Post #1078: W&M St. Patrick’s Day outbreak, updated

Posted on March 27, 2021


William and Mary updated their COVID-19 dashboard at close-of-business yesterday (Friday, 3/26/2021).  They normally don’t update it over the weekend, so that should be the last count of COVID-19 cases we’ll have until Monday evening.

Yesterday’s update was in line with expectations.  Each new batch of tests results is showing that more-or-less a steady 1.4% of students test positive for COVID-19.  Assuming that holds, by the time all the test results are back, they’ll have had just under 250 known on-campus positives this semester.  Unless the lab they are using closes for the weekend, that’s about how things should stand on Monday.

I didn’t check the W&M dashboard every day (and as far as I can tell, there’s no public file that shows the historical data), so my observations have some days missing.  But here’s what I recorded, along with a little arithmetic on the numbers.

On-campus/off-campus split.  In an 3/26/2021 email, the W&M administration said that about 80% of the positive cases were living off-campus.  Just off the top of my head, I think that means that the on-campus population really didn’t see any change at all in the underlying rate of new positives.  But let me check that.  William and Mary’s website says that 75% of students live on-campus.  If I assume that, in the end, 1.4% of students will test positive in this round, the algebra says this:

(To orient this table to the last one, note the total of 90 new cases on the bottom line of the last table.  That’s the same 90 cases that show up on the top line of this table.)  The upshot is that 4.4% off off-campus residents will have tested positive, but just 0.4% of on-campus residents.

In fact, assuming this represents about a week’s worth of new cases, total, you would have expected to see 11 cases in the on-campus population in any case, based on the current rate of incidence among the Virginia age 20-29 population.  Compared to the projected 18 that W&M will actually see.

So this really is almost entirely an outbreak among students living off-campus.

Finally, there will be a fair number of infectious students still in circulation that were not identified by the testing.  It’ll be interesting to see how quickly W&M can find those and bring the new infection rate back down to (say) the Virginia young-adult average.  They’ve already said they’re going to re-test the entire off-campus population.

Extras for experts:  Why bars and parties are common places for significant COVID-19 spread. 

COVID-19 spreads primarily via droplets and aerosols that people emit as they talk, shout, sing, cough, sneeze, or, to some extent, merely breathe.  Conventionally, droplets are larger (greater than 5 microns), aerosols are smaller (under 5 microns).  Droplets rapidly fall to the ground (hence, 6′ social distancing — that gets you out of droplet range), but aerosols do not — they float.

In a nutshell, if you drink and talk, you emit vastly more droplets and aerosols than if you just talk with your mouth dry.  If you want the details, including reference to the scholarly research that showed that, to go Post #723.

And if you talk loudly, you emit far more droplets than if you talk softly.  I think that’s in the same piece of research, but has now been demonstrated in a number of different pieces of research.  (And, not too relevant, singing produces as much aerosol emissions as coughing does (via this reference), which is why most mainstream churches have stopped singing during services for the time being — see Post #682).

The upshot is that a crowded room full of maskless people, drinking and talking loudly, is very close to a worst-case scenario for COVID-19 spread.

This is why it’s unsurprising to see this same “off-campus social event” outbreak scenario occur on campus after campus.  And that’s why everybody hates bars, in their state COVID-19 re-opening plans (Post #737).  In state after state, bars are the first thing to be closed and the last thing to re-open.  For good reason.

The only thing that could make it worse is the presence of a “superemitter”.  Some people, for reasons unknown, emit an order-of-magnitude more droplets than others. Evidence for that can be found in this this fairly readable article in Nature.  If you, by chance, end up with an infected supermitter in that crowded party — that’s the worst-case scenario.