Wednesday, November 11, 2020


An interesting study came out identifying which areas are the highest risk for getting CoVid, and at first glance the numbers are remarkable.  I will not link to the news piece, for reasons that will become clear at the end. I will keep its internal links.

Restaurants, gyms, hotels and houses of worship are among the 10 percent of locations that would appear to account for 80 percent of the infections, according to research published in the journal Nature on Tuesday. (Eighty percent.  Unexpectedly high. Ed. note)

“These are places that are smaller, more crowded, and people dwell there longer,” said study co-author and Stanford University Professor Jure Leskovec at a media briefing on the research.

"Reducing the establishments’ capacity to 20 percent, as opposed to shutting them down entirely, could curb transmissions by 80 percent, the prof said.

“Our work highlights that it doesn’t have to be all or nothing,” Leskovec said...

Yes, that makes sense, and it's useful information. However...

Still, the study did not track people to locations with potentially large populations such as schools, prisons and nursing homes

Oh. Well. So the whole exercise, though still of some value, just fell precipitously in usefulness.  If you step back and think of the data, though, you can see how it could become a killer headline - and it did.


PenGun said...

Your pandemic is just running wild. You are not doing anywhere near enough to bring it under control. You lost 1330 dead today. This is with the numbers not going through the CDC, so its probably far more.

Unknown said...

I'll note that prisons and care homes would necessarily not be represented in phone GPS mobility data, at least for the residents housed in them. And I'm suspecting that schools were closed or 'remote learning' for much of the time-period covered in the data.

There might be answers that we want on certain types of location that are different from what their POI data were able to tell us, but their editorial point seems to be that cities can get very close to the effect resulting from general lockdown by using more targeted non-lockdowns that don't completely close anything, and leave most businesses (those with less customer duration on-site and less crowding) alone.

A really interesting aspect of the paper to me is that the pre-publication reviewer comments are published:

I couldn't resist typing in your first italicized quotation from the article into Google, which lead me to the article on Marketwatch, copying from the NY Post, who quote from CNN, who quote from a press-briefing on the Nature article. They all purport to quote from the Nature article, but I'd bet those quotes don't include anything that wasn't already a pull-quote laid out for them in their press-pack. At least CNN includes comment from the "Science Media Centre" in the UK, a group who I've found to be mostly pretty good in finding several people independent from the study authors but deeply knowledgeable about the field to comment.

engineerlite said...

What’s missing, for me, is some test data about why these areas are bad. Is it because these are areas where people don’t wear masks? Or, is it because they have surfaces that multiple people touch without being sanitized between touches? Or, is it because the virus spreads as an aerosol, and these are places where people dwell for an extended time in poorly ventilated spaces, so the virus concentration builds up over time? Wouldn’t it be funny if the primary method of transmission were aerosol, and masks and sanitization and social distancing had no effect at all?

It would be interesting to get data from air travel. That’s another small, closed space, which should fit their criteria, and good data should be available. Yet, I’ve not heard of a single case of covid transmission traced to airplanes.

I’m concerned that one could publish this and presume to offer economic solutions. I seriously doubt that the scientists or the editors have any expertise in business or economics, so to propose opening at 20% capacity, with no idea if that is economically viable for the business is not useful. Instead, if the scientists could identify a cause to address, a businessman to propose a workable solution.

GraniteDad said...

engineerlite, I have a pilot friend who advised that airplanes actually circulate the air every 3 minutes or so. Comes in the front, is bled out the back to maintain pressure. But it doesn't recirculate like I'd always imagined, so it's very clean.

David Foster said...

"It would be interesting to get data from air travel. That’s another small, closed space, which should fit their criteria, and good data should be available. Yet, I’ve not heard of a single case of covid transmission traced to airplanes."

Here's a Defense Department study on Covid-19 transmission in transport airplane:

I see they have added a disclaimer at the beginning, possibly fearing that they will be accused of heretical thinking.

engineerlite said...
This comment has been removed by the author.
engineerlite said...

Thanks, David. There's a lot of good data in that study. And some good references which I have not had time to explore.

From their data, it looks to me like aerosol transmission in aircraft is minimal or non-existent, and masks/no-masks and distance make no difference. However, they also point out that airplanes are a best case for air exchange, and homes and hospitals can be much worse (and I can only assume that restaurants, hotel rooms are also worse).

They were careful to point to out that they did not evaluate other forms of transmission. However, their anecdotal data from a few populated airflights of no transmission in the presence of known, contaminated passengers, would infer that transmission from big, heavy droplets and surface contamination was minimal.

james said...

I considered us fortunate to have booked a motel room with both a front and rear door. When we arrived we opened both, and sat on the porch and let the stiff breeze off the lake air out the room for about half an hour. Door County was rather a hot-bed of Covid at the time.

engineerlite said...

James, interesting to think about how one can address contamination, if the real problem is aerosol transmission. Even if your room were immaculately sanitized, and no one was in it when you arrived, it could be covid dangerous if the person who cleaned it an hour ago was contagious. Maybe we don’t need social distancing in space, but in time.

james said...

Some of both. And really good ventilation for dilution. We spent almost all of that mini-vacation outdoors, with snacks and camp stove and hiking.