[conspire] When to get a covid test, brains, and a weird case, and a cardiology vs. immunology joke

[Rick Moen]

Quoting Ivan Sergio Borgonovo (mail at webthatworks.it):

> On 8/7/20 10:41 AM, Rick Moen wrote:
> 
> >You keep reiterating the 'docs' mailing list mantra, the one I got so
> >tired of you guys blitzing each other with many times each day that I
> >muted the entire mailing list's delivery stream, that 'There are only
> >two possible ways forward, either herd immunity or a vaccine', and
> >keep missing the obvious point -- even though I keep mentioning it --
> >that New Zealand had neither herd immunity nor a vaccine, yet stomped
> >the infection completely out, nationwide.
> >
> >Maybe the reason you 'docs' regulars (who by a _totally_ strange
> >coincidence seem to all be Trumpistas) keep reiterating the mantra
> >endlessly is so you can avoid thinking about the third obvious route
> >forward:  competence.
> 
> 
> Can somebody explain what heard immunity without a vaccine really means?

WebMD's description will serve well enough:

  Herd immunity, or community immunity, is when a large part of the
  population of an area is immune to a specific disease. If enough people
  are resistant to the cause of a disease, such as a virus or bacteria, it
  has nowhere to go.
  [...]

  You can develop resistance naturally. When your body is exposed to a
  virus or bacteria, it makes antibodies to fight off the infection. When
  you recover, your body keeps these antibodies. Your body will defend
  against another infection. 
  [...]

  Vaccines can also build resistance. They make your body think a virus
  or bacteria has infected it. You don’t get sick, but your immune system
  still makes protective antibodies.

Without a vaccine, the percentage of people with resistance to the
pathogen that will be required to achieve herd immunity (in a given
place) is related to the R0 ('R naught') figure, its infectiousness
(in the jargon, its basic replication number).  R naught is the average
number of people who'll contract an infectious disease from a person
with that disease, during the time before that person is no longer
infectious.  E.g., one of the most infectious diseases known, measles,
has an R naught value about 18.  So, if you work the math, you can
figure that if 17 out of the 18 persons a measles victim exposes have
already had measles, then what results at the end is _one_ new person
has measles as the sick person comes to the end of the infection, ergo
the number of people going around with the measles doesn't rise.  That
would mean that 17/18 of the population, or 94%, must have existing
measles resistance.  The resistance can be either from having had
measles already, or from taking the MMR vaccine prior to exposure.
(This is why public health agencies work hard to achieve 95% coverage
with MMR.)

In the context of COVID-19, when people speak of 'achieving herd
immunity', they mostly mean the nasty way, in the absence of a vaccine,
by having had a sufficient number of people walking around with
resistance to the virus because they've already had the disease.  
(This theory rests on the assumption that having had COVID-19 gives
lasting immunity to getting it again, which has come to be seen as
somewhat questionable, but let's ignore that awkward problem for the
moment.)

Calculating how high a percentage of already-had-COVID people would be
required necessitates knowing R naught -- just as we needed it to
calculate the 94% value for measles.  So, what's the number?  Damned
good question.  Like much else, this is unclear and debated.  Some say
it's around 2-3.  Estimates differ widely.

(For clarity, R naught is more an abstract modeling concept than a
useful figure, and the infection potential of a pathogen, the notional
concept being measured, inherently differs from situation to situation,
and thus is not a fixed, inherent figure.)

If we assume R naught is 3, then, to prevent the disease from expanding
exponentially through a population, and instead _not_ have a greater and
greater number of people ill over time, 2 out of 3 people must be caused
to have had prior disease exposure leading to pathogen resistance, or
67%.  Applied to the USA as a whole, this notion holds that all we need
to do is give the disease to 200 million Americans.  Once we've done
that, the disease will not expand exponentially, engulfing  the remaining
100 million.  

If R naught is 2, then infecting 50% of the population, 150 million,
would do the trick.


> Waiting people die and get ill, just slower? or faster?
> 
> Slower is better than nothing since it won't put too much stress on
> health care infrastructure and will probably lower death rates
> etc...

Quite so.  Of course, the saner people who speak of the concept aren't
literally saying it'd be a good thing to infect 150-200 million
Americans, let alone doing so all at once.  The saner ones merely say
that it's the only way short of a vaccine for the virus to diminish and
recede.  Which is sort-of tautologically true, except that the
observation obscures important differences such as what was done in New
Zealand, mentioned elsethread.

> But considering that at this point there is no reliable scientific
> evidence that the virus is "weakening", up to my understanding "herd
> immunity" is just another way to say: let it be.

You will be disputed if you say this, but it's arguably exactly that,
yes.  Or, as Mr. Trump put it in the disastrous Axios interview when the
reporter brought up the awkward fact of 150,000 Americans already killed
by COVID-19 while he stood around and did nothing, 'It is what it is.'

> If you check the death rates of the most hit zones (eg. Bergamo
> where you could have reasonably reliable data and a reasonably good
> health care system), and you project them worldwide you're talking
> about millions of dead people.

Well, we're working on that.  ;->