Hacker News Reader

Comment by gridlockd | Hacker News Reader

> Please understand that airborne has a scientific meaning, and that is the terminology that scientists use. There is no actual clinical evidence of the airborne transmission of COVID.It is spread through large droplets.

At this point, it's expert consensus that airborne transmission of COVID is a significant if not the major route of transmission[1].

https://www.pnas.org/content/117/26/14857

> If it was airborne, those cloth and surgical masks that people wear would be useless, so you might as well not wear them.

Not quite useless, but nowhere near as effective as people would like to believe them to be.

View on HN

Please explain on what basis you are assigning a paper "expert consensus" ? Are you an infectious diseases researcher? If so, your opinion would matter more to me than the paper.

I read dozens of studies every day as part of my job in biotech, and most papers are junk and/or contain overly enthusiastic claims that few can replicate. I know somehow on HN linking to a paper is seen as a mic-drop, but that is not the case in the real world.

I don't read that many studies (I'm envious you can do that ;) as part of my (science) job, but as far as I read on airborne transmission, most of the "evidence" presented is actually modeling, or some IMO questionable studies, like the one - was it Lancet, or NEJM? I don't remember - about potential airborne transmission in a hospital setting. I was left with an "and then?" thought after I finished reading.

> most of the "evidence" presented is actually modeling

If the model evidence-based, falsifiable, and tested, applying modeling to a set of conditions is a way of deriving evidence-based conclusions about the likely outcome of those conditions.

> deriving evidence-based conclusions

No, it's a way to derive evidence based hypotheses. A model that does not include a confounding variable (of which there are many) cannot give any conclusion about that variable - at most a hypotheses.

I've spent some time looking at the models, all of them fail - some spectacularly so (those that inform policy much more so, unfortunately). To the point of being useless, I might say, because their stated assumptions usually apply almost everywhere, but their prediction (and back testing results) match only a small sample (e.g. a few specific countries or cities).

> Please explain on what basis you are assigning a paper "expert consensus" ?

The source is for the "if not majority" part of that sentence. The other part is just me making an admittedly unsubstantiated claim.

https://www.nytimes.com/2020/07/04/health/239-experts-with-o...

> I know somehow on HN linking to a paper is seen as a mic-drop, but that is not the case in the real world.

Sure, but this is HN. You often have to source stuff just to not get downvoted.

FTR, the letter linked in the NY Times article actually advocates the precautionary principle (which implies there isn't enough evidence, only suggestions of such).

Not enough evidence for what? Quote:

"The evidence is admittedly incomplete for all the steps in COVID-19 microdroplet transmission, but it is similarly incomplete for the large droplet and fomite modes of transmission. The airborne transmission mechanism operates in parallel with the large droplet and fomite routes [16] that are now the basis of guidance."

By using arbitrarily high standards for evidence, I might as well claim there's no evidence that COVID is transmitted by large droplets or by smear infection. Where's the randomized trial that proves beyond any doubt that coughing in each other's faces will spread COVID?

I have not bookmarked them, but I found in the past very convincing evidence for significant fecal-oral transmission -- possibly of comparable standard to airborne and droplet transmission.

I'm kind of surprised that there isn't much of a discussion about that, because if it is indeed a significant route (say 30%), all the masks and distancing in the world won't help until this is also addressed.

Reply on news.ycombinator.com