A simulation study on the use of response-adaptive randomized designs

Response-adaptive designs have been proposed in randomized clinical tri- als to achieve ethical advantages by using sequential accrual information collected during the trial to update probabilities of treatment assignments. We propose the use of a response adaptive design based on urn models in a simulation study on a randomized clinical trial on the efficacy of home enteral nutrition in cancer patients after major gastrointestinal surgery. We compare results with the adaptive design with those previously obtained with the non-adaptive approach.