Sunday, April 12, 2020

The Numbers Are Just NOT Adding Up - UPDATED

[UPDATE - April 15, 2020]

The current case in SC - as of 4/14/2020 - are 3,553 cases confirmed, 97 deaths. Death rate still under 3% - not great, but not OMZ, we're all gonna die!

Four counties have more than 100 cases, all others are less; my own home county, York, is less than 50. As we are within a half hour drive to Charlotte, NC, and many residents work there, we would be high risk for infection.

And, yet, we are not. This is primarily an epidemic of crowded, unsanitary places - major cities with crowded public transportation, nursing homes, and a few other hot spots.

Not generally affecting the rest of the USA. So, a good argument for opening up America for business.

Steps:

  1. Trump releases the states to make their own decisions about lockdowns. IF they chose to continue with full shutdown beyond this point, money will NOT be available from the federal government - EXCEPT for the major hot spots, and ONLY with medical needs.
  2. Nursing homes will be strongly encouraged to offload any patients not already infected into other, clean facilities. Figure out the payment changes later.
  3. No school for the remainder of this year, K12. Schools are to direct their staff to come up with a plan for continued education, in the event parents don't want their kids returning next year. THEY MUST SUPPORT HOME-SCHOOLING EFFORTS IN EVERY WAY, WHETHER VIRTUAL CONFERENCING, REMOTE HOOKUP ON A REGULAR BASIS, OR SOME OTHER COMBINATION. ALL STUDENTS ABLE TO PASS THE YEAR END TESTS WILL BE PROMOTED.
  4. Use social pressure to promote reasonable distancing, use of masks in crowded places, and replacement of many routine visits to doctors and other places with teleconferencing, whenever possible.
  5. Put a limit on how long states can collect additional federal assistance. Offer some additional assistance to states on a SHARED basis - they have to take some risk for the loans - consider tying that sharing to their employee state pension funds. THAT will put the onus on them to not fund the sleazy operators.
  6. Re: 5 - start making it a regular thing for states to have some skin in the game. For every student admitted on financial aid, who is below the level of normal acceptance, the state has to cover 1/2 the cost of the aid given. Same with federal grants - they should not be "free money" but the states should share the burden.
  7. I originally agreed with the idea of having those returning to normal life given a "tested normal" certificate. I changed my mind. Too easy to counterfeit, too easy for government to use as a bludgeon against dissidents. Businesses and private organizations may require testing for return, but that's up to them. Caveat Emptor.


Understand, I'm looking at this from my own experience in South Carolina.

The National Numbers for COVID-19


That US Cases stat works out to just a smidge under 4% death rate. That's less than the World Case death rate of slightly over 6%.

But, in South Carolina, a relatively poor state, the death is less - MUCH less.



That works out to just under 2.5% death rate.

Guys, we're a poor state, with many areas of high poverty. Heck, even our White people aren't all that well off. We're loaded with those co-morbidity diseases - heart troubles, high blood pressure, obesity, and diabetes.

So, why are we showing up lower than the national average?

I know what I'm thinking, but it borders on conspiracy theory.

Put your thoughts in the comments.

11 comments:

edutcher said...

Our betters in the CDC have told us they are "advising" paying) hospitals to list any death possible connected to Winnie The Flu as the cause of death; ie, they're inflating the numbers.

Not only do they want deaths listed as the Chinese complaint, but Medicare is paying hospitals to list admissions that way, 3 times as much if they go on a vent.

It's doubtless the reason Trump wants the country opened up ASAP.

Rick T said...

A number of reasons come to mind.
1. MUCH lower population density and use of mass transit in the cities.
2. Fewer people travelling long distances, so no one is importing the virus from a distant hot spot
3. Data being captured #1: There are reports that EVERY non-trauma death is being recorded as from the Wuhan Flu, even without a positive test.
4. Data being captured #2: Reports were Italy doesn't do death certs for patients dying at home, only for hospitalizations. We aren't comparing the same statistics.

Here in Orange County CA the Public Health department is reporting hospitalizations and ICU usage each day. There are 25 hospitals in the county that could report but they NEVER show data from all 25 nor to they show the data by hospital so you could do real trending by hospital. The number they show is IMO garbage for that reason.

FredLewers said...

Liars can figure but figures don't lie...
There's lies, damn lies & statistics...
Always ask cui bono...

daniel_day said...

Should you not be comparing "recent" deaths (during the last 2 or 3 days, define "recent" however you wish) to the newly confirmed infections 10-14 days ago? Granted that this overestimates the death rate, since the "confirmed" number of new infections undercounts the actual number of newly infected people, most of whom did not get tested.

Punzdeleon said...

Most US cases have not run their course. I believe the correct way to calculate mortality is to add the number of deaths to the number of recovered to give the number of completed cases. Divide the sum into the number of deaths and multiply by 100. Of course as other readers have pointed out any calculation falls victim to GIGO.
Perhaps folks who live in rural areas have self-selected for social distancing.

Paul Bonneau said...

Don't believe anything that comes from the mouth of a politician. Doctors and college professors are equally suspect. Everything results from self-interest.

http://strike-the-root.com/self-interest

Unknown said...

Yesterday I ran calculations on mortality trends for several European countries and every state using data from https://www.worldometers.info/coronavirus/. I divided the deaths by the sum of deaths and recoveries, since those who are still sick have as yet unknown outcomes. This gives a reasonable upper bound for mortality, although there are a couple of reasons to think it will overstate it (unknown number of asymptomatic and undiagnosed carriers and the possibility of front-loading the numbers if the weak die quickly while the relatively healthy take longer to recover). Anyway, I found numbers all over the board. Think about this: San Marino, Italy, France, and Belgium are all clustered between 34% and 40% (yeah, it's that high!) while Germany and Austria are running 4.7% and Liechtenstein a mere 1.8%. Meanwhile, Switzerland is about 9% - which is interesting as it is bounded on different sides by those other two regions. The Asian Rim is running low single digits, except for Japan at 14%.
The states vary wildly. Hawaii is running 2.59% on over 300 completed cases while New York is running numbers closer to Europe. Interestingly, it has been suggested that NY was infected from Europe, but it seems likely that Hawaii was infected by Asia.
The lowest death rates I found were in Wyoming (0%) and the upper Midwest (MT, ND, SD. MN), a central area of Arkansas and Tennessee, the extreme Northeast (Maine and New Hampshire), West Virginia, and DC. All of those regions are surrounded by other states with much worse experience.
I can't explain any of this. I'm hoping that somebody else will take a look at it, but I don't think all of this can be written off to demographics or health care quality. There are geographic trends that seem to cross jurisdictional lines. It might be due to environmental factors, but what if it had to do with multiple strains? If it's the latter, then my mind immediately wanders to the question .... was this by design?

Unknown said...

Yesterday I ran calculations on mortality trends for several European countries and every state using data from https://www.worldometers.info/coronavirus/. I divided the deaths by the sum of deaths and recoveries, since those who are still sick have as yet unknown outcomes. This gives a reasonable upper bound for mortality, although there are a couple of reasons to think it will overstate it (unknown number of asymptomatic and undiagnosed carriers and the possibility of front-loading the numbers if the weak die quickly while the relatively healthy take longer to recover). Anyway, I found numbers all over the board. Think about this: San Marino, Italy, France, and Belgium are all clustered between 34% and 40% (yeah, it's that high!) while Germany and Austria are running 4.7% and Liechtenstein a mere 1.8%. Meanwhile, Switzerland is about 9% - which is interesting as it is bounded on different sides by those other two regions. The Asian Rim is running low single digits, except for Japan at 14%.
The states vary wildly. Hawaii is running 2.59% on over 300 completed cases while New York is running numbers closer to Europe. Interestingly, it has been suggested that NY was infected from Europe, but it seems likely that Hawaii was infected by Asia.
The lowest death rates I found were in Wyoming (0%) and the upper Midwest (MT, ND, SD. MN), a central area of Arkansas and Tennessee, the extreme Northeast (Maine and New Hampshire), West Virginia, and DC. All of those regions are surrounded by other states with much worse experience.
I can't explain any of this. I'm hoping that somebody else will take a look at it, but I don't think all of this can be written off to demographics or health care quality. There are geographic trends that seem to cross jurisdictional lines. It might be due to environmental factors, but what if it had to do with multiple strains? If it's the latter, then my mind immediately wanders to the question .... was this by design?

Historian said...

I'll point out a few assumptions that may have some bearing:

Comparisons between regions, or even states, assume that the treatment regimens are the same. Differing treatment regimens make a big difference in clinical outcome. Oxychloroquin/Azithromycin/zinc sulfate given early in the course of infection appears to show excellent results. Given later in the progression, not much effect in some locations.

Some locations appear to be testing more widely than others. In the state where I work, only 'symptomatic' patients are tested. In south Korea, they test anyone who wants a test. The difference in outcome is obvious, but the point is that limiting testing skews the data.

Comparisons of outcomes between regions, or even between different hospitals in the same county, assume the quality of care is uniform. It is emphatically not the same, and quality of care makes a difference in outcome. You will not get the same care from Bellevue today as you might from Cheyenne General, or from UVA hospital.

It is known that this Chinese virus is highly mutable, as one might expect from a RNA virus, and that different clades vary significantly in speed of onset and infection severity. Assuming that all areas of the globe, or even the US, have the same clade of this disease, would be a mistake, another argument strongly favoring more testing and genetic sequencing, to tell clinicians what clade they are dealing with. Further, my understanding is that genengineered virii tend to be more unstable than their naturally evolved forebears. the virus infecting patient A may not be the same virus once A infects B,C,D, E and F. In fact, each person is their own petri dish for evolving viral clades. Many pathogens evolve towards lower lethality, but not all, anthrax being a prime example.

Lastly, genetics of the human population in an area make a difference, and so do the human demographics. It is known that some ethnic groups show more susceptibility to this damnable disease.

All of the above factors play into how long it takes for a case to resolve, one way or another. In some locales, with overwhelmed hospitals and staff facing large outbreaks, deaths may come quicker than in places where staff are on their top form, having the chance to tailor care to the individual based on specifics of the clinical presentation as assessed through thorough lab testing and possibly even genetic sequencing.

Then there are the political aspects. We have no idea what the real numbers are from China, for example. We know that they routinely lie about everything, and especially about infectious outbreaks, but we have no way to know what the real numbers are. We know that the death stats from NYC are probably understating their death toll, while at the same time undercounting cases is going on to a variable extent worldwide.

We are not likely to know what the 'real' CFR is until well after the fact, if ever, and variations based on the above factors, and likely others, should be expected. Such information takes seroprevalence studies, genetic sequencing of known cases, and enough time for all the cases to resolve, or at least to be able to set a timeline for date of infection, and so forth. At best, it is likely that this will take years, not weeks or months.

In short-
this Chinese virus is infectious as can be, is mostly not a problem for those infected, but if you are the unhappy recipient of a severe case, it can be very bad news, depending on how good, and how well rested and how well informed, your physician and care team are. Ebola it ain't, praise be, but anyone who has seen a previously healthy younger person who is moderately ill go downhill and die in the space of 12 hours is not likely to dismiss it as "not as bad as the flu." How bad is it? Wait a few years for the furor to die down and see.

,

Redd Stater said...

To Historian, you point out that there are many clades of this virus, which I think is part of the reason for regional variability in outcomes not discounting the other factors you mentioned, but this raises a couple of questions. First, could the existence of multiple clades account for the reinfection problem that some have reported? Would antibodies for one clade of the virus confer resistance to all others? If not, what is the real likelihood of developing an effective vaccine against it? Finally, your comment suggests that you think this might have been engineered. I think that's possible. It was evident to me back in February that the Chinese government was lying about their mortality, since the reported mortality trend outside of China was already diverging sharply from their reported numbers. I'm not sure why anyone in our government trusted them as long as they did. In fact, I think they might have engaged in a deliberate disinformation campaign to suppress Western mitigation efforts. In effect, even if this wasn't released on purpose as a weapon, it has been weaponized after the fact. In any case, I have been paying attention to the progress of this thing since it was first reported. I'm interested in the health risk it poses, of course, as that will help inform everyone about the best strategy for reopening the economy, but I am interested in the genesis of this thing, too. If it was engineered, then who did it? Why was it made? Was it for research or was it intended as a weapon? Was the release accidental or deliberate? If this is deliberate, what's "the grand game design?" And how much of this can I glean out of the sketchy epidemiological data?
Incidentally, on the matter of testing and data quality, the source I use reports testing frequency. At first I suspected that a higher testing frequency would generate nominally better outcomes, as that turn up more mild cases of the disease, but that didn't appear to be consistently true on a cursory glance at the numbers. I will take a closer look at that.

Historian said...

@ReddStater:
Keep in mind that while I work around medical professionals, my expertise is not in that area. So, keep that well in mind.

First, could the existence of multiple clades account for the reinfection problem that some have reported?
H>Yes, it could. There certainly could be other explanations as well. It is possible that different clades express at different rates, or that an incompletely surpressed infection mutated in the body to become active again. There is much about this novel disease we do not know.

Would antibodies for one clade of the virus confer resistance to all others?
H>Anecdotal evidence suggests that the answer to this is "no!" but again, the correct explanation for the reinfection may be otherwise. There are all sorts of possible explanations. Evidence from South Korea suggests that some former positives have very weak antibody production, but the reason for this is not known. It is early days yet. This may be a testing artifact, and not real. It depends on what portion of the virus the antibody is reacting to, and whether that portion is mutable or not. If the targeted bit of viral anatomy is one which cannot be changed without rendering the virus nonfunctional, then the body would likely retain effective response.

If not, what is the real likelihood of developing an effective vaccine against it?
H>Hard to say with any authority. On the one hand, you have some very bright Chinese Communist researchers who thought they had this contained, and whatever their ultimate objective was, I'd bet that this did not include devastating their own country. The evidence so far is that this particular coronavirus has a lot we do not know about it. We do know that other coronaviruses have proven immune to vaccines- the common cold, for example. On the other hand, I am generally optimistic about human ability to overcome these challenges. If there is sufficient motivation, I expect that a vaccine will be developed that will reduce, if not completely eliminate, the potential impact of this disease. I also expect that research will show which people are more at risk. Both of these developments are likely to take years, if not decades.

Speculations in advance of data may suggest reasonable hypotheses to be tested, but cannot themselves be conclusive. This problem is compounded by the recent epidemic of 'made-up science' by people whose lack of moral fiber and basic truthfulness corrupt the scientific process. It is very hard to sort out the truth from the corruption; there is a very great deal we do not know.

There are, however, a few things we do know; We do have a reasonably effective treatment regimen, and we are gaining understanding on why it works, politically motivated naysaying notwithstanding. That is a step forward. We know testing helps, and rapid testing is becoming more available, another step forward. Like most diseases, turning the Chinese Communist plague from a potentially lethal threat into an annoyance is an ongoing process made up of a number of small steps, not a single massive development. It will take time.