Anna Minkina wrote a superb primer on COVID testing last spring. Her latest looks at the PCR-is-too-sensitive narrative, considers the dangers of this simple story, and asks what we can do to implement widespread testing going forward. I agree with much of it; well worth reading.
I would like to stress that even low-sensitivity (as low as 50%!) proactive screening tests can take a big bite out of transmission, so long as they are not misused for individual diagnosis or clearance.

I'd like to see a separate FDA track and labeling for such.
I'd also like to stress that low-specificity tests don't have to lead to false positive quarantines; they can be used to triage people for high-specificity tests.

But in general, I think Anna makes very important points about the political expediency of certain narratives.
Before dragging @Anna_Minkina or me, consider her core message, which I think is great comms advice.

"Fast, frequent, and ubiquitous testing would be a game-changer...But striving to get there by eroding public trust in the only reliable test we currently have is short-sighted."

• • •

Missing some Tweet in this thread? You can try to force a refresh
 

Keep Current with Carl T. Bergstrom

Carl T. Bergstrom Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @CT_Bergstrom

14 Oct
Heartbreaking as we head into the third wave.

nytimes.com/live/2020/10/1…
Three heterodox scientists, a think tank known for climate change denial, and a Hoover fellow willing to tell the president what he wants to hear.

That's all it took to trump 8 months of blood, sweat, and tears from tens of thousands of scientists and medical professionals.
Read 4 tweets
13 Oct
So this is interesting.

When searching google for medical info, the "fact box" at the top sometimes links to papers in _Medical Hypotheses_.

Readers outside of the field won't know that this journal publishes highly suspect claimsand wasn't even peer reviewed until 2010.
The paper above is a silly example, in which the authors ignored all confounds and still failed to achieve statistical significance (see below).

But with a bit of effort you may find other examples where google promotes more dangerous material.

For example, here a Google query about an antivax trope highlights one scientist's controversial—and to the best of my knowledge, unsupported—hypothesis linking elevated blood histamine from vaccination to shaken baby syndrome.
Read 4 tweets
11 Oct
In an effort to make me look like an anti-masker, a statistics professor who has blocked me and to whom I cannot respond is retweeting a post of mine from March 5th—a post that I deleted because I was wrong.

In that post I suggested that N95 masks require professional fitting.
I want to stress here that I wholeheartedly support the use of masks.

My post was not helpful because it seemed to suggest people should not use masks at all. This is false.

People absolutely should wear masks. The primary benefit is that wearing a mask, you protect others.
I could go on about the context of what we knew on March 5th, about how we did not believe there as asymptomatic transmission at that time and so we were thinking about masks for receiver protection rather than source control.
Read 5 tweets
8 Oct
Earlier tonight I speculated that Pence's unilateral conjunctivitis was a troubling sign that he might have COVID.

Is that a reasonable inference to draw? We can use Bayes' rule and a bit of Fermi estimation to figure it out.
Conjunctivitis may be a fairly rare symptom of COVID. One meta-analysis puts it at 1% of COVID cases. ncbi.nlm.nih.gov/pmc/articles/P…

So does it really tell us anything to observe that Pence had it?

Turns out it does.
You can plug in your own numbers of course. But the bottom line is that because (1) Pence has been exposed to COVID recently, (2) unilateral conjunctivitis is rare in healthy adults and (3) it is less rare in adults with COVID, the observation substantially increases our concern.
Read 4 tweets
8 Oct
Yesterday this toady, @CDCDirector Redfield....
...abused his office to write the Vice President a permission slip...
...that included disingenuous arguments about Pence's lack of close contacts with COVID carriers. For example Mike Lee, shown, tested positive last Friday but had been seated immediately behind Pence for over 15 minutes at the Rose Garden.
Read 6 tweets
5 Oct
“A metal detector that misses 10% of weapons—you’d never, ever say that’s our only layer of protection for the president,” -@ashishkjha

Dr. Jha is right. This has been a dreadful national security lapse.

Moreover, and I welcome corrections, my understanding is that the Abbott ID Now platform that the White House used to screen guests for the Rose Garden event had *not* been granted authorization by the FDA for use in this fashion.

Here is the latest EUA: fda.gov/media/136522/d…
The EUA was explicitly revised on June 1 to clarify that the ID Now platform was only for use in symptomatic patients within 7 days of onset. This excludes use for screening to find pre- or asymptomatic carriers.

That condition is retained in the Sept 17th update linked above.
Read 5 tweets

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal Become our Patreon

Thank you for your support!

Follow Us on Twitter!