Online Program Home
My Program

Abstract Details

Activity Number: 436 - SPEED: Tests, Trials, Biomarkers, and Other Topics in Biometrics
Type: Contributed
Date/Time: Tuesday, July 31, 2018 : 3:05 PM to 3:50 PM
Sponsor: Biometrics Section
Abstract #332843
Title: Hybrid Cluster-Individual Randomization Allocation
Author(s): Yi-Fan Chen* and Jonathan Yabes
Companies: University of Illinois at Chicago and University of Pittsburgh
Keywords: group intervention; unequal allocation permuted block randomization; biased coin design; cluster randomization

In typical clinical trials, the unit of randomization is either at the subject- or at a group-level but not both. However, when only the treatment arm receives group-based intervention, group-level and subject-level randomization is preferred in the treatment and usual care arm respectively. In this design, we propose a hybrid randomization scheme (Hybrid I) based on unequal allocation permuted block randomization. The proposed method allows cluster and subject allocations at once for different arms, while maintaining unpredictability and balance. Groups are also assembled more quickly, thereby minimizing drop-out in the treatment arm due to long waiting time from randomization to receipt of intervention. Also, we refine our approach (Hybrid II) by adding biased coin design to preserve these features as a trial is terminated early. The performance of the proposed methods is evaluated via simulations. Results show that Hybrid I&II outperform cluster randomization and Hybrid II is better than Hybrid I, in terms of allocation concealment and allocation imbalance. Moreover, with Hybrid I & II smaller standard deviation and better power are observed.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2018 program