Kentucky COVID: Lies, Damn Lies and Statistics.

  • Why the 6% is important, but not what you think.

  • Why cases are a failed measure of risk

  • What’s next?

“Only 6% of reported COVID deaths are exclusively from COVID” This is a very important piece of info, but most people are missing the point.

First, the 6% thing isn’t new. It’s been tracked on the CDC site almost since the beginning of this. Second, people over 65 often have other things wrong with them when they die of the thing that causes death. And those things might contribute to how fast they die or even if they die. If I’m fat and have COPD, COVID is probably going to be more severe for me and if I die, it will be with and significantly because of COVID. Saying the other 94% didn’t really die of COVID is like saying Barry Bonds hit all those homers without the help of Steroids.

2.6 Bad Things

So, please don’t say only 6% of COVID deaths are a real number. It weakens the quality of the argument and there is a quality argument to evaluate here. The fact that 94% of people dying with this have 2.6 other bad things going on is good news. It means if you are healthy, younger than 65 or both, the chances of this being a material threat to your life or those around you in that category is less than the material threat of dying in a car on the way to the Wal-Marts. Or, said another way, the worst pandemic of our lives requires 2.6 other bad things to be wrong with you to have even a 5 or 6% chance of dying if you are among the 1% (probably 2% by the time this is over) that get it.

Maybe not as scary as the anecdotal and emotional stories you hear during Andy’s old-time radio hour, huh? The real problem is lack of transparency. Andy doesn’t share the ‘died of’ v. ‘died with’ numbers every day, because he knows that people without the 2.6 will start to evaluate the relative risk differently and be less scared out of their mind. He doesn’t promote the importance of a healthy lifestyle and healthy decisions as the best tactic to prevent serious illness from COVID (and for that matter lots of other things) because he wants you to think a vaccine will save the day (many months from now and even then it’ll be a miracle to work half the time and not have its own side effects).

Inflated Death Counts?

The “6%/94%” concept does actually point to an inflated death count. Not a “94% of people aren’t dying from this” message, but more like “could 10-15% of these deaths be classified wrong”. Over 2019 numbers, KY Non-COVID respiratory deaths are down 14%. Diabetes down 17%, Nephritis down 20%. All in: 542 deaths. Avg Age: 78. What is the likelihood all of these are down this much at the same time? If 100 of these deaths are classified wrong, that’s a big deal.

Then there is the issue blatant miscoding of cause of death. If I get hit by a bus and had COVID—I didn’t die of COVID. If I shoot myself and am PCR positive, it was the bullet that did it not the Rona. Here is an interesting data set on accidents YTD with the primary cause of death being COVID across the country:

And knowing that there is a $20k Medicare reimbursement incentive for providers to classify a COVID death combined with all these other points…kinda makes you think there is noise in the death number at a statistically significant level.

The 6% thing is important, but often used out of context and wrong—just like Andy uses data. But the real implication of it means that you need to have a few bad health things at play to be super worried about this. And if you do, you might be more worried than most about lots of health things other than COVID. Also it means that we might be overstating the death count—not by 94%–but by a number significant enough to demand better reporting and more transparency.

“Case growth is scary”

Ok. Where to start? First, cases are a problem–or not–only if you can trust what a ‘case’ is. Well, it’s a positive test, right? Kinda.

I could go into a long overly technical description and if you want one you can find it here. But the plain english version is simply, we made the tests (PCR Tests) too sensitive. And it’s picking up ‘positive’ cases of people with barely have enough of this in them to have a symptom or be contagious. I’d call that a false positive, but technically it’s a positive case and it could be happening…wait for it…as much as 90% of the time, though the real number is probably more like 60%…but still?!? Additionally, new research shows the true false positive rate of PCR tests is around 2.3%. Ooooh…fun…let’s do some math.

In Kentucky we’ve administered 826,000 PCR tests. 2.3% of those would be around 19,000. 19,000 of our 49,000 ‘cases’ in this model or 39% would be straight false. Now add in how many cases are positive but with barely have enough of this in them to have a symptom or be contagious. The entire daily 4:00 narrative is based on cases, positivity rate (cases and testing) and the ‘White House Guidelines’ (Cases and Incident rate). Despite reported case growth, measures of serious illness have been relatively stable since June, making this look more like a casedemic than a pandemic.

Now let’s talk about probables!! All of a sudden a few weeks ago, the number of ‘probable’ cases became a statistically significant number of the daily reported cases in KY. So, not only is the accuracy of a case in question, but now we are adding in a large number of probable instances of those bad measures. September 1 showed 23% of all cases were probables and probable trending is moving higher, but before math–a definition. One definition from the state website of what a probable is:

  • Meets vital records criteria with no confirmatory laboratory testing performed for COVID-19 unless additional epidemiologic or clinical evidence exists to refute the vital records criteria


Huh? Here is the graph (thanks to Probables in Blue.

The new information on issues with these ‘positive cases’ all but proves the point that cases are a failed measure of relative risk. Or said another way, the chances of rising cases leading to terrible outcomes is relatively small compared to all the good things I’d give up to hide from becoming a ‘case’.

So now what?

Why are we not evaluating detailed information on the measure of serious illness from COVID? To measure relative risk, we need current (daily) data on Hospital admissions, ICU, Vents, Capacity, Deaths of and Death with—all by age, location and LTC. The current reported hospitalization data lags enormously (CDC), is incomplete or highly questionable to the point of incompetence or fraud or both. If we’re in the worst pandemic of our lives, how can we not have accurate info on that daily? It’s like we are killing ourselves to measure an atmospheric pressure anomaly to determine if it’s raining outside, rather than seeing if the porch is wet.

If we did get to the accurate info (which we don’t really have much of), probably what we’ll find is that people over 65 with other bad stuff wrong with them should be extra cautious. People that regularly engage with those in that category should be more cautious. For everybody else—be not afraid—and we should start living our lives and demanding the freedom to do so. You could die of lots of things way before COVID gets you. And if you’re still ‘concerned’ it’s ok. Just stay home.

All data sourced through the State’s COVID website, the COVID Tracking Project, the CDC or the articles linked through the piece.