Thanks for your addition to this topic. Obviously,
Post# of 154003
(Total Views: 843)
Posted On: 08/06/2020 6:17:56 PM
Thanks for your addition to this topic. Obviously, I don’t know the numbers for the placebo deaths. But you are right that the effect of sample size is greater at the higher placebo death percentages.
According to a fairly accessible document from MoreSteam “The power of a statistical test gives the likelihood of rejecting the null hypothesis when the null hypothesis is false.” In other words, one of the things we do not want a statistical test to say is that there is no difference between two conditions (in this case placebo or Leronlimab) when in fact there is a difference. This is called a Type 2 error. It is dangerous in that it would cause everyone to give up on Leronlimab when they should not. Power can mainly be increased by one of more of the following: “1. Increase the effect size (the difference between the null and alternative values) to be detected 2. Increase the sample size(s) 3. Decrease the variability in the sample(s) 4. Increase the significance level (alpha) of the test.” Generally, we cannot increase the alpha. As you know, the p’s in your table need to come out less than the alpha for statistical significance. I assume the alpha is set at 0.05 in this trial. #1 is what your observation about the % death rate is about. I don’t know if we know anything about that for certain at this point. Ultimately, it is not something we can change. It is what the trial is about. We can do nothing about #3, the variability. That leaves us with only increasing the sample size (#2). It appears obvious that someone (the FDA or CYDY) plugs the numbers into a power calculator to determine what the sample size needs to be to get sufficient power to lower the possibility of a Type 2 error to a reasonable level. From what we are being told that number is 195. As the paper states, “calculating the power of a test beforehand will help you ensure that the sample size is large enough for the purpose of the test. Otherwise, the test may be inconclusive, leading to wasted resources” and I may add a rejection of the use of Leronlimab when it should not be rejected. Normally, decisions about all this is made before the trial is done. That is the way it should be. I understand that some of the rules have been violated in the push to use another of the therapies. That is why many on this board and at CYDY believe that therapy is weak. We do not want the same statistical voodoo to be used to bring about the use of CYDY. For one thing, a small company like CYDY does not have the clout to get the FDA to change its rules like some of the BP are able to do. Secondly, I don’t want CYDY to waste its best chance to be recognized as a viable treatment.
I sincerely understand that everyday between now and beginning of widespread use of a better therapy means additional needless deaths, this trial needs to be done correctly.
According to a fairly accessible document from MoreSteam “The power of a statistical test gives the likelihood of rejecting the null hypothesis when the null hypothesis is false.” In other words, one of the things we do not want a statistical test to say is that there is no difference between two conditions (in this case placebo or Leronlimab) when in fact there is a difference. This is called a Type 2 error. It is dangerous in that it would cause everyone to give up on Leronlimab when they should not. Power can mainly be increased by one of more of the following: “1. Increase the effect size (the difference between the null and alternative values) to be detected 2. Increase the sample size(s) 3. Decrease the variability in the sample(s) 4. Increase the significance level (alpha) of the test.” Generally, we cannot increase the alpha. As you know, the p’s in your table need to come out less than the alpha for statistical significance. I assume the alpha is set at 0.05 in this trial. #1 is what your observation about the % death rate is about. I don’t know if we know anything about that for certain at this point. Ultimately, it is not something we can change. It is what the trial is about. We can do nothing about #3, the variability. That leaves us with only increasing the sample size (#2). It appears obvious that someone (the FDA or CYDY) plugs the numbers into a power calculator to determine what the sample size needs to be to get sufficient power to lower the possibility of a Type 2 error to a reasonable level. From what we are being told that number is 195. As the paper states, “calculating the power of a test beforehand will help you ensure that the sample size is large enough for the purpose of the test. Otherwise, the test may be inconclusive, leading to wasted resources” and I may add a rejection of the use of Leronlimab when it should not be rejected. Normally, decisions about all this is made before the trial is done. That is the way it should be. I understand that some of the rules have been violated in the push to use another of the therapies. That is why many on this board and at CYDY believe that therapy is weak. We do not want the same statistical voodoo to be used to bring about the use of CYDY. For one thing, a small company like CYDY does not have the clout to get the FDA to change its rules like some of the BP are able to do. Secondly, I don’t want CYDY to waste its best chance to be recognized as a viable treatment.
I sincerely understand that everyday between now and beginning of widespread use of a better therapy means additional needless deaths, this trial needs to be done correctly.
(5)
(0)