Gonna hijack your comment section VP for some blogstacks… hope you don’t mind
I think it’s important to realize just how powerful social media has been in allowing the pseudoscience of community masking to thrive. Just as there could be no Witch Hysteria in the 15th-18th centuries without the printing press, there could be no mask hysteria without Twitter and Facebook. Mark Twain famously said “A lie can travel around the world and back again while the truth is lacing up its boots”, and that was before we had social media to speed up the transmission speed of flawed ideas. [0]
I propose that in the distant future Metaverse archeologists will realize that the sudden mask craze of the early 20's originated spontaneously from three circular sources:
First was the PNAS study "Identifying airborne transmission as the dominant route for the spread of COVID-19" [1] published June 11 2020, which made the causal claim that wearing masks in New York City, Wuhan, and Italy caused cases to drop. It was tweeted about 31,000 times directly, picked up in over 400 news stories (of which those were tweeted a million times), cited 400 times, and read nearly 2,000,000 times.
Nearly every op-ed from that point forward, when invoking the holy word "Mask", would immediately follow-up with the responsorial hymn "which we know work", with this article hyperlinked as the evidence.
The article was so terrible, there was an immediate retraction request [2] authored by dozens of epidemiologists, scientists, and physicians who pointed out the obvious - correlation doesn't equal causation (note that nearly all the authors of the retraction letter believe that masks work, just that this study wasn't evidence of that). The retraction request was ignored both academically and socially, garnering a few hundred tweets before it died in silence. Perhaps they should have done what Adam Pearce would later do - recreate the regression tool so you too could make it tell you whatever you wanted by moving the arbitrary dates around [3]. I've long wondered how Wakefield made it into Lancet over two decades ago, I think I better understand now.
Second was on July 3 2020 when the ironic twitter handle “@nuanceORDEATH” provided an n of 70 tweet storm to settle the science [4] where they cited "…*SEVENTY* papers...Includes 31 from 2020 alone (!!)". This tweet became the de-facto response anyone would get when they dared question how we could have missed the science on this for over 100 years. He didn’t seem to realize having 31 papers appear essentially out of thin air should give cause for concern (Ioaniddis comes to mind: “The likelihood that a claim will hold up is inversely proportional to the initial attention that it gets from other scientists and the media. Large, fast-moving, "hot" fields, which can yield large financial payoffs, tend to have the worst records.)
For anyone who took the time to actually go through the entire list, they would have found the 70+ articles fell victim to common pseudoscience fallacy of "counting your hits and forgetting your misses".
It leaves out any study which doesn't support masking (roughly 30 by my count?), even going so far as to cite Meta studies like "The Role of Facemasks and Hand Hygiene in the Prevention of Influenza Transmission in Households: Results From a Cluster Randomised Trial; Berlin, Germany, 2009-2011," (tweet 45/71), then re-cites 5 of the 7 the studies cited by the meta which support masks, but leave off the 2 which didn't find masks worked. ("Surgical Mask to Prevent Influenza Transmission in Households: A Cluster Randomized Trial" and "Findings from a household randomized controlled trial of hand washing and face masks to reduce influenza transmission in Bangkok, Thailand" were both excluded in the 70).
The tweet storm also included 3 of Raina MacIntyre's studies from 2008, 2009, and 2020, but excluded her study 2017. Is it possible the 2017 study was excluded because it noted "Finally, a recent study examined the efficacy of cloth masks compared to medical mask and control groups, and found that cloth masks may increase the risk of infection in HCWs"? [5]
In summary, what @nuanceORDEATH did was deliver the fallacy of “countless counterfeits” packaged into a convenient tweetstorm that now anyone could “mic drop” whenever some pesky skeptic wondered how it could be possible for us to not realize masks could stop the flu for the last 100 years (never mind apparently we already had 40+ studies showing masks worked, which is odd, considering around 7 million people have died from the flu in the 21st century alone). If it was hard to point out how flawed the PNAS study from Zhang et al was to the credulous masker, now there 70 potential counterfeits to sort through.
There was one problem from @nuanceORDEATH’s 1/n collection going fully viral – he isn’t an expert. He has no PhD, MPH, or DPH. And that leads us to…
Third, on November 5, 2020 social media personality Katelyn Jetelina aka “Your Local Epidemiologist” repackaged @nuanceORDEATH’s tweetstorm into the most shared Facebook posts amongst her 380,000 followers.[6] Now a certified public health expert with a massive public following had given the validity to the full set of 70 articles, uncritically repeating the high level findings of the models such as “(masks) stopped more than 200,000 COVID-19 cases in the US by May 22, 2020.”… “reduces the dose of virus a wearer might receive, resulting in infections that are milder or even asymptomatic” (I completely forgot this used to be a viable theory)…and “(if) 80% of the population wore masks, this would do more to reduce COVID-19 spread than a strict lockdown”.
She concluded “I’m not entirely sure why the efficacy of mask use is still up for discussion. But, nonetheless, this simple, low-cost intervention has the potential for a large impact.”
But in a classic case of CYA / Defensive Medicine, she closes with the “Swiss Cheese Model” noting “…it’s important to recognize that mask wearing it’s not the cure all. It has to be combined with other public health efforts for ultimate success”, which cements the unfalsifiability of community masking – when masks are worn and cases rise, we can blame other pieces of cheese not being combined with masks. Which we know work.
Note – I enjoy the writing of YLE and am a founding supporter of hers on substack, I just think on this particular issue she has missed the mark and played an unwitting role in spreading the pseudoscience of community masking more than possibly any other individual other than Donald Trump. And to be fair YLE isn’t always credulous when it comes to reviewing, she does on occasion take a deep dive to pick apart studies. [7]
___________________________
[0] Mark Twain didn’t actually say that after all [8]
Thank you for your analysis. MMWR reports have so many methodological limitations. It is appalling the CDC keeps accepting this "junk" and promoting it. Sad that the CDC has one agenda--nothing else matters besides getting vaccinated.
It is morbidly fascinating how inefficient the human mind is, even if trained in science. So my New Years resolution is to be just a little less cognitively dissonant than last year.
This is an important review for many of us. Brings back that one Hippocratic oath - do no harm. Pity some have lost sight of that as they throw caution to the wind. Can't blame good intentions so much as ignorance in understanding complex information. So many refuse to admit to uncertainty.
Gonna hijack your comment section VP for some blogstacks… hope you don’t mind
I think it’s important to realize just how powerful social media has been in allowing the pseudoscience of community masking to thrive. Just as there could be no Witch Hysteria in the 15th-18th centuries without the printing press, there could be no mask hysteria without Twitter and Facebook. Mark Twain famously said “A lie can travel around the world and back again while the truth is lacing up its boots”, and that was before we had social media to speed up the transmission speed of flawed ideas. [0]
I propose that in the distant future Metaverse archeologists will realize that the sudden mask craze of the early 20's originated spontaneously from three circular sources:
First was the PNAS study "Identifying airborne transmission as the dominant route for the spread of COVID-19" [1] published June 11 2020, which made the causal claim that wearing masks in New York City, Wuhan, and Italy caused cases to drop. It was tweeted about 31,000 times directly, picked up in over 400 news stories (of which those were tweeted a million times), cited 400 times, and read nearly 2,000,000 times.
Nearly every op-ed from that point forward, when invoking the holy word "Mask", would immediately follow-up with the responsorial hymn "which we know work", with this article hyperlinked as the evidence.
The article was so terrible, there was an immediate retraction request [2] authored by dozens of epidemiologists, scientists, and physicians who pointed out the obvious - correlation doesn't equal causation (note that nearly all the authors of the retraction letter believe that masks work, just that this study wasn't evidence of that). The retraction request was ignored both academically and socially, garnering a few hundred tweets before it died in silence. Perhaps they should have done what Adam Pearce would later do - recreate the regression tool so you too could make it tell you whatever you wanted by moving the arbitrary dates around [3]. I've long wondered how Wakefield made it into Lancet over two decades ago, I think I better understand now.
Second was on July 3 2020 when the ironic twitter handle “@nuanceORDEATH” provided an n of 70 tweet storm to settle the science [4] where they cited "…*SEVENTY* papers...Includes 31 from 2020 alone (!!)". This tweet became the de-facto response anyone would get when they dared question how we could have missed the science on this for over 100 years. He didn’t seem to realize having 31 papers appear essentially out of thin air should give cause for concern (Ioaniddis comes to mind: “The likelihood that a claim will hold up is inversely proportional to the initial attention that it gets from other scientists and the media. Large, fast-moving, "hot" fields, which can yield large financial payoffs, tend to have the worst records.)
For anyone who took the time to actually go through the entire list, they would have found the 70+ articles fell victim to common pseudoscience fallacy of "counting your hits and forgetting your misses".
It leaves out any study which doesn't support masking (roughly 30 by my count?), even going so far as to cite Meta studies like "The Role of Facemasks and Hand Hygiene in the Prevention of Influenza Transmission in Households: Results From a Cluster Randomised Trial; Berlin, Germany, 2009-2011," (tweet 45/71), then re-cites 5 of the 7 the studies cited by the meta which support masks, but leave off the 2 which didn't find masks worked. ("Surgical Mask to Prevent Influenza Transmission in Households: A Cluster Randomized Trial" and "Findings from a household randomized controlled trial of hand washing and face masks to reduce influenza transmission in Bangkok, Thailand" were both excluded in the 70).
The tweet storm also included 3 of Raina MacIntyre's studies from 2008, 2009, and 2020, but excluded her study 2017. Is it possible the 2017 study was excluded because it noted "Finally, a recent study examined the efficacy of cloth masks compared to medical mask and control groups, and found that cloth masks may increase the risk of infection in HCWs"? [5]
In summary, what @nuanceORDEATH did was deliver the fallacy of “countless counterfeits” packaged into a convenient tweetstorm that now anyone could “mic drop” whenever some pesky skeptic wondered how it could be possible for us to not realize masks could stop the flu for the last 100 years (never mind apparently we already had 40+ studies showing masks worked, which is odd, considering around 7 million people have died from the flu in the 21st century alone). If it was hard to point out how flawed the PNAS study from Zhang et al was to the credulous masker, now there 70 potential counterfeits to sort through.
There was one problem from @nuanceORDEATH’s 1/n collection going fully viral – he isn’t an expert. He has no PhD, MPH, or DPH. And that leads us to…
Third, on November 5, 2020 social media personality Katelyn Jetelina aka “Your Local Epidemiologist” repackaged @nuanceORDEATH’s tweetstorm into the most shared Facebook posts amongst her 380,000 followers.[6] Now a certified public health expert with a massive public following had given the validity to the full set of 70 articles, uncritically repeating the high level findings of the models such as “(masks) stopped more than 200,000 COVID-19 cases in the US by May 22, 2020.”… “reduces the dose of virus a wearer might receive, resulting in infections that are milder or even asymptomatic” (I completely forgot this used to be a viable theory)…and “(if) 80% of the population wore masks, this would do more to reduce COVID-19 spread than a strict lockdown”.
She concluded “I’m not entirely sure why the efficacy of mask use is still up for discussion. But, nonetheless, this simple, low-cost intervention has the potential for a large impact.”
But in a classic case of CYA / Defensive Medicine, she closes with the “Swiss Cheese Model” noting “…it’s important to recognize that mask wearing it’s not the cure all. It has to be combined with other public health efforts for ultimate success”, which cements the unfalsifiability of community masking – when masks are worn and cases rise, we can blame other pieces of cheese not being combined with masks. Which we know work.
Note – I enjoy the writing of YLE and am a founding supporter of hers on substack, I just think on this particular issue she has missed the mark and played an unwitting role in spreading the pseudoscience of community masking more than possibly any other individual other than Donald Trump. And to be fair YLE isn’t always credulous when it comes to reviewing, she does on occasion take a deep dive to pick apart studies. [7]
___________________________
[0] Mark Twain didn’t actually say that after all [8]
[1] https://www.pnas.org/content/117/26/14857
[2] https://metrics.stanford.edu/sites/g/files/sbiybj13936/f/files/pnas_loe_061820_v3.pdf
[3] https://roadtolarissa.com/regression-discontinuity/
[4] https://threader.app/thread/1279144399897866248
[5] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5705692/
[6] https://www.facebook.com/permalink.php?story_fbid=202002698114314&id=101805971467321
[7] https://www.facebook.com/permalink.php?story_fbid=206293624351888&id=101805971467321
[8] https://freakonomics.com/2011/04/quotes-uncovered-how-lies-travel/
Thank you for your analysis. MMWR reports have so many methodological limitations. It is appalling the CDC keeps accepting this "junk" and promoting it. Sad that the CDC has one agenda--nothing else matters besides getting vaccinated.
It is morbidly fascinating how inefficient the human mind is, even if trained in science. So my New Years resolution is to be just a little less cognitively dissonant than last year.
This is an important review for many of us. Brings back that one Hippocratic oath - do no harm. Pity some have lost sight of that as they throw caution to the wind. Can't blame good intentions so much as ignorance in understanding complex information. So many refuse to admit to uncertainty.
Loved this. Thank you. I made my 15 year old read this great link on Action Bias. We dont teach our kids cognitive biases or heuristics. https://thedecisionlab.com/biases/action-bias/