3 ms·

Testing isn't a delay, unless you think being reckless is a scientific virtue.

by __jem 6y ago

zaroth 6y ago
Phase 1 testing had to be done. However, gambling on efficacy would have been a very reasonable approach and could have gotten tens or hundreds of millions of people vaccinated by now.
- giantrobot 6y ago
  Even with tens of millions vaccinated COVID is too contagious and deadly to avoid social mitigation like restricting businesses, large gathering, mandating masks, and even lockdowns. The long incubation period and asymptomatic carriers makes for easy spread of the virus. It's hospitalization rate means a wide spread will literally shut down hospitals in whole regions. At current infection rates we're seeing hospitals being overwhelmed at regional levels. Even if vaccines had been delivered after a Phase 1 trial of "no one died immediately" in the trial group, we wouldn't have enough vaccinated people to avoid social mitigation. We'd be in a similar situation as we're in right now. It will take something like 75% of the population to be immune (by vaccine or previous infection) to have effective herd immunity. Even if vaccines had started distributing months ago it would still be a long time before we hit that 75%. Distributing vaccines after the phase 1 trials might ultimately get an immune population to that 75% sooner but it wouldn't stop any of the damage happening until that number is reached. Even if no one died from COVID the "long hauler" after effects can be crippling. Some 10-20% of COVID patients end up "long haulers". If the virus were to spread to significant portions of the population because no one practiced social spread mitigation because they assumed the vaccine would fix everything you're still in a terrible situation.
  - lbeltrame 6y ago
    > The long incubation period and asymptomatic carriers makes for easy spread of the virus Long? The median is between 4 and 6 days. The 14 days often quoted is the 99th percentile. Can we stop perpetuating this myth? > asymptomatic carriers makes for easy spread of the virus Actually, most of the spread comes from superspreading events. Most people with the virus don't spread the virus at all, or spread it very little. True asymptomatic people are far less contagious than symptomatic people (up to 10 fold or so less). Severity of the symptoms is also associated with higher chance of spread. > It will take something like 75% of the population to be immune (by vaccine or previous infection) to have effective herd immunity. This is often calculated considering an equally susceptible population (which is not the case) with uniform mixing (same), so it's a very pessimistic estimate. The actual figure might be lower (though not as low as 20% as some "skeptics" say).
    - JamisonM 6y ago
      I am looking for 14 days quoted in the parent comment and not finding it. > Actually, most of the spread comes from superspreading events. I don't know if we can say this for certain at this point. I was very firmly on the "superspreader" train some time ago but I find it hard to believe given the spread we are seeing in places like Canada where most of what we thought were superspreader circumstances are now illegal. I am sympathetic to the idea but being overconfident about this stuff is also not helpful.
      - lbeltrame 6y ago
        > I am looking for 14 days quoted in the parent comment and not finding it. 2 to 6 days incubation time is not "long". > I am sympathetic to the idea but being overconfident It's not being overconfident: it's what the currently available data says. The largest contact tracing study I am aware of, in India, reported that the largest cause of spread were superspreading events.
        JamisonM 6y ago
        Saying that you are not being overconfident with that level of confidence seems like a red flag. If COVID-19 is primarily driven by superspreader events that are largely not happening anymore in the 2nd wave that is causing case increases across nearly every jurisdiction in the world I have doubts.
        lbeltrame 6y ago
        > Saying that you are not being overconfident with that level of confidence seems like a red flag. Were it just me, I'd agree with you, however it is the data that we must look at: https://science.sciencemag.org/content/early/2020/09/29/science.abd7672 https://science.sciencemag.org/content/early/2020/09/29/scie...
        JamisonM 6y ago
        In the spirit of open discussion I would like to raise to you some salient facts. The article in question does not address the fact that I raised, and also you shouldn't make claims the authors are not making: "Our findings, based on comprehensive surveillance and contact-tracing data from the Indian states of Tamil Nadu and Andhra Pradesh, provide insight into the epidemiology of COVID-19 in resource-limited populations." Specifically: "Whereas incidence declined steadily at ages older than 30 to 39 years in the two Indian states, incidence increased at ages of ≥65 years in the United States." So their finding might be applicable if the general trends matched but the study specifically says they don't, in the way that I generally pointed towards. Also the study in question points to "superspreading individuals" but not "superspreading events" and that difference is significant. A "superspreader" who attends an event that can cause superspreading event, but a superspreader that goes to many places and meets just a few people at those places can super-spread! It's different! The study does NOT support your implied claim that asymptomatic spreaders can't be superspreaders.
- JamisonM 6y ago
  Gambling on efficacy would not have been a reasonable approach. I understand the temptation to think that this would be a good way to proceed but developing a complete supply chain for a dud would have been a total disaster for everyone involved and would have taken resources away from viable vaccines. All of the big vaccine producers had full faith in these candidates and they poured money into their production in advance of phase 3 trials and we are still supply constrained. Making the same bet a month or so earlier (the largest gain possible) would have had a small incremental gain but an enormous downside.
  - ketamine__ 6y ago
    A vaccine distributed month earlier would have saved ~40,000 lives?
    - JamisonM 6y ago
      A vaccine scaled up to 1 billion doses 2 or 3 months later would by that logic cost 120k lives. You are assuming that there are no costs associated with one choice vs the other, but we know that is not the case.

Testing isn't a delay, unless you think being reckless is a scientific virtue.

by __jem 6y ago

Testing isn't a delay, unless you think being reckless is a scientific virtue.

Phase 1 testing had to be done. However, gambling on efficacy would have been a very reasonable approach and could have gotten tens or hundreds of millions of people vaccinated by now.

giantrobot 6y ago
Even with tens of millions vaccinated COVID is too contagious and deadly to avoid social mitigation like restricting businesses, large gathering, mandating masks, and even lockdowns. The long incubation period and asymptomatic carriers makes for easy spread of the virus. It's hospitalization rate means a wide spread will literally shut down hospitals in whole regions. At current infection rates we're seeing hospitals being overwhelmed at regional levels. Even if vaccines had been delivered after a Phase 1 trial of "no one died immediately" in the trial group, we wouldn't have enough vaccinated people to avoid social mitigation. We'd be in a similar situation as we're in right now. It will take something like 75% of the population to be immune (by vaccine or previous infection) to have effective herd immunity. Even if vaccines had started distributing months ago it would still be a long time before we hit that 75%. Distributing vaccines after the phase 1 trials might ultimately get an immune population to that 75% sooner but it wouldn't stop any of the damage happening until that number is reached. Even if no one died from COVID the "long hauler" after effects can be crippling. Some 10-20% of COVID patients end up "long haulers". If the virus were to spread to significant portions of the population because no one practiced social spread mitigation because they assumed the vaccine would fix everything you're still in a terrible situation.
- lbeltrame 6y ago
  > The long incubation period and asymptomatic carriers makes for easy spread of the virus Long? The median is between 4 and 6 days. The 14 days often quoted is the 99th percentile. Can we stop perpetuating this myth? > asymptomatic carriers makes for easy spread of the virus Actually, most of the spread comes from superspreading events. Most people with the virus don't spread the virus at all, or spread it very little. True asymptomatic people are far less contagious than symptomatic people (up to 10 fold or so less). Severity of the symptoms is also associated with higher chance of spread. > It will take something like 75% of the population to be immune (by vaccine or previous infection) to have effective herd immunity. This is often calculated considering an equally susceptible population (which is not the case) with uniform mixing (same), so it's a very pessimistic estimate. The actual figure might be lower (though not as low as 20% as some "skeptics" say).
  - JamisonM 6y ago
    I am looking for 14 days quoted in the parent comment and not finding it. > Actually, most of the spread comes from superspreading events. I don't know if we can say this for certain at this point. I was very firmly on the "superspreader" train some time ago but I find it hard to believe given the spread we are seeing in places like Canada where most of what we thought were superspreader circumstances are now illegal. I am sympathetic to the idea but being overconfident about this stuff is also not helpful.
    - lbeltrame 6y ago
      > I am looking for 14 days quoted in the parent comment and not finding it. 2 to 6 days incubation time is not "long". > I am sympathetic to the idea but being overconfident It's not being overconfident: it's what the currently available data says. The largest contact tracing study I am aware of, in India, reported that the largest cause of spread were superspreading events.
      - JamisonM 6y ago
        Saying that you are not being overconfident with that level of confidence seems like a red flag. If COVID-19 is primarily driven by superspreader events that are largely not happening anymore in the 2nd wave that is causing case increases across nearly every jurisdiction in the world I have doubts.
        lbeltrame 6y ago
        > Saying that you are not being overconfident with that level of confidence seems like a red flag. Were it just me, I'd agree with you, however it is the data that we must look at: https://science.sciencemag.org/content/early/2020/09/29/science.abd7672 https://science.sciencemag.org/content/early/2020/09/29/scie...
        JamisonM 6y ago
        In the spirit of open discussion I would like to raise to you some salient facts. The article in question does not address the fact that I raised, and also you shouldn't make claims the authors are not making: "Our findings, based on comprehensive surveillance and contact-tracing data from the Indian states of Tamil Nadu and Andhra Pradesh, provide insight into the epidemiology of COVID-19 in resource-limited populations." Specifically: "Whereas incidence declined steadily at ages older than 30 to 39 years in the two Indian states, incidence increased at ages of ≥65 years in the United States." So their finding might be applicable if the general trends matched but the study specifically says they don't, in the way that I generally pointed towards. Also the study in question points to "superspreading individuals" but not "superspreading events" and that difference is significant. A "superspreader" who attends an event that can cause superspreading event, but a superspreader that goes to many places and meets just a few people at those places can super-spread! It's different! The study does NOT support your implied claim that asymptomatic spreaders can't be superspreaders.
JamisonM 6y ago
Gambling on efficacy would not have been a reasonable approach. I understand the temptation to think that this would be a good way to proceed but developing a complete supply chain for a dud would have been a total disaster for everyone involved and would have taken resources away from viable vaccines. All of the big vaccine producers had full faith in these candidates and they poured money into their production in advance of phase 3 trials and we are still supply constrained. Making the same bet a month or so earlier (the largest gain possible) would have had a small incremental gain but an enormous downside.
- ketamine__ 6y ago
  A vaccine distributed month earlier would have saved ~40,000 lives?
  - JamisonM 6y ago
    A vaccine scaled up to 1 billion doses 2 or 3 months later would by that logic cost 120k lives. You are assuming that there are no costs associated with one choice vs the other, but we know that is not the case.