Hacker News Reader

Comment by jk700 | Hacker News Reader

> We have lots of RCTs on mask wearing (mostly in healthcare professionals) and they struggle to find a benefit.

This is false. If you look around and read actual studies, not demand other people to do it for you and not cherry pick garbage studies, you'll find even real life "experiments" of healthcare workers in the US during COVID-19 not wearing masks with rapidly rising number of COVID-19 cases then mandated to wear masks with subsequent rapid drop of cases.

What you won't find though is studies supporting the assertion that sick person wearing a mask protects others, not the wearer. Those are the studies that show little to no benefit and always did actually. You protect others with a mask by not catching the virus yourself, it's a second order effect.

View on HN

I've read the actual studies. They don't show benefits.

> not demand other people to do it for you and not cherry pick garbage studies,

The reason I ask you to provide the studies you think are persuasive is so we avoid the tedious discussion about cherry picking. I cannot cherry pick articles that you supply.

If you want my link: https://www.thelancet.com/journals/lancet/article/PIIS0140-6...

Lancet is a reputable high impact publication. The paper is peer reviewed.

> Transmission of viruses was lower with physical distancing of 1 m or more, compared with a distance of less than 1 m (n=10 736, pooled adjusted odds ratio [aOR] 0·18, 95% CI 0·09 to 0·38; risk difference [RD] −10·2%, 95% CI −11·5 to −7·5; moderate certainty); protection was increased as distance was lengthened (change in relative risk [RR] 2·02 per m; pinteraction=0·041; moderate certainty). Face mask use could result in a large reduction in risk of infection (n=2647; aOR 0·15, 95% CI 0·07 to 0·34, RD −14·3%, −15·9 to −10·7; low certainty), with stronger associations with N95 or similar respirators compared with disposable surgical masks or similar (eg, reusable 12–16-layer cotton masks; pinteraction=0·090; posterior probability >95%, low certainty).

For masking they find only "low certainty" evidence, and that's with N95s or 12 to 16 layers of cotton. (Most people are wearing 3 to 4 layers).

Low certainty comes from GRADE and means "The true effect might be markedly different from the estimated effect" -- masks may be markedly more, or markedly less, effective than they found.

> What you won't find though is studies supporting the assertion that sick person wearing a mask protects others, not the wearer. Those are the studies that show little to no benefit and always did actually. You protect others with a mask by not catching the virus yourself, it's a second order effect.

Oh-ho, so here we see a clearly shifted goalpost. Everyone else in this thread is saying that the science is obvious and that you wear a mask to prevent your infection from spreading to other people. So, does the science support "source control" or not?

It seems outrageously wrong to me to summarize that Lancet article by saying "We have lots of RCTs on mask wearing (mostly in healthcare professionals) and they struggle to find a benefit." or "I've read the actual studies. They don't show benefits".

The Lancet article looks at dozens of studies of the relationship between mask-wearing and infection with coronaviruses. (Some SARS, some MERS, some COVID-19.) Those studies fall, without exception, into the three following categories.

1. A few saw no cases of infection with or without masks, and therefore tell us nothing about whether masks help and how much.

2. One saw no cases of infection with masks and just two without masks (in a fairly small sample). It somehow manages to report a relative risk of 1.03 for wearing masks (!), which I can only assume is the result of some statistical adjustment that doesn't quite make sense in this case. (The real conclusion from this one is: "This study doesn't provide enough data to draw any conclusions.")

3. In every single one of the others, wearing a mask was associated with lower probability of infection. Some of the studies weren't powerful enough for the difference to be (on its own) statistically significant, but all of them pointed in the same direction and the underpowered ones had relative risk point-estimates similar to those of the not-underpowered ones.

Sure, they report this as "low certainty" because, e.g., none of these studies was a proper randomized controlled trial. Sure, we would do well to get more data. But I see no possible way to characterize what it says in the ways you have.

They don't "struggle to find a benefit". They readily find a very considerable benefit. They don't "not show benefits". They absolutely do show benefits. There are issues: we don't have e.g. nice big RCTs where someone has somehow contrived to have two otherwise identical large populations of the general public make different mask-wearing choices, and that means that there is inevitably some uncertainty in applying the results to the question of what benefit there is from getting the public to wear masks. But it 's pure grade-A bullshit to say that the studies "don't show benefits".

Reply on news.ycombinator.com