4 ms·

The point is if the two groups behave differently wrt risk then the trial is probably ruined. Maybe you're assuming the vaccine will be 100% effective; experts

The point is if the two groups behave differently wrt risk then the trial is probably ruined. Maybe you're assuming the vaccine will be 100% effective; experts I've heard are saying to expect something like the flu shot.

bluGill 6y ago
Experts need to control to control expections. The flu shot like 50% effenceny is a possibility, and we know anti vaxxers latch onto that with confused statistics. The flu shot is poorly effective because the flu mutates though, covid is more stable so odds are that the vaccine will be more effective, but we are still playing odds of something we don't know, so it is best to assume the worse.
- DiogenesKynikos 6y ago
  The flu vaccine is more effective at preventing serious disease than it is at preventing infection. Even a vaccine that only prevents you from getting seriously ill can be incredibly useful. The inactivated polio vaccine, for example, does not prevent infection. It just prevents disease. That means that it cannot confer herd immunity (vaccinated people can act as asymptomatic carriers, leaving unvaccinated people at risk), but for most people who are vaccinated, it eliminates the most serious consequence of polio infection - paralysis. The SARS-CoV-2 vaccines might end up being the same, conferring protection from disease but not from infection, and even that would be important.
cwhiz 6y ago
Well in this scenario I am suggesting that a placebo is a waste of time and we don't need two groups. I am not assuming the vaccine will be 100% effective, or even effective at all. But in order to get meaningful data you can't have all your participants taking the most extreme precautions. The vast majority of American citizens have not, and will not, get Covid.
- DiogenesKynikos 6y ago
  A few percent of Americans have been infected with SARS-CoV-2. In a study of 30,000 people, enough people will get infected within a few months to give good statistics. At the current rate of confirmed cases in the US, you'd expect about 10 people in the study to get infected each day (fewer if half of them are vaccinated and the vaccine works). Without a placebo group, how will you measure effectiveness and safety? You need to monitor a statistically equivalent group of people to determine the difference in rates of infection and disease, and to look for side-effects.

bluGill 6y ago

Experts need to control to control expections. The flu shot like 50% effenceny is a possibility, and we know anti vaxxers latch onto that with confused statistics. The flu shot is poorly effective because the flu mutates though, covid is more stable so odds are that the vaccine will be more effective, but we are still playing odds of something we don't know, so it is best to assume the worse.

DiogenesKynikos 6y ago
The flu vaccine is more effective at preventing serious disease than it is at preventing infection. Even a vaccine that only prevents you from getting seriously ill can be incredibly useful. The inactivated polio vaccine, for example, does not prevent infection. It just prevents disease. That means that it cannot confer herd immunity (vaccinated people can act as asymptomatic carriers, leaving unvaccinated people at risk), but for most people who are vaccinated, it eliminates the most serious consequence of polio infection - paralysis. The SARS-CoV-2 vaccines might end up being the same, conferring protection from disease but not from infection, and even that would be important.

cwhiz 6y ago

Well in this scenario I am suggesting that a placebo is a waste of time and we don't need two groups. I am not assuming the vaccine will be 100% effective, or even effective at all. But in order to get meaningful data you can't have all your participants taking the most extreme precautions. The vast majority of American citizens have not, and will not, get Covid.

DiogenesKynikos 6y ago
A few percent of Americans have been infected with SARS-CoV-2. In a study of 30,000 people, enough people will get infected within a few months to give good statistics. At the current rate of confirmed cases in the US, you'd expect about 10 people in the study to get infected each day (fewer if half of them are vaccinated and the vaccine works). Without a placebo group, how will you measure effectiveness and safety? You need to monitor a statistically equivalent group of people to determine the difference in rates of infection and disease, and to look for side-effects.