Determining the appropriate sample size is perhaps the most important design decision for a randomized trial. It is therefore wise to consult a statistician for advice at an early stage, not only to check that software has been used correctly, but also to advise on whether there might be helpful modifications that can be made to the basic design including, for example, unequal randomization. A good statistician - particularly one with experience in the disease area - should also, with the best of intentions, challenge the assumptions that have been made to ensure that the study design is robust to variations in these assumptions.

0 0

Post a comment