COVID update, December 2021
When science beckons to you follow her, though her ways are hard and steep.
I have several exciting blog-related plans for the new year. For this remainder of this month, probably just this post today and one post next week (possibly about railroads).
Today’s topic: COVID - because it is still here.
But first, a Reddit discussion of the Internet’s Grandchildren. “AITA for yelling at my mom that I hate Harry Potter and to LET ME LIVE MY OWN LIFE”. As with all content on Reddit, you should assume it is fiction.
Some recently published statistical data, as presented by a CNN reporter:
This, unfortunately, is not age-adjusted, and not suitable to detailed re-analysis. Fortunately, the main confounding works with the tooth1 here; older people are more likely to be boosted, more likely to die, and yet there is still a substantial benefit to the boosted pool.
If you are vaccinated, you are much more likely to be “asymptomatic”, and far less likely to die. I estimate that by the end of March, the half the population that is vaccinated will only have 1% of the deaths.
There is a tension between “vaccines where the benefit comes from eliminating outbreaks at a societal level”, and “vaccines where the benefit comes from eliminating risk to you, personally”. As I discussed earlier, the topic is too complex to discuss here in detail.
What is clear is that the benefit of a COVID vaccine today is almost entirely “eliminating risk to you, personally”. As the saying goes, “follow the science”. The science says that we can’t stop an outbreak without draconian measures, but with vaccines and Paxlovid that level of fatalities is acceptable enough2 to not justify draconian measures.
Nate Silver comments on “emergency” measures:
This is fairly clearly true. Many of the “emergency” measures used over the past two years are completely reasonable as emergency measures - yet cannot become permanent.
For example, cancelling student loan debt (through an indefinite series of short-term zero-interest payment holidays - and yes, Biden kicked the can down the road again) and cancelling eviction (through an indefinite series of short-term eviction moratoriums) are the types of policies that must not3 be implemented by claiming a permanent emergency.
Another take on the “emergency”:
This is a less-ideological vote against there being a COVID emergency. Democratic governor Jared Polis has made this case.
Most of the remaining arguments for school closures are pure hysteria. Hysterical parents expressing concern that their kids don’t have booster shots. Hysterical “immuno-compromised” people expressing concern that they are more likely to get sick. Hysterical teenagers expressing concern that their grandpa is 87 and doesn’t want a shot and might die.
Listening to hysterical people expressing concern is one of the best ways to reliably come up with policies that will certainly fail4. Of course, that does not mean that doing the opposite of what they say will work.
Good news on the treatment front:
As a rough estimate, it is estimated to decrease deaths by 90%. The pill is both expensive and somewhat supply-limited currently.
It is also certainly more effective than ivermectin. I have shouted in vain about the dramatic errors in the scientific process in assessing the various claims about ivermectin - and still think it probably decreases deaths by 20%. There is no good reason to think that pairing IVM with Paxlovid is a particularly useful idea. From a pharmacological perspective, it has graduated to the realm of being considered a folk remedy5 rather than a critical medical treatment.
And finally, as an exercise to the reader: when forced to make a choice, should hospitals prefer to give the Paxlovid pill to a vaccinated person or an unvaccinated person?6
When does a pile become a heap? When does a grain of sand become a mountain? When we deal with certain types of inference about statistical data, we sometimes have two systems that interfere with each other. We sometimes have two systems where the effects of one swamp the other.
Sometimes, the transition state between the two types of systems is clear - it is clear you are in a transitory state. Other times, the transition state is sharp.
I envision this as the difference between traveling in the two directions on a sawtooth surface. In one direction, you slide up slowly and occasionally drop. In the other direction, you drop slowly and occasionally run into formidable walls you cannot climb.
Nobody should ever accuse “the science” of being politically correct. Please spare me from listening to the “no level of fatalities is acceptable — even one death is an unbearable tragedy” chorus.
The words CAN, SHOULD, MUST, CAN NOT, SHOULD NOT, and MUST NOT are not intended to be formally defined in the ways they often are defined during an RFC process.
Speculating as to how badly they will fail is an exercise of staring into the abyss. Will it cause civilization to collapse? Will it make the planet Earth completely uninhabitable? Will it simply make billions of people miserable for no substantial benefit? It is best not to reach the point of needing to debate those details.
Remember, lots of “folk remedies” actually work! Not as well as technologically fancier treatments, but very often better than “doing nothing” or doing harmful things.
This is both a riddle and a scissor statement. The riddle is noticing that a more useful piece of data would be which patient is older.