Hacker News new | past | comments | ask | show | jobs | submit login
Ten-Minute Coronavirus Test for $1 (bloomberg.com)
446 points by montalbano 13 days ago | hide | past | web | favorite | 183 comments





From their website [0] my guess is their test is ELISA [1] based or similar method of detecting if the human body has had an immune reaction to the virus in the amount that is detectable. Where as the more expensive test is PCR based [2] that requires the heating-cooling cycles and machines to do it.

The PCR-based test basically take the RNA of the virus converts it to DNA and then doubles the amount of DNA every heating cycle. Assuming we are talking about real-time PCR the PCR cycle of double is what takes time.

I have not seen the data on the incubation period and the immune response, to know if these immune detection tests will be able to provide early detection for asymptomatic case . It's probably for symptomatic cases and wanted to ensure it's not flu or another infection and it's COVID-19. This is not to say the test is useless. It actually great inexpensive way of testing the population with symptoms and using more expensive and timely test for early detection. Reducing the overall burden.

worth mentioning are other initiatives that are focusing on cheaper and faster PCR process (though with potential of higher false negative and positive results) [3]

[0]https://mologic.co.uk

[1]https://en.wikipedia.org/wiki/ELISA

[2] https://en.wikipedia.org/wiki/Real-time_polymerase_chain_rea...

[3] https://twitter.com/JMRothberg/status/1238115465467133959


RT-qPCR is indeed the common detection method right now. RT (reverse transcriptase) converts RNA->DNA, PCR (polymerase chain reaction) doubles DNA by replicating it over and over in heat cool cycles. To a first approximation, you get 2^n copies of the RNA (it was converted to DNA) where n is the number of heat cool cycles. By adding a fluorescent tracer, you get the q (Quantitative), so you can see how much DNA exists at each cycle. You can then fit a curve and see how much RNA (via its DNA complement) existed in the original sample. It's not actually that slow, you can do this in a couple hours if you have the right setup. The bottleneck is you do need time for the heating and cooling cycles. You also need to extract RNA from the sample first, which could be another bottleneck. I'm not totally clear on the relative cost, a good ELISA would eventually be cheaper, but qPCR itself isn't very expensive. The benefit is that qPCR is extremely sensitive, you WILL see if there is virus in the bloodstream. You can even use swabs and avoid blood altogether. Also, the minute you have a viral genome, you can make a qPCR test, you just need the sequence. That said, if you cleared the infection you will be negative under RT-qPCR.

An ELISA works differently. In this case you would present a part of the viral protein, the patient's antibodies will bind to it, and you then use another antibody (this one is conjugated to some sort of readout method, fluorescent or biochemical) which binds to generic human antibody. Now you have a stack: viral protein <> patient antibody <> readout antibody. You read out the signal provided by your readout, usually some sort of colorimetric thing (see home pregnancy tests, also an ELISA, slightly different configuration though). This works fast, can be manufactured in bulk, BUT requires do you have a protein(s) (in this case the viral protein) that is/are universally recognized by patient antibodies. It's a little trickier to develop. You also need to produce that protein at scale which can take a little time as well. Major benefit: with a good ELISA you WILL see not only whether someone IS infected but whether someone WAS infected for some period after illness regardless of symptoms. This is essentially the only way you'll get a really good number for baseline infection rate. That said, you likely actually need a blood draw.

In short, we need both, we probably could do a lot better than we are doing with qPCR in testing volumes and we could also really use an ELISA.

Source: PhD biochemist, have personally run these assays in various forms.

EDIT: Acronym expansion, source


Maybe one should note, that there are two possible Elisa based strategies. Yes, You can present some viral protein to detect antibodies produced by the body after serum conversion, as you said. This serum conversion takes at least a week to happen and you need a blood sample. On the other hand you can also immobilize antibodies specific to the viral proteins and directly detect the virus. This is possible immediately after infection and fluids like sputum or saliva can be used. Both strategies are under development for quick tests right now.

This is all true, but detecting the viral proteins directly with reasonable sensitivity and specificity is much more difficult. The reason is that there is just a lot less of it to detect. After seroconversion, the amount of antibody against the virus can be very high and relatively easy to check.

+1 the major advantage of RT-qPCR is it's a massively amplified test. Tiny amounts of virus can be detected. When you're detecting a patient's antibodies to the virus, the patient is producing massive amounts of it. That said, after clearing the infection that production will decline, but still should be detectable at least for a while.

Direct detection of viral particles is substantially harder because for N viral particles you basically get O(N) signal (to put it in CS terms). There are some caveats, your readout antibody can generate an exponential signal for example.

You still need to be presenting the right part of the capsid and have the right material. In the direct ELISA scenario, you need 2 antibodies. One to bind the capsid to your surface, the other to bind it again and either have a readout attached directly or be bound by yet another antibody with the readout (likely the latter). Fortunately, viral capsids are repetitive so you might be able to use the same antibody for both. You ALSO need those antibodies to have low cross-reactivity with sputum or whatever other rough sample you have. When you are looking for a patient's antibodies, the patient's immune system has already taken care of that for you. This is all tricky, obviously it's not THAT hard, but it's not trivial.

RT-qPCR is directly a O(2^N) signal. It's extremely sensitive, requires very little assay development, can be made very specific, can work with little sample prep, so it's a much better front-line, first-to-go test than direct ELISA. Still, we'd all love an ELISA, particularly to patient antibodies.


I definitely agree about the superiority of RT-PCR and the drawbacks of direct detection. Nevertheless, seroconversion can take between 5-12 days from symptom onset [1]. Therefore it's important to outline the different ELISA strategies (Thank you for doing so!), when discussing diagnostic tests.

[1] https://doi.org/10.1101/2020.03.05.20030502


> RT-qPCR is directly a O(2^N) signal.

The signal is actually more like N*2^C where C is the number of cycles (usually around 30). So a huge amplification, but still linear in the starting amount, i.e. O(N). That's actually more useful, generally, since ratios of amplified material stay approximately constant during amplification.


Correct, got my math a little off in haste. Constant gets pretty huge which give massive signal, thanks for the correction :)

On a related note, UC Berkeley is looking for volunteers to help them set up a test center using these techniques - https://innovativegenomics.org/covid-19/

Yes, there are of course drawbacks. But the direct detection is applicable at the same stage of infection as a rt-pcr and therefore usefull for immediate disease diagnosis. Indirect tests after seroconversion are the easier alternative, and will probably be used extensively in the aftermath, to get solid epidemiological numbers.

One disadvantage to RT-qPCR I forgot to highlight, and an advantage to direct detection ELISAs (and ELISAs in general) is that ELISAs are relatively robust when fully developed. RT-qPCR requires a lot more gear. See pregnancy tests. Sits on a shelf for ages, all you need to do is pee on it, works reliably, needs no extra equipment.

Are the tests used in [1] (real-time RT-PCR) the same?

The false negative reported several times from the same patient is interesting, but may be more due to testing source (upper respiratory fluid vs. sputum or lower respiratory sources or blood).

[1]: https://wwwnc.cdc.gov/eid/article/26/6/20-0299_article


Yes, the "q" is often used instead of "real-time" to avoid confusion with RT, which can also be "reverse transcriptase". RT in these cases is the enzyme that converts RNA to DNA.

What's the approximate setup and validation time once you have a sequence?

(that is, what's the scale of the fuck up that the machine testing is still being brought online)


I used to set up and run a large plate of qPCR (we didn't use the RT step in our particular use case) in an hour or so. Mind you this was for something like 200 samples. It took 2-3h to run the actual machine. If I was setting up such an assay, I'd order a bunch of different primers (you need DNA to make DNA) and run a bunch of plates with as many samples as I could to see what primers worked best. You'd include a bunch of negative controls, ideally some controls from patients/cultures infected with similar but not identical virus (for example a different coronavirus). To develop an RT-qPCR assay would take a week or two (rough guess), I think it took about a week to develop the WHO assay.

From what I can tell (mostly anecdotal and other data from friends I trust) the CDC assay used bad primers, they were noisy, showing a bunch of false positives. They also used a potentially dodgy fluorescence readout. I used to use the same readout, but our scenario was full of internal controls and the input sample was far more consistent. For human sample data a different fluorescent method that is more reliable should have been used.

Important ways this can be screwed up: it's helpful to use the same machine, the same protocol every time. This is because qPCR is an exponential assay, which means small errors can have big repercussions.

TLDR: Scale of fuckup here is massive. This really isn't that hard, and we could just have used what the WHO/others were using. I have no idea why they decided to go their own way.

more info: https://www.propublica.org/article/cdc-coronavirus-covid-19-...


Five days ago, it was reported that a German biotech company had already produced 4 million tests.[0]

"The tests use what’s called the polymerase chain reaction, a diagnostic method recommended by the WHO that amplifies the virus’s genetic code so it can be detected before the onset of symptoms. The kit comes with two vials: a primer to help detect an infection, and a synthetically engineered piece of the virus, which labs use to produce a surefire positive match to ensure their machines are working correctly. A lab technician combines these ingredients with a patient’s mucus sample—usually from a throat or nasal swab—and results are usually available in a few hours."

Coincidentally, I recently discovered the burgeoning biohacking scene in which many folks are already using affordable PCR thermocyclers in their home labs to do DNA amplification and crispr-cas experiments. I do fungal sequencing for phylogeny myself.

Given the simplicity of the testing process using PCR, why does it seem like the US is moving so slowly with getting tests in peoples hands? Is it the primer or control virus synthesis? Is it the manufacturing process? Is it red tape? If so, which specifically? Is it poor communication? Is it a confluence of factors?

Why isn't every local university or lab working together to at the least, copy the Germans protocol and get to work on manufacturing ASAP? Why isn't there a line of people outside these labs ready to drop-off their sample and get the results online? Where is the American version of the aforementioned biotech startup?

This epidemic and the general response has only reinforced my belief in the future of citizen science and the widespread accessibility of indie biohacking tools and methodologies. There are just too many misaligned incentives in pharma research, leaving huge gaps in the ability to bring new tests and drugs to the market. Look no further than the antibiotic resistant drug problem.

[0]https://www.msn.com/en-us/finance/companies/a-berlin-biotech...


As far as I can tell, and admittedly this is conjecture based on piecing together a bunch of off-hand comments by official spokespersons and a few industry insiders I talk to regularly, it's because the U.S. decided that everyone needs to use the same exact test. There is some validity to that approach in that it makes the data a lot better for comparison tracking. But, they compounded that somewhat defensible decision to require that everyone uses the same exact test kit. This turned into a disaster when the chosen supplier couldn't produce the reagents in the required time/amount.

This an aside, but this seems like a textbook example of government regulation of healthcare not only failing but getting in the way and actively causing harm. There are literally 1000s of generally certified labs that they could have said very early on "go to it if you have this certification" and instead they plodded through this top-down heavy handed mess and made this pandemic many times worse.

In the US you have this combination of heavy over-regulation and lack of public provisioning. It's like the worst of both combined. Between this, cannabis, ketamine, and the opioid crisis, I absolutely do not understand why there isn't an uproar over overregulation in healthcare getting in the way.


None of those generally certified labs have any clue how to develop a new test. Not even a test that is exactly the same as a test they already have. There are 1000's of them, because you need very little training to push some buttons and enter data into forms. Those generally certified labs did what they are supposed to do. They looked to the government for guidance on what to test and how. The gov't made many mistakes: disbanding of the CDC pandemic response team. Putting anti-science Pence in charge. Forming a CDC team that was heavy on science and management but light on epidemiology. Not following the WHO's lead on their already established and working test. Listening to pharma's explanation about what they will be able to do in 12-18 months and confusing that with a statement about what they can do now. These are all mistakes that didn't have to happen and happened because morons are in charge.

As others have mentioned, false negatives and false positives are both annoying problems. The standard diagnostic kit for example comes with a positive control, which frankly is absolutely essential. Synthesizing a primer is trivial, a positive control is not as trivial (especially if it's an RNA sample). It's also an RT PCR, which means it's extremely easy to contaminate samples among each other (so you can never be sure if even one of your samples actually has the virus whether ALL of the samples have it or not). Further, you can't trust regular gel results too much, you need a qRT-PCR cycler which most labs would have but probably not these hackers.

Also were talking about a virulent pathogen sample. Do you truly want to stand in line and collect random snot samples with covid virus without proper training and PPE?


There's no stopping anyone from getting the primers (small pieces of DNA) and doing a qPCR yourself. You'll also need reverse transcriptase (the RT, converts RNA->DNA) and a way to purify RNA from a mix of cells/spit/junk. These are all relatively inexpensive, standard molecular biology reagents. There are some shortages of RNA extraction kits, but a good molecular biologist can do without one.

What you might be missing is consistency, which is what the medical system is worried about. You want to avoid false negatives AND false positives when it really counts. Getting that consistency is where the regulations come in. Nowadays I think it's appropriate to relax them a little in the interest of the emergency.


It's going to be hard to get the assay working and trusting without a positive control though

Correct! Though making that template isn't that difficult either, you can order the DNA version and generate the RNA from that.

> Given the simplicity of the testing process using PCR, why does it seem like the US is moving so slowly with getting tests in peoples hands?

This has already happened in the US, with 1.5 million tests ready to ship, with emergency approval from the US government. In April they will be producing 5 million tests per week. https://ir.thermofisher.com/investors/news-and-events/news-r...


Ready to ship and April are much different than the countries who have been, and continue to test, upwards of 10Ks+ folks a day. As of today, the CDC is reporting only N=4225 completed tests in CDC labs and N=20907 in other US labs.

General description of Mologic/Sherlock Bioscience technology:

"Mologic said it will invest a part of an expansion to its $4.9m grant from the Bill and Melinda-Gates Foundation to finance access to Sherlock's INSPECTR™ platform, which the Cambridge-based US company licences earlier this year from the Wyss Institute.

Under the agreement, Mologic will provide its lateral flow-based platform CARD, which has been demonstrated to provide a 1,000-fold improvement in sensitivity over current technology, and its immunoassay platform ELTABA for enzyme activity detection. Sherlock Bioscience will provide its synthetic biology-based INSPECTR™ platform, which consists of a DNA hybridization-based sensor that can be easily programmed to detect target nucleic acids (DNA or RNA) with single base pair specificity, coupled with a paper-based synthetic gene network that translates the sensor’s detection into a bioluminescent signal that is easily visualised or captured on instant film. Crucially, this process can be done at room temperature and does not require any instrumentation."

https://european-biotechnology.com/up-to-date/latest-news/ne...


"What is SHERLOCK™?

SHERLOCK is an evolution of CRISPR technology, which others use to make precise edits in genetic code. SHERLOCK can detect the unique genetic fingerprints of virtually any DNA or RNA sequence in any organism or pathogen. Developed by our founders and licensed exclusively from the Broad Institute, SHERLOCK is a method for single molecule detection of nucleic acid targets and stands for Specific High Sensitivity Enzymatic Reporter unLOCKing. It works by amplifying genetic sequences and programming a CRISPR molecule to detect the presence of a specific genetic signature in a sample, which can also be quantified. When it finds those signatures, the CRISPR enzyme is activated and releases a robust signal. This signal can be adapted to work on a simple paper strip test, in laboratory equipment, or to provide an electrochemical readout that can be read with a mobile phone."

https://sherlock.bio/technology/


Video of paper-based bioluminescence detection (Wyss Institute)

"If the sample contains the RNA of the Zika virus, the test area turns purple."

https://www.eurekalert.org/pub_releases/2016-05/uot-rld05061...


An alternative: "Anitoa systems, a Silicon Valley biotech company has demonstrated a solution for rapid on-site nucleic acid test of the 2019 novel coronavirus (2019-nCoV). Anita combines a high performance ultra-portable Real Time Polymerase Chain Reaction (qPCR) instrument called Maverick, with an innovative “one-step” reverse transcription qPCR (RT-qPCR) reagent for testing 2019-nCoV."

http://www.anitoa.com/docs/press-ncov.pdf


> test for the virus in Nigeria, Africa’s most populous nation, with about 200 million people.

> its saliva and finger-prick kit could be ready for sale by June for less than $1 apiece

> Mologic and the Institut Pasteur have joint capacity to produce 8 million tests a year

every little helps, and a low-cost test would obviously be a game-changer. but let's keep things in perspective and be realistic. it sounds quite hypothetical right now. there's no indication it works reliably yet.


Decision Diagnostic a blood sugar test developer company is developing a handheld device and the cost of each test would be less than $7. They are hoping to deploy 200K unit by May. https://finance.yahoo.com/news/decn-updates-corona-virus-tes...

We are royally screwed if this thing hasn't played out by June.

Economy is screwed either way, but for the people it would be better if it goes slowly - then our healthcare systems can deal with it.

If it just goes unchecked exponentialy for 2 months then significant fraction of world population will die. Like 2-digit percentage.


That does not line up with any data collected before.

Not only that, demographically the hardest hit population (80+) are not in the time of their lives where they spend the bulk of their money nor are they in the most productive time of their lives.


How could a 2-digit percentage die when the fatality rate so far is substantially lower (~3% last I saw)?

Because that is the fatality rate range (0.9-5%) when everyone gets a ventilator/ICU bed when needed.

From what I've heard 12% have complications that would benefit from medical help, and 5% become "critical" and are likely to die without help.

See https://youtu.be/BYTFk34nhoI from the 10 to 12 minute mark.


I believe fatality rate is calculated using diagnosed cases, while the real number is probably way higher so fatality rate is probably close 0.9-1%.

Depends how much we're able to slow it down. For example Canada had first case in January, and 2 months later is only up to ~500 confirmed cases. It is still exponential but much slower. Hopefully the new measures put in the past week will slow it even more.

At which point, once we have fast and reliable testing, we can actually start having people go out more and reduce the isolation, since we can count on better testing to stop the spread.


For example Canada had first case in January, and 2 months later is only up to ~500 confirmed cases.

My niece and great nephews in Victoria, BC just turned up with symptoms for COVID. They were told to self-quarantine and were not tested.

You cannot believe the numbers from any country which does not have a robust testing regime. That emphatically includes Canada along with the USA.


Yeah, in Denmark yesterday simple math says that 18% of people tested was infected.

That's not to suggest that 18% of the population is infected. Merely suggests that positive tests is a useless metric because people are not tested.

Number of people admitted to hospital might be a better indicator -- though this has a lot of delay.


In Canada, growth was slower for a while, but the number of (detected) cases has doubled twice in the past 5 days, which just about matches what we saw in Italy and the US once they hit 100 cases.

Sounds like what happens when one of the criteria to get tested (which is expensive) is "having been in presence of known existing COVID-19 case".

It's a symptom of massive under-testing and would be greatly alleviated by an inexpensive test.


Absolutely. Canada still only tests people with symptoms AND (have traveled recently OR have been in contact with a confirmed positive case).

Yet the messaging until recently was "no evidence of community spread, risk remains low" with very lukewarm response. Of course there is no evidence if you don't look for it.

I understand they don't want to incite panic but anyone who's been paying attention to other countries can see that community spread is inevitable unless you have strict measures and robust testing, very early on. I'm pissed our politicians have ignored lessons from other countries.


Exactly. Starting from the first case is not useful. Starting from a baseline of 100 offers a more consistent comparison. And for that, almost all countries are following the same path.

I think this site may be paywalled unfortunately (but there is just a $1 for the first month charge):

https://www.ft.com/content/a26fbf7e-48f8-11ea-aeb3-955839e06...

The second graph in there shows Canada is tracking exactly the same path as most European countries and the US.


In the White House press conference today, the head of the NIAID said don’t expect things to return to normal until July or August.

I really don't see how anyone expects this to be over by June.

Exponential growth at the current rate would have it burning through the global population by the end of June, but since exponential growth becomes logistic growth after an appreciable fraction of the population is infected, June is probably closer to the halfway point than the end date.

And that's assuming zero curve flattening. The entire point of flattening the curve is to draw it out so that hospitals continue to function for longer. But this also makes the pandemic last longer.

A vaccine isn't expected for over a year, and even then, I'd expect at least a few months of logistical difficulties getting enough manufactured, distributed, and administered, especially given the state that hospitals will likely be in.


Yes, this was what I had in mind when I made this comment.

We would be screwed, IMHO, based on the current mitigation strategies around curve flattening that are not being communicated realistically to the public. The crop of shutdowns over the weekend were billed as 2-3 weeks.. Err, what? How are we even going to know what effect that's having without massive blanket testing? Then, nobody seems to believe anything less than 8 weeks is enough; and that's full-blown lock-down not fractions of half measures.

But as you say, the just spreads it out. So what happens in 8 weeks when we still can't let up the restrictions because if we did that the number of cases would just explode again among the uninfected population? If we aren't turning the corner by June and getting the 40% of adults who can't afford $400 in an emergency back to work..

I'm sympathetic to the UKs strategy and hope the best for them.


> The crop of shutdowns over the weekend were billed as 2-3 weeks.. Err, what?

Exactly. People seem to be mistaking the incubation period (~two weeks) for the length of time the pandemic will last (months at minimum).

It's impossible to have a perfect quarantine, so if we were to all go back to normal after a month, the pandemic growth curve will pick up where it left off. Exiting the lockdown will need to be done super carefully to avoid this outcome.

As far as I can tell, this is probably going to be the new normal until we have a vaccine.


Yup. And the worst-case scenario is infections peaking next winter to coincide with regular 'flu season.

An Imperial College paper [1] out today predicts this under its most stringent suppression scenario; see the chart on page 10.

[1] https://www.imperial.ac.uk/media/imperial-college/medicine/s...


The 40%/$400 item comes from a fed survey about liquidity. It includes 17% of households with $100,000 in income. It is an outlier, other surveys put it the percentage at 20%.

That's insane, I always assumed it was for much lower household incomes. I consider 100k to be upper middle class in the US.

The flattening strategies are like holding your breath, you can't do it forever. But vaccines are not the only way in which we get better at dealing with the virus. Long before we get a vaccine we will get better not only at treating Covid-19 (which will increase the effective medical capacity) but also at keeping the gears turning without too much viral spread. With a bit of luck and a few logistical winning moves we might be able to let this play out safely not that much later than current unrealistic expectations.

Not sure why your score is so slow, you're spot-on. This is literally the global pandemic we've been scared of. It just came out of China instead of the Democratic Republic of Congo.

actually we're more screwed if it has -- that means it would have burned out world wide in 6 months and 10s of millions would have died.

You realise this isn't going to play out by June, right?

Do the math on those "flatten the curve" graphs and work out the X-axis given the projected infection rates (20%-60% of population), hospitalisation rate (15%-30%), days in hospital required (6-9) and the number of beds available.

This ends with a vaccine or a mutation that gives (partial?) immunity and a lower death rate.


How so ? Care to elaborate ?

A lot of people will have died and our economies really aren't built for long term social isolation either.

Would be interesting to use a technique like this:

https://www.google.com/amp/s/phys.org/news/2019-02-origami-d...


"masks and test kits". Epidemiologists say masks are practically useless against virus. The one that is effective with special training is useless without.

Epidemiologists say to stop buying masks only because there are not enough of them for the healthcare system and general population.

Western countries need to spend more energy on mask productions.

I heard the mask production automation lines cost $50K to $100 in China and TW. Now the price for machines and production materials are 5-10 times the normal.

But once setup, 95% - 99% of the process is automated. There are zero pollution and it is neither labor nor capital intensive. Do need the logistic and abilities to source some raw materials. If setup, the next few months would be like printing $$$$.

TW can make 10 millions masks per day now. China has production line making 110 millions mask per day. They were able mobilized the industries like war time - TW actually send enlist Army personnels to the factory to help out.


In general the switch to just-in-time for everything was a fair-weather decision. You goose can your margins when things are great, too bad about your supply chain when things are less than great.

This is probably OK for ice cream and computer hardware. Less so for medical supplies.


Right. And everyone understands this when it's the military: If you can't make more missiles at home, then you are a pawn of whoever you buy them from. We ought to have taken a similar line on a lot of medical supplies, especially those likely to be useful at whole-population scale.

Yup. In mainland China they repurposed car assembly lines to mask production (there's not much car selling anyway). Western countries need to copy at least some of those measures.

Some Chinese manufacturing (or maybe a lot, I don't know) is still based on low wage advantages. Those factories are much easier to repurpose than factories in high wage countries that are optimized like race cars to maximize worker productivity. I wouldn't be surprised if a European car manufacturer would find it challenging to switch an engine factory from diesel to petrol with less than a year run up time.

Would it be a bad idea for the government to have a reserve for healthcare workers for the next time this happens? To slowly sell off what is there, and to restock to create a buffer. And in a time of crisis, have a large amount that can only be bought by healthcare workers.

You're describing the Strategic National Stockpile. Apparently it didn't get restocked after the H1N1 pandemic in 2009. https://abcnews.go.com/US/medical-providers-fearing-equipmen...

The Washington Post said 5m out of 35m masks were expired.

I thought it's more like there's no evidence that show massive use of masks by the public is effective.

It's complicated. It doesn't stop you from getting the virus if you get droplets on your hands and rub your eyes / nose / mouth, which is one of the more common vectors. So I buy that there's no evidence it stops you from getting the disease. But it does limit the droplets you spray out as you're speaking / coughing, etc. So IMO it would be part of a general solution to limit how much you spread the disease. But self-quarantine is easier to be sure you're doing it right. Wearing N-95 masks effectively is harder than just strapping it on, and the untrained are likely to touch their faces more while they're doing so anyway. I'd be surprised if everyone wearing a fresh mask properly and washing their hands properly before touching the mask or face didn't put a major dent in the pandemic. But that's just not going to happen anyway so we can't tell.

I think the big issue is supply for medical staff. There's a limited supply of legit masks. Medical staff need masks more. We need medical staff more. And the masks are more effective with training / other procedures in place. Other than that - if the masks didn't help you, medical staff wouldn't wear them.


This is the right answer. As someone who has the virus, it absolutely helps you not spread it. As someone who doesn't, it may stop a few droplets from getting into your system right at that moment, but said droplets can still get on your hands, and 30m later when you scratch your nose, you'll still get infected.

So yes, with supply being low, it's far more useful for infected people to not spread said droplets.


The masks serve as a reminder to not scratch your nose, since it is covered. Glasses or sunglasses help for your eyes.

Over here, supply doesn't seem to be low. Lots of smoke recently meant lots of masks in the supply chain. I'm surprised I see so few worn in the shops though, especially since I spent a lot of time in Asia where masks were common during flu season.


Exactly: put quite simply the best mitigation measure is a mask, since it prevents most of the virus from getting into the air--the virus's primary medium of transmission.

Even simpler--don't wear a mask for yourself, but for others.


https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2440799/ “Any type of general mask use is likely to decrease viral exposure and infection risk on a population level, in spite of imperfect fit and imperfect adherence, personal respirators providing most protection. Masks worn by patients may not offer as great a degree of protection against aerosol transmission.”

https://www.ncbi.nlm.nih.gov/pubmed/19193267 “We found that adherence to mask use significantly reduced the risk for ILI-associated infection”

https://www.cochranelibrary.com/cdsr/doi/10.1002/14651858.CD... “Surgical masks or N95 respirators were the most consistent and comprehensive [physical intervention] supportive measures.”


I hate to sound anti-science, but with something "spread by droplets" from coughs and sneezes and spit, where we should avoid touching our faces, it sure as hell sounds like even the most basic mask could be hugely helpful.

They're trying to conserve masks for health care professionals, and DEFINITELY don't want people using "I'm wearing a mask" as an excuse to go out when they should just stay home, but I can't imagine they don't help a lot.


It is the black and white thinking. The mask is either useless or perfect. The case where it would lower probability is considered "useless" cause it is not "perfect".

Imo, this all or nothing thinking is contemporary issue that plays out in many contexts and we collectivelly really should get out of it.


For most people, regular hand washing and not touching your face is orders of magnitude more effective than wearing a mask. But as a society we've become addicted to off-the-shelf solutions rather than boring good habits.

Basic masks can be beneficial for a short time. They need to be replaced at least every hour in order for the benefit to persist. This is routine in a hospital, but is simply impractical for most normal people.

(If a mask stops a droplet containing virus, the virus doesn't disappear. Your own breath and fine droplets keeps the mask material warm and humid—making the mask material great environment for the virus to hang out and spread through the fabric until it can be breathed directly into your lung where it can do the most damage.)


With a disease that spreads while asymptomatic, the droplets we really need to stop are the outgoing ones. A high level of even improper or suboptimal mask usage in the general not-yet-diagnosed population would be very helpful, because it would cut the six-foot proximity hazard zone down to inches.

If we only had enough masks, that is.


There is evidence look at Asia. Everyone is wearing a mask and they control the virus. If you want to search for research based evidence there are plenty but focus on the research pre coronavirus - right now there is too much noise. One research suggested that with good handwashing masks reduce the infection rate by 75%.

N95 respirators have a pore size on the order of 100-300 nm. Viral particles can be a lot smaller than that (5-10nm), but this coronavirus seems to have a 80-90nm diameter. That means a respirator might be able to stop the particles IF worn perfectly (if you have stubble or a beard, you're already screwed). They will however stop droplets from someone sneezing or coughing, again IF worn properly. Medical professionals get fitted regularly to ensure proper fit.

Masks are only effective in a fairly immediate radius: someone coughs, that droplet goes a little ways then falls on a surface. Medical professionals are in that immediate radius all the time, that's why they need the masks. People are usually not sneezing directly on each other, so that's less of an issue. What IS an issue is the viral particle that is now on a surface. You touch that, move your hand to your face (perhaps to adjust your mask) and bingo you're infected. Washing your hands obsessively is WAY more useful than any mask.

Most of the time in medicine masks are used to stop the medical professional from contaminating the patient, not the other way around (that's what you see in surgery). If you're with someone sick, it would be good to give them a mask, they'll spread fewer particles that way.

TLDR: Masks are minimally useful for the broader public, very useful for those directly in harm's way. Don't buy masks unless you're sick, or someone close to you is sick. Let the supply chain work for medical professionals who actually need them.


Mask shortage is a red herring. I can tie a bandana or a scarf around my mouth. If it reduces my chances of breathing in a cough droplet by even 5%, that’s great. Why not?

Go for it! Just keep in mind that the benefit there is pretty minimal. Probably not quite zero, but we're talking way less than hand washing. And if you touch your face more as a result of wearing the mask...

That's not to say do or don't do it, just be careful. Handwashing and not touching your face are the single most important things to do.


From https://smartairfilters.com/en/blog/diy-homemade-mask-protec...

> Data shows that homemade masks made with a single layer of cotton clothing, or a tea towel can remove around 50-60% of virus-sized particles.


I thought it was more, they are ineffective because most people put them on wrong.

They are less effective, not ineffective.

But they do also say that masks are difficult for people to use correctly, and if you're not using it correctly it'll increase your risk.

slowhand09 has been unfairly downvoted, because they are factually correct. Epidemiologists are telling the public not to use masks because PPE needs training.


They say that so we, the dumb animals, don't stockpile them like we do with toilet paper (can't find any for 2 weeks straight in my area of Berlin for example).

Of course masks are useful, but they're more useful in hospitals / on health care workers' face, than on the average citizen which won't be in contact with the virus that much now that most of the world is starting to get locked down anyway


> Epidemiologists say masks are practically useless against virus. //

Have you seen any of them cite a research paper showing that? I haven't, I'm open to suggestions if anyone has one.

Same thing with the "keeping schools open won't spread the virus" thing in the UK; lots of establishment scientists saying "trust us" but a dearth of papers detailing how the established beliefs are wrong that schools are prime transmission vectors for ARI (and closing them has helped to hinder influenza trasmissions in the past).

Not being snarky; if you've got backing for "masks don't hinder viral spread" then lets see it.


In general, masks are something that the sick person wears to prevent germ spread. Same idea as why surgeons wear them in the operating theater.

The only kind of mask that's going to filter particulate matter out of the air (i.e., be effective against the virus for someone who is not infected) is something that has an airtight seal.


Yes, so you agree, masks hinder transmission by, in your understanding as presented, decreasing the ability of sick people to infect healthy people. That is, masks reduce transmission.

If they were practically useless, then health care workers wouldn't use them...

Surgical high-proof mask doctors and nurses use are pretty good against viruses, combined with googles of course.

For normal person? They are mostly useless. Only for people, who are already sick, to stop them from spreading virus further. Wearing mask by healthy person doesn't make sense, because there is not much to stop viruses with coming in contact with your eyes. You are always a scratch away from bringing those viruses into your body, negating any benefit from mask. And most of mask, you can buy currently are not dense enough to capture viruses.

OP is correct, unless you are sick or medical worker, you don't need mask. Don't buy them, hospitals need them much more than you.

Is WHO [1] good enough source for you?

[1] https://www.who.int/emergencies/diseases/novel-coronavirus-2...


The source doesn't substantiate anything you wrote in your comment.

They say "If you are healthy, you only need to wear a mask if you are taking care of a person with suspected 2019-nCoV infection."

That assumes you're doing everything else to protect yourself already, like stay away from crowded places.

It also assumes that you know that you're healthy. You could be carrying COVID-19 without knowing it!

One reason health practitioners wear masks is not to infect vulnerable patients with something they might be carrying.

> there is not much to stop viruses with coming in contact with your eyes.

Other than the goggles that you mentioned; yet, nurses in hospitals don't routinely wear goggles, yet do wear masks.

This is probably because your eyes do not suck in air the way your air passages do, and also don't spew viruses into the air.


I believe this is primarily a narrative that was invented to prevent a temporary shortage of masks for health care workers. This may be a noble cause, but it's still a lie. Surgical masks reduce the probability of infecting others if you have it (maybe without knowing), and correctly used N95 masks drastically reduce the risk of getting infected.

There is a reason why South Korea officially distributes masks to the public. Everybody in China and Japan wears masks, too.


>They are mostly useless. Only for people, who are already sick, to stop them from spreading virus further. //

Um, this is kinda what ordinary people want!!

If you stop it spreading, then most of us don't get it!

Yes, it doesn't stop you getting it, but if it stops you spreading it then we hinder/stop the epidemic [but we're way past that now].


The thing with this virus, you typically don't know you are infected until it's too late and you already infected a number of others. The goal with masks is not to protect you, but to protect the general population. Any basic mask will do for that purpose.

Surely that's why it was crazy when ordinary, uninfected people, wanted to wear masks to tell them "well that won't help".

I can't see how reducing transmission "isn't helping".


Because masks don’t reduce transmission. Communication on this has been very clear. Masks don’t help. Frequently washing hands, avoiding contact with others, keeping your surroundings clean, and not touching your face does help. Masks give a false sense of security and reduce supply for those that need them.

I wonder why doctors and nurses use them then. They usw them in most situations. I wonder why they complain about lack of them. Since they are useless.

Sure, they are not as good as respirators. And we as general public should not bulk buy them now.


I wonder why in China and Korea, in counties most experienced in current situation, everyone is required to use them.

Common sense says that any kind of barrier between cough or sneeze droplets and my nose or mouth will reduce the probability of me breathing it in. I could tie a scarf around my mouth. It may not be perfect but even if it reduces transmission by 5%, Why not?

https://smartairfilters.com/en/blog/diy-homemade-mask-protec...

Has more info on studies related to this. I believe their claim is that tying a cloth around your nose/mouth is 50% as effective as a proper mask.


Could you cite me a research paper showing face masks don't reduce transmission?

> Surgical high-proof mask doctors and nurses use are pretty good against viruses, combined with googles of course.

Maybe my country is outlier, but doctors in non surgical situation use mask (not respirator) and no googles. It is not airtight either.

I don't think they do it because it would be useless.


Citation please? This seems like overt misinformation.

They don't say it useless. They just don't recommend it.

https://www.cdc.gov/coronavirus/2019-ncov/faq.html#protect

CDC does not recommend that people who are well wear a facemask to protect themselves from respiratory illnesses, including COVID-19. You should only wear a mask if a healthcare professional recommends it. A facemask should be used by people who have COVID-19 and are showing symptoms. This is to protect others from the risk of getting infected. The use of facemasks also is crucial for health workers and other people who are taking care of someone infected with COVID-19 in close settings (at home or in a health care facility).

Only wear a mask if you are ill with COVID-19 symptoms (especially coughing) or looking after someone who may have COVID-19. Disposable face mask can only be used once. If you are not ill or looking after someone who is ill then you are wasting a mask. There is a world-wide shortage of masks, so WHO urges people to use masks wisely.

WHO advises rational use of medical masks to avoid unnecessary wastage of precious resources and mis-use of masks (see Advice on the use of masks).

The most effective ways to protect yourself and others against COVID-19 are to frequently clean your hands, cover your cough with the bend of elbow or tissue and maintain a distance of at least 1 meter (3 feet) from people who are coughing or sneezing. See basic protective measures against the new coronavirus for more information.

https://www.who.int/news-room/q-a-detail/q-a-coronaviruses

Only wear a mask if you are ill with COVID-19 symptoms (especially coughing) or looking after someone who may have COVID-19. Disposable face mask can only be used once. If you are not ill or looking after someone who is ill then you are wasting a mask. There is a world-wide shortage of masks, so WHO urges people to use masks wisely.

WHO advises rational use of medical masks to avoid unnecessary wastage of precious resources and mis-use of masks (see Advice on the use of masks).

The most effective ways to protect yourself and others against COVID-19 are to frequently clean your hands, cover your cough with the bend of elbow or tissue and maintain a distance of at least 1 meter (3 feet) from people who are coughing or sneezing. See basic protective measures against the new coronavirus for more information.

They don't say it useless, just they not recommend it


If using mask helps lower probabilities then you can sew your own. Which means you can wear it without buying one. It might not be as great as official one, but still probabilities.

> Disposable face mask can only be used once.

They can be cleaned. It's not ideal, but very likely better than nothing. Sorry, can't find link, but they tried a variety of different techniques to hit on one that worked.


I own several cloth facemasks, but it is important to remember that reusing a washable can sometimes be worse than nothing, as it can trap and hold the virus and lead to greater exposure than no mask at all.

It takes you 5 min to learn how to put on and take off a n95 mask correctly. Even if you do it wrong you are still protecting other people against your potential infection.

Hmm, I think NHS courses last longer than that. The mask makers sell a [very expensive!] testing kit you can use to check the mask is fitted too, I gather.

According to Singapore Health https://www.youtube.com/watch?v=zoxpvDVo_NI

And not a single detail on what they are...

Antibody tests are more suitable for test strips, and work under 30m, but they need blood test.

Few companies in China are already marketing 30 minute PCR systems. One from Xiamen, another from Nansha.


> Using technology from home pregnancy and malaria tests, its saliva and finger-prick kit could be ready for sale by June for less than $1 apiece.

Based on this, I'd guess they are talking about an antibody based test. There are point of care Influenza tests that work on the same principle. Probably not as sensitive as a PCR test, but definitely cheaper and faster -- if you can get a good antibody.


A direct antigen test can work if enough virus is shed. For example, there is a coronavirus fecal antigen test for dogs. The technique is ELISA, like for antibody tests.

Agree, tried to figure out how they got an rt-pcr machine that small, then realized the whole article is bullshit. I guess we are in for a lot of this nonsense.

It's real tech - miniaturization with microfluidics allows shorter cycles -> faster results. But it's not clear whether any of these companies are yet making enough machines to help, or if it's just PR. Probably the latter.

https://www.chinadailyhk.com/article/121312


I'd just like to note that, while we should be conducting many more tests, we should also be aware of the danger of false positives, and the burden that places on individuals and systems. Some of these tests are just not very accurate, according to an MD I know.

Always a concern, but under the present circumstances, a false positive (someone quarantines for two weeks who doesn't need it) is far better than a false negative (someone has false confidence in attending public places).

Unlike other false positives that could lead to unneeded/harmful interventions (surgery, chemo), erring on the side of false positives would actually be desirable for COVID-19.


I’d rather not be quarantined with a bunch of sick people for two weeks if I’m not actually sick. The risk of me getting the virus is much higher there than among the general populace. So I disagree that there are no downsides or risks for a false positive here. Especially as the disease has a chance to kill me.

Self-quarantine means they send you home and tell you not to leave for 2 weeks.

Unless you are in such an extreme case that you need a respirator or something, in which case they will hospitalize you and you might be around other diseased victims, but if you are in that condition then it is probably not a false positive. There is a shortage of hospital beds worldwide right now, so they are sending everyone home unless hospitalization is considered life-saving.


You quarantine at home.

All the restaurants and bars of California and New York are closing or are closed and all the people that work there are hourly wage. How how long do you think they will still have homes? Not trying to be alarmist but I also don’t see how people making an hourly wage are going to continue living in their very expensive states without income.

Unless the government backstops it, they won't. It is and will be a huge problem.

Some states are banning evictions for a few months, so at least there is that option. But it just means they'll be homeless in June instead of April.


Most NYC landlords have agreed to a 90-day hold on all evictions. So... about 90 days I guess.

You may have misheard the MD. The current tests have a very low false positive rate. That means if the test says you have Coronavirus, you have it. There's no "burden".

Where the tests are "not very accurate" is the false negative. The false negative rate can be as high as 50-60%, depending on tests and how it's administered. Sometimes they will conduct the test twice to improve accuracy.

The Roche ones recently approved have a lower negative test rate (10%?)

Edit: I removed my mention of 0% false positive rate because nothing in life is absolute.


> The current tests have a 0% false positive rate.

Need a citation on this. In order to even make a strong statement about the false positive rate you need to have some sort of alternate confirmation mechanism, usually clinical diagnosis or another more comprehensive test. That kind of work takes time and we haven't gotten to the point where we know this.

False positives in the test are one possible reason for the resurgence in Singapore; the initial screenings were overreporting the number of cases leading to a false sense of security, followed by a surge in real cases as the spread continued.

EDIT: A quick look at some of the tests available indicate that currently providers are not publishing specificity/sensitivity numbers in general under the accelerated release guidelines. One I did find was biomedomics test [1], which claims "The overall testing sensitivity was 88.66% and specificity was 90.63%", so ~10% of negative cases will show as positive.

[1] https://www.biomedomics.com/products/infectious-disease/covi...


That's an antibody-based test not approved for diagnostic use in the US, which is faster but less accurate than the RT-PCR ones currently being used.

It's just the only one that I could get numbers on. So... show me sensitivity/specificity numbers that show a 0% false positive rate. The CDC PCR test [1], for example, has some data and no false positives, but the sample sizes are so small (107 known negatives tested) that the best we can estimate is ~96% specificity.

[1] https://www.fda.gov/media/134922/download


"Clinical characteristics of 24 asymptomatic infections with COVID-19 screened among close contacts in Nanjing, China." - https://link.springer.com/content/pdf/10.1007/s11427-020-166....

It seems asymptomatic patients can test negative multiple times on RT-PCR assays but be diagnosed other ways.

I've read somewhere that at least in China, the specificity of the PCR tests is only 60-70%, leaving ~30% undetected.

From the Discussion section of this paper: https://pubs.rsna.org/doi/10.1148/radiol.2020200642:

"According to current diagnostic criterion, viral nucleic acid test by RT-PCR assay plays a vital role in determining hospitalization and isolation for individual patients. However, its lack of sensitivity, insufficient stability, and relatively long processing time were detrimental to the control of the disease epidemic. In our study, the positive rate of RT-PCR assay for throat swab samples was 59% (95%CI, 56%-62%) which was consistent with previous report (30 - 60%) (6). In addition, a number of any external factors may affect RT-PCR testing results including sampling operations, specimens source (upper or lower respiratory tract), sampling timing (different period of the disease development) (6), and performance of detection kits. As such, the results of RT-PCR tests must be cautiously interpreted."


A 0% false positive rate is not possible. What if they switch your sample with someone else's while it's being handled by a human?

Edit:

>That means if the test says you have Coronavirus, you have it.

This line implies the parent comment is talking about the false positive rate of the entire testing process that results in you being told you're infected, not PCR as a platonic ideal, theoretical process.


That's still not a false positive rate of the test.

But I agree that 0% false positive rates are going to be nearly impossible to achieve (for just about anything that isn't mathematics).


From the FDA itself. Note how firm they are in regards to a positive, vs how cautious they are in regards to a negative. If you have COVID-19 RNA in your body...you have the virus.

What does it mean if the specimen tests positive for the virus that causes COVID-19? A positive test result for COVID-19 indicates that RNA from SARS-CoV-2 was detected, and the patient is infected with the virus and presumed to be contagious

What does it mean if the specimen tests negative for the virus that causes COVID-19? A negative test result for this test means that SARSCoV-2 RNA was not present in the specimen above the limit of detection. However, a negative result does not rule out COVID-19 and should not be used as the sole basis for treatment or patient management decisions. A negative result does not exclude the possibility of COVID-19.

https://www.fda.gov/media/135662/download


> A positive test result for COVID-19 indicates that RNA from SARS-CoV-2 was detected, and the patient is infected with the virus

You still haven't addressed the possibility of a protocol error such as a switched sample. The lab test itself may never produce false positives, but that emphatically does not mean that if the lab test comes back positive there is a 100% guarantee that the patient is infected.


Considering the context that I replied to is "we should also be aware of the danger of false positives", I think given the FDA source I posted (and what I've read elsewhere) I can say that there is absolutely no need to be aware of a danger of false positives in terms of disrupting people's lives.

If you want to argue on the fact of life that nothing in life is absolute. Fine. you win. If administered properly, and the test kit detects a presence of COHVID-19 RNA in your body, and the person telling you the results is not drunk, then the results is as close as 0% as practical.


That’s a protocol fault and has nothing to do with the accuracy of the test

Still better to have as one of many screening options

Cheap and fast screening tests for a number of health conditions are commonly available. However the sensitivy and specificity of these tests tend to be lower than the PCR, for example. Once a person tests positive in the first screening test, a second confirmatory tests is then ordered. This test should thus be one such fast screening test and would certainly help in a place like Africa.

So, how does the PCR-based testing work? Is it just the usual gene amplification steps?

* Get DNA out of cells and unwrapped from their histones.

* Cleave the DNA sample with restriction enzymes at boundary sites which should mark the start/end of a stretch of viral DNA.

* Amplify the sample with PCR, using primers which match stretches of viral DNA.

* Run a gel electrophoresis and look for a mass of DNA that matches the expected size of the stretch of viral DNA.

Is that how things work, or are these kinds of diagnostics done differently?


It's an RNA virus, so you start by isolating RNA and then using a reverse transcription reaction to make cDNA from the RNA. Then you use viral-specific primers in a traditional PCR. Most of the tests that I've seen presented have then be real-time qPCR tests as opposed to gels, but for the most part, you're not far off. Also, restriction digests aren't required.

Note: I haven't seen any of the protocols, just what I've been able to learn from press reports.

Edit: Here's a link to the CDC's panel protocol: https://www.fda.gov/media/134922/download

And to add Wikipedia's page on real-time PCR (not to be confused with RT-PCR, which is also used here): https://en.wikipedia.org/wiki/Real-time_polymerase_chain_rea... . However, the crux of it is that the reaction contains a chemical that fluoresces in the presence of double-stranded DNA. As the PCR reaction progresses, if there is a valid template in the original mixture (i.e. if there is virus present), then there will be a detectable amount of DNA present after a number of PCR cycles. The lower the number of PCR cycles required to detect this fluorescence, the more viral RNA was present in the initial sample.


Neat, thanks - it's amazing how quickly sequencing technologies have evolved.

This isn't sequencing... you'd use sequencing to identify novel mutations or strains of the virus. This is a set of known PCR primers that amplify the virus's RNA. You could (in theory) also use a similar approach to identify specific strains of the virus, but this shouldn't be the priority at the moment.

While you could have an assay that used sequencing, it wouldn't be nearly as fast or cost efficient.


This CDC document was an interesting read. I found this part particularly fascinating.

>RNA isolated and purified from upper and lower respiratory specimens is reverse transcribed to cDNA and subsequently amplified in the Applied Biosystems 7500 Fast Dx Real-Time PCR Instrument with SDS version 1.4 software. In the process, the probe anneals to a specific target sequence located between the forward and reverse primers. During the extension phase of the PCR cycle, the 5’ nuclease activity of Taq polymerase degrades the probe, causing the reporter dye to separate from the quencher dye, generating a fluorescent signal. With each cycle, additional reporter dye molecules are cleaved from their respective probes, increasing the fluorescence intensity. Fluorescence intensity is monitored at each PCR cycle by Applied Biosystems 7500 Fast Dx Real-Time PCR System with SDS version 1.4 software.


Right, sorry...I sort of think of PCR as 'grep for genomes', even though it doesn't really 'read' the DNA.

You mean in the last 20 years? Yeh, it was super fast.

Yes, that is what I mean.

I can't tell if you're being sarcastic or not, but 20 years is a very short amount of time if you look back on the history of the advancement of engineering and our understanding of the world. It's very exciting, especially when you look at how little the basic operating concept of computers has changed in the same timeframe.


As someone outside the industry it _appears_ [emphasis there] to have gone from "we can take a few weeks and test you for a specific disease susceptibility" to "place your finger here and then grab a coffee ... sorry for the wait, OK we have details on 100 different diseases now" in a matter of 2 or 3 years.

qPCR doesn't use a gel, and you get specificity by using primers to the viral dna (which are what tell you the part you start and end amplification at), not restriction enzymes, but yes conceptually similar sort of idea. You set up reaction conditions so that at the end of the run, if there was any viral dna at the start, there's now a lot more for you to detect, while if there wasn't, you just see the background since nothing else should have been replicated. I believe the reagent issue with the original run of tests from the CDC had to do with the controls you set up to show that nothing amplifies if you don't have any virus (since you can't just run one tube and see if there's a signal, you've got to run a few for reundancy and run samples where you should and should not see signals to make sure you are seeing signals where you should, and the reaction worked, so your signal from the patient samples are accurate)

Sort of. There are now reagents that can fluoresce as the strand is elongated allowing instantaneous reading.

Also, for an RNA virus, it starts with reverse transcriptase.

https://en.m.wikipedia.org/wiki/Real-time_polymerase_chain_r...


These kinds of engineering/science solutions to a hard problem are the bread and butter of a community like this. Where can engineers pool their resourcing and work together to assist in the fight?

It’s relatively easy to develop a detection method, however the most important thing for any detection is the false negative rate, considering current situation. If the false negative rate is high, such a test might encourage spread of the virus. Some Chinese company already developed antibody-based detection method a while ago, But never saw they change the detection method in clinic. Elisa use the same basic principle as an antibody based test.

What is enabling them to be the only people able to develop this far better test? Is the rest of the world unaware how home pregnancy and malaria tests work?

If this was worth reporting on, there should be some kind of data showing how effective and cheap the test is when compared to the competition. I wonder what led Bloimberg and other financial news to pick this story up.


"What is enabling them to be the only people able to develop this far better test?"

They're far from the only ones who've developed a far better test.

Without specifically looking for them, since the outbreak began I've seen 4 or 5 reports of different companies developing far better tests.

Unfortunately, I don't know if any of them have actually been put in to use, or what the obstacles to doing so are.


If you are a company that wants to sell a test, you will definitely say it's a "far better test". Whether it actually is a better test is a much different story.

I think the biggest obstacle is actually doing the testing of the tests to see if they're far better. Putting an ineffective or highly false positive/negative test out there could do more harm than good.

>I've seen 4 or 5 reports of different companies developing far better tests.

Were any of the reports more detailed than "hey, we developed a better test. Give us money."


$1 would be nice. Some company in Japan is selling kits $250 for a 10 test kit

https://monoist.atmarkit.co.jp/mn/articles/2003/13/news055.h...

Google Translate Translation

https://translate.google.com/translate?sl=auto&tl=en&u=https...



Someone sent me this link recently. Can anyone provide some color on what's actually being sold here? Is this just fraud or are there cheap tests already available in China?

https://www.alibaba.com/trade/search?fsb=y&IndexArea=product...


Assuming tests aren't available -- is there nay prescribed care one can provide to others or do for ones-self if they contract the virus? I realize some conditions may worsen to the point that you'd require medical attention --- but what's the best course of action assuming that professional care is unavailable?

I feel like a question like this is better served doing a Google search and finding results from reputable sources...rather than asking a question on HN.

Having said that, Coronavirus is consider a repository illness...so you'd probably perform the same home treatment as any other respiratory illness.


>you'd probably perform the same home treatment as any other respiratory illness.

Notably, you should not take ibuprofen or other NSAIDS. Stick to acetaminophen/paracetamol instead.

Guidance from France's health minister:

https://www.theguardian.com/world/2020/mar/14/anti-inflammat...

https://twitter.com/olivierveran/status/1238776545398923264


This is good advice generally, especially for severe infections like COVID-19 but there's little evidence _specific_ to COVID-19 and a contraindication with NSAIDs.

The Snopes [1] article covers the general gist although there is an early report around that the minister likely read and passed on in the form of a tweet.

Research is changing constantly right now so evidence may or may not show NSAIDs contraindicated with greater risk than other infections.

[1]: https://www.snopes.com/fact-check/covid-19-nsaids-ibuprofen/



by June when it will be available I'll be either dead or immune, same with most of the population

A test for $1 can be a test for free.

A test for $1 can be afforded by anyone.

I'd say "can be afforded by the World for everyone". Sure, the average GNI (gross nat. income) for the poorest countries is about $2 (USD) per day, but affordability is more complex. However, if we taxed the World, and paid according to ability of nations to pay ... I mean USA could probably pay it for everyone and only have to not build an aircraft carrier (https://en.wikipedia.org/wiki/Gerald_R._Ford-class_aircraft_... says that is $12B, money to spare!).

[Yes, finance isn't that easy, the knock on effects would make the cost possibly greater, maybe less(!)?]


The logistics of getting goods such as food and medicine into third world countries have hardly anything to do with their manufacturing price. Even once you get them there, distribution is a nightmare. Obviously I'm not referring to countries with no actual economy to speak of.

The "$1" in a $1 test has very little to do with manufacturing costs, too.

For another novel detection method which may be suitable for low-cost implementation see this post https://news.ycombinator.com/item?id=22592173

1st, SARS-CoV2 ELISA already exists. 2nd, the CDC did compare RT-PCR vs ELISA SARS-CoV back in 2005, and ELISA performed really poorly, and that was for acute cases. So, that is why we should stick with RT.

What's the current best false negative rate of any coronavirus test?

Wouldn't a far less specific test do the trick as well?

If we had a cheap test that tells us whether we have some kind of illness, corona-flu included, then I'd say that would be tremendously useful.


This sounds like something we could really use in the U.S., too.

Here is a list of potential diagnostic tests for COVID-19

http://www.finddx.org/covid-19/


How many can they make?

it's either that and containment South Korea style or total lockdown China style. too much risk to be in between, though some were trying.

Why are flu tests so much slower and more expensive?

Probably because "the flu" is hundreds of different viruses and the entire world is not currently dumping massive amounts of resources behind developing them.

Then again, the rapid flu test only takes about 5 to 20 more minutes than this, and doesn't really cost much more than any routine diagnostic.


I'm sure the are different types of flu tests, but the last one I had at the doctor's office about 2 years ago was two swabs of my nose, which the doctor put into a test kit, and a few minutes' wait for the results to develop in the kit. No idea how much it cost, though.

In the same lines - Apple Watch or Samsung watch sensing it through sensors would be so cool. I'm not a doc, so don't know the practicality of it

They have some LEDs that shine light into your skin. You're not detecting a specific pathogen by shining UR and visible lights.



Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact

Search: