Masks Don’t Work: A Review of Science Relevant to Covid-19 Social Policy | Hacker News Reader

Masks Don’t Work: A Review of Science Relevant to Covid-19 Social Policy

rcreader.com

11 pointswalterbell22 comments

This is a particularly insidious form of advocacy because it mimics the writing of a scientific article while presenting a political opinion.

I stopped checking after the 4th linked journal article. All 4 had their results were misrepresented in the article. Not one suggested that masks didn’t work. 2 showed they did work, 1 showed there wasn’t much difference between N95 respirators and surgical masks and 1 was a small sample size result that showed family spread amongst 2 families, 1 who wore masks outside and one who didn’t.

Further they were a mix of respiratory illnesses mentioned. Colds, flues and SARS were all mixed and matched.

Thats before we get into the dubious claim that there was a 100% infection rate if any aerosolized droplets got through the mask.

"This is a particularly insidious form of advocacy because it mimics the writing of a scientific article while presenting a political opinion."

"Not one suggested that masks didn’t work. 2 showed they did work."

You're right, if I hadn't been reading the research papers on the matter over the past few weeks I reckon I'd likely be fooled (at least in some parts).

Dr Rancourt points out that: "The main transmission path is long-residence-time aerosol particles (< 2.5 μm), which are too fine to be blocked, and the minimum-infective dose is smaller than one aerosol particle."

What he fails to mention is how N95/P2 masks actually work. From my reading, these masks are made by stretching the plastic sheet until they start to part and form into string-like strands/fibers with gaps between them. He's right, the gaps are wider than 2.5 μm (somewhere around 5 μm) but the hidden factor is that the stretching of the plastic causes electrostatic charges to permanently form on the plastic through deformation somewhat akin to the way the charge is formed on an electret microphone and it is these charges that attract the aerosol particles which then collect on the plastic strands.

It seems Dr Rancourt's search of the literature was insufficiently deep (seems it's another instance of a non-expert in the field stepping above his station).

The most disconcerting aspect of Rancourt's report is the statement that it only takes one virion particle to infect a human. It was my understanding that the figure was considerably higher than this. If his quoted figure is correct then N95 mask won't work as they're only about 95% effective.

Until we get more definitive information on this I'll continue to wear my N95 masks in public.

Applying this article's standards:

Seat belts don't work, because they don't prevent all fatalities.

Don't use a safety rope while climbing, because your anchors can break and you can still fall.

Safety glasses don't work, because foreign debris can still enter through the sides.

Don't wear a life jacket, because you can still die in cold water.

Don't get your teeth cleaned, because it doesn't prevent every instance of tooth decay.

This is ridiculous. If wearing masks reduces transmission rates by 10%, it's still easily worth it. The cost of masks is tiny compared to the enormous costs of reduced economic activity (not to mention infection-related healthcare).

"This paper was originally published at Rancourt's account on ResearchGate.net. As of June 5, 2020, this paper was removed from his profile by its administrators"

http://activistteacher.blogspot.com/2020/06/covid-censorship...

A few german cities could reduce the number of new infections to 0 early in the pandemic after making masks (including community masks) mandatory. The university town Jena in particular was frequently cited for its pro-mask campaign. I thought they had written a paper about their results, with statistically corrected comparisons to cities of similar size; but can’t find it. Those studies indicating that masks might not work were known in the public and political discourse at that time (e.g. they were discussed in the coronavirus update podcast). The city governments went for the policy anyway.

Japan and South Korea are also still doing much better than nations where masks are not commonly seen...

Neither the author nor the publication seem credible sources for information on this topic. The author is a Physics professor who was fired for giving all of his students A+ in a course after coming under fire for putting a course description in the catalog to get students to enroll, and then teaching something completely different in the course. (They called it academic squatting)

I’m writing this with the caveat that it takes an order of magnitude more effort to refute bull than to produce it.

The way I understand the current mask guidelines is that your mask isn’t there to help you, but to help other people. I clicked through a few of the studies and they looked at the effectiveness of masks in preventing the wearer from getting infected in uncontrolled situations, not the effectiveness in preventing the wearer from spreading it to other people.

But this doesn't hold up in the real world.

Particularly since viral load matters wrt severity of infection for COVID-19.

Do you want to get shot with 100 count bird shot or 70 count with a shorter effective distance? An individual pellet probably won't do much harm, but multiple pellets damaging the same organ will cause non-linear increases in damage and severity of outcome.

Do you mean viral load after being infected, or viral load in a hypothetical asymptomatic person who is showing no symptoms, not quarantined, and the primary threat vector being mitigated by masks? Are there studies on viral load in pre-symptomatic or asymptomatic individuals with Covid-19 or other viruses?

The data for the reduction in viral load came from and Italian experiment with symptomatic patients. A simple surgical mask caught 30% of viral particles when they coughed vs no mask.

There might be some level of virus that your body can handle without even developing antibodies / becoming symptomatic, but beyond that threshold, you actually get sick.

Anecdotally, we have this for asymptomatic spreaders and potential casualties.

> 2 infected hairstylists

> Masks worn by everyone

> 0 of 140 clients infected

https://www.livescience.com/hair-stylists-infected-covid19-f...

What do people think about the article suggesting that R0 is influenced by humidity? Is it time to add giant humidifiers everywhere?

https://news.yahoo.com/sunlight-destroys-coronavirus-quickly...

> ... experiment that was carried out at the National Biodefense Analysis and Countermeasures Center in Maryland. It showed that the virus's half-life -- the time taken for it to reduce to half its amount -- was 18 hours when the temperature was 70 to 75 degrees Fahrenheit (21 to 24 degrees Celsius) with 20 percent humidity on a non-porous surface. This includes things like door handles and stainless steel. But the half-life dropped to six hours when humidity rose to 80 percent -- and to just two minutes when sunlight was added to the equation. When the virus was aerosolized -- meaning suspended in the air -- the half-life was one hour when the temperature was 70 to 75 degrees with 20 percent humidity. In the presence of sunlight, this dropped to just one and a half minutes. Bryan concluded that summer-like conditions "will create an environment (where) transmission can be decreased."

Ridiculous. What does “not work” even mean?

If you block 90% of a surface, that area is blocked. Of course, not all of it.

I mean: photons are even smaller, and they happen to be blocked. Yes, this is not a good example but it is not totally wrong.

Ah yes, I too get my scientific news from sites called "River Cities Reader" that publicly solicit donations at the end of articles. Clearly, an institution we should trust when it comes to applying the scientific method.

I also don't trust NPR. They're always begging for money.

False equivalence fallacy. NPR isn't comparable to rcreader, which has a circulation of 19,700; NPR gets 52 million impressions per month.

NPR articles also generally aren't comparable to this one with direct sources. Are you really going to accuse me of a fallacy while committing your own Argumentum ad populum?

So like NPR?

A more useful rebuttal might be to find a single counterexample to this statement:

> No RCT study with verified outcome shows a benefit for HCW or community members in households to wearing a mask or respirator. There is no such study. There are no exceptions.

EDIT: That said, there's this note at the bottom:

> This paper was originally published at Rancourt's account on ResearchGate.net. As of June 5, 2020, this paper was removed from his profile by its administrators at Researchgate.net/profile/D_Rancourt. At Rancourt's blog ActivistTeacher.blogspot.com, he recounts the notification and responses he received from ResearchGate.net and states, “This is censorship of my scientific work like I have never experienced before.”

It's certainly possible this was just plain censorship, but I'm more inclined to believe that there was something actually wrong with it.

Fortunately, the hyperlink was invented to mitigate issues of transitive trust. The following institutions are cited by this article. Thanks to technology that is merely decades old, a single click is sufficient to consult these primary sources:

  American Journal of Infection Control
  Epidemiology and Infection
  Influenza and Other Respiratory Viruses
  CMAJ (Canadian Medical Association)
  Clinical Infectious Diseases
  JAMA (published since 1883)
  JEBM (Journal of Evidence-Based Medicine)

I can also cite papers in various journals. However, it only matters if the citations are relevant. In this case the author is largely citing papers that show how wearing masks can prevent the wearer from becoming infected, not preventing other people from becoming infected.

Do you know of studies on the effectiveness of masks on prevention of other people from becoming infected?

Reply on news.ycombinator.com