For a those that still don't understand the signif
Post# of 154722
(Total Views: 770)
Posted On: 07/22/2020 9:21:36 AM
For a those that still don't understand the significance of yesterday announcement:
Lets assume we are only studying SAE (which, of course is not the case as we need to quantify all the events with different degrees of severity) Remember:
0= no symptoms
1= mild
2= moderate
3= severe
Per CYDY announcement:
We should count total SAE (no patients) because if a patient has two (or three) SAEs they all count in the calculation of the total mean that wll be used to evaluate efficacy.
So, for 28 patients and 11 SAE compared to 56 patients and 8 SAE we have a binomial p-value of 0.00784
Once again, this is not the way data will be analyzed as we need to account for mild and moderate events as well. THe p-number will change and can go either way depending on mild and moderate frequency in both arms (and when they occurred). We only know that there were a total of 14 in placebo and 19 in Leronlimab arm with at least one event.
However, the significance of the safety data cannot be ignored.
Now, what happens if we analyze only total events (without discriminating severity): 28 patients with 14 and 56 with 19 ?
p-value of 0.07879. This would not be statistically significant.
Please note: this is NOT the way this data will be analyzed. Just playing a little bit with the numbers to form an idea of what to expect.
Lets assume we are only studying SAE (which, of course is not the case as we need to quantify all the events with different degrees of severity) Remember:
0= no symptoms
1= mild
2= moderate
3= severe
Per CYDY announcement:
Quote:
39% (11 SAEs in 28 Patients) in Placebo Arm as Compared to Only 14% (8 SAEs in 56 Patients)
We should count total SAE (no patients) because if a patient has two (or three) SAEs they all count in the calculation of the total mean that wll be used to evaluate efficacy.
So, for 28 patients and 11 SAE compared to 56 patients and 8 SAE we have a binomial p-value of 0.00784
Once again, this is not the way data will be analyzed as we need to account for mild and moderate events as well. THe p-number will change and can go either way depending on mild and moderate frequency in both arms (and when they occurred). We only know that there were a total of 14 in placebo and 19 in Leronlimab arm with at least one event.
However, the significance of the safety data cannot be ignored.
Now, what happens if we analyze only total events (without discriminating severity): 28 patients with 14 and 56 with 19 ?
p-value of 0.07879. This would not be statistically significant.
Please note: this is NOT the way this data will be analyzed. Just playing a little bit with the numbers to form an idea of what to expect.
(3)
(0)