High false negatives due to low viral load in samples would be problematic for diagnosis and treatment (except there is no specific treatment) but for
public health purpose it is just as important to test for infectiousness. Imagine if we have a quick and easy way to self test saliva with high false negative rate due to viral load in the saliva, we can feel more confident about not infecting others through talking even if we can't be sure we are disease free.
Roughly 1/3 covid tests are falsely negative[1]. I'm not sure any have been cleared/approved by the FDA yet, but some are authorized for emergency use due to the circumstances.
"Testing was performed using the cobas(R) SARS-CoV-2 test. This test was developed and its performance characteristics determined by LabCorp Laboratories. This test has not been FDA cleared or approved. This test has been authorized by FDA under an Emergency Use Authorization (EUA). This test is only authorized for the duration of time the declaration that circumstances exist justifying the authorization of the emergency use of in vitro diagnostic tests for detection of SARS-CoV-2 virus and/or diagnosis of COVID-19 infection under section 564(b)(1) of the Act, 21 U.S.C. 360bbb-3(b)(1), unless the authorization is terminated or revoked sooner."
1) Presumably quite accurate considering that they received the FRA approval
2) Not clear for this one, but FDA authorized another lab to do saliva-based tests 10 days ago [1], and that test takes 24 hours.
3) It appears this test will be performed in the labs of a partnering health network, RWJBarnabas, and not for a wider distribution (from the article). Their goal is to get to tens of thousands of tests per day. They cover the New Jersey area. My guess is that you take the saliva sample, send it to one of their labs, and you get the answer in 24 hours.
4) Again, sufficiently robust so the FDA grants them the approval.
This test has not been FDA approved and only has an Emergency Use Authorization which does not test the efficacy as an FDA approval would. Even the linked article says "this test has not been FDA cleared or approved."
As I understand it, the material supply chains for testing are limited in two key ways: The sample collection material (swabs), and the reagents that the tests need to function.
It sounds like this helps address the sample collection material supply chain, but how do we scale the reagent supply chain? The entire world desperately needs test capacity to increase by a factor of 10 or more, but is that actually possible?
Every single expert says mass testing is the only way to get ahead of this before a vaccine arrives, but I'm very worried that mass testing on the scale needed is simply impossible, which leaves us with only two choices - stay in lockdown for over a year waiting for the vaccine, or accept that herd immunity is the only way out, and do what we can to minimize the damage as the virus spreads, e.g. mandating masks in public, keeping the elderly population isolated while the rest of the population is gradually exposed,etc.
As I understand it, the goal of "flattening the curve" was never to reduce the overall number of people that contract the virus. Which, to me, means that neither herd immunity nor hiding were ever really the plan for covid. Giving manufacturers the time to shift into overproducing PPE and researchers the time to develop/test short term treatments so the hospitals could shift into treating it as a part of routine coverage. I don't think you need herd immunity to be prepared.
I can't tell if that's what you meant by "manage the flow of bodies." As I understand it, lots of the deaths have been a result of timing and lack of preparedness, not necessarily overall lethality of the virus itself.
The virus has a reported 12-20% hospitalization rate. Once they overfill in any given region, the death rate and the hospitalization rate become the same.
We can argue about small percents for CFR, which only apply given healthcare resources, but if we do not stay under the hospital thresholds by some means, we get a catastrophic death rate that is comparable to the 1918 Spanish Flu, which had something in the neighborhood of a 10-20% fatality rate.
Unless we somehow find an astonishing number of people are true asymptomatic, this will hold. The Iceland survey does not yet count, as they surveyed and tested and found a goodly number of people positive, but we haven't seen the followup of how many develop symptoms where it can take 11.5 days in 97.5% of all cases to develop.
So, we are in a Class 5 pandemic, one of similar severity to the 1918 Spanish Flu. We just have oxygen and ventilators now for pneumonia. Once we run out of those resources, we get to "manage the flow of bodies". If we get the infection rate of the 2009 swine flu at 60.8 million cases, we stand to lose up to 7 million lives in the United States alone.
You're right that flattening the curve wasn't about reducing the overall number of people that are infected -- it was keeping the amount of people that needed treatment below the hospital capacity for the state/country/region. This necessarily means that the number of people that are exposed will be roughly the same as without exposure, but in a longer time-frame. The WaPo simulations do a good job of showing the effects of flattening the curve on the total number of infected at any one time.
And the point of keeping these total number of infected was to not overwhelm the hospitals. When the hospital systems are overloaded, then you start to see deaths from normally treatable causes where the system couldn't keep up with COVID, so couldn't treat an otherwise survivable condition. These numbers aren't being readily reported, but the absolute death rate in places has gone up. Italy has reported some of these numbers and while the COVID death rate is fairly low (but still substantial enough to try to avoid), the impact it has had on the overall death rate of the population is striking.
But, if you extrapolate out the WaPo simulations, there are three possibilities: 1) Total burn where the virus does what it's going to do and whoever survives is (hopefully) immune. Thankfully we didn't go this route. 2) A moderate degree of infection where most of the population has been exposed by the end. This will take a long time to achieve this level of herd immunity and you still have to deal with the death rate of the virus itself. 3) Slowing the infection rate so much that the outbreak burns itself out. This takes time as well, and a lot of discipline. And at the end, there is still a large percentage of the population that is potentially at risk when the virus flares up again. This is likely the scenario we are headed towards.
If we are going to start to remove restrictions before we have a vaccine (which is all but a given), then I interpret "managing the flow of bodies" as controlling where people can travel. Or more likely, contact tracing where we monitor where people travel in near-real time so that if they end up getting infected, then we can retrospectively contact-trace who they may have been in contact with so that they could also be tested and/or quarantined.
Yes, I agree with everything you just said. I took "Managing the flow of bodies" as a facetious comment implying the path we're on meant catastrophic death, which is why I attempted to clarify.
Look into group testing for one mechanism for increasing testing with the same amount of reagents. It doesn't save swabs, and it also potentially increases latency. it makes the most sense if the percentage of positive tests is low, such as you might get if you had much more widespread testing.
Perhaps silly to ask, but: why does the FDA regulate this?
It’s not a food, it’s not a drug, if it’s purely a saliva swab it is non-invasive — so where does the FDA come in?
In a university context, there is often a blanket IRB exemption for certain kinds of research where the risks are self-evidently minimal, such as taste tests.
If the FDA’s role here is to provide a stamp of approval indicating “this test ‘works’ according to an accepted standard,” then I would think something closer to how food supplements are monitored would be a better fit, for example, the test could come with text saying “this language has NOT been approved by the FDA” or equivalent for approval.
I am failing to see risks here besides the risks of misdiagnosis, but I just didn’t think it was the FDA’s mandate to approve or not approve based on efficacy.
The legislation that created the FDA (the Food, Drug & Cosmetic Act) defines drugs to include "articles intended for use in the diagnosis, cure, mitigation, treatment, or prevention of disease in man or other animals". See https://www.law.cornell.edu/uscode/text/21/321, section 321 (g)(1)(B).
The FDA maintains it is within the administration's purview to regulate tests--including laboratory-developed tests, or LDTs--although the current state is that the FDA often elects to exercise its discretion to permit LDTs sans regulation-as-medical-device provided the lab adheres to CLIA and has any required accreditation in order.
Dietary supplements are more or less unregulated, as long as they don't claim to diagnose or treat a disease. This is not exactly a great situation for consumers. Also, the current administration has taken sort of a "safety over efficacy" tack, but I don't know, therapeutic failure seems pretty bad, so maybe there isn't that much meaningful difference between the two.
Back to the testing concept, if not for the FDA to come down on operations like Theranos, then who?
> if not for the FDA to come down on operations like Theranos, then who?
Theranos was committing fraud - selling investors a product that didn't exist. You don't need the FDA for fraud to be illegal, ordinary civil and financial law covers this.
If Theranos weren't defrauding investors and actually had a product that worked, it'd be up to the market to determine if that product were worth paying for. Since different customers have different risk profiles there's no one-size-fits-all best balance between safety, efficacy and cost - let people buy different products that choose different tradeoffs.
Theranos was in fact testing people, who received both false negative and false negative results.
From the article I link below:
> At its height, Theranos operated 40 “Wellness Centers” in Walgreens stores in Arizona and a single location in California, which were the source of much of its revenue. USA Today reported the metro Phoenix-area centers alone sold more than 1.5 million blood tests, which yielded 7.8 million tests results for nearly 176,000 consumers.
> Theranos was committing fraud - selling investors a product that didn't exist
Well, the FDA is not the SEC--and Theranos was also providing patients with erroneous lab reports! This is more than a bit worse than wasting the time, money, and reputation, such that it may be, of Tim Draper, George Shultz, Henry Kissinger, et al.
> Dietary supplements are more or less unregulated, as long as they don't claim to diagnose or treat a disease. This is not exactly a great situation for consumers.
If dietary supplements were regulated, they would probably 1) cost 10x as much 2) be prescription-only 3) only be affordable through health insurance 4) there would be 1/10th the innovation compared to what we see now.
I'm a consumer who consumes a basic set of supplements (vitamin d, vitamin k2, magnesium, multivitamin) and I'm very happy with how things are now.
I'm glad that the extreme level of waste, red tape, and corruption that affects the rest of the medical industry does not affect dietary supplements.
> I'm a consumer who consumes a basic set of supplements (vitamin d, vitamin k2, magnesium, multivitamin) and I'm very happy with how things are now.
Fundamentally, no you aren't. What you are is a consumer who thinks they consume a basic set of supplements naively without oversight. Without a regulatory body verifying the safety and efficacy of the products, you actually have no idea that they contain what they claim and that they aren't tainted by e.g. lead or arsenic.
> There are regulatory bodies overseeing generic prescription drugs
This is also sadly wrong. There are regulatory bodies overseeing only a very tiny fraction of generic prescription drug production (ask how often they inspect and test batches) because of a systemic drive to dismantle and defund regulatory oversight exactly as championed by the person I was replying to.
Oversight in the face of rampant fraud requires testing continuously, not once every few years, not sometimes after something goes wrong, and it requires funding to hire overseers.
Anyone know what mechanism this test uses? I didn't see it in a skim of the article.
Saying this because viral load seems to be peaking in the upper respiratory on _day one_ of symptoms, and quickly decreasing from there as it moves to the lungs.
But maybe if this is an antibody test, that won't matter?
The decisive absence of this information in the article leads me to assume that it has all the same properties (and reagent demands) as existing PCR tests, just aimed at a simpler sample acquisition process. If there were further advantages they surely would have written about them. Perhaps the RNA signatures the test is looking for are shorter or something like that, to tune the test sensitivity, but I could just as well imagine that it's purely a paperwork difference. Wasting test capacity on samples acquired in a non-approved way would be reckless when done at grassroots level, but if the manufacturer can clear a relaxed sampling process it can be a valuable improvement.
Paperwork-heavy engineering disciplines are full of examples where the exact same hardware got re-rated to higher performance levels once the necessary experience was accumulated. It's not necessarily a bad thing.
@dang, is there value in changing the title from its current state ("FDA Grants Approval for First Saliva Based Coronavirus Test") to "FDA grants Emergency Use Authorization for Saliva Based Coronavirus Test"?
---
Edit 2:33pm EDT: calling Rutgers to clarify.
Edit 2:36pm EDT: the appropriate contact for the article has received my message and (as best as I know) is clarifying the release.
Edit 3:01pm EDT: thanks for changing the title, dang!
They are using lay terminology, but it doesn’t matter for regulatory affairs: authorization, approval, and clearance are legally distinct and companies are fined for using the wrong word.
Easy english could be a reasonable excuse if the precise terminology would be indecipherable to laypersons, but I don't see that here. "FDA grants emergency use authorization [...]" would give the exactly same diffuse "they allowed something on some level" information to people like me who are unaware of the specifics, but without ruining it for those in the know.
