Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
Covid-19 outbreak associated with air conditioning in restaurant in Guangzhou (cdc.gov)
52 points by rsecora on April 19, 2020 | hide | past | favorite | 27 comments


Title is misleading. This is an article in a journal published by the CDC, not a statement by the CDC.

The evidence that the air conditioning was responsible is circumstantial at best, and I'm somewhat skeptical. If the index patient had been located at Table C then I would find the claim more credible, but they're claiming the air conditioner both pushed aerosolized droplets away and pulled them toward it. According to the figure the other air conditioner circulates air over a large diffuse area covering 9-10 tables, but the air conditioner in question magically only covers 3 tables and E and F get no air circulation or something. It seems like the air circulation pattern is drawn to fit the hypothesis rather than based on any principled analysis.

They entirely ignore other possibilities like "someone from Table C walked past Table B on their way to the bathroom/stairs/whatever, and the index patient at Table B coughed around that time."


Typically, in this context, 'aerosolized' means that the virus is persistent in the air. Droplet is different, it doesn't persist.

Also, much of the point of the article is to bring attention to the circumstantial evidence.


If you read the paper it turns out there is no actual evidence for the Air Conditioning route other than "the airflow direction"

As a matter of fact "air conditioning" is not even involved, the same airflow could be just as well caused by a fan or the waiters moving about carrying the air around them.

It is ridiculously sloppy.


Yeah. The 6' safety distance applies to non-moving situations. If you have either moving people (you need more than 6' when following someone moving at a decent pace) or airflow (wind or air conditioners) it changes. This should be obvious, the only reason for research would be to figure out how much it increases the needed distance.



Spending one hour in a small restaurant room without windows sitting next to a family who just escaped from Wuhan on January 24th is the reason for the contagion, not any AC issue. Bad title. Astute but useless study by the CDC.


I agree with your comment, but how does "astute and useless" go together?


Astute + useless = pedantic.


Most of the people that spent time in the room didn't get infected.


How about a bit better title? - "COVID-19 Outbreak Associated with airlines and public transportation to Restaurant(s)"


Not surprising as many air conditioning units don't filter the air at the level required to remove viral agents.

Also been reports about them and virus's in the past.

But more recently https://www.telegraph.co.uk/news/2020/03/06/air-conditioning...

But like many things, even common sense needs the data to back it up so people take note.

Coz we have had experts in the field of air-con say they don't, and also experts say they do:

https://www.straitstimes.com/singapore/health/coronavirus-tu...


Office buildings I've worked at in the past barely had any filtration. Within a few days, the untouched parts of your desk would have a layer of dust on it. I'm now (well, not now-now since everyone is WFH) in a LEED Gold building and the filtration is much better, and I only have to clean the monitors every 5 or 6 weeks.

As good as it is, it still won't catch virus sized particles - that's a biosafety level 3 or 4 facility (like in Andromeda Strain) and you're not going to find that in an commercial building.


Coronavirus requires droplets to spread. Those droplets can be very small, and carry a ways through the air (which is what this article is proposing, not that the virus was carried through the ducting), but they are orders of magnitude bigger than the virus itself, and would absolutely be caught by a typical filter.

There are viruses that can spread that way. Measles, for instance, is known for doing that. Coronavirus can't (as far as anyone knows, and we know quite a lot about that).


>that's a biosafety level 3 or 4 facility As dramatic as that sounds, you can get commercial and home aircons that will filter at the level to kill or remove viruses.

Though for COVID-19 you would want something that can filter at 0.125 micron as that is the size of this virus.


They're not saying the virus went through the AC, but that the stream of air leaving the AC likely blew the virus from one downstream table to another.


Yes, though the point being that most will not filter out the air that does pass thru them enough to remove any virus level material in the airstream. Hence if they went in in/out or just got blown about by the aircon, it would make no difference to the virus due to its size.


Did you read the abstract? It explains a theory of how the movement of droplets throughout the restaurant is consistent with the airflow from the air conditioner. Nothing to do with filters.


Yes, did you read my comment about having read many papers about this in the past?


Gotta wonder at who the specialists are who deal with this sort of thing and where they come from. They'd need a complicated background of medicine crossed with physics or engineering. It seems like an obviously useful thing to specialise in so they must have a few in-house experts but it seems unlikely that a medical background would prepare people for modeling the motion of small water droplets in air. That sort of thing isn't necessarily hard but it also isn't trivial.


This "airstream" study is really interesting as it could change much of the response and protection put in place in the next few months.

Another thing interesting to study would be how it spread on the cruise ships.


Wouldn't it be more accurate if the headline said "A Covid-19 outbreak..."

As it's written, it sort of sounds like it's implying the global pandemic was the fault of a restaurant air conditioner.


My personal hypothesis is nearly all (ie. 99%) of infections are via aerosolised droplets, and none or nearly no infections via surfaces or lack of hand washing.

Looking at XRays of lung tissue showing clear infection hotspots suggesting the infection has been breathed in, rather than come via the bloodstream (which would typically cause the whole lung to get infected at roughly the same time).

The fact that disease transmission has been dramatically slowed in places implementing facemask requirements backs this up.

If only we could collect more reliable data confirming or denying the above hypothesis, and we managed to confirm it, then simply saying "you must wear a mask at all times outside your house" would probably allow lifting all other disease spread measures.


> Looking at XRays of lung tissue showing clear infection hotspots suggesting the infection has been breathed in, rather than come via the bloodstream (which would typically cause the whole lung to get infected at roughly the same time).

Poor argument, since most infected people are apparently asymptomatic, so what you see on X-ray are cherry-picked worst cases, not representative of the infection as a whole.

This being said, your hunch may be right, but so far there are still many things we don't understand clearly about how the infection occurs.


Help me understand this a bit.

You're saying that if someone takes in the virus via a droplet on their finger it leads to a less severe infection than if they breathe it in.

You then say that people should wear masks to help prevent them breathing infected droplets.

But if I have a less severe infection, and I'm wearing a mask, I am shedding the virus into my breath and when I breath out droplets are held in my mask - held in front of my nose and mouth. And then when I breath in I'm breathing these droplets.

> The fact that disease transmission has been dramatically slowed in places implementing facemask requirements backs this up.

That feels a bit premature.

https://apnews.com/9140ddd7283d534d8464778d9c4bd92a


Not sure whether masks alone are sufficient to prevent spread of the virus, but they certainly do seem to help. At this point, all the other measures seem like an incredible amount of effort just to avoid mandating wearing of masks.


Do Chinese not kiss and hug as a greeting for close family? It seems far more likely that it happened that way than air-borne droplets dispersed via AC.


> On January 23, 2020, family A traveled from Wuhan and arrived in Guangzhou. On January 24, the index case-patient (patient A1) ate lunch with 3 other family members (A2–A4) at restaurant X. Two other families, B and C, sat at neighboring tables at the same restaurant.

Three families, not one.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: