I really wish people would develop some terminology to differentiate b/w the antibody screen and the viral test. A cheap and rapid viral test would significantly change the course of the pandemic but this is not it.
So because this tests for antibodies, are you saying someone could test positive, but it's because they got it like a month ago? (e.g. this is more similar to those tests you have to take before you start at a hospital; so they can see what booster vaccines you need?)
Yep, the fundamental problem is that the antibody test only triggers positive well after you would already be infectious probably for days, and most likely symptomatic already at least. So it can't curtail the outbreak, though as others are pointing out it can be very useful to clear people after they are infected.
Just knowing: "yeah you caught it and recovered, get back out there, you're low risk" would be super helpful. Would allow you to go volunteer at a minimum.
Yes, that is more like the breakthrough we need. The problem I guess is that anything new like that is a huge long way from being readily scaled up, mass produced and distributed at the level necessary. Still, it's not out of the question which is why its particularly frustrating to see hundreds of rapid test articles pop up but no way to differentiate the ones that would really make a difference.
There was another thread about someone at a University making a testing kid and there were replies of scientists explaining all the materials needed.
My question about these. This one specifically says it tests for "COVID-19 IgM / IgG" anti-bodies. Could these kits also test positive for other non-SARS-cov2 Coronaviruses in the family, or are they accurate enough they only test for SARS-cov2?
One suggestion: put figure 1A from the test kit study on your page in huge / reactive size. Many stakeholders concern themselves with the mechanism of the test — if you make it easy for them to know how it works, then they’re more likely to buy. Way to go
The tests are designed to only test for SARS-CoV-2, we've trialed them against other coronaviruses and common viruses in general. Will differentiate.
Will reveal tech on swab, but it's not the swab itself, it's combined with a reactive reagent in the kit to show clear results.
Test differentiation is clear to medical professionals, but perhaps not general public, this product is only to support professionals. We are in development of a cheap and rapid antigen test that can detect ASAP before symptoms and buys us a few days headstart. It's coming.
Testing for antibodies is still very important as people continue to shed a viral load post recovery in many cases and this should be a clear enough reason for that person to enter isolation of some description.
Working the hardest with what we have and will keep moving forward.
That's the rtPCR expensive test for early detection. Self-swab of the front of the nose is now possible without a nasopharyngeal swab.
A 15 minute test cannot possibly do rtPCR which takes 45 minutes minimum, so this could only be an antibody test for who has recovered or is recovering. It's totally worthless in a clinical setting.
In Norway if you have symptoms but not been tested you are allowed in the shops and work again one day after symptoms subside as "the chance of infecting someone drops drastically at the end of the infection" [1].
If you have a positive test you still have to stay in for seven days but they don't tell you why since the argument for the non-tested should also apply to the ones with a positive test.
However there is no hard data here but I think these anti-body tests are better for statistics than prevention. Maybe you could even get a "corona passport" so that when you have had it and become immune you have more freedom.
Only people without symptoms and with proof of antibodies should be allowed to move freely. Everyone else who isn't an essential worker should be isolated until there is a vaccine. It's the only way to make H0 stay below 1.0 and stop this crisis fastest. There are too many people flaunting the isolation orders and risking the health of everyone for their arrogant, selfish recklessness.
This is meaningless. rtPCR is the gold standard, which this obviously isn't. Testing for antibodies is clinically-worthless except for who is now immune or had it in the past and doesn't need a vaccine.
If we have the virus dna, why can't we just have a pcr or similar kit to home test a swab? Someone can make their own pcr for $25 and synthesize significant segments of cov2 for a couple hundred dollars. I don't understand why cov2 testing is so difficult.
What I meant is that I agree that one can build a simple
thermal cycler (I doubt it can be done for $25 but does not matter). I know how PCR works, what I do not understand is the claim of synthesizing reagents at home, especially template and primers.
The test proposed on the website is NOT currently approved for use in Australia, and is unlikely to be approved for self-test use.
Quote from TGA:
"NOTE: In Australia, the supply of self-tests for most serious infectious diseases, including self-tests for COVID-19, is prohibited under the Therapeutic Goods (Excluded Purposes) Specification 2010. Testing for serious infectious diseases is best conducted in conjunction with a healthcare professional who can provide appropriate advice and treatment if required."
>The test proposed on the website is NOT currently approved for use in Australia
I can see it right there in the table. So it's definitely approved. On the self-test bit, of course not, even the manufacturer stresses this is for professionals to use. Not lay people.
There are many companies coming out with antibody test kits. They're useful, but not for diagnosis. They're for telling who's had the disease and is probably now immune. Virus negative and antibody positive should indicate that someone is no longer infected, now protected, and need no longer isolate. Hopefully.
My limited understanding of this is that IgM antibodies appear against viral infections quite early in the infection. So this could still be useful for identifying people with active infections. Correct me if I am wrong!
Can someone more experienced with clinical work give their thoughts on the sensitivity and accuracy of this test? The group's paper reports:
>sensitivity was 88.66% and specificity was 90.63%
My understanding is that this test will miss 12 in every 100 covid-19 infected individuals, and it will return a false positive for 10 in every 100 patients.
My intuition says this is more than acceptable for a rapid test. But I would love to see the opinion of someone with some clinical experience.
Remember, there’s probably a similarly high false negative rate. Also, take into account the number of people who won’t be going home, but will instead be flooding into ERs.
If you want to cause true mass hysteria and quickly overwhelm our already strained medical infrastructure, tell 10% of the population that that have SARS-CoV-2.
You need to stay home. There’s no way around it—certainly not for $20.
I don't think the quote you provided says anything about false positives; at least, I really hope it doesn't. A 10% false positive rate would be ridiculous. If 0.1% of the population has SARS-CoV-2, a test has a 10% false positive rate, and you get a positive result on that test, you only have about a 1% chance of actually being infected.
That's a little unintuitive, but here's how it breaks down, without taking into account false negatives:
* 10% of the population will receive a false positive
* 0.1% of the population will receive a true positive
Which category are you more likely to fall into--10% of the population or 0.1%? You're 100 times more likely to fall into the false positive category.
Can't find the article now but there was two fast tests that got forbidden in Sweden yesterday because they miss a large percentage of the cases making it meaningless in best case or dangerous to others in worst case. You think you don't have corona so you go to work anyway even though you feel a bit sick.
A single false negative is worse than a single false positive, but false positives affect far more people with equivalent percentages. They’re both problematic, but in different ways.
Our medical infrastructure is already strained. If you tell 100x the number of actually infected people that they are infected, everything will grind time a halt.
It’s important to find a balance. A single false negative could be disastrous—but so could 1000 false positives. If the false negative and false positive rate are both 10%, and 0.1% of the population is infected, there will be 1000 false positives for every false negative. Both of those are going to cause huge (but very different) problems.
Nope, that is exactly what they are saying. It really is 10% false positive. They tested 128 negative blood samples and got a positive result for 12 of them. There is a frequency table at the end of their paper (0). So I guess you are right that if you sample a population of 1000 people (and only 1 actually sick) you would get 100 positive results, and 88% chance of detecting the one person who is actually sick.
I only know the rate for the current German PCR Sequencer. They are at the industry standard of 3%, and require two tests per swab/person to get the higher accuracy required to avoid false negatives.
Why does this company sell N95 masks as well? For a company selling 3 products that's a strange combination - unless this company is just some sort of distributor. In the latter case, why is this news?
Yup. Two days ago I received an email from someone asking to advertise hxxp://www.masktouch[.]com/ on our site. Compare the masks on there to the masks on the site posted here--they're nearly identical. Compare the source code--same style, similar comments. Even the red buttons on the sites are similar.
I went through all the "certification" docs on both sites. I found nothing that actually certified either of them for anything. Anything that would've been remotely meaningful was censored (blurred).
If I'm making incorrect assumptions based on circumstantial evidence, what choice do I have? There's nothing on these sites substantiating their claims. If this is legitimate, you need to publish a complete paper trail that readers here can verify.
Edit: To clarify: Calling it a "scam" might be a bit harsh, but it's definitely misleading. There's nothing special about these tests, and there's a reason they're cheap and not in widespread use--they're not going to do what most people expect. This post sums it up nicely: https://news.ycombinator.com/item?id=22642421 All these sites are probably just dropshipping the same items from the same factories at heavy markups.
This is probably a scam, but I'm on the look out for the reason why the market went up like this this week. It's either the typical bounce, or there is some information to come that insider traders moved on.
Look at the “our team” section. No one on the team has any medical or biological experience. If they did, I’d be researching them further and contacting them independently to verify. But there’s no one worth following up. Flagging this as a scam.
This company seems to have a history of producing such tests (for Ebola, Influenza and other pathogens) and there are numerous pubmed publications about them...
Chinese Bioeasy also offers those, Spain and Poland tried using them, instead of 80% sensitivity its more like 20-30%. Works if you are willing to burn 5 tests per person.
53 comments
[ 3.0 ms ] story [ 93.7 ms ] thread(And the antibody test is also useful to figure out which nurses/docs are safe to send back in?)
https://www.straitstimes.com/singapore/health/singapore-scie...
My question about these. This one specifically says it tests for "COVID-19 IgM / IgG" anti-bodies. Could these kits also test positive for other non-SARS-cov2 Coronaviruses in the family, or are they accurate enough they only test for SARS-cov2?
Real quick response.
The tests are designed to only test for SARS-CoV-2, we've trialed them against other coronaviruses and common viruses in general. Will differentiate.
Will reveal tech on swab, but it's not the swab itself, it's combined with a reactive reagent in the kit to show clear results.
Test differentiation is clear to medical professionals, but perhaps not general public, this product is only to support professionals. We are in development of a cheap and rapid antigen test that can detect ASAP before symptoms and buys us a few days headstart. It's coming.
Testing for antibodies is still very important as people continue to shed a viral load post recovery in many cases and this should be a clear enough reason for that person to enter isolation of some description.
Working the hardest with what we have and will keep moving forward.
Thanks for your interest and support!
A 15 minute test cannot possibly do rtPCR which takes 45 minutes minimum, so this could only be an antibody test for who has recovered or is recovering. It's totally worthless in a clinical setting.
Any hard data on how long this is the case?
If you have a positive test you still have to stay in for seven days but they don't tell you why since the argument for the non-tested should also apply to the ones with a positive test.
However there is no hard data here but I think these anti-body tests are better for statistics than prevention. Maybe you could even get a "corona passport" so that when you have had it and become immune you have more freedom.
[1] https://www.fhi.no/nettpub/coronavirus/rad-til-personer-som-... (Norwegian)
Ok
> and synthesize significant segments of cov2 for a couple hundred dollars.
Please tell us more
https://www.tga.gov.au/covid-19-diagnostic-tests-included-ar...
Advice re Legal supply of COVID-19 diagnostic tests in Australia
https://www.tga.gov.au/legal-supply-covid-19-diagnostic-test...
Quote from TGA: "NOTE: In Australia, the supply of self-tests for most serious infectious diseases, including self-tests for COVID-19, is prohibited under the Therapeutic Goods (Excluded Purposes) Specification 2010. Testing for serious infectious diseases is best conducted in conjunction with a healthcare professional who can provide appropriate advice and treatment if required."
I can see it right there in the table. So it's definitely approved. On the self-test bit, of course not, even the manufacturer stresses this is for professionals to use. Not lay people.
>sensitivity was 88.66% and specificity was 90.63%
My understanding is that this test will miss 12 in every 100 covid-19 infected individuals, and it will return a false positive for 10 in every 100 patients.
My intuition says this is more than acceptable for a rapid test. But I would love to see the opinion of someone with some clinical experience.
False negative would be more bad or am I missing something?
If you want to cause true mass hysteria and quickly overwhelm our already strained medical infrastructure, tell 10% of the population that that have SARS-CoV-2.
You need to stay home. There’s no way around it—certainly not for $20.
That's a little unintuitive, but here's how it breaks down, without taking into account false negatives:
* 10% of the population will receive a false positive
* 0.1% of the population will receive a true positive
Which category are you more likely to fall into--10% of the population or 0.1%? You're 100 times more likely to fall into the false positive category.
A false negative is worse than a false positive.
Our medical infrastructure is already strained. If you tell 100x the number of actually infected people that they are infected, everything will grind time a halt.
It’s important to find a balance. A single false negative could be disastrous—but so could 1000 false positives. If the false negative and false positive rate are both 10%, and 0.1% of the population is infected, there will be 1000 false positives for every false negative. Both of those are going to cause huge (but very different) problems.
[0] https://rapidward.com/assets/downloads/Li_et_al-2020-Journal...
What we need are PCR tests...
There is literally NOTHING to be found about "rapidward" before 2 days ago.
I'm very skeptical to your added value, and not just a middle man with a marketing team, I don't care that it's non-profit.
You got 2 directors, a head of marketing, a head of Comms & PR, and a social media impact advocate...
I went through all the "certification" docs on both sites. I found nothing that actually certified either of them for anything. Anything that would've been remotely meaningful was censored (blurred).
If I'm making incorrect assumptions based on circumstantial evidence, what choice do I have? There's nothing on these sites substantiating their claims. If this is legitimate, you need to publish a complete paper trail that readers here can verify.
Edit: To clarify: Calling it a "scam" might be a bit harsh, but it's definitely misleading. There's nothing special about these tests, and there's a reason they're cheap and not in widespread use--they're not going to do what most people expect. This post sums it up nicely: https://news.ycombinator.com/item?id=22642421 All these sites are probably just dropshipping the same items from the same factories at heavy markups.
> Australian firm Rapidward said that five days ago it began supplying Covid-19 blood tests direct from China to Switzerland"
...meaning 7 days ago. But company "RapidWard" was only registered in Australia in the last 1 or 2 days: https://abr.business.gov.au/ABN/View?abn=67639947970
That article also says:
> it is registered with the Medicines and Healthcare products Regulatory Agency (MHRA) in the UK
This is easily checked:
https://medicine-seller-register.mhra.gov.uk/search-registry...
> 0 results found.
>Raw Whois Data
Domain Name: rapidward.com
Registry Domain ID: 2504220709_DOMAIN_COM-VRSN
Registrar WHOIS Server: whois.godaddy.com
Registrar URL: http://www.godaddy.com
Updated Date: 2020-03-17T05:43:49Z
Creation Date: 2020-03-17T05:43:49Z
Registrar Registration Expiration Date: 2021-03-17T05:43:49Z
Registrar: GoDaddy.com, LLC
...
Registrant Name: Registration Private
Registrant Organization: Domains By Proxy, LLC
Registrant Street: DomainsByProxy.com
Registrant Street: 14455 N. Hayden Road
Registrant City: Scottsdale
Registrant State/Province: Arizona
Registrant Postal Code: 85260
Registrant Country: US
The test in question has been fast tracked and approved by European regulations.
Here is their product sheet: https://www.corisbio.com/pdf/Products/SARS-COVID-19_20200324...
Here is their page: https://www.corisbio.com/Products/Human-Field/Covid-19.php
It seems to be quite similar but on this one you can see their test results and how effective it is.
Evaluation 1(n=99),95%,confidence interval:
-Sensitivity,60%,(40.7 to 76,8%)
-Sensitivity threshold,85.7%,on sampling with CT under 25(62.6 to 96.2%)
-Specificity,100%,(93.4 to 100%)
-Positive predictive value,100%,(78.1 to 100%)
-Negative predictive value,85.2%,(75.2 to 91.8%)
-Agreement,87.9%(87/99)
Evaluation 1(n=1 32),95%,confidence interval:
-Sensitivity,60.3%,(48.1 to 71.3%)
-Sensitivity threshold,76.7%,on sampling with CT under 25(61 to 87.7%)
-Specificity,98.3%,(89.7 to 99.9%)
-Positive predictive value,97.8%,(86.8 to 99.9%)
-Negative predictive value,66.7%,(55.7 to 76.2%)
-Agreement 77.3%(102/132)
It was developed by the ULB university in Brussels. It has been tested on 250 patients in two of Belgian's main university hospitals.
https://www.brusselstimes.com/all-news/belgium-all-news/1027...
This company seems to have a history of producing such tests (for Ebola, Influenza and other pathogens) and there are numerous pubmed publications about them...
For initial infection, rtPCR is the gold standard and takes at least 45 minutes as of writing.
https://www.surescreen.com/products/covid-19-coronavirus-rap...