Exact Properties of Restricted Randomization Procedures
Date
2015
Authors
Shao, Hui
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Restricted randomization is employed in clinical trials to achieve balanced or near balanced treatment assignments, but is associated with a loss of randomness. For example, the permuted block design, probably the most commonly used restricted randomization procedure, randomizes patients block by block and achieves perfect balance within each block. However, there is at least one predictable treatment assignment within each block. To avoid deterministic treatment allocation in the permuted block design, many clinical trialists prefer randomizing the block sizes as well. While such a procedure is rarely formalized, it is generally assumed that the design will be less predictable.
Description
Keywords
Statistics, Clinical trials, Imbalance, Predictability, Random block design, Restricted randomization, Selection bias