Abstract:
|
Minimization randomization method has been used widely for decades in clinical trials when there are a larger number of known prognostic factors that need to be balanced among treatment groups. It has the advantage of minimizing the imbalance with respect to predefined prognostic factors, as well as for the number of patients in each group, and thus to improve the efficiency and accuracy of the comparisons. However, concerns about the use of minimization have been raised since the method was introduced. Is a standard test for statistical inference that assumes random treatment assignment still appropriate for data analysis since this assumption may not hold? Are the concerns merely type I error inflation, loss of power, and stochasticity reduction---or all of them plus others? Under what circumstances should it be used? How should it be used appropriately in a so-called less well-experienced adaptive design trial? The discussion leader will stimulate lively discussion on these issues.
|
ASA Meetings Department
732 North Washington Street, Alexandria, VA 22314
(703) 684-1221 • meetings@amstat.org
Copyright © American Statistical Association.