COVID19 Testing: Tennessee Versus Virginia

Tennessee drive-through test location

Much has been written about the paucity of COVID-19 testing in Virginia. One place where the shortfall is most evident is in Bristol, a city bifurcated by the Virginia-Tennessee state line. On the Tennessee side of the border, anyone who wants a COVID-19 test can get one, reports the Roanoke Times. On the Virginia side, tests are reserved for the very ill and those who care for them.

In Tennessee, the health department has set up testing centers where people can come Saturday and Sunday to be tested if they are worried they have the disease. They do not have to be pre-screened.

In Virginia, testing centers require pre-screening and a doctor’s orders.

“We are scouring the country looking for equipment to help our institutions increase testing. We haven’t been successful because of a national shortage,” said Virginia Secretary of Health and Human Services Daniel Carey.

The lack of testing capacity will have severe real-time impact on Virginia’s economy. Governor Ralph Northam said in his Friday news conference that more widespread testing is needed before Virginia can relax its social-distancing restrictions.

“We need to be able to test, we need to be able to track and we need to be able to isolate individuals,” Northam said. “We have no national guidance on testing,” he said. “Every governor is having to establish our testing protocol and our supplies on our own. While that improves every day, we are not there yet.”

Question: How come Tennessee can do more testing than Virginia? It was dealing with the same national shortage as Virginia. What did Tennessee do to build its testing capacity that Virginia did not do? Is our situation different somehow? Or have Tennessee’s public officials just been more aggressive,, competent or far-sighted?


There are currently no comments highlighted.

51 responses to “COVID19 Testing: Tennessee Versus Virginia

  1. Brilliant research and the question of the year.


    More detail. In Tennessee it seems to be the test process where a swab is stuck up into your nose, not one of the rapid result tests.

  3. Do we need testing to identify who is currently positive for Covid 19, a snapshot, or testing to identify who has antibodies? If the latter, who decides which testing mechanisms are most accurate and able for wide deployment?

  4. JAB – did you get permission from the WaPo and other defenders of the Democratic Party to investigate this issue, much less to write about it? Ever since I was told by a WaPo reporter that the editorial board pressured the reporter not to write anything critical of Tim Kaine, I believe news suppression is part of the Post’s business strategy. Democracy Dies in Darkness my ***.


    Check out this March 5 article from the Tennessean. March 5! Expanded and apparently coordinated testing. Do we have that yet in VA? As I feared/predicted a couple of weeks back, the testing disaster will be the focus of the “after action” reviews….

    • Meanwhile, the University of Virginia claimed to have started coved-19 testing weeks ago after finding a million dollar private gift and now is scrambling backwards after its tiny testing start.

      Instead, now UVA claims to be focusing on “long haul” with yet another modeling project. So much for UVA’s world class research and its addressing real problems in Virginia, ones that now threaten and take the lives of Virginians daily.

      Hence, on this coved-19 crisis, as with most else it does, UVa appears to be little more than a huge advertising and propaganda machine designed to raise other peoples’ money to spend on itself with little benefit to anyone else, whether in Virginia or otherwise. This obviously phony UVa. PR campaign has been going on in plain sight for a decade now. Yet few in Virginia have bothered to complain, much less do anything to fix the real problem.


      Likely because the money kept rolling into UVA for things like $10,000,000++ remodeling jobs on UVA President Ryan’s house, and $60,000,000 spent on Board of Visitors board Room, formerly known as the Rotunda, and into vast social justice and race baiting jihads, and much else make work, that does little more that enrich those who run the University and those who have captured it for their own private advantage, at the expense of everyone else.

      • Was posting on this below before I saw yours, Reed.

        • Carol –

          Thanks for posting that follow on detail to my comment, detail that I was unaware of as regards this latest UVa. event. It reaffirms and drives home my above comment that elaborates on an even earlier one made. UVa. is shameless obviously. This most recent event follows a long stream of similar claims and assertions of UVa. over the last decade that never pan out, are half truths, cover ups of truth, outright falsehoods, posturing, and/or braggadocio often childlike in nature. Many of these are documented in detail on this blog over past 9 years. We should collect them and lay them all out side by side on this blog. The result would reveal and devastate.

  6. While I appreciate the apparent discrepancy between the two states, I guess I’d like to know what we are testing for.

  7. Some questions. How much has UVA followed through on what it said March 30th when they announced that a “$1 Million Grant to UVA Health Boosts Ability to Produce COVID-19 Tests.” That “funding is expected to help increase the numbers of tests produced daily from 200 to more than 500…will allow testing in a second shift, significantly expanding capacity. The funds will also be used to order a second robotic instrument which, when installed, will further increase capacity.”

    “The Qualitative Foundation grant will also be used to implement a new testing platform that will allow test results to be delivered in as little as two hours.” Anyone see a further press release on any of this or how many tests they are processing now?

    How much is the Division of Consolidated Laboratory Services work on “building a library of genetic information from the positive tests DCLS gets, as well as those from private labs, health systems and university systems in Virginia” (announced by Governor Northam on April 6) impacting their ability to test more Virginians?

  8. Are they? Are they really doing more, or more harm?

  9. The whole testing discussion is a mess and we know it is when we say “test” and we don’t know what it means in that context.

    “testing” only lasts as long as it takes for someone to come into contact with others. Once that happens , they could be infected if someone in the group they had contact with is infected.

    That’s why contact tracing is so important. Even contact tracing would NOT find everyone but it will find a lot – and it will isolate them.

    Many seem to be looking for a silver bullet type answer, akin to finding those that have immunity. There are none. Every path has less than 100% satisfaction.

    What’s even more amazing is that the ones in the streets don’t seem to care at all about testing… much less one kind – just “open up”.

    But it would be good to know what exactly Tennessee is actually doing with testing… just testing without contact tracing is not as useful although
    many employers would be happy with just being able to test their employees – but again the problem is they’d have to test them almost every day or certainly more than one a week and if one did test positive – they’d then want to test all the employees they had had contact with.

    There is no silver bullet here.

    • These are some of the some thoughts I have been having. A positive test is just a snapshot. If everyone is not being tested, with those with positive results and everyone that person had contact with being isolated or quarantined, testing for the sake of testing, unless done repeatedly, has minimal value.

      • But if it is more likely to be a false positive than a true positive then you are overburdening the track and trace resources.

        At least, the results, if properly understood, can be used to get an idea of the spread, assuming they are collecting some geographical information about the subject and his home/work/movement.

        • re: “false positives”

          if true, what’s the better testing approach?

          • It’s not a matter of a ” better” approach, but what is done with the results.

            The test may very well be the best they can do. But, if they start with an assumption that the, say, neighborhood that the person came from has a 2% background infection probability, then they can use the positives and negatives, conditioned by the device specifics to do what is called a “Bayesian Update” to better estimate the neighborhood infection probability.

            You may have heard of a Kalyan Filter. The cool thing about it is that even if the measurements you take are really bad, very noisy, the KF will produce an estimate of the state with a smaller and smaller error with the more measurements you take.

            The Bayesian Update does the same sort of thing with the apriori probability density. So, you can get a better idea of the distribution of the disease, even with poor measurements.

            What you cannot do is accept the results are certain, as I showed previously as Dean Wortmier.

          • Kalman … stupid autocorrect. At least it learns.

          • Kalman Filter ?

            I understand what you are saying and yes… it’s more than just the test result – which is just raw data that has to be put into a process like you are alluding to of which there could be several but is a discussion with the folks that know the options.

            And agree, the average person is looking for a simple yes or no type test answer.. and there is more to it than that – it really is a mathematician’s realm…

            No one here cares but we did a “W” matrix when we had multiple error sources – linear algebra – and that would take the error sources and combine into an equation that then yielded one answer for overall error… when you have a a 3-d body moving through a geometric sphere – all kinds of things affect it… but the folks who know math can put that THANG right on where it’s aimed at – within an expected error !!

          • Yes Larry, those guidance and control algorithms use Kalman Filters. I think the W matrix to which you are referring is the Wronskian. It describes the change in state wrt the measurement. Mathematically, it determines if a state is observable or controllable, that measurements can be used to improve the state estimates, or that controls can be used to make the state stable.

            Fortunately, this is true for modern aircraft. Unfortunately, lose the computer and the pilot and crew will wind up a pizza on the cockpit canopy.

          • head is starting to hurt again like it used to… 😉

          • Why am I worried that Dean NN understands this math better than anybody working at the Virginia Department of Health…..

          • It’s really a dual discipline field. Epidemiology by itself limits knowledge to the nature of infectious diseases. In order to be able to take raw numbers of the characteristics to knowledge, it takes the math and the mathematical analysis of the numbers that the epidemiologists know and provide.

            It’s sorta like having a bunch of engineers working on a missile [- they know the hardware and operating environment, etc, but they don’t really know how to process the info into things like probabilities and risk – that’s the province of mathematicians, physicists, etc.

            You shoot that missile 20 times and it gets 20 hits not dead on target… how do you understand that and make changes to improve accuracy when the errors themselves are not easily quantifiable into anything more than seemingly random data?

            So you have a plethora of different kinds of tests – each with it’s own characteristics to include it’s error rates.

            How do you decide what the testing regime should be – based on the choices you have from the various tests? There is no one silver-bullet test – it’s much more complex than that.

            NN uses abbreviated shorthand to say this… she/he says a LOT in a very few words… but not everyone really catches it.

            And we have a bigger problem with society that has more than a few who do not trust science nor believe scientists. Witness that in the protestors – they’re not calling for a particular path forward, nope, they don’t care, they just want to re-open.. and
            let it all hang out… impatient with the folks who say ” we need to think this through” – nope – we don’t need no stinking thinkers.

          • “Why am I worried that Dean NN understands…”
            Don’t be. I assure you that there are those in Richmond who know far more than I.

            Given Northam’s reluctance to give the unwashed masses their pacifier of “tests for everybody”, I can only assume he’s been told that will really screw things up.

            This I can tell you, SC, GA, FLA, and any other states that open in the next week, or waste valuable diagnostic tests on Joe Sixpack’s curiosity will sorely regret the month of June.

  10. re: ” … testing for the sake of testing, unless done repeatedly, has minimal value.”


    Let’s say an employer wants to insure all his workers are virus-free.

    So he/she would test them all – and they’d all test negative.

    How long would the employer be assured that no one got infected after that?

    One day, a week?

    It would all depend on how many others they came in contact with.

    So someone in an ER at a hospital – you’d need to test them daily.

    but how about a restaurant?

    this is why contact tracing is so important.

    Everytime someone tests positive, you need to track down everyone they were in contact with and test them and every positive one needs to be isolated.

    This is not a new thing – this is exactly how VDH dealt with infectious diseases before – like tuberculosis.

    This is a race. Can we find infected people – faster – than they would infect others?

    and this is going to go on for a while or until some percentage of the population has been infected and/or a vaccine is developed.

    In the middle of all of this – there is a question about those who have developed immunity and therefore would, in theory, not have to be further tested. That idea has issues and will need some effort to decide whether or someone can be immune, for how long, and could they still be a carrier even if immune?

    Most of us are fed up and frustrated but the reality is this is what it is. It’s not what we want – but it is what we got.

  11. The doctor who conducted the Santa Clara County study said that it required a finger stick blood sample and cost in the $200k range.

  12. Let’s be honest – our governor might as well be Forrest Gump or Ralph Wiggums as Ralph Northam.

    The Democratic machine foisted a dud on the state.

    I’ve been calling out Virginia’s testing failure for a month. At first the Northam apologists blamed Trump. Now that it’s obvious that many (all?) other states are running a more effective testing program the Northam apologists insist that testing isn’t important.

    This isn’t that hard. When you find yourself with a dud as your leader you look elsewhere for competent ideas. Jim looked at Tennessee. He found what sounds like testing on demand. I look at Maryland and do you know what I find? 500,000 test kits from South Korea. Hogan’s wife is apparently a Korean-American so she made some calls. Voila. Half a million test kits.

    From the article: Hogan said adequate testing is one of the “four critical building blocks of our Maryland Strong Roadmap to Recovery.”

    So, a competent governor thinks testing is critical and his state has a roadmap to recovery. First things first, how about we call Gov Hogan and ask for a peak at that roadmap. Yeah, we’re going to outsource the management of this crisis to Maryland. At least we’re going to outsource any of the deep thinking required … like having a plan. Then, let’s ask around the General Assembly and see if anybody knows anybody in Germany. You know – the other modern western country that used broad based testing to get past the peak. Yeah, that’s the country. The one that is reopening its economy right now. Maybe they can spare half a million test kits or so.

    • Maryland governor: ‘Simply not enough supplies’ on hand to tackle coronavirus

      Hogan not competent?

      • Nice try with an article from March 17. Yesterday was April 20 Larry. Hogan solved Maryland’s problem.

        • Trump attacks a Republican governor for following his coronavirus testing advice
          Gov. Larry Hogan bought test kits from South Korea. Trump isn’t happy about it.

          Gov. Larry Hogan (R-MD) followed President Donald Trump’s advice and took coronavirus testing into his own hands. Trump attacked him anyway.

          Trump began Monday’s White House coronavirus briefing by criticizing Hogan — chair of the National Governors Association — for turning to a foreign source to buy coronavirus tests.

          “The governor of Maryland didn’t really understand,” Trump said, describing a call that Vice President Mike Pence had with governors earlier in the day to encourage them to do more to increase coronavirus testing on their own. “He didn’t really understand what was going on.”

  13. Food for thought: mononucleosis, tuberculosis, herpes simplex, chicken pox, HPV… SARS COV2?

    New criteria for testing in Virginia: known contact, symptomatic, 1st responders and medical personnel, and hospital and nursing home entrants.


    • NN, Known contact was dropped from VDH guidelines for testing from March 24 forward.

      4/17, for State lab, they added:
      Symptomatic staff, residents, entrants to all congregate settings; symptomatic with underlying conditions; symptomatic un or underinsured. Newborn of diagnosed mother is only one not required to be symptomatic to be tested.

      For clinicians for private lab testing as resources are available: they added critical infrastructure workers with symptoms, individuals not meeting other categories with symptoms. Healthcare workers and first responders. Individuals with mild symptoms in communities with high hospitalizations.
      NON-PRIORITY Individuals without symptoms.

      • Carol, BR’s fact guru, says VDH guidelines for testing changed on 4/17:

        “4/17 Symptomatic staff, residents, entrants to all congregate settings; symptomatic with underlying conditions; symptomatic un or underinsured.”

        Now count it up on your fingers, 1, Jan., 2. Feb. 3. March, 4. April – 1, 2, 3, 4 – that was April 17, that was FOUR DAYS AGO. That’s so bad it makes Inspector Clouseau in the Pink Panther look competent as Albert Einstein.

        • Reed. Sorry for incomplete info: 4/17 addition was entrants to congregate settings, as well as the symptomatic with underlying conditions or symptomatic un/underinsured.

          3/24 they updated from symptomatic nursing home or long term care residents or staff to all symptomatic congregate setting staff or residents.

      • Those were the criteria Northam stated in his conference yesterday. If he errored, well, so be it. More importantly, is the oft repeated “with symptoms ” — increases the probability that it is not a false positive and we won’t have 60x asymptomatic as confirmed using unvalidated devices.
        Non-priority aka good luck getting a test

  14. No, the Santa Clara test, whose subjects were invited via Facebook, cost about $200 k. Similar tests have been conducted more recently by Dr. Bhattacharya for support employees of 27 major league baseball teams. Those results have not yet been published.

  15. Prediction: large numbers of those who tested positive in high testing areas, like Santa Clara, will begin to mysteriously reinfect.

    • Yeah, so far those finger stick tests are not proving out. One or two that are more reliable may rise to the top. I think the basic conclusion is valid, because it is being found in other ways in other places: lots of folks have been infected, far more than the stats report.

      • I was pleased to see that the governor expanded, only slightly, the criteria on obtaining a test to persons with symptoms, hospital and nursing home entrants, and medical workers.
        So, perhaps it’s wishful thinking, but maybe someone has told him of the futility of drive-thru tests at the CVS.

  16. “Nancy, I don’t undsertand that stupid equation you put up. I’m a Conservative. I believe in God and truth and right. I don’t believe in mathematics.”

    “Besides, we will test everyone so what does it matter?”

    Okay, let’s test everyone using that device. Let’s go into a community of 100,000, say Roanoke, and look at what we have, and might get.

    Roanoke– population 96,700, COV2 cases confirmed 24.
    Let’s assume there are absolutely none unreported.

    Well, if we test with a machine that detects 97% of infections, you will get 24×0.97=23… meh, call it 24. Got ’em all.

    It will also get false alarms. Let’s be generous. Let’s say the false alarms are not 5%. Let’s say it’s 1/2%, 0.005×96,676. That’s 483 cases.

    “Okay so what does that mean?”
    “Well, they’re false alarms. Those people were NEVER sick.”
    “What? You mean asymptomatic?
    “No, I mean never sick.”
    “But, but, but, that means they never were immune. They could get sick!”

    • Nancy , if you test the same population twice with the 97% machine how many will you find? How about 3 times?

      • Assuming a single queue? Well, there’s the guy in front of the sick person, and the guy standing behind him, who are at risk of turning up positive on the second test….

        • well, assume more than a single queue… enlighten me…

          • Well, depends on the time between tests. Let’s assume I could run the tests faster than people could become infected. Every time I ran the test, I would expect a crowd of around 500. In the 2nd test, I would expect to see 25, 26, 27 faces I saw in the first test, the rest would be strangers. In a third test, I might again see 26 faces. The “true positives” at 97% detection would most probably appear in all 3 tests. Something like 4 to 6 people would probably appear in two of the 3 tests. The exact probabilities can be worked out.

            But given 100,000 test kits, a better “experiment” might be to randomly select, 5,000 Roanoakeans(?) and test the same group 20 times over maybe 2 months. As the disease spread, you’d see more and more positives. The true positives would be seen in lots of experiments and the false positives would not. You would be able to then get a good estimate of the infection rate in the city.

      • Why would you retest the entire population? You would only retest the 483 + 24 that tested positive. That’s 505. Even if the false positives were 5% you’d get 4,830 false alarms + 24 truly sick people requiring 4,854 retests over a total original testing regime of 100,000 tests.

        • You’re trying to also find the ones who are infected in the general population that show no symptoms and the tests themselves are only as good for the day you were tested if you have mixed with others who may have been infected.

          If you are an employer with a workforce. How often do you want to test them – even though non show symptoms?

  17. “Nancy, I don’t understand that stupid equation you put up. I’m a Conservative. I believe in God and truth and right. I don’t believe in mathematics.” Wow. Now you are venturing into my usually useless college degree in religion, and I’ve always thought Galileo got it right about math being the language God gave us to understand the world.

    • 👍. God gave us mathematics and science as well as religion and philosophy so that we might answer our questions.
      He gave us math and science to seek the answer to “how” questions, e.g., how did life come to be?
      He gave us religion to answer “why” questions, e.g., why is there life?
      I will readily accept whatever religious answer to your “why” question if you will just leave it out of my “how” answer.

Leave a Reply