Revised my projections for possibilities: 1) p va
Post# of 153800
(Total Views: 820)
Posted On: 10/06/2020 8:04:48 PM
Revised my projections for possibilities:
1) p value was off due to Excel analysis tool pack's issue with samples not actually being independent.
2) Added corrections and better details to figures. (Thanks to those who responded)
3) Added more cases to reach 30% in the LL arm (based on Dr. Yo)
Overall I think 30% would be quite hard to achieve, though, I don't have enough experience to dispute if makes sense or not. We would be decreasing deaths by 80% at that point.
I do think the mortality rate in the PL wing will be higher than we expect (25%) but it might be balanced by the improving SOC over the course of the trial. In the end we will know when we know. This table is just to help visualize the possibilities. If I have time I will try to use a binomial distribution as has been suggested for calculating p and add it to the table.
Also, after re-reading this I want to add that I initially used morbidity vs deaths and after thinking about this it makes sense to use deaths as it makes it all the more real what we are dealing with and the seriousness of the situation. While I do hope for LL to make a difference it does not keep me from hoping for *any* other therapeutic or vaccine to cross the finish line and save lives.
1) p value was off due to Excel analysis tool pack's issue with samples not actually being independent.
2) Added corrections and better details to figures. (Thanks to those who responded)
3) Added more cases to reach 30% in the LL arm (based on Dr. Yo)
Overall I think 30% would be quite hard to achieve, though, I don't have enough experience to dispute if makes sense or not. We would be decreasing deaths by 80% at that point.
I do think the mortality rate in the PL wing will be higher than we expect (25%) but it might be balanced by the improving SOC over the course of the trial. In the end we will know when we know. This table is just to help visualize the possibilities. If I have time I will try to use a binomial distribution as has been suggested for calculating p and add it to the table.
Also, after re-reading this I want to add that I initially used morbidity vs deaths and after thinking about this it makes sense to use deaths as it makes it all the more real what we are dealing with and the seriousness of the situation. While I do hope for LL to make a difference it does not keep me from hoping for *any* other therapeutic or vaccine to cross the finish line and save lives.
(11)
(0)