X Close

How far away are ‘immunity passports’? There are huge hurdles facing us before we can allow people out and start normal life again

Sample testing devices used in diagnosing the coronavirus are checked on a production line in South Korea (Photo by ED JONES/AFP via Getty Images)

Sample testing devices used in diagnosing the coronavirus are checked on a production line in South Korea (Photo by ED JONES/AFP via Getty Images)


April 7, 2020   5 mins

We need antibody testing. Everyone agrees with that ‚ÄĒ the question is how it can be done.

There are two main kinds of test for Covid-19: PCR tests which tell us who has the disease right now; and antibody or serological testing, which can tell us who’s had it in the past. The Government wants to build testing programmes that can give a reliable answer from a pinprick test in under 20 minutes.

I’ve been banging on about antibody testing in recent pieces. It’s vital for understanding how far the disease has spread throughout the population; without knowing that, we can’t know how deadly it is, or exactly how alarmed we should be.

But other people have been making another point: that if someone has already had the disease, then ‚ÄĒ hypothetically, hopefully ‚ÄĒ they ought to be immune to getting it again. If they‚Äôre immune to getting it again, they can‚Äôt spread it, so they should be allowed back out into the world, to help get the economy started again.¬†

This is the idea behind mooted ‚Äúimmunity passports‚ÄĚ for people who‚Äôve tested positive. Tony Blair seems to be talking about something similar when he writes, in the foreword to a report for his Institute for Global Change, that unless we can ‚Äútrack who has the disease and who has had it‚ÄĚ, we will remain locked in a situation ‚Äúin which everyone has to be isolated so that we don‚Äôt miss the small number of cases which need to be isolated‚ÄĚ.

Immunity passports sound like a good idea. But I wanted to sound a note of extreme caution. I think antibody testing is absolutely vital ‚ÄĒ for understanding the spread of the disease, at a population level. But using it to establish individual immunity is very different and potentially much more dangerous.

Let‚Äôs assume that people who have had the disease really are immune. Now, imagine that you have a test that is 95% accurate ‚ÄĒ that is, if you use it on someone who has had the virus, it‚Äôs 95% likely to say they have had the virus; if you use it on someone who hasn’t, it‚Äôs 95% likely to say they haven‚Äôt”. The US FDA has just issued emergency approval for a test with reported numbers pretty much like that (to be precise, 94% and 96% respectively).

So you are given the test, and it comes back positive. How likely is it that you’ve had the disease?

The answer, for the record, is not 95%. The answer is: you don’t know. Unless you know what percentage of people in the population have had it, then you simply don’t have enough information to answer the question.

Let’s look at why. Imagine that 3% of people in this country have had Covid-19, which probably towards the upper end of most reasonable estimates at the moment. Now let’s imagine that you test a million people, completely at random. 

Of those million, about 30,000 will actually have had the disease. Your test will correctly identify 28,500 of them. And 970,000 people will not have had the disease. Your test will correctly identify 921,500 of them. So there will be about 1,500 people who are wrongly told that they haven’t had it, and who are forced to remain in unnecessary isolation. That’s bad. But it’s not the real problem.

The real problem is that 48,500 will be told they have had the disease and are now immune. That outnumbers the true positives by almost two to one. If you issue immunity passports on this basis, barely a third of the people you give them to will actually be immune. ‚ÄúThere‚Äôs nothing peculiar about this statistically,‚ÄĚ Kevin McConway, an emeritus professor of statistics at the Open University, told me. ‚ÄúIt‚Äôs just Bayes‚Äô theorem.‚ÄĚ The likelihood of you having had Covid-19, if you‚Äôve had a positive test, depends not just on the accuracy of the test but on the prevalence in the population you‚Äôre looking at.

This is a huge problem in diagnostic screening for other diseases, such as cancer and dementia, and can mean that screening tests do more harm than good. It’s why there are follow-up tests. With Covid-19, the concern is that if you just test the same person twice with the same test, there’s a good chance it’ll get it wrong for the same reasons.

This is not to suggest that immunity passports are impossible. For instance, McConway points out, it would be possible to narrow the group you‚Äôre testing, so that it contains more people who‚Äôve probably had the disease. It might be, let‚Äôs imagine, that people who‚Äôve had the main symptoms of the disease ‚ÄĒ a persistent dry cough and a high fever ‚ÄĒ are 60% likely to have had it. So you ask people if they‚Äôve had those symptoms, and you test a million people who have.

In that case, your million tests would send you back 570,000 true positives and only 20,000 false ones; you could be 97% confident that a positive test was correct.

Of course, some tests will eventually be better than 95% accurate ‚ÄĒ some of those in development claim 100% specificity (that is, they never give false positives). If that‚Äôs real, great, although I agree with McConway when he says that no test can be literally perfect.

And this blog post by Sir John Bell at the University of Oxford explains the difficulties of making an antibody test for a new disease: you need something that can be rapidly rolled out to millions of people; you need to validate it on people who you know tested positive on a PCR test, at least 28 days ago, to check for false negatives; and you need to test it on blood taken from people before the outbreak, so you can check for false positives. 

The Oxford post says ‚ÄúSadly, the tests we have looked at to date have not performed well. We see many false negatives (tests where no antibody is detected despite the fact we know it is there) and we also see false positives.‚ÄĚ None of those tests have reached the standards that they and the Medicines and Healthcare products Regulatory Agency have agreed.

This does not mean that antibody tests are of no use! Far from it ‚ÄĒ antibody tests are essential. If you have a well-validated 95% accurate test, you can‚Äôt confidently say that any individual is or isn‚Äôt immune ‚ÄĒ but you can use it to get a sense of how many people have the disease. The more people you test, the more accurate an answer you can get; tens or hundreds of thousands of tests would give you quite a precise estimate with quite narrow confidence intervals.

That would, as I keep saying, give you much more information about the disease itself. But it would also give you a better chance of using antibody tests for immunity passport purposes. As we’ve seen, if you don’t know the prevalence in the population, you don’t know what a positive test means for the individual. But if you knew that some high percentage had already had it, and were able to narrow it down further by asking about symptoms, and then could use a highly accurate test, then you could start to get towards the sort of precision we need; where you might be able to let people out into the streets again.

In the end, that’s going to be a horribly cold-blooded calculation. If you let people out when they’re 90% likely to be immune, that means one person in 10 is going to be at risk of getting and spreading the disease. Is that risk a price worth paying for reducing the real costs (economic, social, physical, mental) of isolation? I don’t know and I’m glad I don’t have to work it out. But someone has to. And they’ll have to start by getting a reasonably effective test, and testing hundreds of thousands of people, to see how many of us have had it. 

 


Tom Chivers is a science writer. His second book, How to Read Numbers, is out now.

TomChivers

Join the discussion


Join like minded readers that support our journalism by becoming a paid subscriber


To join the discussion in the comments, become a paid subscriber.

Join like minded readers that support our journalism, read unlimited articles and enjoy other subscriber-only benefits.

Subscribe
Subscribe
Notify of
guest

33 Comments
Most Voted
Newest Oldest
Inline Feedbacks
View all comments
Daniel Smallwood
Daniel Smallwood
4 years ago

Am I the only person thinking that there’s a paradox here? That is, assuming for a moment that the tests are accurate, that a person adhering to the ‘rules’ and staying at home, and never contracting the infection, will never be ‘released’ from house arrest. So anyone who doesn’t consider themselves to be in a particularly vulnerable group should rush out and infect themselves. Which I doubt is the intention of these tests, and possible ‘passports’.

Helen Barbara Doyle
Helen Barbara Doyle
4 years ago

No, you are not the only one, the responsible will be punished it seems.

nickandyrose
nickandyrose
4 years ago

If anybody thinks I’m staying a prisoner in my home indefinitely, you can foxtrot oscar.

Nigel Blumenthal
Nigel Blumenthal
4 years ago

As soon as there’s a document, or a card, or a passport that gives some people access to something that the rest of society is denied, that document will be copied, counterfeited, and resold by people whose only interest is to find a new source of revenue. Once that happens, of course, the value of the document is lost, and the infection (in this case) can easily be spread by people pretending to have immunity who in fact do not. This would be the most dangerous step.

roslynross3
roslynross3
4 years ago

Health will never come from a vaccine. The best defence is a robust immune system. That will not come from a pill or a needle.

Some countries demand Yellow Fever vaccination before entering. It would be impossible to have a vaccine for everything and impossible to have every human on the planet vaccinated. Immunity passports would simply be another form of fascism. Those who are vaccinated still get the diseases and those who are not don’t necessarily get the diseases.

As to antibodies, no doubt antibodies from naturally acquired disease are different but those from vaccines count for nothing. From an NCBI article.

Severe tetanus in immunized patients with high anti-tetanus titers.
Crone NE1, Reder AT.
Author information
Abstract
Severe (grade III) tetanus occurred in three immunized patients who had high serum levels of anti-tetanus antibody. The disease was fatal in one patient. One patient had been hyperimmunized to produce commercial tetanus immune globulin. Two patients had received immunizations 1 year before presentation. Anti-tetanus antibody titers on admission were 25 IU/ml to 0.15 IU/ml by hemagglutination and ELISA assays; greater than 0.01 IU/ml is considered protective. Even though one patient had seemingly adequate anti-tetanus titers by in vitro measurement (0.20 IU), in vivo mouse protection bioassays showed a titer less than 0.01 IU/ml, implying that there may have been a hole in her immune repertoire to tetanus neurotoxin but not to toxoid. This is the first report of grade III tetanus with protective levels of antibody in the United States. The diagnosis of tetanus, nevertheless, should not be discarded solely on the basis of seemingly protective anti-tetanus titers.

Nigel Clarke
Nigel Clarke
4 years ago

Hmm…so people who have been asked if they have had symptoms then get an anti-body test, with the resulting proportion of false +/-.
I may be missing something here (and almost surely am), but should we not be testing randomly throughout the whole population rather than selecting a sub-group to test who have been asked if they have had symptoms and are subsequently tested on the basis that they have in all likelihood had the virus? Won’t this result in an incorrect finding that a higher % of those that think they have been infected will in fact test positive, because they are not selected randomly from the population?
I thought we needed to discover the number of people who are asymptomatic, and thus discover the R0, which then leads to directions on which policy can be used to bring the lockdown to an end.

sh
sh
4 years ago
Reply to  Nigel Clarke

You need a random sample if you are making inferences about the population.
So, if a study aimed to determine the % of people in the UK who have had the
illness, then a random sample would be correct.

If the purpose is to identify people who have had the illness, say
to allow them out of lockdown. And, given a test which has some % false
positives, then to reduce that number from overwhelming the true positives, you
want to apply the test to people with a higher prior probability of having had
the illness, such as have had symptoms.

Michael Baldwin
Michael Baldwin
4 years ago

As several people have raised objections to my comment below, I’d like to reply to them all at once here.

My critics all seem to have overlooked the figure I gave of the San Francisco Bay area, in which 26% tested positive of what appeared to be a random sample, but was not that much different than the roughly 20% figure I was claiming for the UK, based on the admittedly insufficient UK data I acquired via the Guardian, but which therefore looked at least approximately correct.

Both figures suggested something like a 12 to 15 million currently infected or recovered number of the UK population, and therefore the current something like 6,000 covid-19 total death rate at this point in the year, was little different than the average annual mortality rate for seasonal flu of 17,000.

Thus not in any way able to justify this deeply dangerous and tyrannical lockdown, dangerous both to the mental and economic health of most of the population, and an anti-libertarian act of unprecedented and outrageous proportions, with no parallel in human history at any place or any time, except perhaps regarding the slave populations who built the Egyptian pyramids.

The point being, even if the figures are out by a factor of even ten, these kinds of infection and mortality rates are showing clearly this is not in anywhere near the league of the 1918 Spanish flu or bubonic plague or similar.

I’m also quite sure there are quite a lot of people if they thought they were carrying covid-19, would not dare go near a hospital or wish to be tested, for fear of the possible consequences on their liberty if they were found to test positive.

So once again, there are strong reasons to think the number of carriers is even greater than I’m suggesting – the Oxford University study suggested up to half the entire population may have it or already have had it.

And I’m afraid it’s not good enough to say we need to do a lot more testing before we can be sure. Because we have actually made – I mean the government has made – this unprecedented in human history decision to lock up its whole population without such clear data, with possibly catastrophic economic and social consequences.

Consequences, which if the consensus of economic experts are correct, would see not only massive job loss and business closure, but possible bank collapses even.

I think you’ll find for example if you examine your bank terms and conditions, they are not guaranteeing to reimburse anybody if a collapse occurs for more than about √ā¬£54,000 or something, even if you’ve got more than a million in (it varies from bank to bank, but I think not very much), so even millionaires may end up near penniless by their own standards.

And very likely the insurance companies that we all rely on to bail us out in extraordinary times, may also go to the wall.

I wouldn’t be at all surprised if they’d attempt to evade their duty to pay out by framing the current events as “an act of God.”

And of course, as likely the NHS budget will have to be massively slashed as a fall out of the economic disaster we may face, this could lead to hundreds of daily deaths merely of people unable to get hospital operations in time, as the waiting lists may go from months (you know, including for things like cancer treatments) to years, which will then be an effective death sentence upon them.

So if we don’t stop this lockdown soon, we may be facing a 30s style recession, in which even former university professors after the crash came, were found to be begging on the streets merely in order to eat.

And my guess is, if this carries on long enough – likely even another month would do it – the government may have no choice but to tax people across the board “till the eagle screams”, such that doctors or other professionals on for example a salary of say 80k-√ā¬£100k a year before the lockdown, may end up with effective “Third World” levels of pay, like those in Poland or South America, not even equalling our current – we may have to call that former, shortly – national average wage.

Finally, I would for example like the critics to picture the life situation of a retired person, who does not know if he or she will live another summer after this one, with or without the threat of this virus.

And as such, may face – and this will undoubtedly be true in numerous thousands of cases – being robbed of the last summer and year of their lives, unable to travel outside freely, see their friends and relatives, take holidays, and so on, before they die.

Thus many who may be in that situation, would understandably like to be allowed to take the risk, and not be placed effectively in prison under house arrest for possibly what little remains of their lives, when they have committed and are guilty of no crime.

So as it appears that the support for this lockdown is coming mainly from the relatively young and healthy – such as the politicians and media members who are mostly supporting it – it appears they might consider what they are doing to their own parents or older relatives, effectively blackening what little freedom and quality of life may remain for those many who most certainly will be facing the last year of their lives, with or without the presence of covid-19.

Pat Duignan
Pat Duignan
4 years ago

Norway has tested 2.2% of it population and found that less than 0.2% have Covid19. That is the available evidence. You assert and Oxford assert that very high percentages of the population have had Covid19 without any evidence to support that. It is totally irresponsible to do so. As a 70 year old currently locked down in New Zealand I can assure you I wish to enjoy my grandchildren for many years to come and together with 90% of New Zealanders – evidenced by a reputable polling organisation – support the lock down.

Michael Baldwin
Michael Baldwin
4 years ago
Reply to  Pat Duignan

duplicated so deleted

Michael Baldwin
Michael Baldwin
4 years ago
Reply to  Pat Duignan

“Norway has tested 2.2% of it population and found that less than 0.2% have Covid19. That is the available evidence.”

Is it really?

Not according to this site, which says there are 6,298 confirmed cases and 123,170 test = 6,298/123,170 x 100% = 5.11%

Which is over 25 times as high a percentage as you are suggesting.
https://corona.help/country

(I checked the figures here by the way with the John Hopkins University site, and they were almost identical for deaths/positives, but the JHU site does not show testing rates, but I would therefore safely assume to be correct).

Not only that, by a similar simple calculation it can easily be found the infected percentage of those tested so far in other countries is 23.6% UK, 19.7% USA, and France a staggering 37.5%.

https://corona.help/

So it appears to me it is you who are being irresponsible with your statistics, as are many in general who support the lockdown policy.

So by all means if you are scared of this virus, please feel entitled to lock yourself way in hiding but please don’t continue to believe that given the it appears to me non-existent statistical basis for this lockdown, now we are actually starting to see that the death rates are in fact very low, and not significantly worse than a bad year of seasonal flu, you are entitled to lock the rest of us in prison for what may turn out to be the rest of our short lives, and in either case, condemn us to live out of our life in something like a 1930s style depression era, which you should perhaps consider you may also thereby bequeath to your children if banks go bust and so on, if the economy is not restarted again very soon.

Peter Kriens
Peter Kriens
4 years ago

Fat chance more than 3% of your readers understand Bayes … Although there will be a lot of false positives! Bayes, and its little sister Simpson’s paradox, are extremely hard to understand without doing some real mental effort, they are just so counter intuitive. That said, great work you’re doing trying to make people understand the reality.

cally hill
cally hill
4 years ago
Reply to  Peter Kriens

The reality of what? What you assume to be true? And assuming that most readers don’t understand things in the article and you do is laughable.

Peter Kriens
Peter Kriens
4 years ago
Reply to  cally hill

Bayes theorem and Simpsons paradox are very counter intuitive. I’m an engineer. Although I can do the math, they still feel wrong. Pretty sure 50% of the population has the mental capacity to understand them but most people do not bother to spend the effort. Especially women seem to have the tendency to skip the effort because they don’t like math.

routine1
routine1
4 years ago

Seems to me that equating “the disease” with having the virus might mislead. You can test positive for the virus without any symptoms, I believe. Big difference for the individual concerned, if not so much for national action.

Gerald gwarcuri
Gerald gwarcuri
4 years ago

The answer to the “paradox” is not in the poorly-defined concept of “accuracy” that most people have in mind.

I spent my career in healthcare as a clinical laboratory scientist. I performed literally thousands of laboratory tests during that time. Strict quality control was, of course, imperative to being able to rely on the test results for clinical diagnosis and monitoring.

But accuracy and precision are just the beginning of understanding the limits of the test and therefore the limits of its clinical utility.

Every test is designed with two attributes in mind: sensitivity and specificity. These attributes are usually mutually incompatible. The more sensitive a test is, the more likely it is to yield false positives. The more specific it is, the more likely it is to yield false negatives. Those two competing tendencies must be balanced against one another to yield a practical, clinically useful test. It isn’t easy.

In the case of a serological test for a highly contagious disease, you would want to err on the side of false negatives, i.e., you would design your test antibody to be as specific as possible, accepting a higher rate of false negatives as a consequence. Some people who actually had the disease might test negative, but you would be less likely to include people who had not had the disease, and falsely give them a clean bill of health.

Janie Doebuck
Janie Doebuck
4 years ago

It used to be that Humpty-Dumptification applied only to language but apparently it has been pressed into service in these times where everything and everyone must bend the knee to the Great Corona. A number means whatever the CDC/WHO says it means

In the CDC 2019-Novel Coronavirus (2019-nCoV) Real-Time RT-PCR Diagnostic Panel (https://www.fda.gov/media/1… ) page 2, paragraph 2 it reads “Positive results are indicative of active infection with CoVid 19 but do not rule out bacterial infection or co-infection with other viruses. The agent detected may not be the definite cause of disease. Laboratories within the United States and its territories are required to report all positive results to the appropriate public health authorities.”¬Ě In other words, CoVid may or may not be the cause of death but the medical community is required to report it as such.

The second death in my state was a woman who went to the local Emergency Room in cardiac arrest. She died. She was posthumously tested for Covid. Positive results. Reported cause of death: Covid 19. Did she die from CoVid or with it? Big difference.

For a very clear analysis of why this distinction is critical, read Dr. John Lee’s recent articles in The Spectator.

More shenanigans from the CDC (and in case you aren’t aware, the CDC/WHO are not just sleeping with Big Pharma…it’s more like CDC/WHO/Big Pharma version of a Roman orgy)
At any rate, it appears that the CDC has given up all pretense of living in anything other than la-la land which is where most of my fellow citizens in the US are currently residing because now it is possible to record a death as CoVid caused “whether or not there’s actual testing to confirm that’s the case.”¬Ě “COVID-19 should be reported on the death certificate for all decedents where the disease caused or is assumed to have caused or contributed to death.” Never mind that the decedent was suffering from COPD or cardiac problems or any number of other chronic debilitating diseases. https://www.cdc.gov/nchs/da

No need to actually think about anything. And there is most definitely no need or reason to question what our betters have to say. CoVid 19 has relieved of us the tiresome burden of critical thinking. And yes, bring on the microchipping of the entire planet that Bill Gates so ardently advocates because, after all, it is “only for convenience and safety”¬Ě and it’s just a variation “on traditional cattle branding”¬Ě.

An extremely timely quote from Josef Goebbels: “Propaganda works best when those being manipulated are confident they are acting of their own free will.”¬Ě

Gates and Goebbels “‚Äú a new public relations firm? Has a ring to it, don’t you think?

roslynross3
roslynross3
4 years ago

Many people have Strep bacteria in their throat and never develop disease. It is time perhaps to move beyond mechanistic materialist reductionist medicine. The human body is too complex to be reduced to number-crunching and generic diseases which don’t exist.

andy young
andy young
4 years ago

Surely the reasons for the false results are of paramount importance? If there’s something about individuals which means they always give false results, then you’re stuck with the third of positive results being false scenario; however if it’s purely random, a function of the testing process, then a second test reduces the error to 0.05 squared, i.e 0.0025, or 0.25%, which is obviously much more useful.
Does anyone know which of these options is correct? Is false testing associated with individuals or a random problem associated testing?

Nigel Clarke
Nigel Clarke
4 years ago

Can I ask why I am being pre-moderated?

Brain Unwashed
Brain Unwashed
4 years ago

Note:- Not my research but thought people might find it interesting and VERY enlightening.
________________________________________________________________________________________________-

Chicago Tribune, April 3: “A new, different type of coronavirus test is coming that will help significantly in the fight to quell the COVID-19 pandemic, doctors and scientists say.”¬Ě

“The first so-called serology test, which detects antibodies to the virus rather than the virus itself, was given emergency approval Thursday by the U.S. Food and Drug Administration.”¬Ě

“The serology test involves taking a blood sample and determining if it contains the antibodies that fight the virus. A positive result indicates the person had the virus in the past and is currently immune.”¬Ě

“Dr. Elizabeth McNally, director of the Northwestern University Feinberg School of Medicine Center for Genetic Medicine”¬¶’You’ll see many of these roll out in the next couple of weeks, and it’s great, and it will really help a lot,’ said McNally, noting doctors and scientists will be able to use it to determine just how widespread the disease is, who can safely return to work and possibly how to develop new treatments for those who are ill.”¬Ě

Got that? A positive test means the patient is now immune to the virus and can walk outside and go back to work.

NBC News, April 4, has a somewhat different take: “David Kroll, a professor of pharmacology at the University of Colorado who has worked on antibody testing, explained that the antibodies [a positive test] mean ‘your immune system [has] remembered the virus to the point that it makes these antibodies that could inactivate any future viral infections’.”¬Ě

“What the test can’t do is tell you whether you’re currently sick with coronavirus, whether you’re contagious, whether you’re fully immune “‚Äú and whether you’re safe to go back out in public.”¬Ě

“Because the test can’t be used as a diagnostic test, it would need to be combined with other information to determine if a person is sick with COVID-19.”¬Ě

Oops. No, this really isn’t a diagnostic test, it doesn’t tell whether the patient is immune and can go back to work. Excuse me, what??

Business Insider, April 3: “The world’s leading industrial nations have so far failed to identify any coronavirus antibody tests that will be accurate enough for home use, according to the UK’s Health Secretary Matt Hancock.”¬Ě

“The UK and other nations are currently examining plans to use antibody tests to allow individuals with immunity to COVID-19 to exit their national lockdowns early through the use of a so-called ‘immunity passport’.”¬Ě

“Spain was recently forced to return tens of thousands of rapid coronavirus tests from a Chinese company after they were found to be accurate just 30% of the time.”¬Ě

“Some tests have demonstrated false positives, detecting antibodies to much more common coronaviruses.”¬Ě

“Scientists also remain unsure about the extent to which a past infection could prevent reinfection and how long an immunity would remain.”¬Ě

Hmm. So the new antibody test has very serious problems, and it hasn’t been cleared for public use.

Medicine Net (undated): “Researchers at the Mount Sinai Health System say they’ve developed a test that can find out if you already have had or were infected with the new coronavirus.”¬Ě

“The test is called “serological enzyme-linked immunosorbent assay,”¬Ě or ELISA for short. It checks whether or not you have antibodies in your blood to SARS-CoV-2, the scientific name of the new coronavirus that causes COVID-19.”¬Ě

“Researchers say ELISA works like antibody tests for other viruses, such as hepatitis B. It will show whether your immune system ” the body’s defense against germs ” made contact with SARS-CoV-2, even months before.”¬Ě

“The test could help scientists fight the pandemic in several ways. It can give researchers a more accurate measure of how many people had the new coronavirus. It would also let health care workers who were ill with COVID-19 symptoms, but were never tested for the disease, return to work ” confident that they are now immune.”¬Ě

So wait, it’s a great test. Right? A positive test result indicates immunity, and people can return to work. What??

Science News, March 27: “The United Kingdom has ordered 3.5 million antibody tests, which would show whether someone has been exposed to COVID-19. Such tests, which just take a drop of blood, could help reveal people who have been exposed to the virus and are now likely immune, meaning they could go back to work and resume their normal lives.”¬Ě

“Science News spoke with David Weiner, director of the Vaccine and Immunotherapy Center at the Wistar Institute in Philadelphia, and Charles Cairns, dean of the Drexel University College of Medicine, about how antibody tests work and what are some of the challenges of developing the tests.”¬Ě

“Cairns: ‘The big question is: Does a positive response for the antibodies mean that person is actively infected, or that they have been infected in the past? The tests need to be accurate, and avoid both false positives and false negatives. That’s the challenge’.”¬Ě

Oops again. Cairns is saying the new test, if it reads positive, might mean the person is infected now. Or it might mean they were infected”and are now presumably immune. Figuring out which is the challenge. No kidding. It’s the difference between sick and healthy. So a positive test result means the patient is sick OR healthy.

As a reference, let’s look at how this same antibody test has been used in the past. For example, in the case of hepatitis A:

URMC Rochester (undated): This test looks for antibodies in your blood called IgM. The test can find out whether you are infected with the hepatitis A virus (HAV)”¬¶If your test is positive or reactive, it may mean: You have an active HAV infection”¬¶You have had an HAV infection within the last 6 months.”¬Ě

In other words, a positive antibody test could mean you’re sick now, or were once sick but are presumably immune now. Wonderful.

Medscape comments on the meaning of a positive antibody test for the Zika virus: “”¬¶immunoglobulin (Ig) M and neutralizing antibody testing can identify additional recent Zika virus infections”¬¶However, Zika virus antibody test results can be difficult to interpret because of cross-reactivity with other flaviviruses”¬¶”¬Ě

Two things here: no word about a positive test result revealing IMMUNITY from Zika; and a warning that a positive test might not have anything to do with Zika at all”that’s what “cross-reactivity”¬Ě means.

Medlineplus, referring to a Zika “blood test,”¬Ě which would include antibody testing, states, “A positive Zika test result probably means you have a Zika infection.”¬Ě Not immunity.

And there you have it. The official word on the COVID antibody test from official sources. It’s yes, no, and maybe. Public health officials can SAY whatever they want to about antibody tests: a positive result means you’re immune, it means you have an infection, it means you’re walking on the moon eating a hot dog.

Generally speaking, before 1984 a positive antibody test was taken to mean the patient had achieved immunity from a germ. After 1984, the science was turned upside down; a positive result meant the patient “had the germ”¬Ě and was not immune. Now, with COVID-19, if you just read news headlines, a positive test means the patient is immune; but if you read down a few paragraphs, a positive test means the patient is maybe”¬¶maybe not”¬¶immune. Maybe infected, maybe not infected. Maybe sick, maybe not sick. And, on top of all that, antibody tests are known to read falsely positive, owing to factors that have nothing to do with the virus being tested for.

Brain Unwashed
Brain Unwashed
4 years ago

Note:- Not my research but thought people might find it interesting and VERY enlightening.
________________________________________________________________________________________________-

Chicago Tribune, April 3: “A new, different type of coronavirus test is coming that will help significantly in the fight to quell the COVID-19 pandemic, doctors and scientists say.”¬Ě

“The first so-called serology test, which detects antibodies to the virus rather than the virus itself, was given emergency approval Thursday by the U.S. Food and Drug Administration.”¬Ě

“The serology test involves taking a blood sample and determining if it contains the antibodies that fight the virus. A positive result indicates the person had the virus in the past and is currently immune.”¬Ě

“Dr. Elizabeth McNally, director of the Northwestern University Feinberg School of Medicine Center for Genetic Medicine”¬¶’You’ll see many of these roll out in the next couple of weeks, and it’s great, and it will really help a lot,’ said McNally, noting doctors and scientists will be able to use it to determine just how widespread the disease is, who can safely return to work and possibly how to develop new treatments for those who are ill.”¬Ě

Got that? A positive test means the patient is now immune to the virus and can walk outside and go back to work.

NBC News, April 4, has a somewhat different take: “David Kroll, a professor of pharmacology at the University of Colorado who has worked on antibody testing, explained that the antibodies [a positive test] mean ‘your immune system [has] remembered the virus to the point that it makes these antibodies that could inactivate any future viral infections’.”¬Ě

“What the test can’t do is tell you whether you’re currently sick with coronavirus, whether you’re contagious, whether you’re fully immune “‚Äú and whether you’re safe to go back out in public.”¬Ě

“Because the test can’t be used as a diagnostic test, it would need to be combined with other information to determine if a person is sick with COVID-19.”¬Ě

Oops. No, this really isn’t a diagnostic test, it doesn’t tell whether the patient is immune and can go back to work. Excuse me, what??

Business Insider, April 3: “The world’s leading industrial nations have so far failed to identify any coronavirus antibody tests that will be accurate enough for home use, according to the UK’s Health Secretary Matt Hancock.”¬Ě

“The UK and other nations are currently examining plans to use antibody tests to allow individuals with immunity to COVID-19 to exit their national lockdowns early through the use of a so-called ‘immunity passport’.”¬Ě

“Spain was recently forced to return tens of thousands of rapid coronavirus tests from a Chinese company after they were found to be accurate just 30% of the time.”¬Ě

“Some tests have demonstrated false positives, detecting antibodies to much more common coronaviruses.”¬Ě

“Scientists also remain unsure about the extent to which a past infection could prevent reinfection and how long an immunity would remain.”¬Ě

Hmm. So the new antibody test has very serious problems, and it hasn’t been cleared for public use.

Medicine Net (undated): “Researchers at the Mount Sinai Health System say they’ve developed a test that can find out if you already have had or were infected with the new coronavirus.”¬Ě

“The test is called “serological enzyme-linked immunosorbent assay,”¬Ě or ELISA for short. It checks whether or not you have antibodies in your blood to SARS-CoV-2, the scientific name of the new coronavirus that causes COVID-19.”¬Ě

“Researchers say ELISA works like antibody tests for other viruses, such as hepatitis B. It will show whether your immune system ” the body’s defense against germs ” made contact with SARS-CoV-2, even months before.”¬Ě

“The test could help scientists fight the pandemic in several ways. It can give researchers a more accurate measure of how many people had the new coronavirus. It would also let health care workers who were ill with COVID-19 symptoms, but were never tested for the disease, return to work ” confident that they are now immune.”¬Ě

So wait, it’s a great test. Right? A positive test result indicates immunity, and people can return to work. What??

Science News, March 27: “The United Kingdom has ordered 3.5 million antibody tests, which would show whether someone has been exposed to COVID-19. Such tests, which just take a drop of blood, could help reveal people who have been exposed to the virus and are now likely immune, meaning they could go back to work and resume their normal lives.”¬Ě

“Science News spoke with David Weiner, director of the Vaccine and Immunotherapy Center at the Wistar Institute in Philadelphia, and Charles Cairns, dean of the Drexel University College of Medicine, about how antibody tests work and what are some of the challenges of developing the tests.”¬Ě

“Cairns: ‘The big question is: Does a positive response for the antibodies mean that person is actively infected, or that they have been infected in the past? The tests need to be accurate, and avoid both false positives and false negatives. That’s the challenge’.”¬Ě

Oops again. Cairns is saying the new test, if it reads positive, might mean the person is infected now. Or it might mean they were infected”and are now presumably immune. Figuring out which is the challenge. No kidding. It’s the difference between sick and healthy. So a positive test result means the patient is sick OR healthy.

As a reference, let’s look at how this same antibody test has been used in the past. For example, in the case of hepatitis A:

URMC Rochester (undated): This test looks for antibodies in your blood called IgM. The test can find out whether you are infected with the hepatitis A virus (HAV)”¬¶If your test is positive or reactive, it may mean: You have an active HAV infection”¬¶You have had an HAV infection within the last 6 months.”¬Ě

In other words, a positive antibody test could mean you’re sick now, or were once sick but are presumably immune now. Wonderful.

Medscape comments on the meaning of a positive antibody test for the Zika virus: “”¬¶immunoglobulin (Ig) M and neutralizing antibody testing can identify additional recent Zika virus infections”¬¶However, Zika virus antibody test results can be difficult to interpret because of cross-reactivity with other flaviviruses”¬¶”¬Ě

Two things here: no word about a positive test result revealing IMMUNITY from Zika; and a warning that a positive test might not have anything to do with Zika at all”that’s what “cross-reactivity”¬Ě means.

Medlineplus, referring to a Zika “blood test,”¬Ě which would include antibody testing, states, “A positive Zika test result probably means you have a Zika infection.”¬Ě Not immunity.

And there you have it. The official word on the COVID antibody test from official sources. It’s yes, no, and maybe. Public health officials can SAY whatever they want to about antibody tests: a positive result means you’re immune, it means you have an infection, it means you’re walking on the moon eating a hot dog.

Generally speaking, before 1984 a positive antibody test was taken to mean the patient had achieved immunity from a germ. After 1984, the science was turned upside down; a positive result meant the patient “had the germ”¬Ě and was not immune. Now, with COVID-19, if you just read news headlines, a positive test means the patient is immune; but if you read down a few paragraphs, a positive test means the patient is maybe”¬¶maybe not”¬¶immune. Maybe infected, maybe not infected. Maybe sick, maybe not sick. And, on top of all that, antibody tests are known to read falsely positive, owing to factors that have nothing to do with the virus being tested for.

sh
sh
4 years ago

We ran the predictive value numbers last week for four different prevalences of disease and one of the tests claimed sensitivity/specificity:

https://analyse-it.com/blog

jezfisk
jezfisk
4 years ago

I feel that this issue of “Immunity Passports” is being looked at all wrong and should be ditched immediately. It is the wrong use for any antibody tests we may develop, even if these tests get to a higher than 95% accuracy.

Even if we ignore the significant problems highlighted by this article due to false positives and negatives and the fact that many people might be told they are immune who aren’t or vice versa, there are many other problems, some mentioned by others:

1. It immediately creates discrimination in society. Those with passports will be given work, those without won’t. Not only that, but it positively discriminates towards people who have not followed government guidelines and caught the virus and negatively discriminates against those who did. It creates a “master race” of irresponsible (some might say stupid) people and locks up the rest indefinitely

2. It incentivises people to try to catch the virus to gain their passport and join the privileged group, rather than to help society by not catching and spreading it.

3. It would lead to criminals trying to steal or forge immunity passports or the wealthiest bribing people to get them or jumping the queue to get tested.

4. We don’t yet know whether the presence of antibodies now guarantees immunity and for how long this will last, or whether it guarantees that a person can’t still be contagious. We won’t know that for some time after the tests are probably available.

We have ALWAYS had a society where medical records are confidential for precisely the reason of avoiding discrimination. This should stay in place, even for COVID19

Instead, the antibody tests should ONLY be used for gathering population wide data on the prevalence of the disease. Combined with the antigen tests in a random sampling of the population, we should begin to get a picture of the number of current infections and a picture of how many have already had the disease.

The important point is that I would advocate that individuals in the random sample SHOULD NOT BE TOLD THE RESULTS of their tests. Quite controversial I know, but I’ll explain why shortly. Please note that individual test results would be kept – for once we have a vaccine, but not shared with them until then.

Only the overall results should be regularly and widely shared. Sampling will be invaluable for calculating when it is reasonably safe to end the lockdown FOR EVERYONE. There should be no special treatment for those deemed to be immune. Either we all come out of lockdown, or no one does. For example we might say that if fewer than 1 in 10000 currently have the disease then it is safe to reopen schools for a while, particularly if x% have already had it, because the risk levels are low. These risk levels can be discussed openly and agreed and perhaps it could be agreed what could happen when – eg most vulnerable only come out when prevalence is 1 in 500000 or something.

Then, as people start to go back to some sort of a “normal” life, presumably with some sensible social distancing still in place, the random sampling continues. Since no one knows if they are safe, they will ALL continue to behave as if they were at risk of catching it or passing it on. But, we will all know that we are going out at a time when risks are fairly low.

However, it would also be explained that, as soon as the random testing results reach a certain level, we will all be in lockdown again. This will further encourage people to behave responsibly as no one wants lockdown again if we can avoid it. Plus it helps people and govt to plan and make informed choices.

We may go through several lockdown phases before we get to having a vaccine. But, the government would be able to lockdown earlier, with more information, and save a lot more deaths than they have done this time by leaving lockdown so late.

Eventually though we will get a vaccine. Giving a vaccine to 66m people is a huge undertaking. Even at 1million a day, it would take 2 months. So, this is where we finally use individual information. To begin with, we ONLY give the vaccine to people that have not had a positive immunity test. We reward the people that have sensibly socially distanced and kept themselves virus free.

The others should, in theory, be safe if the antibody test they’ve had is correct, but they’ll be at the back of the queue for a vaccine because in theory they don’t need it anyway. Again, if people know well in advance that if they catch the virus they will not only risk getting very sick or dying, they’ll also go to the back of the queue for a vaccine if they survive, people will behave very differently over the next 18 months than the current “take it on the chin” or “I’m safe now” brigade.

So, I say NO TO IMMUNITY PASSPORTS. Create the correct incentives, not the wrong ones.

Andy Capp
Andy Capp
4 years ago

I speculated about this weeks ago calling them ‘Had it’ passes.

Useful even if testing is not very reliable.

Covid-19 isn’t going to be beaten by playing hide and seek or whack-a-mole with it. That is just delaying the inevitable. The only way it ends is with herd immunity. That comes from being infected or possibly a vaccine and we can’t afford to stay in an economically crippling lockdown waiting for a vaccine that may never arrive.

So the vulnerable who are likely to be killed by it regardless of medical treatment need to remain isolated and most of the rest of us need to get infected and become immune. Yes some of the rest of us will die – tough blame the Chinese government for letting it loose on the world.

Most of the rest of us need to get infected at a rate which allows health services to provide care to all than need it. Smart countries will be building those temporary hospitals and ventilators so they can reach herd immunity sooner and so get their economies fully functional sooner.

The aim is not to prevent infection but to control the rate and ‘Had it’ passes even if not very reliable should let us get more of the economy running again while managing the infection rate.

David Radford
David Radford
4 years ago

I totally agree with Nigel Clarke on randomising data and have been trying for several weeks to get a random YouGov survey of the population to estimate key types of respondent into consideration – eg no symptoms, have Covid-19, had it but better – and cross analyse by demographics. Large,sample using off the shelf methodology, done quickly and repeatable regularly at relatively low cost
Waiting for a reply from email to Freddie who has “previous” in this area.

Fraser Bailey
Fraser Bailey
4 years ago

No, there are no significant ‘hurdles’ to allowing us to resume our normal lives again. Others below have citied facts and figures in support of my assertion. All I will say is that we simply need to build up some capacity in the health system, equip medical staff with suitable PPE (something our vastly inefficient NHS seems incapable of doing), and continue to take precautions with the old and vulnerable.

I haven’t seen a doctor since 1980 because I have a strong immune system. As someone points out below, this is the key. Eat well, sleep well and get regular (daily) exercise.

Also, in the US they are basically reporting zero deaths from flu and pneumonia. Anyone who would have died from these conditions is automatically being assigned as a Covid 19 victim. The whole thing is a giant scam. The good news is that the Swedish approach seems to be at least as effective as the total lockdown approach, Saxony is re-opening the schools, and the Austrians plan to re-open shops soon.

Really, what a racket…

Tripod
Tripod
4 years ago

Instead of immunity passports I think we should issue white feathers to all those low risk non-key workers who stayed at home (like the passport office workers currently on strike) instead of doing their bit to protect the economy. They’ll be the first in line to suckle on the tax-payers teat when they find they don’t have jobs to go back to.

cally hill
cally hill
4 years ago

You start your article with the usual BBC type statement ”We need antibody testing. Everyone agrees with that.’ Have you asked everyone? No. You don’t know what everyone agrees so stop assuming you do. You discredit yourself before you even start by stating something which cannot possibly be true. The fact that you yourself know what every single person on this planet wants.

Eva Watson
Eva Watson
4 years ago

Nice Article

Immunity yes. 2020 is very cruel year. so we need more immunity. we need to strong the immunity of our body. covid-19 doesn’t seems to be stop yet.

our doctors are doing great job . PCR Test and testing by Covid-19 IgG/IgM Rapid Test Kit these are using.

Guys please maintain social distance, stay healthy, stay Home. and eat healthy so that our immunity can boost.

Michael Baldwin
Michael Baldwin
4 years ago

Having read the above article and the linked one, I will present some approximate arithmetic, which I compiled using figures given in the Guardian the other day, which said there had been something like 234,000 (thereabouts) tests in the UK at that point and about 50,000 had tested positive and about 5,000 had died.

So that’s 50,000/234,000 infection rate which is roughly 20%.

So that suggests (just speaking roughly again, as I just don’t feel we need to be that exact, it’s a question of orders of magnitude only in my view) if we tested the whole UK population we would expect to get roughly (as I keep saying) based on a population of 60 million, about 20% of them, so about 12 million, testing positive.

So then – to date – we’d be getting only 5000 deaths out of 12 million infected people.

So at a rough guess, if this is something like a 3 to 4 months season, we might get 20,000 deaths in total.

So then really, we’re just looking at something comparable to seasonal flu with average 17,000 a year.

I actually performed this surely very simple calculation a few weeks ago, using a rare Washington Post article that actually listed the testing rates in each country per million at that point. And in a diverse number of countries, I got very similar, almost identical results – something like seasonal flu in the mortality rate.

I later tried this again about a week or two later however and the results were very variable indeed, not showing a consistent mortality rate/IFR.

Which I suspect was mostly because of the revisions of how the deaths were being classified, such as one apparently reliable Italian source saying that only about 12% of deaths attributed to COVID-19 in Italy, were actually believed medically to have been caused by COVID-19.

I don’t doubt similar “revisions” of cause of death had set in elsewhere.

I also found a statistic in the San Francisco Chronicle about a week or so ago, which said in the Bay Area they had tested about 500 people and about 125 (this is roughly again, from memory) had tested positive on the first day of testing, actually I recall it was 26%.

As far as I could see this was actually random testing – the thing it appears could have so easily been done in the UK but mysteriously wasn’t – and later the testing unit focused on those only with significant or serous symptoms.

But likewise, if we assume that 26% of the UK population would also test positive (I mean, that might be a big assumption, as it’s assuming the infection rate in the SF Bay area is similar to that in the UK, but I don’t find it that unlikely), then once again, we end up with 15 million who would test positive (note how near to the 12 million I calculated above based on official UK data supplied to the Guardian by the medical authorities that is), so again we end up with a death rate/IFR similar to seasonal flu.

So I imagine, the article author, TC, is able to judge whether this analysis is approximately right, and I do mean only approximately – it wouldn’t make that much different if it was 5 million or 30 million infected or previously infected, to the decision making process, as it would be assumed to be beyond control.

That is to say, it would be easy enough to find out if those were the “ball park” figures of infected persons we had in the UK (or elsewhere) merely by random testing say 100 people in 20 supermarkets up and down the country.

And if it did turn out that as per the SF Bay area, it was 25% testing positive or more, then the authorities would know as they couldn’t test the entire UK population, and therefore find out who exactly has got it and who not, there would be absolutely no point in trying to lockdown anybody except those who were seriously ill or vulnerable.

And even then, nobody is talking about the several cats and dogs that are known to have tested positive. As when we are told humans caught this off pigs, as cats and dogs do things like sneezing, producing sputum, salivating and so on, it is hard to believe what the biological mechanism is that can allow pigs to pass it to humans but not cats or dogs.

But moreover, we need to know very seriously why this simple random testing of the British public was not done long ago, and I cannot answer that question, but it certainly fuels the fires of a million conspiracy theories, that won’t likely be as easy to dismiss as usual, because it seems to have been an oversight of colossal proportions.

i.e. that without such easy to acquire evidence, governments should mass imprison their own people, in a way that is undoubtedly without precedent in the history of human life on the planet, so literally shall we say, an event that has only happened once in a million years (depending on what we call a human).

Evan Caesars did not dare to do such a thing to their own people.

But leaving the human rights issue aside, in all this discussion of the complexity of modelling this thing, which I fully accept – thus vastly increasing the chances that someone is making a very wayward and misleading or not proven even vaguely reliable estimate and presenting it to government – the awful thing that is still not being much discussed, is what the health and economic costs are going to be due to this lockdown.

So given as TC has explained in this and the linked article how incredibly tricky it is to produce a model just to try to determine the true infection and death rates are nationally, just imagine the number of variables that would be needed to work out an economic and social/psychological/medical model of what’s currently going on – an event with as far as I can see no parallel in known human history.

The way human life and the economy works on this planet is like the most vastly complicated machine, full of almost countless inputs and outputs and interrelated systems of all kinds. Including international ones also of a kind totally unprecedented in human history.

So I suggest it would be virtually impossible to model it, and make reliable predictions, so to make such a fundamental change to that my guess is very fragile system, by effectively more or less shutting the whole system down, suggests it might be dangerous in the extreme.

But the most obvious threat seems to be to small businesses and the millions employed in them, and that on its own would seem sufficient to create a very serious and long lasting economic disaster, potentially creating millions more employed, just in the UK.

So we seem to be stacking on one side of the scales, trying to prevent about the same number or order of respiratory tract infection disease deaths that usually die annually anyway, around 17,000, regardless of how we try to stop it (which frankly, we usually don’t much, as compared to this current near infinitely zealous effort), and on the other scales we have the lives, hopes and dreams of about 60 million, which now are all being put seriously at risk, for all we know.

The calculations of deaths and risk of deaths also don’t take into account the mass panic and fear that has been caused by these lockdown measures.

Frankly, I think it is fair to describe it as terror for many – especially those most at risk from an virus, let alone COVID-19 – the old and vulnerable.

Who are the most terrified of this unprecedented event, in which they are now scared in their millions (there are millions of people over 60/70) of instant death, of never seeing their relatives or friends again, or even the world outside their doors, likely only to briefly see even that if they are carted off to hospital and taken into an ICU, not allowed relative visits and so on, and assuming in terror they are probably going to die.

I mean I wish Boris Johnson a speed recovery as much as anybody else, but I find it hard not to believe that the enormous stress of having to make decisions about this matter is not at least partially responsible for his hospitalisation.

Decision of almost Biblical proportions, more or less forced on him by so called experts, who are not even all united in their beliefs/thinking – e.g. the Oxford study – and Mr Johnson is being forced to decide between a few thousand dying who may or may not anyway, or maybe millions having their lives destroyed if he carries this unpredictable experiment on.

I mean that is probably the point at which I take issue with Peter Hitchens – I don’t think anybody yet knows what the economic damage is.

And I suspect if they did, and it is going to be very bad, I think most of the population if they could see that future, in which even well known journalists might be getting furloughed and never come back, almost everyone concerned might decide this should be stopped immediately – I mean really tomorrow morning, life should be 99% what it was 3 weeks ago, due to the terrifyingly destructive effect on millions of lives (probably 90% of the population at least) this may have if it is continued even another day.

Daniel Smallwood
Daniel Smallwood
4 years ago

It’s my impression that most testing in the UK is of people presenting themselves at hospitals with Covid-19 like symptoms, in which case it’s surprising that the positive test rate is not much more than 20%. This suggests large numbers of Covid-anxious testees. Of the wider population, uninfected, asymptomatic or mild, I don’t think we currently have any idea at all. The only answer would be decent random testing of the population.

Pat Duignan
Pat Duignan
4 years ago

You fail to recognise that the UK’s appalling failure to implement testing biases the results. The UK until very recently had the ability to process only 12,500 Tests per day and so drastically limited them. New Zealand with 1/10th the population of the UK had the capacity to process 5,000 test per day at that time! At the time when the UK had undertaken a total of 50,000 tests New Zealand had undertaken 29,000. Naturally the very limited test number undertaken in the UK resulted in a high proportion of positives because the testing was confined drastically to those where there was a high likelihood that they had Covid19. Norway has large scale testing and it’s testing indicates a far smaller level of Covid19 has been experienced by the population to date.