Thread by @DiseaseEcology

Aerosol vs droplet transmission for #COVID19
(How to over-interpret data?)

Article by @apoorva_nyc & thread by @BillHanage, who is quoted in article, are strangely very different from each other, and make overall story very confusing.
Thread.

https://twitter.com/BillHanage/status/1279578191745683457

Background: "Aerosol" transmission is differentiated from "droplet" transmission for good reason: droplets usually refers to larger droplets that gravity quickly drags to the ground; this type of transmission is short-range (thus the 6'/2m rule) and quick - droplets don't linger

Aerosols generally refer to smaller droplets that can travel further and persist in the air for longer. The 6' rule wouldn't (fully) protect against aerosols.
See here for additional discussion of the two (but paper conclusion overstates evidence):
nejm.org/doi/full/10.10…

#SARSCoV2 is much more likely to be transmitted at close range than farther away (Fig, paper) which supports droplets as key mode of transmission.
But CAN virus also be transmitted by aerosols (at distances >6')?
doi.org/10.1016/S0140-…

Headline of the NY Times article seems to say YES:
"239 Experts With 1 Big Claim: The Coronavirus Is Airborne"
But is that really it? Does "Airborne" = "Aerosol"? It seems so, but quotes in story paint different story.

One quote says: "There is no incontrovertible proof that SARS-CoV-2 travels or is transmitted significantly by aerosols". @BillHanage agrees with this quote in his thread.

So what is actual evidence NYT article is based on?
(Same Q for letter signed by hundreds of experts)
(Note: No link to letter in NY Times article - why do story on it then???)

Most people are familiar with several spreading events that people & article say provides evidence that transmission can span >6' distances (not just droplets):
WA Choir
cdc.gov/mmwr/volumes/6…
Restaurant w/ air conditioning
doi.org/10.3201/eid260…
medrxiv.org/content/10.110…

Except the actual paper from the choir investigation indicates that v close contact (droplet transmission) could explain all transmission. Singing could have produced aerosol & contributed but lots of close contact. So clearly not definitive.

Restaurant example is also not nearly as simple as first appears. 1 person-A1- is treated as index case but whole family at Table A went to Wuhan &A2 got sick only 2d after A1 (could have been infector). Thus all infected cases are <2m from possible infectors.

Also note substantial variation in susceptibility to infection - 1 person at B table did NOT get infected as well as 5 at Table C and all those at table E (which paper above showed were w/in aerosol plume).

One more example that comes up in twitter threads is S Korea call center. But if you read that paper carefully it provides no information about who interacted with who and thus all infections could have arisen from in-person close contacts.
wwwnc.cdc.gov/eid/article/26…

So where is this large body of evidence showing aerosol transmission? NYT article links to one more example which is just a news story of German meat factory with no actual analysis or data.
So why does NY Times article try to paint such a clear picture of aerosol transmission?

The quotes in it includes argue against the title (and the 239 signature letter) and suggest that evidence is not strong. @BillHanage and others suggest that we should consider it possible but not main mode of transmission (close contact via droplets that mask would help stop).

This matters because key policy/economic decisions hinge on importance of aerosol - indoor dining & movie theaters for example. If aerosol substantial risk then bad idea to do either. But if relatively small risk, then 15M restaurant employees in USA might justify indoor dining.

NJ & NY have decided NOT to re-open indoor dining based on risk posed. But I don't know of evidence besides weak data above showing transmission >6'/2m. Most press stories (and NJ gov tweet) indicate lots of breaking of 6' rule which highlights additional challenge - compliance.

This issue could be solved by public health depts making case investigation data available which would show whether aerosol occurs frequently or so rarely it can be considered very low risk. Data w/ & w/out masks (for movie risk) very important; less so rest. (can't eat w/ mask)

Same data would also help address surface contamination as risk. I know of very little data supporting this as risk; see recent paper but w/ issues:

https://twitter.com/DiseaseEcology/status/1279097818351726592

Addendum: Even if aerosol risk shown to be low, safe indoor dining would obviously require 6'+spacing & 1 household per table. Violation of either of these would lead to transmission. Also, would be good to know if ventilation can make substantial difference.