 0%.

In stats, only sensitivity (chance of testing positive when you do have the disease) and specificity (chance of testing negative when you don't have the disease) are measurements of accuracy of a test. False positives and false negatives are not because of their dependence on the prevalence of the disease. That's why the rapid HIV test which has a 99% accuracy (both sensitivity and specificity) have a 50% false positive rate.

 

It did say in your quote as of April 14th

 

so now we are going to have more daily deaths, which might mean, hey, lets pump the brakes on re-opening the economy, great.

 

You have been seeing this adjustment for the last week. Also Worldmeter has said where they can, they will adjust previous daily counts(prior to 4/14) to reflect proper distribution.

 

as i have posted before, i am against the, for lack of a better phrase, the random dumping of deaths w/o a real cause as from covid19. just create a colume for " cause of death unknown " and leave it at that, no one knows how they died, covid 19 or not.

 

The issue with just the most vulnerable self isolating is many don't live by themselves or just in a group by themselves. It's one thing to close off a nursing home with elderly residents, but there are plenty of households out there that are generational, with grandparents living with their kids and grandkids. Or people with immune-compromised people who live with a spouse and children. Unless you expect the elderly that live with younger people and immune-compromised people to self isolate in their homes and have zero contact with their housemates, opening up the economy raises the risk for these people. The more people who live in the house come in contact with outside the house, the higher the risk they get sick and infect those most vulnerable.

Take the elderly person living with kids and grandkids. Grandma may stay home as the rest of the family goes to school, work, restaurants, etc., but unless Grandma stays 6 feet and farther away from everyone, and constantly wipes every surface she touches with Clorox wipes, her risk of infection is much higher than if everyone stayed home except to go out for absolute necessities. In simple terms, let's say under lockdown, the family only comes in close contact with 20 people a week at Wal-Mart. But open up and the family comes in close contact with 200 people a week, then risk of household infection jumps by a factor of 10.

Patience and required, and if we open up the country prematurely, we'll quickly get back to where we are now, with the only difference being more people sick and dead than necessary.

 

while we are at it, why not add tens of thousands to the total cases and the recovered for those who we all know most likely had very mild cases, that would really drop the death rate numbers,if you add unknown deaths, add unknown recoveries and total cases also.

 

2499 now.

 

To our right wing friends here. Don’t bet your life on what Rush, hannity. claim this is nothing to be concerned about.

Dad killed by coronavirus after dismissing 'bulls***' lockdown as a ‘political ploy’

 

Drop the death rate and raise the spread numbers.

I expect that this will happen once we have a more thorough understanding of the disease. It is how we figure out how many had the flu each year.

 

I disagree. IMO, the principle reason not to include them is optics (politics?). That comes at the expense of spread and mortality information which may be needed for planning and allocation of resources. I think we are better off with the error being in over counting vs under.

 

Is your position that these numbers are being inflated to push an agenda?

 

Late reply, but that is absolutely 100% ok since we know Romney's "guilty" vote was done out of personal animosity and that he is not a fan of Trump's. I'm still missing the problem here. Everybody knows the history between Romney and Trump. It's not like Romney has been respectful.

 

I was listening to a podcast today that near 20% of Covid cases show cardiac damage. This includes a lot of cardiac arrest deaths actually linked to Covid damage. I have no idea if these are being counted, but it is the first I had heard of that.

 

@tilly I think you're missing the demanding scientific basis for being extremely skeptical of big or positive claims as a normative matter (actually, it's *any claim* since critical reviews of science is to pick apart methodology).

Under these circumstances, the big findings in non-peer reviewed studies have been put out in public by the researchers or their Universities themselves. People should be properly skeptical always, but even more so when a study hasn't even been through peer review. And that bar for validity should be ridiculously high, and perhaps higher with bigger claims.

I don't write about this much, but I've peer reviewed over sixty scientific manuscripts in my career. In grad school, I was taught to go directly to the data/methods section. Pick the methodology the heck apart. Look for every single flaw because it's a waste of time reading an entire manuscript if the author's can't get the methodology to pass peer review. Point is, this isn't a picnic with Peggy Sue. It's about a standard of validity that is/should be ridiculously high and we're seeing that standard play out in public because the stakes are enormous.