Looking at a 50% low risk, 50% high risk scenario, we can only save 50% of what we could save if we started in a 50% high risk scenario.
I don’t think this is right.
It’s worth noting that the 14x difference between the risk for the first kid in the house and the second is a (noisy) lower bound on the degree to which the risk depends on dose. For example, this data is consistent with the toy model that the risk vs dose is a step function going from 0% to 100% at a certain point; it would just require that the kid bringing the disease into the house is 14x less likely to have crossed that threshold. With intentional inoculation we don’t know how much lower the risk can be driven, nor do we know how much of the exceptionally bad cases are simply due to exceptionally high initial dose of the virus. It’s entirely possible that even in your “50% low risk 50% high risk” scenario, all or most of the low risk people that die are because of unusually high viral loads given their risk grouping, and that with careful titration of dosage we can do much better than shifting people from one crude grouping to another.
I’m also quite skeptical that we should find the absence of more evidence to be particularly damning. Where would it come from? Ethics boards aren’t going to be happy about intentionally infecting people with deadly diseases, and it’s hard to get more than a very crude guess at the initial dose of cases caught “in the wild”. Furthermore, if you’re going to intentionally inject people with non-potent virus in order to build antibodies, you’d normally want to go all the way and do an actual vaccine. How many people have been thinking about what to do in case there’s a pandemic where you can’t wait for a real vaccine, and how many of them have been studying variolation? I wouldn’t think many.
To me, it sounds like taking volunteers to empty cruise ships sounds like an easy and potentially big win. There are plenty of young people who aren’t concerned and aren’t at high risk to begin with, and you can offer them both a lower risk (because they’re likely to get it anyway), a free party, and a way to feel like they’re helping instead of hurting other people. In return we get data, a step towards herd immunity, and workers which can safely treat other patients or run nursing homes. Once we need to scale up we can start thinking on how to triage people who are wanting to take this risk. For now, it seems like we just want to rush to get this idea accepted and tried somewhere.
I don’t think this is right.
It’s worth noting that the 14x difference between the risk for the first kid in the house and the second is a (noisy) lower bound on the degree to which the risk depends on dose. For example, this data is consistent with the toy model that the risk vs dose is a step function going from 0% to 100% at a certain point; it would just require that the kid bringing the disease into the house is 14x less likely to have crossed that threshold. With intentional inoculation we don’t know how much lower the risk can be driven, nor do we know how much of the exceptionally bad cases are simply due to exceptionally high initial dose of the virus. It’s entirely possible that even in your “50% low risk 50% high risk” scenario, all or most of the low risk people that die are because of unusually high viral loads given their risk grouping, and that with careful titration of dosage we can do much better than shifting people from one crude grouping to another.
I’m also quite skeptical that we should find the absence of more evidence to be particularly damning. Where would it come from? Ethics boards aren’t going to be happy about intentionally infecting people with deadly diseases, and it’s hard to get more than a very crude guess at the initial dose of cases caught “in the wild”. Furthermore, if you’re going to intentionally inject people with non-potent virus in order to build antibodies, you’d normally want to go all the way and do an actual vaccine. How many people have been thinking about what to do in case there’s a pandemic where you can’t wait for a real vaccine, and how many of them have been studying variolation? I wouldn’t think many.
To me, it sounds like taking volunteers to empty cruise ships sounds like an easy and potentially big win. There are plenty of young people who aren’t concerned and aren’t at high risk to begin with, and you can offer them both a lower risk (because they’re likely to get it anyway), a free party, and a way to feel like they’re helping instead of hurting other people. In return we get data, a step towards herd immunity, and workers which can safely treat other patients or run nursing homes. Once we need to scale up we can start thinking on how to triage people who are wanting to take this risk. For now, it seems like we just want to rush to get this idea accepted and tried somewhere.