Skip to main content


Thoughts On a New Coronavirus Test (And on Testing)

Word came yesterday that Abbott received an Emergency Use Authorization for a new coronavirus test, one that is faster and cheaper than anything currently out there. The two types of tests that we see in use now are RT-PCR, the nasal-swab test that detects viral RNA, and various antibody tests, that tell you if you have raised an immune response due to past exposure to the virus. This one has features of both, but its main use is more like the RT-PCR test: it will tell you if you are actively infected. It does that by detecting a particular antigen, the nucleocapsid protein (Np) of the coronavirus. It’s a key part of its structure, and in an actively replicating infection you can be sure that there will be plenty of that one floating around.

The test itself is one of Abbott’s “BinaxNOW” assays, and they have a whole line of these already as tests for malaria, RSV, various bacterial infections, and so on. It’s a lateral flow assay, which will be familiar to anyone who’s seen a pregnancy test, and I explained the general principles of those (as antibody tests) in this post. This new test is a sort of flipped version of what I described there, though. In this case, a nasal swab is taken, and several drops of solvent are used to put that sample onto the beginning of the absorbing strip inside the card. As it soaks up along the length of the strip, the sample will encounter a zone of antibodies that recognize the Np antigen, and these antibodies are also attached to nanoparticles of gold. This gold-antibody-Np complex is carried along in solution further along the strip until it runs into another antibody zone, one that’s immobilized on the solid support and which will bind the gold-antibody-Np complex molecules tightly. That stops them in their tracks and allows the gold nanoparticles to pile up enough to be visible as a pink or purple line. Along the way, the sample has also crossed a zone containing another soluble gold conjugate species as a control, which gets carried along until it runs into another separate zone of immobilized antibodies specific to it. The presence of a pink control line means that the test has been performed correctly; absence of such a control line means that the whole test has been messed up somehow and needs to be run again.

I had described earlier a test that looks for antibodies to the coronavirus by running them past gold-conjugated antigens on the test strip, but this one looks for antigens by running past gold-conjugated antibodies. Developing a test like this involves a lot of work to find the right antibodies, to make sure that they’re attached to the gold nanoparticles in ways that don’t impair their function, to find the right second immobilized set of antibodies that will develop that test line, and to make sure that the control line system is compatible with the test itself. You’ll also need to work on the composition of the test strip and the solvent that’ll be used to take the patient’s sample into it: these need to allow as much of the antibody complex to flow down the strip in a controlled fashion as you can get, and to do so in the same way every time. And finally, you need to validate the assay with a lot of coronavirus patients and controls, to see what your false positive and false negative rates are.

For this assay, those come out to a sensitivity of 97.1% (positive results detected when there should have been a positive) and a specificity of 98.5% (negative results when there should indeed have been a negative). Flipping those around, you’ll see that about 1.5 to 3% of the time, you will tell someone who’s infected that they’re not, or tell someone who’s not infected that they are. That’s about what you can expect for a test that sells for $5 and takes 15 minutes to read out with no special equipment, but such tests (if used properly) can be very valuable. Flipping that around, you can also infer that if used improperly, they can be sources of great confusion.

What’s proper? The FDA’s EUA is for testing people that show up with symptoms to see if they really do have SARS-CoV2. I think that’s appropriate, because you’re more likely to have a higher percentage of those folks who are really infected. If you tried to deploy this test across a large asymptomatic population with a very low true infection rate – everybody in New Zealand, for example – you would create turmoil. New Zealand’s real infection rate is vanishingly small, but Abbott’s quick $5 test would read out a false positive You Are Coronavirused for 1.5% of the whole country, never lower, which would be a completely misleading picture that would cause all sorts of needless trouble.

On the other hand, if you’re testing symptomatic people in a community where the virus is already known to be spreading, you can do a huge amount of good. Let’s imagine you test 1000 such coughing, worried patients under conditions where you expect that 10% of them really do have the coronavirus. In the course of testing all thousand, you’ll run those 100 positive folks through, and you’ll correctly tell 97 of them they they need to go isolate themselves immediately, which is a huge win for public health. Three of them, unfortunately, will be told that they’re negative and will go out and do what they do, but that’s surely far fewer than would be out and around without the test. You’ll also run the 900 other people through who actually have a cold or flu or something and not corona, and you’ll tell maybe 13 of them (900 x 0.015) that they’re positive for coronavirus and that they should isolate as well. That’s not great, either, but it’s worth it to get the 97 out of 100 real infectious coronavirus patients off the streets. And meanwhile you’ve correctly told the other 890 people in your original cohort that they do not have coronavirus, which is also a good outcome. But remember, with that 98.5% specificity you’re going to send 15 people out of every thousand you test home to quarantine even if no one really has it at all. If 1% of your sample of 1000 people is truly infected, you’ll probably catch all ten people who are really positive. . .but you’ll also tell 14 or 15 people who don’t have it that they do, crossing over to finding more false positives than there are real ones.

And let’s not forget the other really good aspects of this test: it’s cheaper than anything else out there but best of all, it’s fast. The delays in the RT-PCR testing have been killing its usefulness in too many cases – what good is knowing that you tested negative sometime last week, really? Far worse, what good is knowing that you tested positive last week if you didn’t isolate yourself because you weren’t sure if it was the coronavirus or not? But an answer in fifteen minutes, that’s actionable. As long as this test is deployed correctly, it can be very useful.

Addendum: I’m well aware that the CDC seems (controversially) to be changing its testing recommendations in general. This “only test if there are symptoms” guidance seems to apply to RT-PCR testing as well – and turnaround problems aside, that test still has far higher selectivity and specificity than this new 15-minute one and is far more appropriate for use in a broader, largely asymptomatic population. We need to be addressing the delay problems in RT-PCR, because we need to be doing a lot of those tests – not closing our eyes and whistling a happy tune instead. This appears to me, and to many others, to be political interference from above. What else is one to think when administration officials have suddenly started referring to the pandemic in the past tense? So here’s something I never pictured myself saying: it is my hope that this CDC guidance will be ignored. It’s a hell of a situation to get to, isn’t it? Update: I am very happy to report that the CDC appears this morning to be walking this one back. Good.

110 comments on “Thoughts On a New Coronavirus Test (And on Testing)”

  1. Fraud Guy says:

    A hell of a situation indeed.

  2. Fraud Guy says:

    A hell of a situation, indeed.

  3. Etu Brutus says:

    WHO are we to criticise the CDC for lack of disinterest?

  4. G Man says:

    I’m still dumbfounded that we don’t have cheap paper strip tests that can be handed out by the millions to everyone, every school, every job site, etc.
    Something Amazon could mail a box full to every address.
    Why don’t we have this? The technology exists!

      1. Jason Gratt says:

        Visit Dr. Michael Mina’s website to see what actions you can take to make cheap at-home paper strip tests a reality.

    1. John Wayne says:

      The technology to create the test exists, but actually creating a new test and manufacturing large quantities of it still takes time. Science isn’t magic, you still have to do the work of actually making a thing that works.

      The pregnancy test is a good example of how accurate and cheap you can make something if you spend decades working on it.

    2. Cole J. Batty says:

      This is essential and it is dumbfounding that we aren’t doing it. We could quickly get back to something close to life as normal while keeping the virus largely under control using a testing regime like this. See the links in this tweet:
      to learn more about the effectiveness of this strategy.

      I think this post is also significantly overplaying the downsides of widely implementing testing in the asymptomatic population. See Nicholas Weininger’s comment below. The test is $5 and we have better PCR-based tests. If you have a positive you can either 1. take another $5 test to confirm or 2. Confirm with a rigorous PCR-based test. High frequency of testing will take care of false negatives. The key to this strategy is widespread implementation, and a post saying that such implementation would cause ‘turmoil’ is actively working against this implementation.

      1. ILikeMike says:

        Dr. Mina’s argument makes a lot of sense to me but hey, what do I know?

      2. Wilhelm Cody says:

        This staged testing was also the standard for HIV testing. Do a quick test then a slower more expensive confirmatory test if positive. While waiting for the result, try to avoid situations where you might infect someone else. From the CDC website on HIV testing:

        “What does a positive result mean?
        :If you use any type of antibody test and have a positive result, you will need another (follow-up) test to confirm your results.
        “If you test in a community testing program or take a self-test and it’s positive, you should go to a health care provider to get follow-up testing.
        “If your test is done in a health care setting or a lab and it’s positive, the lab will conduct the follow-up testing, usually on the same blood sample as the first test.”

    3. Here is a website advocating for that: These people are in contact with Dr Mina of course.

      The biggest problem with mass testing is the ambiguity. As the analysis in the article above shows with mass testing you end up with lots of false negatives and false positives and the authorities don’t know what to tell people about that. There is a bit of truth in the fears – but it is so paternalistic – as if we were not living with lots of other fuzzy indicators like thermometers. Maybe it just needs to be done first on smaller scales – so that people have time to learn how to react.

      1. Wilhelm Cody says:

        The key to rapid,, inexpensive testing is to do it every day someone enters a situation where they could infect someone outside of their small social bubble: entering a school or work or a shop or wherever. There is a simple logic tree to follow that compensates for the situations Dr. Lowe presents. In fact, this testing regime will be much more effective than our current one.

        -If positive, go immediately for a slower but more accurate test then go into quarantine.
        -If a true negative, the result will come back within a couple of days (especially if the nucleic acid testing capacity is no longer swamped by using it as a screening test) and you can get out of quarantine.
        -However you still get tested every day you enter one or more social situations.
        -if a true positive, discuss details further with your physician and, if you have symptoms, treat accordingly. Also, really really quarantine. Also, discuss contacts with contact tracers

        -If negative, continue practicing physical separation, hand washing, face mask wearing. That will reduce any risk if you were a false negative and help protect you if you were a true negative.
        -The next time you enter a social situation, you get tested again.
        -If you were a false negative because you were in the first hours of shedding virus, a time when virus levels are low. That low viral level seems to be the primary cause of a false negative. Between the first and second test, the virus levels have climbed to become very infectious and also very detectable even by this assay. With the positive result, go back to the action above for positive results.
        -Because you were diligent between the last negative and first positive result, it is unlikely you would infect anyone else outside your bubble.
        -If you were a false negative because you were in the last stages of infection, you are even less likely to have a positive result with the rapid test.
        – It is likely that the active virus count is only a fraction of the antigens produced, based on research results showing low infectiousness even with a positive nucleic acid test at this stage.
        -Because you were diligent, by physically distancing, washing hands, and wearing a mask, you were unlikely to have infected anyone outside your bubble over the previous days. Anyone inside your bubble who did get infected will be discovered either by symptoms or by being tested when they go into a social situation.

        1. eub says:

          Yeah, building capacity so that PCR testing could be used after antigen testing — so PCR testing is fast — seems to make a hell of a lot of sense. I hope it happens in a systematic way.

    4. RAD says:

      This is a type of “paper test”. The question I have is why Abbott’s existing BinaxNOW platform did not come up in the discussion and coverage following Michael Mina’s revelations in his NYT OpEd and the TWiV podcast coverage.

    5. Ezra Abrams says:

      Dear G Man
      Part of the answer is that “quality” takes time and a lot of money
      eg Antibodies cost money to make and QC
      The gold particles are not, AFAIK, that easy to make
      You need trained people to do QC testing

      also, you don’t want people just testing; if the test has a false pos rate of say 1 in 500, if you have millions of tests, that is a lot of false positives, which means people getting really upset and calling their doctor and wasting a HUGE amount of time and money in the medical system

  5. Alia says:

    This would be a great thing to use this fall at the PCP’s office to screen all those people coming in with flu-like symptoms.

  6. WCMR says:

    Here is an explainer video of the concept of antibody testing and false positives/negatives which may be helpful for understanding:

    Thanks for the great reporting Derek!

  7. Nanochem says:

    And what about running two tests per person? That would reduce the number of false positives (and negatives). If there are enough tests available, the price will double but is still low.

    1. aairfccha says:

      This assumes the false positive and false negative results are truly random and at least false positives might not be (cross-sensitivity with other types of coronavirus).

      1. Nanochem says:

        True, I see.

        1. Nanochem says:

          A doubt concerning sensitivity/specificity: how are them determined?
          I mean, will the rate of false negatives/positives be the same now than in other seasons when there are other types of coronaviruses in higher numbers?

      2. Eric Nuxoll says:

        Maybe. If they’ve already got approval for the single test, then I hope they’re already doing the trial to see what the sensitivity and specificity of a double or triple test would be. It wouldn’t be that difficult to do, and it *could* result in a test that’s still comparatively cheap and fast with a far great scope.
        Or maybe they already did (it seems pretty obvious) and discovered that the errors really are mostly correlated and just aren’t talking about it.

        1. FoodScientist says:

          Each test is a continuous band, so it’s like a million tests.

      3. Isolating and not spreading other coronaviruses is not a big cost.

        1. Charles H. says:

          Isolating’s not a big cost for me. I’m retired. For many people isolating is such a high cost that it’s effectively impossible. For others it’s a high, but bearable cost. Not everyone is in the same situation. If you live from paycheck to paycheck, then isolating probably means losing a place to live. (If not now, then in a month or two.) Also losing your job. And any health insurance you may have.

          1. My point was that cross-sensitivity to other coronaviruses is not such a big problem, because if you have *any* coronavirus infection than the most appropriate reaction (barring some extreme special cases) is to isolate yourself.

            There might be special cases when isolation is truly costly – but at least for the 99% cases isolation when infected with any virus is the right thing to do. So that particular kind of error, the cross-sensitivity, is not very costly.

          2. Thomas says:

            Zbigniew, one kind of common cold is a coronavirus too.
            No reason to self-isolate for that. If anything it might even bring some cross-immunity. (Or perhpas some ADE, who knows).

  8. Giannis says:

    The solution to the takeover low specificity is to include some strips that detect another viral antigen eg Spike protein. That way only if you test positive for both you consider the result as true positive.

  9. First of all, thank you Derek for such a clear description of this type of test, and presentation of false positives and negatives in a way that should be very understandable to non-specialists.

    I agree that this type of test is better used in a symptomatic, point-of-care setting where it will be a powerful tool in identifying people with an *active* viral infection, with all the good that will flow from that.

    I am a little concerned that it will lead to problems when people see discrepancy between this test being negative but a molecular test being positive, but that essentially concurs with your comment on such tests having their place in the picture.

    One of the issues with turn around time in the molecular testing has been lack of reagents generally – suddenly everyone wants Qiagen extraction kits. This type of supply-line issue is never reported by the mainstream media. I know of one smaller diagnostics company that is resulting in 15 hours from receipt, but they can only handle five to six thousand tests a day.

    This new test will, as you say, be extremely useful, *but* will require a change in mindset away from testing the asymptomatic with a screening philosophy, to testing the symptomatic with a diagnostic philosophy. This in turn will require that national and state health officials will need to make the same change. Undoubtedly therefore, this will inevitably become part of the politicization of this virus leading up to the election.

    One can only hope that (what I consider to be) common sense will prevail and that the availability of this test will encourage the lifting of the still-draconian restrictions in place in many states. It will be interesting to see also how this modifies people’s general acceptance of flu in the workplace or school. Will little Jane or Johnny be allowed to sit coughing and sniffing when their 15 min CoV-2 test comes back negative?


    1. David G Whiteis says:

      Are we assuming mandatory testing for students and workers? I wouldn’t mind that, but I don’t think it would be accepted by a significant chunk of the American public.

      How effective would the tests be in stemming the pandemic, if asymptomatic carriers (many of whom are children like little Jane and Johnny) were still untested, i.e., still out there spreading contagion?

      Also, would people have to be tested every day, every few days, etc., at least until (hopefully) the pandemic subsided? That would mean a steady stream of upwards of 320 million kits, consistently delivered and on hand, over a period of time. Is this feasible?

    2. Calvin says:

      Excellent points raised. I do like the comment around screening versus diagnosing. It is often forgotten that the two are not the same, particularly when it comes to infectious disease. I think it’s unfortunate that diagnostics are seen as a diagnosis tool, when really they only do that sometimes. Personally, I’d like to see this test used as a screening tool (with the caveat that it can only be used in a place with reasonably high disease prevalence). It then is used as a filtering tool for a more robust diagnosis test, eg PCR. Mass screening would actually allow you to identify the infected population to isolate (still on the fence as whether the specificity of this tests is good enough). And as long as you keep testing, you’ve got a chance to really narrow it all down. As ever, prior to a vaccine/treatment/prevention, the only way out of this is test, trace, isolate…….

    3. wubbles says:

      My understanding is that proteinase K can be used instead of other RNA extraction methods. And of course there is the old phenol-chloroform method. The supply problems are really problems of flexibility with method and the astonishing lack of a push by government to make the reagents. The proteins grow in yeast, and we have companies that are very good at growing lots of yeast every day.

  10. exGlaxoid says:

    If they used the quick Abbott test as a prerequisite for the PCR test, that might be enough to make it worthwhile also. For colleges or other places trying to test people a few times a week, that would eventually catch most positives, and if you retest the positives, you can catch most false positives as well, for way less cost, and the lower number of PCR tests would allow faster turn around times for those as well.

    The problem I see with the PCR test is that the CDC originally seemed to want everyone to use the same materials, reagents, etc, which sounds good until you realize that they were in very limited supply. They would have been better to try to get everyone to use different tools, so as to not create such a huge bottleneck. Just like having 100 vaccines in development with many different methods may allow us to find the better ones faster than just having everyone use the same sequence target and method (DNA, RNA, viral vectors, peptide fragments, attenuated, etc).

  11. Bell4 says:

    Is the sample to be taken from the front or back of the nasal cavity? The latter is more accurate but often uncomfortable and is best done by a professional.

  12. Nicholas Weininger says:

    I think you’re underplaying the usefulness of this as a screener test for high-risk working populations where lots of people have to be in the same indoor space for a long time– say people who work in a meatpacking plant, or nursing home, or school. Suppose you screen such a population regularly with these tests and if someone tests positive, you test them again, and if they get a second positive, they go home and isolate till they can get a PCR test. And suppose the baseline prevalence of COVID in this population is close to zero (e.g. suppose you’re testing *only* asymptomatic people because anyone symptomatic is staying home anyway).

    What happens? If false positives are random uncorrelated noise, the second test will take care of all but one in 10K or so of them. If at the other extreme they are perfectly correlated, i.e. the same unlucky 1-2% of people falsely test positive again and again: well, you lose those people from the workplace. But having the other 98% be able to work together without fearing an outbreak, because you’re going to catch that one asymptomatic superspreader 98% of the time, may well be a large enough benefit to justify having the 2% stay home! And in reality false positives are probably only partially correlated, so you lose less than 2% in practice.

    1. Andrew Richards says:

      I was just going to make this comment but you beat me to the punch. Test everyone with the cheap, quick test, isolate any positive until PCR comes back, return to work if the PCR is negative. This maximizes finite testing resources and the public health benefit.

      1. Some idiot says:

        What about the false negatives? They are the ones that slip through the net…

        1. Barry says:

          The false negatives are a real concern. If one (perhaps a staffer at a nursing home, or a crewman on a cruise-ship) slips through (and one will slip through) hundreds of the most vulnerable may be infected .

        2. Cuibono says:

          the false negatives are ALREADY a concern. The NP Pcr has them too and more than we know.
          I had one patient this week test negative twice over three days (while symptomatic) and the THIRD NP PCr came back positive

          1. eub says:

            False negatives are absolutely real, but the individual transmission is not the point (apologies to infected folks), the point is whether we can find a sustainable way to hold Re < 1.0.

          2. ILikeMike says:

            False test reports are a concern. But, equal to if not more of a concern is someone not being tested at all – whether PCR or antigen. In my view, as long as the tests are “point of care” as opposed to a surveillance/rapid at-home test, many people without symptoms won’t get tested, and we won’t get out of this mess. I believe it is established that asymptomatic/presymptomatic spread is partly driving this pandemic. Taking frequent at-home surveillance test in conjunction with a confirmatory PCR test can help resolve the problem. In terms of manufacturing to scale, if necessary, utilize the DPA.

          3. Harvey 6'3.5" says:

            One question for all “false” negatives is whether they are false because it is so early in infection that even the exquisite sensitivity of RT-PCR cannot find a virus nucleic acid to amplify or they are false because the sample was collected properly or the test didn’t operate properly. In the real world, it is a combination of causes, which is why repeated testing over a few days (as is being done at UIUC (Illinois)), helps.

  13. Derek Freyberg says:

    On your Addendum, Derek, Governor Newsom of California stated yesterday that California will ignore the new CDC guidelines and test as it sees appropriate, and my local paper (San Francisco Chronicle) has reported this morning that California has just signed a deal with Perkin-Elmer for a facility that will run 150,000 PCR tests/day with a 1-2 day turnaround, more than doubling the state’s present testing capacity. This will not be at full capacity until March, which is rather late. Right now the state has the capacity of around 100,000 tests/day, but the turnaround is an average of 7 days, depending on who’s doing the testing.

    1. Ken says:

      Several universities have set up their own test labs and procedures for their staff and students. The University of Illinois Urbana-Champaign, for example, is testing everyone twice a week, so can process over 15,000 tests a day (one source: They’re also not bothering with the change in CDC guidance
      I think I read that some of these universities are using non-standard (possibly even non-approved?) tests, but I have no details.

      1. Edward T says:

        My daughter just started at U of Illinois at Urbana. She moved in on Aug 20th and already had 3 negative test results. There have been some rollout issues with the test, but in each of the 3 instances she received her results in just under 24 hours, which is longer than the 4 to 8-hour target. They also have about a 5% rate of inconclusive, which are treated like positives in respect of building access. This is a big issue the university is trying to address right now. I was not surprised when my daughter told me about one person, she observed taking a sip from a water bottle while in line, a big no-no as far as the test goes. The FDA gave the test emergency approval. The saliva-based Polymerase Chain Reaction (PCR) test is said to be a new “gold” standard in accuracy, six to 8 times more accurate than the Yale version. It also returns the load level of the virus. Specific details about the test are hard to come by but on a website of another university that uses it, it was reported the test has less than a 0.3% false positive rate. On Monday of the last week they hit a one-day test total of 17,656 tests with conclusive results. On 3 other days of the same week they performed over 15K tests with conclusive results each of those days. The dashboard has been reporting between 50 and 89 new cases each day, but mostly in the 60 range. However, I cannot determine if that includes false positives. With 15K tests, and a 0.3% false positive rate, the number of false positive cases would be around 45, which would account for most of the new cases. It is possible that the university would rather not publish this information, as to not give the students who test positive, an excuse not to follow isolation rules if the actual probability of them actually having the virus is less than 50% or even less than 30%.
        I could see that last point become extremely critical, with mass paper tests in an application on a scale of hundreds of millions and a much, much higher false positive rate. In a scenario where a positive test result translates into a 10% probability or even 5% of having the virus, the trust in the test would quickly erode. Yes, subsequent tests could improve results, but would that improve the public trust, is questionable. This in combination with the number of missed positives could create a nightmare scenario where test results stop being trusted all together. I could see this as a sticking point in any approval of mass paper tests with lower accuracy.

  14. Bill says:

    Seems to me there’s a reasonable alternative explanation supporting restricting testing recommendations. Given the current situation where results lag by up to 10 days, as stated by Derek, the testing might as well not have occurred. Fewer, more timely tests are a better situation than massive testing without timely results.

    When results catch up with tests, go back to higher volume recommendations.

    Not everything has to be political. Only if you have a bias towards finding such links.

    1. Marko says:

      If the CDC had stated that as the reason , you could make that case.

      Only if you have a bias would you dismiss the CDC’s failure here.

    2. The rapid tests are ‘rapid’ – there are no delays.

  15. John Hasler says:

    Lateral flow, not laminar flow.

  16. Nimish says:

    The metric to use here is precision/positive predictive value. At low, sub 1% prevalence, this test stops being all that useful. For every test there’s a “prevalence threshold” below which the test rapidly becomes imprecise.

    The problem here is that isolation of positives implies serious economic and mental health tolls. It’s not a free action, especially if it results in unnecessary shutdowns.

    Excessive testing has long been known to cause poorer outcomes — look at prostate screenings for a disease with far more severe consequences. At what point did we throw the hard-fought knowledge of the dangers of overscreening out the window?

    1. charlue says:

      when there is 25 billion dollars at stake — which is what the combined loss of every other company that makes tests on today’s market.

    2. The problem here is that we now act as if we all tested positive (with some low specificity test) and there are serious economic and mental health tolls.

  17. Philip says:

    Warning, political comment!

    What COVID-19 killed that I will miss the most:
    John Prine, a great singer and song writer.
    Trust in the CDC.
    Trust in the FDA.

    I will miss all three a great deal.

    1. eub says:

      Damn I miss John Prine.

      “Your Flag Decal Won’t Get Get You Into Heaven Anymore” linked from name as relevant to interests

    2. Mammalian scale-up person says:

      Agreed about John Prine. Never did get to see him in concert, and I will regret it the rest of my life.

      As for the other two, I am thoroughly baffled why they accept and follow what their political overlords tell them to do when they know it ain’t right. What kind of ding-dong can’t manage a fool of a manager – like we haven’t all had some idiot telling us to defy the laws of physics? You say “hmmm let’s form a committee to discuss that” and “let’s be sure to involve all the relevant stakeholders to ensure we can make the right decision, and we have to meet every Friday lunchtime with Panera ordered in until we achieve consensus” and “I’ll take it under advisement and get back to you” and slow-roll it until they’ve forgotten about it or the next round of layoffs (in this case new administration) gives you a new manager. If they look like they’re thinking about firing you, well, that’s OK, there’s other jobs in the world and plenty of companies would be delighted to have ex-FDA or ex-CDC working for them.

      1. eub says:

        Literally LOL. I work in a different industry but not much has changed in manager management.

  18. Daniel Barkalow says:

    It would be really helpful to know about the sources of error on these tests. For the false positives, is it that 1.5% of people have something else in their samples that binds to the antibodies? Is it that 1.5% of the tests have antibodies that come loose without binding to anything? For the false negatives, I presume the major sources are (a) the sample didn’t actually collect anything from the patient; (b) the sample didn’t get into the test; (c) the patient isn’t in a phase of the disease where they shed much Np.

    Of course, the source of the errors matters a lot to how useful the test is in certain applications. Does it make sense to retest people, or will you just get the same result? Are there people with endogenous proteins that the antibody in the test will bind to, such that they’ll test positive forever but aren’t actually infected or infectious? Does this misclassify a strain of the common cold, such that you can catch false positives from a patient who gets a false positive?

    1. Charles H. says:

      You left out manufacturing variation. I often think that my glucose test strips have a high rate of malfunction due to manufacturing defects. I could be wrong, of course, since I’ve no way to verify. The couple of times I did test twice quickly, once the reads were very different, and once they were about the same. And nobody will tell me about how much inherent variation to expect. (My blood sugar is reported as ranging between 90 and 150, so it’s pretty safe, but the variations aren’t predictable based on either what I eat or how I exercise. I’ve been trying to make sense of them for about 3-4 years now without success.)

  19. Philip says:

    I do believe that antigen tests have a place in testing asymptomatic people. For example, schools. If you test frequently enough, you will remove almost all of the infectious (I do not mean infected) people from the campus. Those testing positive should be given a RT-PCR test to confirm. That way false positives would miss just one day of school. Not a bad price to pay to keep so many infectious people off campus. False negatives, hopefully are those early in their infections that are not yet shedding virus. False negatives for those late in their infection may not be a problem, because the immune system’s response to the virus is what is causing the disease at that time, not the virus itself.

    Even with daily antigen testing, schools would need to keep mandatory physical distancing and mask requirements in place.

    Don’t let the perfect be the enemy of the good.

    Yes, I drank the Dr. Michael Mina Kool-Aid and found it good.

  20. Daren Austin says:

    I look forward to the self-testing sensitivity test. Or even the teacher-administered swab test. Ability to self-swab will reduce sensitivity from the lab testing (I have a fabulous gag response when I did mine). It is, however, a welcome advance, but I suspect that saliva tests will be the eventual winner here. Everyone can spit.

  21. Barry says:

    I learn now that the quoted numbers for sensitivity and specificity were only when testing a population all of whom were symptomatic. But we know that something like 40% of all infections are asymptomatic. I expect we will see the sensitivity and/or specificity numbers revised when a representative population is tests. My wager is, both will go down a bit.

  22. yuriwho says:

    If you test 2x each time you use the new rapid test, your sensitivity and specificity would increase dramatically and it would be suitable for use as a CLIA-waived home test for populations like New Zealand with low transmission. It would still be cheaper than a RT-PCR test

    1. Barry says:

      testing twice would “increase dramatically” your sensitivity and specificity only if the errors were random. Especially for specificity, that’s unlikely to be the case. There are a lot of other coronaviruses out there to which a subject may have Abs, or even viral fragments

      1. Mic says:

        And the samples independent.

  23. Michael says:

    Another angle: I would think that clinical trials of infected subjects can benefit from rapid tests like this. For example, monoclonal antibody trials have been slowed, if not completely frustrated, by PCR turnaround time — rapid tests might really help assessments of real-time increases/decreases in viral shedding for trial participants at various points of time.

    1. Philip says:

      Any antiviral that I would trust needs to start treatment early in the infection. That is really hard if you are waiting for people to show up at a clinic because they are sick. For effective antiviral studies we need rapid and frequent testing.

      Antigen tests would be a great way to find subjects in congregate living. Nursing homes, the military and prisons would be good places to start.

    2. Marko says:

      “…rapid tests might really help assessments of real-time increases/decreases in viral shedding for trial participants at various points of time.”

      The problem is that the rapid tests don’t give you a sense of viral load dynamics like a PCR test can. You only get a (+) or (-) result.

      I suppose you could do serial dilutions to try to wring some viral load information out of the test, but by that time the PCR probably looks pretty convenient.

      1. Philip says:

        RT-PCR can give you the CT values, which relate to RNA load, which depending on where you are in the course of the infection can relate to viral load. The problem is that most clinical RT-PCR reports do not include the CT value.

        Quick CT value info: For every increase of 1 in PCR CT values, means a decrease by half of the RNA in the sample. So 27 to 30 means there is 1/8 the amount of RNA.

      2. JasonP says:

        Phillip above is correct, we need to look at VIRAL LOAD, i.e. CT values. The beauty of the low specificity tests is that they don’t trigger a “positive” response until the viral load is high enough that transmission is possible. So fast results (15-min) and frequent testing (daily? every 2-days?) should catch the infectious people and allow for effective isolation.

        If you wanted to do PCR testing “correctly” wouldn’t that imply identifying the CT values and rapid (day or two) follow up testing to show if the Viral load is staying the same (relatively low = non infectious) or remaining high (infectious)?

        To those who think false positives and negatives will cause confusion, I think that short changes the ability of the masses to learn or understand. We are going through a lot and adjusting and changing throughout this pandemic. Might be a challenge to put the explanations in terms that meet people where they live and cut out the scientific jargon, but certainly achievable as Derek’s blog demonstrates.

  24. cuibono says:

    “and is far more appropriate for use in a broader, largely asymptomatic population. ”

    Not so fast Derek. If that asymptomatic population has a high prevalance of infection, this is not true.
    Turn around time is INCRDIBLY important. In many places that is up to a week or more, makingthe value of your gold standard test close to ZERO.

    Given what we know about the high proportion of cases being either asymptomatic or presymtomatic, I think this test will be a game changer.

  25. JasonP says:

    ***Preamble: I am writing my thoughts to question thinking. But I suspect the fact that my prose isn’t erudite, I don’t have a bunch of links to journal articles and I refer to a YouTube video (THE HORROR) many will skip to the next comment. But for those who make it to the end I’m interested where this new idea fails, if indeed it does? ***

    First I have to say I get the 95% confidence level (CL) stuff. But again, we are in a war of sorts and we can play by “the book”, or we can be like the Marines and improvise & adapt. Oh sure, we can’t be accused of being wrong if we have 95% CL data, but is that the correct approach when other things can be done?

    My view point was swayed when I saw this video which features a new idea on testing by Dr. Michael Mina of Harvard. (17 min by an MD, with another with a PhD Mina added in. Oh an there are journal articles referenced and even a critical analysis of some of that data.)
    I’ll add this one too, by Mina:

    ###The concept here is SCREENING for infection when the viral loads are high enough to make a person infectious.### What good does it do to isolate a person who is past the point where they can spread the virus, but still have detectable levels of viral particles in their system? None. Second what is the point of an expensive test that takes too long to get results, strains the testing infrastructure? The PCR test is so sensitive that it picks up these people and reports them as “positive.” Apparently, the key is that the time in which one is “infectious” is limited, even though the viral particles that PCR detects remain. So perhaps a test that has 50% specificity, or at test that needs **a viral load that is high enough for the person to be infectious*** before it reports “positive” is what we need to be focused on? If nothing else “cases” should be delineated by VIRAL LOAD and not just viral particle presence.

    Apparently the technology is out there to make a paper strip kind of test, that costs $1-2 and has an accuracy of 50%. Whoa! So far off of the 95% CL that most will ignore this idea at this point. Sigh. And apparently the scientific literature has beaten down this idea. Why? The data in that paper (mentioned at the end of the video) in fact shows that these cheaper tests do in fact report positives that are in the infectious range! That seems to be what we need – faster, cheaper, more frequent testing! So what if it doesn’t hit the 95% CL, it gets the mission accomplished: identifying those who can spread the disease.

    What if we gave school children these tests every morning? You test positive and you stay home from school until you have several negative tests or 14-days? Could this be expanded elsewhere – You want to go to a football game? Get tested on the spot. Positive, go through the door over there, negative, enjoy the game!

    But oh no! Are we so mired in the 95% CL stuff that we can’t see the forest for the trees? Spreading of the infection.

    On another note – Contact tracing isn’t working where it is being employed – too many people flat out refusing to cooperate or refusing to share names. The data I have seen show that of the cases eligible for contact (interview) 8% had no valid contact info, 8% could not be reached and 31% didn’t complete the interview. So a failure rate of 47%!!!! Then of the remain 53% that were indeed interviewed, 21% refused to ‘name names!’ So in essence 58% failure rate on the contact tracing. Of the people’s names collected by the contact tracing, 41% had no valid contact info! Of the remaining 59%, 41% could not be reached. In today’s society, I doubt it will work for various social and political reasons. Seems like ‘the data’ is demonstrative of this! But keep insisting that this proven method is the only way to solve this problem or do we do something else? Why are we wasting resources on this?

    The reason why I bothered to post this is that I have great respect for the author, but believe this is an opportunity for scientists to step out of convention and get behind an idea that can have results. Certainly this post opens the door a crack, but again, I believe the focus needs to shift to transmission/viral load and not just partial presence or antibodies.

    But again, the discussion falls back on hand wringing of, “Oh the test isn’t perfect and negative people are going to be told they are positive and positive people are going to be told they are negative.” Sigh, this is the wrong focus! Instead of positive or negative shouldn’t we be looking at VIRAL LOAD and TRANSMISSIBILITY!??? Wouldn’t more widespread and frequent testing even with an imperfect test be better in controlling the virus?

    We need to change our thinking and approach and since CDC is a science bound org, we need outside scientists to be working to help move the needle. There are several credentialed scientist pounding the table on this, why not more from this realm?

    EOM/ As you were. 🙂

    1. Philip says:

      JasonP, I am going to add one more Dr. Mina YouTube video to your list. A long one, TWiV 640:

      I have also had the Dr. Mina Kool-Aid. It is very tasty.

      With Abbott coming out with a $5 lateral flow antigen test that has an FDA EUA, rapid, frequent testing is a step closer. Not close enough, yet. The Abbott test does not need equipment, but is still a point of care test. We need home/point of entry tests. Which according to E25Bio, are coming.

      Here is a table from Dr. Daniel Griffin (I hope you can make out the data, sorry for the formating):
      RNA copy number Ct value/Testing Modality/Relevance
      1000 36/NAAT Methods/Outside Infectiousness Period
      4,000 34/Abott ID Now/ Outside Infectiousness Period
      50,000 30/Rapid Antigen Tests/ Outside Infectiousness Period
      3,000,000 24/All modalities/Infectious

      I see no way out of our SARS-CoV-2 predicament without rapid and frequent testing to see who is infectious. Well, we can wait until there is a safe, effective and widely available vaccine, that people trust, but I do not expect that for about a year.

      BTW, Derek and the In The Pipeline blog post about the Russian vaccine got a shout out on a recent TWiV. Sorry, I don’t remember the number.

      One more thing about rapid, frequent antigen testing, testing negative does not mean you can skip physical distancing, masks or hand washing.

      1. Not Jim Jones says:

        I also thought of Kool-Aid when reading Dr. Mina, so that makes three of us. Maybe it’s scientists’ learned skepticism of passionate arguments that don’t build on the same shoulders we’re standing on. But maybe he has a point that this time really is different. The type of screening tools that can best control the rapid expansion of a novel epidemic are different than those used for well understood diseases. Perhaps only since HIV have we begun to learn general principles for “rapid response” public health measures toward novel infectious agents, and perhaps we should start judging our options first through that lens, and only later through the lens of better known diseases?

  26. Marko says:

    Just out : “Test Sensitivity for Infection versus Infectiousness of SARS-CoV-2”

  27. Jonathan says:

    Isn’t this test drastically more accurate than RT-PCR tests are?

    I am not talking about perfectly executed RT-PCR tests, but real world accuracy, which have been shown to miss infections up to 30% of the time.

    1. Barry says:

      Oh, in the real world, both sensitivity and specificity for this one will be less, too.

  28. David Eugene Young says:

    There needs to be a high school course on Bayesian Analysis, for all to take, or as part of a math class that explains it along with stage migration (the Will Rogers effect) and why Marilyn Savant was correct in her explanation of the Monty Hall game.

    1. eub says:

      Heh. From the reference to the Monty Hall episode I bet I can date you within +/- 5 years.

  29. LeeH says:

    I agree with David Eugene Young. But thankfully Bayes predicts that the problem will be with false positives, which is an inconvenience, not a public health hazard.

    Say the true infection rate for people walking in to get a test is 1%. With a sensitivity of 97%, if 10000 people walk in, 100 are infected, and 97 of them are correctly identified, and 3 are false negatives (mixed in with the 9751 true negatives). So given a negative test, you’re not doing badly. Given a positive test, though, of the 9900 that are not infected, 148.5 (9900 * 0.015) are falsely labeled as infected. So 97 are true positives, and 148 are false positives. That’s only about a 40% accuracy given a positive test. So there are going to be twice as many people quarantining that don’t need to.

    Just another example of the problem of making predictions on highly imbalanced data.

    1. LeeH says:

      Sorry. For every five people (in this scenario), 3 are quarantining without having to.

      Ironically, this actually improves as the infection rate goes up.

    2. Jason Gratt says:

      The specificity issue can be fixed by having each 30 pack of paper strips contain 5 confirmatory paper strips that detect a different antigen. Out of 10000 people you now have only 2 false positives (148 * .015). Those two people can get a PCR or just stay home.

      1. LeeH says:

        Hi Jason!

        Assuming the errors are random and not systematic, that true. The question is whether the powers that be will want to pay for duplicate tests. I’m skeptical.

  30. Thomas says:

    When false positives are a problem, run an RT-PCR on the samples that come out positive in this test.
    That will reduce the number of RT-PCR tests needed by more than a factor of 10.
    The result well be that most tested people will know that they are without infection in 15 minutes.
    Some will be less lucky and will have to wait longer.
    Sounds like a big improvement over ‘everyone waiting 24 to 48 hours’.

  31. Edward Olejniczak says:

    If everyone was tested every week and 97% of actives identified and 80% of these isolate, wouldn’t this be as good at stopping infections as HERD IMMUNITY or a Vaccine?

  32. A Berman says:

    Given the lab backups and delays in results, it seems reasonable for the CDC to restrict testing to symptomatic people to reduce turnaround time. Fewer tests, but more actionable since they’d get back more quickly.

    1. Bill says:

      That seems obvious to me. You do no one a service by piling more people into a life-boat resulting in it’s sinking. If you want to save more people, FIRST get more life-boats.

      But there’s an election coming up.

  33. The Anti-Fake News Patrol says:

    Now this junk-science snake-oil peddler Derek Rowe is having delusions of “political interference from above” and dishing out half-crocked medical advice to ignore the CDC recommendations. Maybe the interference is from demons in his head. What kind of a Quack would pretend to be a Physician and advise ignoring the CDC?

    Further, any idiot can look at the R naught, as well as the infection and hospitalization rates, and see all are clearly diminishing. It is obvious the China Virus is dwindling, dissipating and disappearing from large swaths of America. The Herd Immunity Threshold has been reached in much of the country. Why wouldn’t administration officials begin referring to the pandemic in the past tense? Shouldn’t our leadership try to encourage the Nation, boost morale, and try to be optimistic?

    It will be great when His Eminence is re-elected and Fraudsters like Derek Rowe are investigated and put out of business.

    1. Another Kevin says:

      Unfortunately, in our time, broad–brush satire doesn’t work. Poe’s Law is too string.

      1. EJ says:

        Until I got to “His Eminance,” I wasn’t really sure.

        The insult frequency is spot on.

  34. Another Kevin says:

    What are sensitivity and specificity like for RT-PCR in the clinic? Having hung around with the lab rats, I’m moderately amazed that it works there at all. (I know that for RNA analysis for surgical specimens, we essentially needed to get the sample in liquid nitrogen on the way to the path lab, or the exonucleases would have eaten all the RNA by the time it was extracted. But that was a tidy few years back, by now.) I know that the viral capsid _should_ protect the RNA, but how good is the RNA recovery when you lyse the viral particles?

    My guess is that a protein test that can be done in-office might turn out to perform better in practice than an RNA test where the samples can degrade in transit.

  35. Wilhelm Cody says:

    Why was SARS-CoV-1 brought under control while SARS-CoV-2 has not been?

    SARS-CoV-1 patients had symptoms at the same time they were infected. A fast, easy, inexpensive screening could be applied and patients isolated: fever detected by brow or ear temperature. The test was applied to everyone entering an environment conducive to transmission. If people have fever they could be forced to quarantine and get medical care while a slower, more expensive, but more specific test could be applied for confirmation.

    SARS-CoV-2 patients often have symptoms much later than when they become infectious to others. The tests we apply are slow, laborious, and expensive. Not everyone can be tested every time they enter an environment conducive to transmission. They often will not quarantine while waiting for the results.

    This test is fast, easy, and acceptably inexpensive given the economic damage being incurred. If positive, then a slower, more laborious, and expensive nucleic acid test can be applied while the person is forced into quarantine. With reduced demand on nucleic acid testing, the person will know results in a day or so and can be freed or kept in quarantine.

    The result will be forcing R0 well below 1 while vaccines are developed. That is why this test is needed now. We need 150 million such tests but 10 or 100 times that number over the next 6 months or so, but even the 150 million will help enormously now compared to the 75 million tests run over the last 5 months.

    1. Bill says:

      Talk of hundreds of millions of tests boggles the mind. But now you’ve reached into the billions. Do people believe we have the infrastructure and logistics to address such a task?

      If it requires the building of new industries, how long do you figure that will take?

      When thing get huge, doing the math is a bitch. Ignoring the math is worse.

  36. James Millar says:

    If someone gets a false positive or negative and then immediately takes another test, will they have the same odds of an erroneous result? Or it is something specific to them?

    Ie, do two consecutive tests have a 0.03 false negative rate or a 0.03^2 false negative rate, or somewhere in between?

  37. Jakub Narębski says:

    > The two types of tests that we see in use now are RT-PCR, the nasal-swab test that detects viral RNA, and various antibody tests, that tell you if you have raised an immune response due to past exposure to the virus.

    I have recently watched good video on The Thought Emporium channel about various tests for SARS-CoV-2 / COVID-19: “How Does COVID-19 Testing Actually Work?”

    There beside RT-PCR / qPCR and antibody tests he mentions LAMP / RT-LAMP test (Reverse Transcription Loop-mediated Isothermal Amplification). Do you count it as RT-PCR test, or is this type of test not used as much?

    BTW. I really liked the mention of PCR being Pipette-Cry-Repeat 😉

  38. Barry says:

    Testing only symptomatic cases (Abbott’s sensitivity and specificity numbers are only from symptomatic cases) can’t stop the spread
    “In this study, the authors estimate that 85 infected children (93%) would have been missed using a testing strategy focused on testing of symptomatic patients alone,”

  39. Kaleberg says:

    Now we need to solve the problem of so few people being able to afford sick days and potential job and home loss so they CAN isolate.

  40. TallDave says:

    good news

    better testing will become even more helpful as we find more effective treatments to direct at COVID patients… a terrible shame we don’t have Phase 2 synthetic antibody results by now, but at least AstraZeneca has joined the field

    1. TallDave says:

      btw note the CDC recommendations specifically defer to local authorities… everyone was already free to ignore them

  41. donorcure says:

    Based on experience with antigen-based RDTs for other respiratory diseases such as influenza, in which affected patients have comparable concentrations of influenza virus in respiratory samples as seen in COVID-19, the sensitivity of these tests might be expected to vary from 34% to 80%.

  42. RonH says:

    Am I missing something on the specificity/sensitivity here, compared to RT-PCR currently in use? The sensitivity I keep coming across for RT-PCR is 2 – 29% (for instance here and in reference 6 therein: Maybe the RT-PCR is more accurate in principle but in practice, due to errors in swabbing and testing, it’s not? This sounds much simpler and more accurate.

  43. Michael says:

    Scott Gotlieb, MD has opined that at-home Covid tests may be available sometime this year, and perhaps as early as the fall.

  44. john says:

    Maybe this has been discussed, but the number of cycles to detect a positive PCR seems to be a big deal to me, as it can change the number of positive tests by 70% if the number of cycles to detect positivity is reduced from 35-40 to 30 or so.

    I’m surprised that people are just beginning to realize this now.

  45. anon says:

    john, that is a very startling link that you provided! 90% of the positive tests for COVID are positive merely due to a non-clinically valid definition of positivity?

    Michael, at home, cheap, and convenient COVID tests remains the game changer we are waiting for. It will be amazing to see these home tests hit the shelves and the pandemic
    immediately stop. Removing ~97% of infectees on each testing round would be a profoundly
    effective method to halt COVID.

    1. DataWatcher says:

      How close are we to seeing the paper lateral flow assays becoming available?

      I am not as confident as some people seem to be that asymptomatic or mildly symptomatic people will self-test at home, and /or that those who test positive at home will follow up and report / quarantine responsibly (even countries like New Zealand, where the “social contract” remains a lot stronger than it is in the U.S., have been forced to take people out of their homes and ensconce them in supervised quarantine centers to ensure compliance). I agree with @Bill, above, that distribution will be an issue, as well. Hundreds of millions of kits would need to be produced in the USA alone, they’d have to reach people in remote areas as well as major cities (many of which have sizeable low-income communities with little access to supermarkets or even a drug store), and even at a few dollars per unit, they wouldn’t necessarily be affordable to many of the most high-risk people.

      Nonetheless, they will do a lot of good when they finally appear. At the very least, they could help reduce the R0 to a level where even a reasonably effective vaccine might be sufficient to play a major role in gettting us back to “normalcy” (meet me at the pub — I’ll buy the first round!)

    2. Dark Day says:

      ” . . . 90% of the positive tests for COVID are positive merely due to a non-clinically valid definition of positivity?”

      That’s a mixed blessing. It could be good news, in that the number of actual cases might be significantly lower than previously thought; it might be bad news in terms of moratlity rates, because it reduces the denominator.

    3. DataWatcher says:

      Here’s another example of why so many people get confused by apparently conflicting news reports. At the same time we’re hearing this, we’re also hearing estimates from the CDC that the true coronavirus case rate is actually likely to be at least 10 times HIGHER than reported:

  46. anon says:

    Dark Day, I continue not to understand why testing has been so much neglected. Why isn’t the primary focus on mass testing? In my community, testing of 2 per 1,000 is thought sufficient. Yet, every once and awhile a new wave appears to be brewing. Having the capability for population scale daily testing would be such a relief. At least then we would know that we would have the ability to stop COVID at will. As it is now, we anxiously wait and hope things will not get worse again. Yet, it seems almost inevitable that a second wave will arrive.

    Vaccines can take months and months to develop and approve. Indeed realistically this process usually would take years and years, though everything is now on triple time (or more). With the testing approach, it is only a question of how long would it take to make a few billion of them; how could these tests not be approved if they could demonstrate reasonable efficacy. 97%? That is beyond the minimal expectations. No push back if safety or efficacy concerns as might arise with a vaccine. Testing seems to be a near slam dunk strategy. Considering this is election season, it is surprising that this has not been enthusiastically embraced. Stopping this pandemic cold would almost be magical; it would confer a sense of high competency on whoever could associate themselves with such a powerful solution to COVID.

    1. Dark Day says:

      Well, inexpensive paper-based lateral flow assays that can be self-administered at home are apparently on the horizon. I don’t think they’ll necessarily be able to “stop the pandemic cold” (they’ll require significant self-discipline, and compliance, which are things American’s aren’t very good at), but they should certainly go at least a way toward the dirction you’re suggesting.

  47. anon says:

    Dark Day, this is such an exciting moment. I am not aware of any other instance in which population scale testing has been attempted before to stop a pandemic. Our response to date to COVID is not that much different from that of the 1918 influenza epidemic. Why not see how effective mass testing might be? Why not innovate?

    This is the perfect instance in which real world science could produce a decisive conclusion. Removing ~95% of COVID infectees daily should reasonably have a rapid and dramatic effect. A negative COVID test could be made a requirement to enter mass transit or buildings. Science could be shown to be an overwhelming force against the virus. Once we get this down, perhaps we could apply a similar strategy to the annual flu epidemics.

    1. Dark Day says:

      Let’s hope so! Let’s hope my rather pessimistic view of human nature (at least as it applies to mainsream Americans) in terms of [1] trusing science, and [2] doing what’s necessary on an individual basis to help science work proves false. If this can help us get out of this nightmare, I’ll meet you at the (now re-opened) bistro of your choice, and the first round is on me.

Leave a Reply

Your email address will not be published. Required fields are marked *

Time limit is exhausted. Please reload CAPTCHA.

This site uses Akismet to reduce spam. Learn how your comment data is processed.