Hacker News Reader

That's not what I have been reading about the study. The evidence seems to be on the pro-mask side, but it does not appear to be _that_ clear. I'm not an expert in masks, health, or even statistics, so I'm not here to say you are right or wrong. I'm just pointing out that there is a debate. That itself is enough to want to look for more studies, don't you think?

To answer your question, the point of more studies is to increase confidence in our opinions and decisions, make better findings, realize errors, etc. And you are quickly jumping to conclusions with "folks like you are just going to dismiss them out of hand".

Now, my question would be: what do you mean by "folks like you"?

Have you looked at the study itself? From the abstract: “Mask distribution and promotion was a scalable and effective method to reduce symptomatic SARS-CoV-2 infections.”

The existence of “debate” doesn’t mean anything. Anti-maskers are always going to dispute studies like this, no matter how clear the evidence

Have you looked at the study itself? It found that cloth masks do nothing and surgical masks help a tiny bit. (control: 8.60%; treatment: = 7.63%) I suppose you could parse this as "masks work", I parse this as "if you are really worried about airborne virus, wear N95."

Page 12: "no statistically significant effect for cloth masks"

https://www.poverty-action.org/sites/default/files/publicati...

Page 12 paragraph for context:

"We find clear evidence that the intervention reduced symptoms: we estimate a reduction of 11.6% (aPR = 0.88 [0.83,0.93]; control: 8.60%; treatment: = 7.63%). Additionally, when we look separately by cloth and surgical masks, we find that the intervention led to a reduction in COVID-like symptoms under either mask type (p = 0.000 for surgical, p = 0.066 for cloth), but the effect size in surgical mask villages was 30-80% larger depending on the specification. In Table S9, we run the same specifications using the smaller sample used in our symptomatic seroprevalence regression (i.e. those who consented to give blood). In this sample we continue to find an effect overall and an effect for surgical masks, but see no statistically significant effect for cloth masks."

> Have you looked at the study itself?

Please don’t do this here. This is against the HN guidelines. https://news.ycombinator.com/newsguidelines.html

Please don't comment on whether someone read an article. "Did you even read the article? It mentions that" can be shortened to "The article mentions that."

Don't get me wrong - the Bangladesh study suggested that masks were a good public health initiative and should be used. What it didn't do was reassure me that they were a strong individual NPI.

In much the same way that not masking was recommended early in the pandemic as a public health initiative to reserve the masks for first responders - wasn't particularly good advice for the individual.

The world has 8 billion people on it. Shouldn't we care about what helps everyone instead of what only helps you specifically?

If masking helps everyone, then you should do it, even if it somehow magically didn't help you.

So - several threads here. It’s not unreasonable to try to understand what helps globally as well as individually. Early on, when supply chains for masks were stretched or non-existent, we were instructed not to wear masks. This was good advice globally, as it reserved the limited supply for those who could most benefit from them (first responders) - but bad advice individually. If you were looking for guidance to protect yourself and those you loved, following global advice was not the best path.

Likewise, the current guidance that wearing masks - particularly if they are well fit FFP3/99%+ filtering, is good advice globally, but it’s unclear if it’s sufficient to reduce the risk down to <.1% for the individual (the point at which I would pretty much not worry about social distancing and avoiding stores, workplace, etc… I’m <65 and vaccinated, So I would be comfortable with a 1/1000 risk of catching Covid, particularly as I can usually avoid those scenarios)

The question isn’t whether masks are an important NPI - they absolutely are - particularly if we could convince people to avoid going indoors unmasked and ensuring they are wearing (and not taking off) well fit FFP3s, the question is whether that is sufficient to mostly eliminate the risk of catching Covid.

One lesson I learned here is that you really need to expand on your thoughts and not assume everyone else is already on the same wavelength as yourself.

It "produced clear evidence that masks work"...if you think a 10% reduction in the speed of transmission is "working". That's a pretty minimal return for a large intervention (30% increase in mask-wearing).

Plus, they had to throw out 26 villages with 0 infections in order to get that much of an effect. Now there is a certain logic to that, but I wager that if most of those 26 villages had been ones with the intervention, they would not have thrown them out; I'm not aware of that part of the analysis being pre-registered.

They probably don't have a non-zero impact, but it appears to be pretty small, even with a massive intervention like the one in Bangladesh.

Here's a breakdown of that Bangladesh study:

https://www.skirsch.com/covid/BangladeshMaskReview.pdf

Please don't cross into personal attack when posting to HN.

https://news.ycombinator.com/newsguidelines.html

Lazy dismissal of scientific work is something of its own epidemic on HN these days. "Ah, but this study doesn't address this laundry list of things I just made up, and I haven't bothered to look at other sources, and I probably haven't bothered to even read this study, actually." It's especially common on COVID-related studies amongst the anti-maskers and anti-vaxxers on the site, but I have been seeing the same kind of lazy dismissal on lots of other posts as well. It's effectively an evolution of the Gish gallop.

I find it both dreary and frustrating, personally, and it's not terribly surprising that people like the poster above call people out on it.

I agree. Every HN post is full of "well actually" non-experts who love dismissing the work of experts with a few sentences of layman talk.

It's endemic on HN because it's endemic on the internet because it's endemic to humans. There's no "lately" about this.

The question is, what do we do about it? Posting even worse comments, like the GP, is obviously not helping.

I like to believe that HN has at least been partially successful in holding itself to a higher standard than the internet at large, through a combination of community guidelines, overall community belief in the guidelines, and thoughtful moderation.

You could, of course, say the same thing about personal attacks: They're endemic on the internet and to humanity, but we've mainly kept them out of HN using the above measures.

Lazy dismissal is often the top comment on any given post, COVID or not. It's harmful to discussion and I would argue antithetical to the hacker spirit: If we only allow perfect airtight products/proofs, we miss out on discovery and exploration. Perhaps some guideline to avoid lazy dismissals (or something along those lines) would go a long way.

I think my comment:

I keep looking, and I find it kind of incredible that we're two years into a pandemic, and except for one study in Bangladesh with masks (that wasn't really that clear) - there is nothing to say that any of these masks are working.

Was what was seen as a lazy dismissal. I should have spent more time expanding on why I didn't think it was really clear, but the point I was really trying to get at - is that there has been a lack of experimentation, trials, research - despite the millions of lives lost and many billions of dollars spent (Trillions?) - and I was really hoping someone would respond with a pointer to such a RCT that we could all read.

Unfortunately the takeaway wasn't that we should spending more time to demonstrate that masks work, but that I was dismissing that they did. The absolute opposite of what I was attempting to do. The failure in communication is clearly mine, and I'll try and be a little more expansive in the future.

For the record, what I really should have written was,

"To date, there have been no studies that have reassured me that I could go indoors into a crowd of infected, unmasked people, confident that as long as I kept my well-fit 3M 8210 N95 mask on, that I ran < 0.1% chance of being infected by Covid. The Bangladesh Study that showed 8.6% of people in the control villages reporting Covid symptoms and 7.6% in the experimental villages, did not provide me that kind of reassurance, thought I think it provided statistically significant results that masks as a general public health NPI are effective at reducing transmission of the Coronaviruses under study at the time."

Of course - Omicron essentially reset all expectations for transmissibility.

Yeah, that's exactly where I read to, which registered as more anti-masker nonsense. Thanks for the explanation(s), and sorry for the noise.

But - I was super excited by the Bangladesh study, and spent the better part of a half hour or so trying to see if I could tease out some clear evidence that masks work. I'm insanely pro-mask, and was way ahead of the curve back in february 2020, when a lot of the mainstream was suggesting masks don't work (If that's the case, then why were we trying to reserve them for first responders in hospitals?) - and encouraged everyone I know to wear them. I've spent hours reading the Japanese Meta-Analyses of mask usage in influenza scenarios, looking for differences in Coronavirus and Influenza Virus behavior.

I have never, not once in 18+ months gone indoors where other people are, without wearing an N95 Mask (after I was able to source them ~ June 2020).

I don't think it's an exaggeration that a lot of the last two years of my life have been dedicated to both encouraging (strongly) everyone I know to wear a high-quality, well-tested N95 from a well-known trusted source, as well as scouring the internet for solid research on the effectiveness of masks.

I'm not lazily dismissing that Bangladesh study - I just came away from it not being particularly encouraged that it provided evidence that the masks / methodology presented there would shut down a pandemic by preventing transmission. I'm eagerly and enthusiastically looking for evidence to confirm my bias that masks work.

The Bangladesh study failed to produce a significant result when simpler nonparametric statistics were used to analyze the data. There were also significant discrepancies in the follow up rate between the control and treatment groups. And even if you ignore those issues, the absolute effect size (20 cases) was tiny. The effect size was expressed in terms of the relative rate of symptomatic seropositives, i.e., normalized by population denominators, which makes it appear larger and is sensitive to the follow up rate discrepancies (see above).

Ben Recht at UC Berkeley has several good blog posts on this issue: https://www.argmin.net/2021/12/01/unblinding/