Imagine if I wrote a paper advocating that everyone wore a rabbit's foot to protect from car accidents, and then a year later I published a paper showing an absolute reduction in car accidents by .0009. Would the Washington Post call this "the nail in the coffin that rabbit foots work?" Or instead would you expect some rigorous skepticism?
What they hell happened to science?
The author of Bangladesh Mask Study, Jason Abaluck, previously authored "The Case for Universal Cloth Mask Adoption and Policies to Increase Supply of Medical Masks for Health Workers" on 4/6/2020 - https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3567438 which concludes:
"The economic case for universal mask wearing is convincing and urgent, but the moral
need to provide adequate equipment to frontline healthcare workers is an even higher
imperative. Enacting policies to increase medical mask production, and concurrently
encouraging the widespread production and use of cloth masks can achieve both
objectives. Public officials should encourage and support universal cloth mask adoption
immediately."
Is it possible that because he started the Bangladesh Mask study already convinced of the conclusion, that masks work, this may have created subconscious bias in his design, gathering of data, and synthesis of results?
Look how his co-authors framed the minimal efficacy findings in their study when given the chance to summarize their findings in the Washington Post.:
(this is typically paywalled but I used the gift option to bypass that)
"We conducted the largest study on masks and covid-19: They work"
Hilarious experts include: "We now have the results, and the answer to both questions is an emphatic yes."..."The results were striking."...."We found that surgical masks averted 1 in 3 symptomatic infections among those aged 60 and older."...."Masks may have even greater potential than our study was able to demonstrate."...."Our study showed conclusively that masks are a cost-effective way to reduce infections and demonstrated that a mask-promotion strategy can work."
Note they didn't bother to point out they found cloth masks didn't work, but instead data-drudged their way to the "1 in 3 cases averted over the age of 60" without a second glance at the ridiculousness of that statement.
The press uncritically regurgitated the hype without ever reading the actual study.
Nature: "Face masks for COVID pass their largest test yet"
"“This really should be the end of the debate,” says Ashley Styczynski, an infectious-disease researcher at Stanford University in California and a co-author of the preprint describing the trial. The research “takes things a step further in terms of scientific rigour”, says Deepak Bhatt, a medical researcher at Harvard Medical School in Boston, Massachusetts, who has published research on masking."
WebMD: "Large Study Confirms Masks Work to Limit COVID-19 Spread"
Former Director of the CDC Tom Friedman: "Masks work. A new study of more than 340,000 adults across Bangladesh found that the more people wore masks, the less spread of Covid there was"
I have spent many hundreds of hours trawling through the data for this study (https://gitlab.com/emily-crawford/bd-mask-rct). According to that data, some individuals were given a serology test at endline despite being "asymptomatic" (that is, not meeting the study criteria for COVID-19 symptoms, which are based on the WHO definition of a probable COVID-19 case). I am not sure whether or not these individuals were randomly selected.
Out of 2375 asymptomatic individuals tested in the control group, 449 were seropositive (18.9%).
Out of 2987 asymptomatic individuals tested in the intervention group, 543 were seropositive (18.2%).
This is a smaller effect size (3.8%) than the one seen in the study, which is already on the edge of statistical significance.
Also notably, the percentage of symptomatic individuals who tested seropositive is 29.9%. So symptomatic individuals were more likely to test seropositive than asymptomatic individuals, but not by as much as one would expect or hope (especially given the authors' claim that they ran the symptom surveys in order to cut down on the number of blood tests they would need to perform to find seropositive results). This is one of many indicators that the symptom survey is unreliable.
We must keep in mind that the surveys themselves are fraught with problems. A researcher went to each enrolled household, asked "has anyone in your household experienced any of these symptoms in the past four weeks?", and read out a list of 11 symptoms. That list of symptoms doesn't even match what's reported in the Science paper. (Here is the full script: https://docs.google.com/document/d/14li6x3Wg0INClpBfuTzhFcALqC6aFyNtYc9DjVurcv0/edit.) This is a far cry from meeting any reasonable standard for COVID-19 diagnosis, including the one used by the FDA for vaccine trials (https://www.fda.gov/media/142143/download).
Can you explain why wearing a surgical mask does not make any difference in transmission ? Not worn properly ? Not worn 24/7 ? It is counter intuitive and needs an explanation, no ?
Imagine if I wrote a paper advocating that everyone wore a rabbit's foot to protect from car accidents, and then a year later I published a paper showing an absolute reduction in car accidents by .0009. Would the Washington Post call this "the nail in the coffin that rabbit foots work?" Or instead would you expect some rigorous skepticism?
What they hell happened to science?
The author of Bangladesh Mask Study, Jason Abaluck, previously authored "The Case for Universal Cloth Mask Adoption and Policies to Increase Supply of Medical Masks for Health Workers" on 4/6/2020 - https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3567438 which concludes:
"The economic case for universal mask wearing is convincing and urgent, but the moral
need to provide adequate equipment to frontline healthcare workers is an even higher
imperative. Enacting policies to increase medical mask production, and concurrently
encouraging the widespread production and use of cloth masks can achieve both
objectives. Public officials should encourage and support universal cloth mask adoption
immediately."
Is it possible that because he started the Bangladesh Mask study already convinced of the conclusion, that masks work, this may have created subconscious bias in his design, gathering of data, and synthesis of results?
Look how his co-authors framed the minimal efficacy findings in their study when given the chance to summarize their findings in the Washington Post.:
(this is typically paywalled but I used the gift option to bypass that)
"We conducted the largest study on masks and covid-19: They work"
https://wapo.st/3DAYmPQ
Hilarious experts include: "We now have the results, and the answer to both questions is an emphatic yes."..."The results were striking."...."We found that surgical masks averted 1 in 3 symptomatic infections among those aged 60 and older."...."Masks may have even greater potential than our study was able to demonstrate."...."Our study showed conclusively that masks are a cost-effective way to reduce infections and demonstrated that a mask-promotion strategy can work."
Note they didn't bother to point out they found cloth masks didn't work, but instead data-drudged their way to the "1 in 3 cases averted over the age of 60" without a second glance at the ridiculousness of that statement.
The press uncritically regurgitated the hype without ever reading the actual study.
Nature: "Face masks for COVID pass their largest test yet"
https://www.nature.com/articles/d41586-021-02457-y
"“This really should be the end of the debate,” says Ashley Styczynski, an infectious-disease researcher at Stanford University in California and a co-author of the preprint describing the trial. The research “takes things a step further in terms of scientific rigour”, says Deepak Bhatt, a medical researcher at Harvard Medical School in Boston, Massachusetts, who has published research on masking."
WebMD: "Large Study Confirms Masks Work to Limit COVID-19 Spread"
https://www.webmd.com/lung/news/20210907/masks-limit-covid-spread-study
Former Director of the CDC Tom Friedman: "Masks work. A new study of more than 340,000 adults across Bangladesh found that the more people wore masks, the less spread of Covid there was"
https://twitter.com/DrTomFrieden/status/1434901885106876416
LiveScience: "Huge, gold-standard study shows unequivocally that surgical masks work to reduce coronavirus spread"
https://www.livescience.com/randomized-trial-shows-surgical-masks-work-curbing-covid.html
"Results from a massive study in Bangladesh unequivocally show that surgical masks reduce the spread of SARS-CoV-2, scientists say. "
HOW??!! How can science be in such a state of disarray where we are so easily fooled? How can journalists covering science fall for such nonsense?
Why did it a bad cat to debunk this and not The New York Times or Nature or The Lancet?
https://boriquagato.substack.com/p/bangladesh-mask-study-do-not-believe
We are going to look at this as a mini-dark age.
I have spent many hundreds of hours trawling through the data for this study (https://gitlab.com/emily-crawford/bd-mask-rct). According to that data, some individuals were given a serology test at endline despite being "asymptomatic" (that is, not meeting the study criteria for COVID-19 symptoms, which are based on the WHO definition of a probable COVID-19 case). I am not sure whether or not these individuals were randomly selected.
Out of 2375 asymptomatic individuals tested in the control group, 449 were seropositive (18.9%).
Out of 2987 asymptomatic individuals tested in the intervention group, 543 were seropositive (18.2%).
This is a smaller effect size (3.8%) than the one seen in the study, which is already on the edge of statistical significance.
Also notably, the percentage of symptomatic individuals who tested seropositive is 29.9%. So symptomatic individuals were more likely to test seropositive than asymptomatic individuals, but not by as much as one would expect or hope (especially given the authors' claim that they ran the symptom surveys in order to cut down on the number of blood tests they would need to perform to find seropositive results). This is one of many indicators that the symptom survey is unreliable.
We must keep in mind that the surveys themselves are fraught with problems. A researcher went to each enrolled household, asked "has anyone in your household experienced any of these symptoms in the past four weeks?", and read out a list of 11 symptoms. That list of symptoms doesn't even match what's reported in the Science paper. (Here is the full script: https://docs.google.com/document/d/14li6x3Wg0INClpBfuTzhFcALqC6aFyNtYc9DjVurcv0/edit.) This is a far cry from meeting any reasonable standard for COVID-19 diagnosis, including the one used by the FDA for vaccine trials (https://www.fda.gov/media/142143/download).
Even more confusion over the impact and protection offered by cloth and surgical masks.
Thank you for all these precious detailed info !
Can you explain why wearing a surgical mask does not make any difference in transmission ? Not worn properly ? Not worn 24/7 ? It is counter intuitive and needs an explanation, no ?