I wrote a few days ago that the data suggests the coronavirus outbreaks in Tennessee and Kentucky are much more similar than the overall number of cases would suggest. This is an update.
Tennessee’s coronavirus death toll has risen substantially over the last few days. There are now 72 people who have died of COVID-19 in the Volunteer State, as of Tuesday, April 7. On Saturday, just 72 hours ago, there had been only 43 people who had died, and the death toll has essentially doubled in the last four days (it rose 48% in one 24-hour period from Sunday to Monday).
But the death toll has also risen quickly in Kentucky. The number of coronavirus-related deaths there has climbed from 40 to 65 since Saturday.
So Kentucky’s death toll has risen 63%% in three days, while Tennessee’s has risen 68% in three days.
As of Tuesday, there were 1,149 coronavirus cases in Kentucky, and 4,138 in Tennessee. It’s an almost 4-to-1 spread, which seems pretty damning on the surface. But, as I’ve maintained all along, much of that difference can be explained by the fact that Tennessee is testing far more people than Kentucky.
Tennessee had tested 52,874 people as of Tuesday, or 778 per 100,000. Kentucky had tested 21,604 people as of Tuesday, or 480 per 100,000. So the Volunteer State isn’t quite testing twice as much per capita, but it’s a fairly significant difference — 62%.
In Tennessee, 7.8% of the tests are returning positive. In Kentucky, 5.3% of the tests are returning positive. It’s a notable difference, but not overwhelming.
One interesting point to note, according to my conversations with Scott County Mayor Jeff Tibbals: Not all of Tennessee’s negative test results are being reported. In Scott County, the state lab tests are being counted, as are the tests at one major private lab — AEL. But the tests at all other private labs aren’t being counted. There’s one major physician’s office in Scott County, Grace Primary Care, that uses AEL. However, the largest primary care provider in the county, Mountain People’s Health Councils, uses LabCorp. The county’s only hospital, Big South Fork Medical Center, also uses LabCorp. It isn’t known how which private labs the other primary care providers in the county use.
I can’t speak to how that plays out across the rest of the state. But in at least some counties, there are significantly more people being tested than are being counted. (All positive test results are counted; it’s the negatives from some private labs that aren’t being counted.) So Tennessee is testing more people than the numbers reflect, but just how many more people remains in question. Perhaps the same is true in Kentucky, but I’ve heard nothing to that effect.
Unfortunately, those unknown numbers can’t be accounted for. So we can’t get a true breakdown of what’s happening with coronavirus in Kentucky vs. what’s happening in Tennessee. But using the numbers we have, we can at least make a halfway-educated guess.
Based on what we know, if Kentucky were testing at the same rate as Tennessee, there would be 1,859 cases of coronavirus in Kentucky. That’s assuming the same rate of positive tests, 5.3%, held true.
Based on that, Kentucky would have 62% more confirmed cases of coronavirus than it currently does, but still fewer than half the number that Tennessee has. However, population differences have to be taken into account to. Tennessee is a state of 6.8 million people; Kentucky is a state of 4.5 million.
If you break those numbers down on a per-capita basis, Tennessee’s infection rate is 61 cases per 100,000 people, while Kentucky’s would be 41 cases per 100,000 people. So the outbreak is still worse in Tennessee, but the gap just keeps shrinking.
Now let’s look at fatalities. Based on Tennessee’s 72 fatalities, the death rate in Tennessee is 1.7%. Based on Kentucky’s 65 fatalities, the death rate in Kentucky is 5.6%.
I’ve argued from the beginning that the death rates are the the surest example that Kentucky isn’t testing enough people. At 5.7%, Kentucky would have the highest death rate in America. To put that in perspective, the death rate in New York is only 3.9%, and in New York City it’s 5.2%. The death rate in Louisiana is only 3.6%, and in New Orleans it’s only 3.7%.
Why is Kentucky’s death rate the highest in America — by a substantial margin?
There are some answers. Diabetes is a big problem in Kentucky; 14.5% of the adult population has type 2 diabetes, compared with 10.5% of the population across the entire U.S. High blood pressure is also a problem; 39.4% of the people in Kentucky have high blood pressure, compared with 32.2% in the U.S. as a whole. Smoking is another issue; 23% of adults smoke in Kentucky, compared with 15.1% across the U.S. as a whole.
But the health differences between Kentucky and Tennessee aren’t that great. In Tennessee, 13% of the population has diabetes. In Tennessee, 38.5% of the population has high blood pressure. In Tennessee, 23% of adults smoke.
So the question remains: Why is the coronavirus death rate in Kentucky 5.6% while the death rate in Tennessee is just 1.7%?
There’s only one logical conclusion: Kentucky’s death rate isn’t 5.6%. It isn’t substantially higher than any other state in the nation.
Because there are more than 1,149 people in Kentucky who’ve been infected with coronavirus.
Let’s back up to the number we derived if we assumed Kentucky were conducting the same amount of tests per capita as Tennessee, with its current number of positives. That theoretical scenario concludes that there would be 1,859 cases of coronavirus in Kentucky. And if you run the numbers again, Kentucky’s death rate would be 3.5%. Still higher than Tennessee, by a long shot — and likely because Kentucky has more than 1,859 coronavirus cases. But more in line with the rest of the U.S. (the nationwide death rate is 3.2%).
Based on these numbers, it really stretches the imagination to assume that Kentucky’s numbers can be taken at face value. Simply put: either there is an alarming number of people dying from coronavirus in Kentucky, or the commonwealth has substantially more cases of the virus than it is reporting.