@PolitiFact apparently fact-checked the comments made about the CDC mask study politifact.com/factchecks/202… written by Bill McCarthy

Here is the study: cdc.gov/mmwr/volumes/6…

Should we fact-check PolitiFact as well...? 🧵
"It's from a very small sample...that may not be representative"

Small sample sizes are susceptible to bias. The smaller they are, the less representative and more biased. Self-reports are also heavily biased and unreliable. Let's remember this.

RATING: True
Ok, so here we learn that the CDC wasn't measuring mask effectiveness (true), and yet we'll see people drawing conclusions about mask effectiveness from this study.

In fact Politifact themselves do this in politifact.com/factchecks/202… written by Noah Y. Kim

RATING: Misleading
Remember. Correlation != causation. Also, the correlation between on-site eating does NOT indicate that masks had anything to do with the spread. It MAY have, but we don't know because the study didn't check that.

RATING: MORE THAN MOSTLY, BUT I'LL BE NICE, MOSTLY FALSE
All we do know from this UNREPRESENTATIVE SMALL SAMPLE size study that is susceptible to SELF-REPORTING BIAS is that CoV+ patients were more likely to have gone to a restaurant at least 1x in the 2wk prior to illness than were CoV- patients.
The last unhighlighted sentence in prev image says CoV+ PTs were less likely to report seeing "almost all" others wearing masks OR social distancing. Bc this stat includes mask wearing AND social distancing together, we can't really say anything about mask wearing itself...
"The rates of people who always wore a mask in each group were not 'statistically different.'"

AKA, there is NO EVIDENCE that mask wearing made a diff. Generally, in order for a result to be statistically significant/meaningful, the p-value should be < 0.05. It is 0.86 here.
In this "fact"-check: politifact.com/factchecks/202… they cite virologist Ben Neuman who says the following. Just in case we get confused: there's also nothing in this study that indicates that not wearing a mask is assoc w/more coronavirus, bc well...they didn't study that remember?
This is apparently what they are fact-checking in this 2nd post. Politifact is correct when they say that the FB post misrepresented the data in the CDC study.

RATING: Alright fine...I'll give you this one, but not so fast...
The 1st img are the claims made by the FB post. The 2nd is Politifact's fact-check claim. The CDC study does not study whether masks collect the virus, or whether the particles stay on our faces vs dissipate...
It does hwe point to data abt correlations btw mask wearing & no mask. It found that there was no statistically significant diff btw the two groups. Basically, we don't know. This is not "opposite" of what the FB post claims. Therefore:

RATING: Wth do you think opposite means?
Here's another PF claim.

No, it does not "probably" indicate this. What it actually indicates is that FREQUENT ON-SITE EATING probably plays a role in the spread of the virus, not masks. Bc mask use+social distancing are grouped under 1 category, we can't tell what did what...
Technically yes, more ppl who tested CoV- reported wearing masks "always" than those who tested CoV+. BUT, remember, p-value should be <0.05 to be significant. It is 0.86 here. Dumb fact-check.

Also more CoV- ppl reported anything than did CoV- ppl 🤷

RATING: Misleading as fucc
Neuman is a real scientist and he greatly points out the issue with self-reporting bias. This makes much of the data in this CDC study of low certainty and thus unreliable. That doesn't mean we throw the entire study out, but it does mean you should take it w/a bag of salt...
Politifact also cites some studies that apparently show mask mandates reduced infections. I haven't gone through those studies yet, but I am willing to bet that this is a classic error of thinking correlation = causation. It does not. The studies on mask mandates...
and how much they contributed to the ⬇️ in infections don't always tease out effect other interventions had, or how much of the effect can be due to mask mandates, & whether ppl even complied, or why mandates failed in the past as well as in many countries right now...
If I had to give this a rating right now without having done a deep dive into those specific studies I would say:

RATING: Mostly misleading, pending review after which it will likely be mostly BS
So let's summarize.

The findings don’t prove masks are effective, either. Mask mandates lead to counteractive behavior & have not been proven effective. There is no clear evidence suggesting mask mandates reduce spread.

...
The correlation does not suggest this. This is a hypothesis. The correlation suggests that going to a restaurant may increase risk. Whether that is because of removing masks or because of lack of social distancing is unclear bc this study did not look at that.
With respect to the first Politifact "fact" check, I also rate Politifact's statement false.

And finally...
I have no idea what Politifact's definition of "opposite" is but this ain't how it's used normally, lol. Studies on mask mandates show no evidence, or evidence of low reliability. Evidence of individual symptomatic patient mask use do show effectiveness. Hwe, see 3rd image.
I rate this post as dumb, stoopid, and dumb huhh #GoobaFacts
The images of the tables and the study text come from this: cdc.gov/mmwr/volumes/6…

It's the same study
The "respectively" refers to (1) restaurant vs (2) bar/coffee shop

• • •

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

Keep Current with Critical Critical Theory

Critical Critical Theory 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 @CCRTheory

17 Oct
If we put this finding along w/findings that show immunity could last up to 4mo:

-nejm.org/doi/full/10.10…
-immunology.sciencemag.org/content/5/52/e…
-medrxiv.org/content/10.110… (PREPRINT but similar results)

along with the hypothesis that...🧵
If the virus was in our population well before we began to detect it, it is curious to wonder how many people had already been infected...

& if these SP estimates are low estimates bc, for ex, we might miss the milder or recovered (a)symptomatics (bmj.com/content/370/bm…)...
..I wonder what that means in context of where we are & how we perceive the threat. It could mean that not only are we underestimating the true # of infections but IFR rate would drop along w/CFR rate

It should make us stop & wonder whether this is an alt path forward, ie GBD Image
Read 5 tweets
16 Oct
"It's really disheartening to see the scientific community suppress ppl who didn't agree w/the folks who control the msging. Fauci and Birx have 1 view of the science, but it is not the only view." - Dr. Jay Bhattacharya

@gbdeclaration

"When people say 'follow the science' what they often mean is censor scientists who don't agree with some scientists, the people who are controlling the policy" - Dr. Jay Bhattacharya

Science is nothing without discourse. Democracy is nothing without debate.
"In the US we don't have a national policy, we couldn't because we have federalism. We have state policies"

This is KEY to understand. Our government doesn't operate like a centralized authoritarian regime. State gov is most responsible for how they handled Covid response.
Read 4 tweets
16 Oct
Just started watching this #TownHall w/Trump. The first 20 min are so frustrating. This interviewer is obsessed w/masks but she's citing data that itself says the results are of "low certainty" and inconclusive bc CORRELATION IS NOT CAUSATION. Even CDCs own site says this...
The research basically says, "well masks got mandated at this date, and the cases went down on this date...so yeah. also correlation is not causation, and we don't know if this is bc of masks or bc of other measures".

People then use this to wrongly justify their own beliefs...
Additionally, this interviewer goes on to say, "estimates say 10% of the country got covid, that means 90% are vulnerable".

This is so stupid. That does NOT mean 90% are vulnerable. That means 90% COULD be vulnerable. The susceptible pop is likely a hell of a lot lower...
Read 4 tweets
15 Oct
What is GBD really promoting? Isolation, targeted testing+tracing+screening, reallocation of resources to bolster the aforementioned efforts, using the ~0.0003 IFR estimated for <49yo in order to allow natural immunity,...
...leverage SP studies indicating IgG lvls up to 90d post-symptoms, prevent unnecessary indiscriminate closures of SBAs, prevent spikes in unemployment, prevent child abuse (), prevent suicides+famine+missed screenings+drug abuse+inc in addictions...
1 diff btw the plans is that Gregg is advocating for universal mask wearing. Ignoring the fact that most studies showing any benefits at the public health lvl are of low certainty (even the one on CDC - cdc.gov/mmwr/volumes/6…), how do you plan on enforcing this? Mandates? Image
Read 10 tweets
15 Oct
Common concern is that not all vulnerable can be isolated bc "ppl make mistakes". This is true regardless of lockdown protocol u use - gen or targeted. Diff is that w/targeted approach, u don't end up accruing all the other damages assoc w/indiscriminate sol'ns...
Other damages: SBA closures, unemployment, child abuse, sex abuse, mental health issues, suicides, addiction, missed medical procedures, famine etc

Efficient tracing/testing in targeted approach in order to isolate what can prevent from spreading seems a better approach...
Clear dissemination of protective protocols for vulnerable & their primary contacts seems important. From what I can tell, all this constant lockdown, reopen, lockdown, reopen is doing is spreading the deaths across a larger window while damaging the economy + ⬆️ health risks...
Read 4 tweets
13 Oct
citeseerx.ist.psu.edu/viewdoc/downlo…

In 2006, the WHO Writing Group said, "Ill persons should remain home when they first become symptomatic, but forced isolation and quarantine are ineffective and impractical.", about influenza pandemic
ncbi.nlm.nih.gov/pmc/articles/P…

What data changed? ImageImageImage
And this is what the WHO had to say about mask studies (from the NCBI paper)

I wonder what data changed so drastically to validate the lvl of politicization of mask efficacy. It doesn't add up. If we were rlly listening to the science, how does this study fit the narrative? Image
We began this pandemic w/a consensus that lockdown was only intended to flatten the curve, which afaik it did. After that point, it seems that we should have switched to a more targeted approach for protecting the vulnerable. That's what the science told us to do.

So why then...
Read 6 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!