The wording of question 6C is a bit confusing. Is this question meant to ask how the effect size measurements in B would be changed if crossover had not been allowed?
Thanks,
Teresa
I interpreted it different as:
I think what you are asking is what is NNT if the mortality rate in control was higher??
can you clarify?
Yes, Question 6C is asking how the effect size measurements in B would be changed if crossover had not been allowed.
You can assume that the control patients who crossed over to ECMO would have had a higher mortality if they had stayed in the control group.
Tom
Hello,
I am confused. Isn't the question in 6B asking us to calculate the effect size measurements BASED on an intention-to-treat analysis (i.e. an analysis where crossover is not allowed)? And in C, isn't it asking how these measurements in B would change if an as-treated analysis was done instead (since it said COMPARED to a trial where crossover is not possible i.e. compared to a trial with intention-to-treat analysis, how would the effect measurements change?)
Or do I have it the other way around? It should be that the effect size measurements in B are calculated on an as-treated basis (i.e. crossover is allowed) and C is asking how these measurements would change if crossover had not been allowed?
Thanks
Hi, Bushra-
You seem to be confused about the meaning of "intention to treat" analysis.
In an intention to treat analysis, subjects are analyzed according to the group to which they were randomly assigned, regardless of what happens (e.g., crossover) post randomization.
TOm
Okay, thank you! This is was helpful.