ABSTRACT
In many phase III clinical trials, patients enter the study sequentially. Response-adaptive allocation designs may be useful in such cases, where the idea is to randomize entering patients to one of two treatments in such a way that a larger number of patients are eventually treated by the more favorable treatment resulting in ethical gain. This can be done by skewing the allocation probability of a patient in favor of the treatment doing better. To achieve this, one needs to incorporate the allocation and response history up to the entry time of a patient, instead of adopting a balanced allocation scheme.
