News

PSI Spotlight 2024: Recent development of advanced randomization designs

May 16, 2024, 10:07 AM
The objective of this session is to disseminate the up-to-date research achievements in clinical trial randomization design and implementation. Randomization is the foundation of clinical trials involving treatment comparison. It helps mitigate selection bias, promotes similarity of treatment groups with respect to important known and unknown confounders, and contributes to the validity of statistical tests. Currently, permuted block randomization is the most popular method for treatment assignment in clinical trial practice. Recent research indicated that the permuted block randomization has a lower allocation randomness compared to other newly proposed restricted randomization designs with the same maximum tolerated imbalance. Several advanced randomization designs recently published provide superior statistical properties, including higher allocation randomness, more accurate allocation ratio targeting, better protection of allocation ratio in unequal allocation trials, and stronger capacity for baseline covariate imbalance controlling. However, the dominance of permuted block randomization in clinical trial practice has not been affected yet. Consequences of using sub-optimal randomization designs include the high risk of selection bias, the low trial efficiency caused by inaccurate targeting allocation, and the incapacity of important covariate imbalance control.

Upcoming Events