Hacker News new | ask | show | jobs
by legerdemain 2128 days ago
I'm a little surprised by the author's tendency of wondering abstractly and suggesting cataclysmic shoot-from-the-hip solutions.

There still hasn't passed enough time for us to make confident statements about immunity in patients who recover from acute COVID, not to mention make grandiose plans, either as a state or as a society of individuals.

A popular blog post or article circulated earlier, plotting very simple mathematical models of infection numbers based on some assumptions about the durability of immunity. But even without visual aids, mathematical intuition suggests that widespread immunity of short duration would lead to a relatively high constant number of new cases, with relatively frequent small waves. In contrast, widespread immunity of longer duration, on the order of a year or two, would lead to lower background level of infection, punctuated by larger, more severe, less frequent outbreaks, made worse if they are seasonal and coincide with the flu.

And these are assumptions, things we don't know. Which of these scenarios should we be making grand, nationwide plans for?

We saw a definitive report about an asymptomatic case of reinfection. That patient managed to get tested twice, despite being reported as asymptomatic the second time. Both times, the viral genome was sequenced, which is why we now believe this is a case of reinfection.

In the US, this patient would likely not have been tested at all. Should we update our expectations about the incidence of reinfection?

Are there differences in humoral and mucosal immunity to COVID? I don't think we have consensus yet. Do asymptomatic carriers infect others? We have some evidence that they do. Other diseases can be spread by asymptomatic carriers. As an extreme example, Typhoid Mary was a real historical figure who infected hundreds over a lifetime.

In this context, suggesting that we ship our most elderly and medically fragile members to some kind of COVID-free zone is... staggeringly whimsical.

1 comments

The case of the person that got reinfected is grossly exaggerated. Don't you think if this was really occurring frequently there wouldn't be only one or two reported incidences of this world-wide? Asymptomatic carriers do definitively infect others. Just because you personally are not sure of things, doesn't mean that they are not known. And this "do nothing" approach you are advocating for- which is what happens if you wait until everything is fully and definitively proven- has resulted in the disastrous results the US has had so far. Also please explain "suggesting cataclysmic shoot-from-the-hip solutions"- I don't think anything I suggested comes close to that. I said that we shouldn't reopen colleges in places with rising cases. That doesn't seem cataclysmic to me.

  > Don't you think if this was really occurring frequently there wouldn't be only one or two reported incidences of this world-wide?
No, for reasons I have already mentioned in my first comment.

The patient we think was reinfected was reported to have a mild case in the spring. The US, at the time, had extremely limited access to tests for anyone except the grievously ill. It is likely the patient would not have received a test in the US, and certainly not had the infection genome-sequenced.

The second time the patient tested positive, he was asymptomatic. The CDC currently recommends against testing asymptomatic individuals. It is likely that, in the US, the patient would once again not have received a test.

Basically, the US tests sparsely and doesn't seem to test longitudinally on any scale at all. Does any country in the world with more than a few thousand cases test longitudinally?

If anything, I think it's more telling that this patient was identified in Hong Kong, which only has 4.5 thousand cases total. Finding seemingly ironclad evidence in such a small pool suggests that this isn't a rare phenomenon.

  > And this "do nothing" approach you are advocating for- which is what happens if you wait until everything is fully and definitively proven- has resulted in the disastrous results the US has had so far.
If I'm advocating anything at all (and I don't think I'm advocating anything in my original comment), it's that we hold off on making aggressive policy decisions based on our Swiss cheese "understanding" of COVID immunity. How about working toward goals we set months ago and have since kept failing to meet, such as ubiquitous access to testing and meaningful case tracing?

On the other hand, let's maybe hold off on jokes about sending Grandma to New Orleans? The photo of the elderly woman wearing sparkly bangles is the joke, but I think your suggestion right before that, that "we could send the highest risk people to the places that are the most immune so they could be shielded" is made seriously, right? Like, even moving an individual, medically fragile, institutionalized elderly person leads to complicated and often negative outcomes in the best of times, simply because of the complex and fragile support systems their well-being relies on. Shipping "the highest risk people" en masse and organizing care for them seems dramatically more complex than, again, following through on testing and mitigation strategies that we, as a state, have been paying lip service to for months now.

It literally says "All joking aside" following that comment. I don't know how much clearer I could have made that. I also advocate wearing masks, distancing, protecting the vulnerable, and not opening schools. I don't know how you can possibly be construing me as advocating for any of the extreme things you are talking about.
The logical connection between our comments is pretty tenuous. I don't think there is a need to continue the conversation. Good luck promoting your point of view!
I have to ask though, have you accounted for the seemingly deliberate lack of testing, and politically-motivated alterations to CDC guidance (the latter has knock-on effects of changing what insurance companies will reimburse for)?

How big a factor is this in the numbers? If the numbers are unreliable, what does this do to your assumptions?

My data comes from worldometer.info which pulls directly from the states and hospital reporting and not from the CDC or HHS. So I think it is reliable.