8 ms·

> And what exactly would that have told you about the safety and effectiveness of the vaccine for the people at risk, those who are neither young nor healthy?

> And what exactly would that have told you about the safety and effectiveness of the vaccine for the people at risk, those who are neither young nor healthy? So if the goal was to reduce transmission, you can save a lot of lives of those at risk by vaccinating the young and healthy. It would be reasonable to basically use challenge studies to get a lower transmission rate, while in parallel using standard trials to prove the efficacy in at risk populations. Getting more data quicker is usually a good way to make more informed decisions, even if the data isn’t exactly what you want it to be to make the perfectly optimal choices.

runako 6y ago
> So if the goal was to reduce transmission, you can save a lot of lives of those at risk by vaccinating the young and healthy. Has this been tested yet? Is there evidence that this is true beyond the broad population surveys like those from Israel? (The accuracy of those data were hotly debated here last week.) Surely a study to determine actual transmission rates would be much more complex than one to determine whether young & healthy people are infected after inoculation (which again: not that useful of a data point by itself)?
- Closi 6y ago
  Well transmission effectiveness is a good example of something that’s much easier to measure in a controlled challenge study and almost impossible to measure pre-rollout without it!
  - runako 6y ago
    How? The vaccinated individuals would come into contact with lots of people who are not vaccinated. How do you track all of them to determine if vaccinated individuals can transmit the disease? That seems like a much larger logistical problem that differs from the existing (and challenge) trials in very material ways.
    - Closi 6y ago
      Well you observe transmission within a controlled environment - ie you keep people in the lab and don’t let them go home. You could take 50 people, vaccinate half and give half a placebo, infect them with Covid, then put them in 50 rooms with 50 other unvaccinated people (so each has two people) for 2 weeks and see what happens. The reason you can’t do it without a challenge trial is exactly because of what you said - you can’t track infections in the community.
      - runako 6y ago
        I wonder how big such a trial would need to be in order to generate any meaningful confidence in the results, especially given the unpredictability of this pathogen. Also note that had this been run last spring you've doubled the size of the trial and therefore the number of hospitalized patients who would not have life-saving therapeutics. By comparison, the BioNTech/Pfizer vaccine had over 40,000 people in its trial. Not sure you could even run a trial that big with even a tenth as many people.
        Closi 6y ago
        You should need less people in a challenge trial - the reason 40,000 was so high is because only a small amount of those people was expected to be infected, e.g. maybe only a few hundred of those actually got infected even in the control group. (Eg in a control group of 20k maybe 1% got infected = 200 people). In a challenge trial you can make people cough in each other’s faces, or literally inject them with Covid depending on the study, which raises your chance of getting people infected and seeing the impact. You would expect the control group to have a very high infection rate, and to see the difference in the study group much more easily. I’m not saying 100 people would be enough, but you would need significantly fewer people fewer to get a good level of confidence.
        sudosysgen 6y ago
        You can't measure immediate response. You'd have to wait 3-6 weeks to see if protection against transmission still works. It would be maybe a 2-3 weeks faster but not yield as useful of data.
- graeme 6y ago
  Every vaccine we have developed in the past had reduced transmission. No exceptions. It is scientific blindness to abandon all of our priors and say “who knows if it reduces transmission?”
  - runako 6y ago
    I'm not suggesting we abandon priors, but Covid does have some novel facts: - apparently high asymptomatic rate - somewhat unpredictable who will get severe case - random neurological symptoms in a respiratory virus And most/all of the vaccines we are using will be the first approved for human use on their respective platforms. It's worth looking at. Also, for a pathogen that is so highly transmissible, it's worth understanding what we mean by "reduced transmission". Is it 10% less? 90%? Surely this would be important to know before following a strategy of vaccinating the young and healthy first. (Especially given the difficulty of doing so in a population that is least likely to suffer serious harm from the pathogen.)
    - graeme 6y ago
      The point is we could have vaccinated the young and healthy summer 2020, had we had challenge trials proving efficacy in that group. How could that not have helped? We vaccinated no one in that period as it is.
      - tinus_hn 6y ago
        We didn’t even bother manufacturing vaccines. Even though they were paid in advance.
        graeme 6y ago
        Because there was no one approved to give them to. Companies could have gone some out if govts had ordered delivery earlier for the age groups in question. It would have sped up subsequent production too.
        6y ago

runako 6y ago

> So if the goal was to reduce transmission, you can save a lot of lives of those at risk by vaccinating the young and healthy. Has this been tested yet? Is there evidence that this is true beyond the broad population surveys like those from Israel? (The accuracy of those data were hotly debated here last week.) Surely a study to determine actual transmission rates would be much more complex than one to determine whether young & healthy people are infected after inoculation (which again: not that useful of a data point by itself)?

Closi 6y ago
Well transmission effectiveness is a good example of something that’s much easier to measure in a controlled challenge study and almost impossible to measure pre-rollout without it!
- runako 6y ago
  How? The vaccinated individuals would come into contact with lots of people who are not vaccinated. How do you track all of them to determine if vaccinated individuals can transmit the disease? That seems like a much larger logistical problem that differs from the existing (and challenge) trials in very material ways.
  - Closi 6y ago
    Well you observe transmission within a controlled environment - ie you keep people in the lab and don’t let them go home. You could take 50 people, vaccinate half and give half a placebo, infect them with Covid, then put them in 50 rooms with 50 other unvaccinated people (so each has two people) for 2 weeks and see what happens. The reason you can’t do it without a challenge trial is exactly because of what you said - you can’t track infections in the community.
    - runako 6y ago
      I wonder how big such a trial would need to be in order to generate any meaningful confidence in the results, especially given the unpredictability of this pathogen. Also note that had this been run last spring you've doubled the size of the trial and therefore the number of hospitalized patients who would not have life-saving therapeutics. By comparison, the BioNTech/Pfizer vaccine had over 40,000 people in its trial. Not sure you could even run a trial that big with even a tenth as many people.
      - Closi 6y ago
        You should need less people in a challenge trial - the reason 40,000 was so high is because only a small amount of those people was expected to be infected, e.g. maybe only a few hundred of those actually got infected even in the control group. (Eg in a control group of 20k maybe 1% got infected = 200 people). In a challenge trial you can make people cough in each other’s faces, or literally inject them with Covid depending on the study, which raises your chance of getting people infected and seeing the impact. You would expect the control group to have a very high infection rate, and to see the difference in the study group much more easily. I’m not saying 100 people would be enough, but you would need significantly fewer people fewer to get a good level of confidence.
        sudosysgen 6y ago
        You can't measure immediate response. You'd have to wait 3-6 weeks to see if protection against transmission still works. It would be maybe a 2-3 weeks faster but not yield as useful of data.
graeme 6y ago
Every vaccine we have developed in the past had reduced transmission. No exceptions. It is scientific blindness to abandon all of our priors and say “who knows if it reduces transmission?”
- runako 6y ago
  I'm not suggesting we abandon priors, but Covid does have some novel facts: - apparently high asymptomatic rate - somewhat unpredictable who will get severe case - random neurological symptoms in a respiratory virus And most/all of the vaccines we are using will be the first approved for human use on their respective platforms. It's worth looking at. Also, for a pathogen that is so highly transmissible, it's worth understanding what we mean by "reduced transmission". Is it 10% less? 90%? Surely this would be important to know before following a strategy of vaccinating the young and healthy first. (Especially given the difficulty of doing so in a population that is least likely to suffer serious harm from the pathogen.)
  - graeme 6y ago
    The point is we could have vaccinated the young and healthy summer 2020, had we had challenge trials proving efficacy in that group. How could that not have helped? We vaccinated no one in that period as it is.
    - tinus_hn 6y ago
      We didn’t even bother manufacturing vaccines. Even though they were paid in advance.
      - graeme 6y ago
        Because there was no one approved to give them to. Companies could have gone some out if govts had ordered delivery earlier for the age groups in question. It would have sped up subsequent production too.
        6y ago