Blind 2Stage [Two-Stage / GS Designs]

posted by Astea – Russia, 2019-05-30 12:37 (2214 d 03:57 ago) – Posting: # 20305
Views: 9,181

Dear Smart People!

Digging forum posts I've found this 9-year old question from Detlew. It turns out that the question can be now easily answered by the author after constructing Power2Stage.
But can you clarify me the meaning of blindness in BE trials?

I naively thought that blind interim analyses should not influence the type-one error. But
comparing power.tsd.ssr(n1=10,CV=0.1,blind=FALSE,theta0=1.25) and power.tsd.ssr(n1=10,CV=0.1,blind=TRUE,theta0=1.25) it turns out that blinding may even worse the situation (5.01% vs 7.36%, TIE). It is connected with the unknown PE (cause for BLIND=TRUE s20s<-mses), but isn't it contrintuitive?

Can it be true for parallel design also? Suppose we want to make a blind interim analyses after first stage with N subjects and recalculate sample size on fixed GMR=0.95 if total CV would be greater than initial suggestion. Will it cause any inflation? How to estimate it?

"Being in minority, even a minority of one, did not make you mad"

Complete thread:

UA Flag
Activity
 Admin contact
23,424 posts in 4,927 threads, 1,672 registered users;
35 visitors (0 registered, 35 guests [including 9 identified bots]).
Forum time: 16:34 CEST (Europe/Vienna)

Medical researches can be divided into two sorts:
those who think that meta is better and those
who believe that pooling is fooling.    Stephen Senn

The Bioequivalence and Bioavailability Forum is hosted by
BEBAC Ing. Helmut Schütz
HTML5