Thursday Throughput: Fetal Gene Edition
[ThTh1] The NYT has an expose on fetal genetic testing. A number of years ago, it was discovered that DNA from fetuses will circulate in the bloodstream of their mothers1. This DNA can be found in a blood samples and tested for genetic conditions. It has the advantage of not being invasive of the fetus, unlike amniocentesis.
The NYT expose covers the downside. For conditions that are not terrifically rare — like Down’s Syndrome — the tests are accurate. But the more rare the condition, the more accurate the test has to be and the more likely it is to produce a false result. For example, Wolf-Hirschhorn syndrome only occurs in one in 20,000 births. So even if your test is 99.9% accurate, you will get 20 times as many false positives as true positives. And the more conditions you test for, the more likely you are to get a false positive. So a lot of soon-to-be-parents are getting bogus scares for conditions their children do not have and getting unnecessary amniocentesis to rule them out.
Moreover, even a negative result is not necessarily meaningful. There are literally hundreds of possible genetic conditions a fetus can have and these tests only rule out a few. So it’s not clear how exactly these tests are helping parents make good decisions.
I don’t think these tests should necessarily be banned. But I do think parents need to be informed up front that most positives are false and a negative screen doesn’t necessarily mean anything. I think the use of these tests would plummet significantly if that information were more available.
[ThTh2] There has been some smug commentary from the COVID-19 skeptic corner on recent changes in CDC pandemic guidance. But these changes make complete sense given what we now know about the dominant omicron variant.
The new recommendation against cloth masks is because, while cloth masks were fine against wild COVID and delta, omicron is far too infectious. Only KN-95’s or surgical masks will do. The changed recommendation on quarantine duration is partly because omicron ramps up and ramps down more quickly. But also because omicron has changed the tradeoff. A longer quarantine period could mean entire industries grinding to a halt because it’s spreading so fast. And the lower virulence means the risks of a shorter quarantine are lower.
Omicron has also — and this is a very good thing — broken the connection between cases and hospitalizations/deaths. Up until now, these three numbers tracked each other very well. But the combination of a milder variant and a much more vaccinated population has broken that connection. We are seeing sharp rises in cases combined with smaller (but still very bad) rises in hospitalizations and much smaller rises in deaths. So how we track the progress of the virus and our mitigation efforts is changing.
Omicron has changed the game in almost every way imaginable. And the next few weeks are going to both very difficult and very confusing. But, given what we’ve seen in other countries, we can expect to emerge from this in February with most of the population resistant to COVID and the latest wave having passed. That doesn’t mean the pandemic is over — the virus will continue to evolve. But it does mean a respite is coming sooner rather than later.
[ThTh3] Robert Malone became the latest vaccine skeptic to be banned from Twitter. You can read up on him here and here.
[ThTh4] Scientific American has fallen far from the standards it once had.
[ThTh5] In looking at things we’ve learned from the pandemic, I think this list from Surowiecki is a good place to start:
There are certain things that people are never going to agree on, like masks for kids. But I did hope that after this was all over, there would be a long list of things we could all agree were true. Instead, it's just going to be "Virus gonna virus, brah!" and "It was the flu!" pic.twitter.com/ZijzmCYWUp
— James Surowiecki (@JamesSurowiecki) December 31, 2021
The biggest and ongoing failure may be the dearth of tests.
[ThTh6] I will never not be fascinated by time-lapse videos.
My original video of the instantly freezing Shannon Creek in Squamish, BC this morning. Sped up 4x. It looks like the creek flows uphill. Weather is truly fascinating! #bcstorm #ShareYourWeather #wxtwitter @spann @AMacOnAir @weathernetwork pic.twitter.com/cuGObVnYeb
— Brad Atchison (@Brad604) December 27, 2021
[ThTh7] You can track the progress of the JWST here. So far — knock on wood — so good. The most fraught step — the deployment of the five tennis-court sized sun shields, went fine. And the secondary mirror had deployed and latched into place. There’s still another week for everything to deploy and unfold.
And then the work really begins.
On the very same day that ThTh4 was published, Scientific American plumbed new depths with this breathtakingly ignorant smear of the recently deceased E. O. Wilson. It’s pretty terrible throughout, as is to be expected, but the author, Monica McLemore, really went above and beyond in bringing shame upon herself, Scientific American, and the UCSF with this gem:
Report
Wow that reads as if it was written by a precocious 9th grader who reads a lot of Tumblr blogs.Report
ThTh2: The recent changes in CDC guidance have also received criticism from the pro-lockdown people. There are a bunch of different forms of the criticisms but the one that first comes to mind is “You can get Covid at the beach, in the park, in restaurants, at concerts, and in grocery stores but office jobs are safe!” (Some wags might point out “Oh, and Mostly Peaceful Protests are okay as well.”)
What should have been done? Well, the CDC’s messaging has been horrible. Absolutely awful. I have no idea how messaging about risk ought to have been done, but it’s difficult to imagine worse ways than the way that it was done over the last couple of years.Report
ThTh4: It’s been a while since I paid for a subscription to it, and articles like this is one of the reasons I stopped.Report
Mike, I know there’s no way for you to know this, but actually, the NIPT tests are a godsend. They are freaking fabulous, I consider them to be the greatest thing since sliced bread, and if companies are at present unable to deliver on predicting some of the more rare conditions, well, technology marches on and I bet the tech will be there within the next 3-5 years. Parents will not stop using these tests because they’re AMAZING and wonderful and create so much more benefit vs potential harm, that I am literally baffled by the existence of this hit piece.
The previous tests they used, the triple/quad screen, were riddled with false positives. Nearly every woman over 35 got a false positive on them and doctors never bothered to explain this to anyone. There were millions of unnecessary amnios and CVS because of those tests. Two of my clients lost genetically normal babies after a false positive on the old tests (due to the unavoidable risks of amnio/CVS). People who did not receive proper counseling had abortions rather than bothering to do amnio and most of those babies were genetically normal. I know of people over 35 who had 2 and 3 abortions before being told that most women over 35 had a false positive on the triple/quad screens. I cannot tell you the number of people who have come to me in despair, convinced that their baby had Downs when they didn’t because of those older tests. They sucked and were terrible and the NIPT replaces them, and does an absolutely phenomenal job of it, BTW. Those old tests killed babies, NIPT tests save them.
The NIPT does away with far, far more false positives and unnecessary amnios than it causes. It does away with far, far more heartbreak and angst than it causes. Whoever wrote this article did not do their homework and was trying to produce clickbait in light of the Theranos verdict. Shameful and absolutely terrible bottom of the barrel scientific journalism.Report
I honestly thought it was from ProPublica until I saw that it was an NYT article.Report
In the end, it comes down to making sure the patients understand the risks of a given test. That’s less on the test and more on the physician.Report
I absolutely agree that the tests are useful for things like Down’s syndrome. I apologize if that wasn’t clear. But for rare genetic conditions, I think the value is marginal, at least at present, because the accuracy you need for a useful test is far better than what they have at present.Report
That is really insightful, thanks.Report
No account of the Covid era in America will be accurate without acknowledging the role of the Republican Party going all in on the anti-vax hysteria. More than any other development, this is what has brought us to this point where such a large percentage of people reject both masks and vaccine and belligerently demand to circulate among other people.Report
What punishment do you think that people who politicized the vaccine should receive?
Do you think that anybody who politicized it should still be holding office?Report
Let’s save ourselves about 30 comments, shall we?
Only one side reacted to the pandemic by using it as a political weapon, and one side only.
And only one side continues to do so and bears responsibility for hundreds of thousands of deaths.Report
If you want to save 30 comments, you could write more balanced analysis. But you don’t get to cast blame then complain if challenged.Report
You’re free to write such an analysis.Report
I wasn’t even challenging him!
Remember the tail end of the televangelist era in the 80’s? Good times.Report
“(Contrast this with my male friend, who is often mistaken for being female. He never seems to find this injurious to his psyche.:
Of course not. He hasn’t battled dysphoria his entire life.
That analogy is like poking me in the stomach and me saying “That doesn’t hurt at all” and then telling the person with the burst appendix they’re just being overly dramatic when they tell you to stop poking them, it hurts.
How is that not obvious?Report
C’mon, you’re not allowed to say that one side is worse than the other, especially on January 6th.Report
ThTh5 — I found Surowiecki’s list to be a reasonable starting point, but I’m struck by a few caveats:
4 – This appears to be true, but I don’t know if the data demonstrates it conclusively. I’m not saying it’s wrong.
5 – I disagree with the second half of this.
6 – While limiting the spread prevented covid deaths, it had a lot of unintended consequences that shouldn’t be ignored.
7 – I’m not comfortable blaming the government without reviewing what was known and possible at any given time.
9 – Did the CDC director say that? I’m glad I wasn’t listening.Report
7 – Short of buying them up from China (assuming China wanted to sell us their supply), I’m not sure what the government could have done. We don’t want our government stockpiling such things (waste of tax payer money, doncha know), and there was no financial incentive to maintain a domestic supply chain.Report
Recall that there was (probably still is) a national strategic reserve of ventilators. Many/most of them were old and had not been properly maintained, so were unusable. Masks have use-by dates. A government stockpile would probably have been largely out of date, absent an ongoing program to replace the ones that had already or were about to age out.Report
Ergo, waste of taxpayer money.
About the best the government could have done is subsidize a local production base so that it could be competitive with the masks from China. But that would require a considerable bit of foresight on the part of the government.Report
IIRC, one local manufacturer of either vents or PPE called up the WH early in the pandemic and asked for help to get another line running. They got ignored.
It would have been slack capacity outside of a pandemic, and basically all they wanted was a contract for the Feds to purchase the new line’s output at a cost they’d basically defray the investment of what would unused capacity for years post-COVID — so like 10% more than their previous costs.
This being the Trump WH, and this at the height of the “It’s a blue state problem, let them die” view from Kush and Co, they were roundly ignored.Report
9 — I don’t recall the CDC director ever saying that the vaccinated couldn’t transmit disease. Biden has said it a few times though:
https://www.politifact.com/factchecks/2021/dec/22/joe-biden/biden-says-vaccinated-people-cant-spread-covid-19-/Report
I guess this is his statement back in March: “We can kind of almost see the end,” Walensky told Maddow. “We’re vaccinating so very fast, our data from the CDC today suggests, you know, that vaccinated people do not carry the virus, don’t get sick, and that it’s not just in the clinical trials but it’s also in real world data.”
CDC clarified a few days later that “It’s possible that some people who are fully vaccinated could get COVID-19. The evidence isn’t clear whether they can spread the virus to others. We are continuing to evaluate the evidence,” a CDC spokesman told the Times.
https://thehill.com/changing-america/well-being/546234-cdc-reverses-statement-by-director-that-vaccinated-people-are-noReport
Where did the CDC change their recommendations with regards to cloth masks? It wasn’t in the link provided. To date I’ve only seen one article on CNN that cited two doctors, one of whom worked for the outlet.
Do we have studies on the effectiveness of cloth masks for Omicron?Report
Care to share the citations you work from?Report
That HEPA study is pre-print, meaning not yet peer reviewed. A number of such studies on a number of COVID related topics in the last two years have had to be retracted after peer review because they don’t hold up. While HEPA filters (properly used) may indeed be part of the solution, this isn’t saying that just yet, and it make no statements on effectiveness of masks as its not a study on masks.
Your mask study – while fully peer reviewed – is over a year old. There have been other more recent peer reviewed studies with different conclusions. And more importantly this study says the masks provide some protection.
Which means we shouldn’t toss them, especially with so many individual states ignoring even modest social distancing guidelines, to say nothing of low vaccination rates in many areas and lack of air filtration upgrades. We need every tool in the tool box because a bunch of partially effective tools are still collectively better then no tools at all.Report
https://www.cdc.gov/coronavirus/2019-ncov/community/ventilation.html#refphf
Effective controls exist that are 99.97% effective.
Why talk about ineffectual ones?Report
To be clear, I’m not questioning the effectiveness of masks.
But I haven’t yet seen an official recommendation that we should abandon cloth masks for K/N95s. The article implies that was offered by the CDC but it wasn’t mentioned in the link provided.
My girlfriend’s school has shifted to requiring the K/N95s and it is causing major issues there, including availability, costs, and parent pushback.
My school has not changed our mask rules.
We are in the same city so our regs aren’t different.Report
An official recommendation? Would a study done by Nature qualify or are you specifically talking about the CDC (or similar)?
Here’s the CDC. It’s okay with cloth masks.
Here’s the study in Nature.
Here’s the paragraph that I excerpted last time:
Is that an official recommendation or not?
(I’ve also heard people argue that the study itself was “sus” and we should therefore ignore it.)Report
Here, I’m asking specifically for the CDC because it is what was offered.
“ There has been some smug commentary from the COVID-19 skeptic corner on recent changes in CDC pandemic guidance. But these changes make complete sense given what we now know about the dominant omicron variant.
The new recommendation against cloth masks is because, while cloth masks were fine against wild COVID and delta, omicron is far too infectious. Only KN-95’s or surgical masks will do.”
Where is this “new recommendation” coming from?Report
If the tests are 99.9% accurate, why not run them twice? What are the odds of two false positives in that scenario?Report
So much depends on what causes a false positive or negative. Was it a mix-up of a sample, or someone sneezing on it, or a lack of sensitivity to a low level of what’s being tested for, or a false positive caused by the presence of something similar? Are you checking for the presence of a thing, or the presence of a common effect of the thing? How much variability is there in natural levels of the thing?Report
So the tests can say 99.9% accurate, but when the underlying condition is rare, you are going to get alot of true negatives, so what really matters is the false positive rate [ false positive / (true positives plus false positives)], which is most certainly not 99.9%
Its entirely possible for a test to have a 75% false positive rate but be overall 99.9% accurate. That’s the issue.Report
I believe that false positive rate is more commonly defined as the likelihood of a false sample testing positive, rather than the likelihood that a positive test was incorrect.
Here it doesn’t matter so much because the term is used as defined in the comment itself, but it’s worth keeping in mind when seeing the term in other contexts.Report
I hate it when a good troll thread with fantastic responses by me get appropriately moderated.
That said, a reminder on masks:
https://www.pnas.org/content/118/4/e2014564118
https://jamanetwork.com/journals/jama/fullarticle/2776536
https://www.bmj.com/content/375/bmj-2021-068302
None of these suggest wearing masks by themselves, but if you live, as I do, in a place where vaccination rates remain low, case counts remain high (and increasing due to omicron), social distancing is non-existent, schools are fully open, and most businesses are as well, then mask wearing needs to be part of your ongoing personal defense against COVID.Report