2021-12-11/12/13 General


This report says that rapid tests are really inaccurate (they compared to PCR). This Twitter thread summarizes, but the brand I just picked up in the US misses half of symptomatic cases and a quarter of asymptomatic cases. (They say you should test multiple times, which if the failures are randomly distributed, should drop the error rate down by a lot.)

The report says that there are very few false positives, so if you see a positive test, take notice, it’s probably correct.

This article says that there are growing calls for more rapid tests to be deployed across Canada, but with accuracy like this report says, I think it’s reasonable to be concerned that might do more harm than good. I worry that people will get a negative result and feel that gives them license to do things that are unsafe for the people around them.


This article says that a woman in Taipei who was bitten by COVID-infected mice developed COVID-19 herself. (Her genome matched the strain that the mice had.) I am guessing that is possible that she picked it up some other way, than from the bite-to-blood way (maybe she sucked the finger that got bitten, introducing it into her GI tract?).

This study (from 9 November 2021) found that:

  • singing produced 77% more aerosol than talking,
  • adults produced 62% more aerosol than minors, and
  • males produced 34% more aerosol than females.

The study says that when they adjusted for volume and amount of CO2 expelled (which I am guessing is a proxy for volume of air exhaled), the age and sex differences were no longer significant.

This Twitter thread talks about the three possibilities of how Omicron and Delta will interact: Omicron replaces Delta, Delta replaces Omicron, or they co-circulate. Based on Omicron’s R0 and level of evasiveness to immunity to Delta, the possibilities look like this:

I am not an epidemiologist, but my hunch is that we’re going to see R0 of 3-4 and cross-immunity of like 20-30%, leading to co-circulation.

Mitigation Measures

This study says that turn-taking is disrupted more over Zoom than one would expect. They had subjects respond verbally to cues given either in-person or over Zoom, and the delays over Zoom were much longer than the estimated audio delay.


This study found that vaccination helped Long COVID.