corona test reliability



I have done  rapid Covid test here in Italy. 

These Rapid tests are used in Italy according to :

  1. l’SD Biosensor Standard Q COVID-19 Ag;
  2. il Coris COVID-19 Ag Respi-Strip;
  3. e l’Huaketai COVID-19 Antigen Detection Kit

These are mainly antigenic tests. I never got information which of these were used here with me. writes about unreliabilities in the results. Like my result could be a false negative result. I read about that before.

I got terribly confused reading news around the world. Like that writes “Nepal’s decision to use rapid diagnostic kits for mass testing may actually be putting more people at risk” writes  Nepal should have gone directly for PCR like Korea and other countries instead of RDTs.

I decided to figure out what rapid test (RDT) really are and why there and risc for false results and when..



Test Types

  • PCR (Polymerase Chain Reaction, a laboratory test that shows if you have the disease). (ed. Detects corona RNA) 
    are highly sensitive and excellent for diagnosing acute infection,
    they are expensive and require the use of special reagents, equipment and specially trained clinical laboratory scientists, all of which have been stretched thin during the pandemic so far. ( )

  • RDT (Rapid Diagnostic Test)
    There are two kinds of RDT:

    1. Antigen test or  RAD (Rapid antigen detection) (ed. Detects corona proteins recognized by antibodies) Are cheaper to produce than PCR tests and can be performed very quickly, making them an option worth considering as a screening tool. ( )
    2. serological test, a rapid blood test that shows if you have IgA, IgM, IGG antibodies). ( )
      Is useful for determining if someone had a previous SARS-CoV-2 infection, but since it takes 1-2 weeks for antibodies to develop, it is not recommended for the diagnosis of acute disease. ( )
  • ELISA (Enzyme Linked Immuno Sorbent Assay, a laboratory analysis of blood showing the amount of antibodies) 

A doctor of Spalanzani in Rome confirm that these 3d generation rapid tests are now secure.

PCR and Antigen Testing in COVID-19

(American Society of Microbiology )writes:

Antigen tests

 While molecular tests can be performed in real-time, supply chain and throughput challenges have made this unrealistic during the COVID-19 pandemic. Implementing antigen testing in the COVID-19 testing algorithm could reduce strain on laboratories and reserve molecular testing for situations where high test sensitivity is required. 

The advantages and disadvantages of antigen testing for SARS-CoV-2 need to be considered to determine if and when it is a reliable method for COVID-19 diagnosis. 


  • Antigen testing is cheap and fast.
  • Antigen testing may be a useful public health tool. Since antigen tests are generally considered very accurate when they are positive, they may be helpful in quickly identifying highly infectious individuals within a community. However,
    it is important to note that antigen test performance in areas with low disease prevalence may be poor and data on test performance in asymptomatic individuals are limited. Since antigen tests are more likely to be positive when a person has a higher viral load,
    super-spreader events or outbreaks may be prevented by identifying these individuals early and isolating them. 
  • Potential for use in the point-of-care setting. The rapid nature of antigen assays, as well as less reliance on expensive equipment and reagents, make them good candidates for point-of-care testing. This may reduce the burden of testing on laboratories and provide rapid results at the location where the patient is tested. …


  • Lower sensitivity than PCR tests. Antigen tests are not as sensitive as PCR tests and false negatives pose a real problem. A negative antigen test should always be treated as presumptive, but can still give the patient and care provider a false sense of security. These tests may not be appropriate in settings where a positive result cannot be missed, such as hospitals or settings with high-risk patients or staff. 
    While they occur less frequently with this test type, false positives are also problematic
  • Less expert interpretation in point-of-care settings. Although point-of-care testing is convenient, it is important to remember that the test is often performed and interpreted by healthcare professionals that are not trained in clinical laboratory science. Appropriate messaging around the interpretation of results is imperative, including the possibility of false negative and false positive results, and the continued use of public health measures, such as masking and distancing, are essential.
  • Limited evidence on performance and use. There is very little literature available …
  • Laboratory space, supplies and staff are still required.

Read more at (Source: )

sensitivity and specificity

Each test for Sars-CoV-2 – be it molecular, serological or antigenic – has in fact two main characteristics: sensitivity and specificity.

Sensitivity corresponds to the proportion of positives to the new coronavirus correctly identified as such: the higher the sensitivity of a test, the lower the likelihood of incurring false negatives (i.e. people who according to the test are not infected, but who in actually they are).

Specificity, on the other hand, corresponds to the proportion of negatives that are correctly identified as negative for the virus: the higher the specificity of a test, the lower the probability of incurring false positives (i.e. people who according to the test are infected, when in reality they are not I am). Read more in


False results – when

I found two youtubes that partially explains the risks of false results.

To understand the unreliability of covid 19 testing it is important to understand how the body react on a virus infection like corona infection.


Body immune responses

To explain this, Daniel Lewis Gerard Medinger uses information from aForbes article was published on May the 27th written by William Haseltine


The body typically has two major immune responses:

  1. an antibody response is also known as a lymphoid response and 
  2. the cell-mediated response also known as a myeloid response 

What is this myeloid response?

Myeloid cells is the collective name of Granulocytes and Monocytes. They are differentiated descendants from common progenitors derived from hematopoietic stem cells in the bone marrow.  ( )

  • Myeloid cells attack the virus and the infected cells directly
  • some myeloid cells engulf and destroy virus particles. 
  • others killed infected cells directly
  • and others still induce a protective inflammatory response by the release of compounds called cytokines
    ( from Greek “κύτος” kytos “cavity, cell” + kines, from Greek “κίνησις” kinēsis “movement”. Wiki

What is it that the SARS-virus is doing differently than other coronaviruses to have this effect?

SARS-virus affects the immune response in two different ways:

  • The lymphoid pathway is muted
  • the myeloid response becomes hyperactive.

This is why some people who recover from Covid-19 have very low, sometimes undetectable levels of anti-SARS-2 antibodies, and others have undetectable levels of “neutralizing” antibodies capable of the inactivating virus in laboratory experiments….

A hyperactive myeloid response, on the other hand, can result in the famous cytokine storm associated with the rapid decline and death of Covid-19 patients.” ( ) 

In dampening the antibody response to infection and ramping up production of chemokines, SARS-2 is amplifying what happens to us naturally as our immune systems age. While our ability to mount an effective antibody and T cell response to new infections declines, the myeloid arm of the immune system becomes overactive. These features of the aging immune system account for both the decline in our response to new vaccines and to an increase in inflammatory auto-immune diseases such as rheumatoid arthritis. ( ) 

To my feeling is that this is what we’re seeing with the people out there the one in ten still suffering brutal fatigue well after a covert infection and potentially the long tail of symptoms to all seems to kind of make sense right. But science needs more than just sense it needs evidence.

Does this mean that elder people may be more easilty get false negative serologic RDT result?

Covid and antibodies

Information from

So let’s take a look researchers at the Rockefeller University in New York looked at blood plasma samples from 149 patients they discovered that 

The majority of the samples they have studied showed poor to modest “neutralizing activity,” indicating a weak antibody response. However, a closer look revealed everyone’s immune system is capable of generating effective antibodies—just not necessarily enough of them. Even when neutralizing antibodies were not present in an individual’s serum in large quantities, researchers could find some rare immune cells that make them

… In 33 percent of donors, the neutralizing activity of plasma was below detectable levels. It’s possible that for many in this group, their immune system’s first line of defense had resolved the infection quickly, before the antibody-producing cells were called in.” (Read more in )

Let’s look at the accuracy of the two primary antibody tests being done in the UK the Abbott test and the rosh test.

Information taken from
We’ll ignore the manufacturer’s figures and instead look at Public Health England’s independent tests here’s the Abbott test the specificity is very high essentially a hundred percent.

This means that the risk of false positive is negligible ie you can trust a positive result


but the sensitivity not so much 93% this means that there is a significant risk of false and negative ie if your test comes back negative.

Tat doesn’t necessarily mean you’ve got no antibodies and we can see here how that figure drops with time for people who caught Covid in early mid March extrapolation would suggest you’d be looking at a seventy-five to eighty percent figure here for a negative result and here’s the rush test also very good at positive results a high specificity but not so much the negative ones overall 15 of the 93 samples tested incorrectly negative interestingly though in contrast to the Abbott test the sensitivity increases as the interval increases that is to say if it’s been a while since you had an infection this data suggests you should go for the rush test not the Abbott one what do these results tell us well to understand let’s look at the way the tests are calibrated in the first place the calibration of test is done with the very sick in hospitals potentially creating spectrum bias and there have been studies done to investigate this that means that the level of Tolerance in these tests may not be set in the right place for those with more mild conditions especially if those tests are done several weeks after the initial infection when antibody levels are started to decline here’s a graph of how the IgG antibody level declines over time in some of these people it may be declining under the test threshold that has been set to high due to the initial calibration using extremely sick patients here’s what I think is going on this is a novel coronavirus a novel is certainly the right word it seemingly attacks organs at will lungs in some people hearts in others pretty much whatever it wants so it’s not unreasonable to assume that given the different ways that the virus attacks the body there may also be a spread in terms of how individuals respond to the virus whilst in any immune response there will always be some combination of the two some people may have a heavier antibody response whilst others may have a heavier myeloid orc cell mediated response my theory is this those that have a heavier myeloid response are both more likely to a test negative for antibodies especially after a longer period of time and B suffer from post viral fatigue now data to test this is somewhat harder to come by so I conducted my own research using both the UK and US kovat long-hauler Facebook groups numbering many thousands of people now of course there are some caveats to this kind of data the sample is both self selecting and self reporting and there is no shortage of intrinsic biases and so on but as the body politic group found in their survey there was no statistical difference between those who had tested negative and positive for kovat in terms of which symptoms were reported ie symptoms are a far more reliable indicator of infection than a swab pretest probability also backs this up and all of this group of long haulers were experiencing symptoms now there were 95 people in my sample the average interval for an antibody test was about 10 weeks after infection the results from the rosh test would predicts a hundred percent accuracy after this time interval the Abbott test with extrapolation would suggest around 80 so let’s split the difference and call it ninety percent out of our 95 sample then that should mean would expect to see 85 positive antibody tests but if long-haul is have a propensity towards a cell mediated or myeloid response rather than a lymphoid or antibody response then this number could be lower it’s a bit like Family Fortunes this you said 85 our survey said 21 that’s right only 21 of the 95 long haulers tested above the detection threshold for antibodies against Tsarskoe v2 that’s 22 percents not ninety percent or even 80 percent the remainder 74 of course were negative for antibodies breaking it down 14 of the 95 had had a positive swab for Ovid 81 had either not been tested or tested negative interestingly a positive swap is no guarantee of a positive antibody test as you can see here as many testing negative for antibodies as positive one small side note here these particular positive antibody tests tended to have happened at a much sooner interval than the unswept or negative swabs are you less far into the infection and this may account for some of the higher positive ratio most of a sample found it very difficult or impossible to get hold of a swab at the time of infection which for most of them was early to mid-march and this is of course consistent with the number of tests available and the testing criteria in both the US and the UK at the time some of them did go on then to have a test in week 7 8 9 or 10 which came back negative and this of course is entirely consistent with what we know about that positive testing window with the swab which usually is 7 to 10 days after infection and as for the brands of tests themselves sample size is too small really to draw any major conclusions but there’s nothing here to suggest rush is super sensitive to long tail infections or late interval tests does this support the theory that the immune response Turk Ovid in this group of long hauls is primarily cell mediated rather than antibody led well I’d say so let’s look at some more data hot off the press this is a study of 200 frontline healthcare workers at University College Hospital in London this data was collected much earlier than most of the antibody tests taken in my so unsurprisingly it shows a high proportion of positive antibody tests in total forty five point three percent of the staff in the sample eventually tested positive for antibodies but let’s quickly pick up on some other interesting findings only twenty one percent tested positive for SARS two via swab in this period now of course some may have already had it but when you’re positive swab rate is half the antibody rate it tells you something about how reliable those swaps are not very that’s reflected in my data – and only 19 percent of those who tested positive on a swab met the government’s definitions of symptoms um what does that tell us about the symptoms the government told us were coded and that nothing else qualified for a test well you can work that out for yourselves so let’s just recap on where we’re up to I’m proposing that there are three categories of people who contract kovat and the ways that they respond to the virus are different so let’s call them long haulers fast clearers those who basically have symptoms for a week or two and then get better and asymptomatic and I’m proposing that long haulers may have a different immune response that may be partly responsible for some of this long hauling that is to say a more significant myeloid response and a less significant lymphoid or antibody response how am i proposing to show this well via the results of antibody tests and the proportion testing positive in each group so 22 percent of our long haulers are testing positive for antibodies what about the staff at UCL H well again I’m going to assume that symptoms are the most accurate indicator of infection compares the number of false negatives we see with swabs so of those reporting symptoms at UCL H 24 of 45 tested positive for antibodies that’s 53 percent now asymptomatic is a bit more complicated some of this group will currently have coded some may have previously had covered and some won’t have had coded at all at the point the survey was taken 27 of the 155 or 17% swapped positive but 37 percent of them had antibodies so it stands to reason that at least 30 7% of them have had kovat but we don’t know what proportion of that 155 actually did have Cove it so the number for a symptomatic is a minimum of 37% if you assumed that half of the total asymptomatic that had Kovac and half had not the percentage showing antibodies would be 74 percent however you look at this number even the minimum is significantly higher than our long haulers so let’s look at our three categories long haulers 22% fast clearers 53% and asymptomatic assuming that’s half of 155 were infected 74% now we don’t know if there are some long haulers hiding in the fast Clara’s data but according to King’s College London long haul is only number about one in ten of those who have symptoms and even if they were in there that only served to drag this number downwards if my theory is correct it’s kind of interesting huh but it does of course raise more questions what is a negative antibody test mean regarding possible immunity and reinfection that’s a bit harder to answer right now William Goldman the famous novelist and screenwriter once said of the movie business nobody knows anything and the same could almost be said of the current understanding of what Kovac does to the body all we can really do is to trawl the data as we get it and try and learn what we can now I doubt my study is going to make the Lancet or BMJ but I think it certainly deserves further investigation I’d love to see a larger cleaner dataset especially when it comes to the category of positive swabs not of that particular effects my opinion over whether someone’s had COBIT or not but it makes the data harder to argue with perhaps as we start to see more studies getting published we’ll find that data out there there’s also no distinction in my study between those who have had ongoing viral symptoms and those who are suffering something closer to post-viral fatigue post-viral fatigue has previously been connected to this hyperactive immune response or cytokine storm so that makes sense but what’s going on with these mad ongoing more viral type symptoms is the virus still active in the body or is the body may be reacting to damage caused by the virus we just don’t know at least the condition is starting to get a little bit more attention here’s professor Williams from King’s College London it may be that people having this overactive inflammatory response are also at greater risk at getting viral fatigue with kovat and that’s one of the things we’re going to look at chronic fatigue is overlooked in general medical training although a new online training module has been devised recently doctors have traditionally been very poor at dealing with this now it’s more important than ever that they are educated about it I think we’d all agree about that so for now all we can really do is look after ourselves but I hope it’s slightly reassuring when a clearer

picture starts to emerge just a question of waiting for the data until then if you’ve got any antibody test result stories you’d like to share then that starts a discussion in the comments and perhaps we can start to get a slightly fuller picture till next time.


Issue 3

“The Hong Kong cocktail”

Hong Kong Doctors See Progress In Treatment Of COVID-19







  • Sources




A pluralist agnostic seeker

Insert math as
Additional settings
Formula color
Text color
Type math using LaTeX
Nothing to preview