ABSTRACT
A clinical trial design with paired data often involves missing observations.
In such a case, the data from the trial become a mixture of paired and
unpaired data. A commonly used approach for the analysis of the trial
data is to ignore the incomplete pairs. Such a treatment of missings data is
not statistically efficient. We will discuss a simple method that will allow
us to use all data, including the incomplete pairs. The method is optimal
in the sense that it minimizes the variance. We will show how to design
classical and adaptive trials with the proposed method for different types
of endpoints with superiority, noninferiority, and equivalence designs. The
method can also be used for meta-analysis, in which, some trials are with
paired data and some are not.