Use this URL to cite or link to this record in EThOS:
Title: Using re-randomisation to increase the recruitment rate in clinical trials
Author: Kahan, Brennan Chaim
ISNI:       0000 0004 9355 4618
Awarding Body: Queen Mary University of London
Current Institution: Queen Mary, University of London
Date of Award: 2020
Availability of Full Text:
Access from EThOS:
Access from Institution:
Re-randomisation trials allow patients to be re-enrolled and re-randomised for each new treatment episode they experience. For example, in a trial evaluating treatments for acute sickle cell pain crises, patients could be re-randomised each time they have a new pain crisis. However, uptake of this design has been slow, likely because of uncertainty around its validity. The purpose of this thesis is to evaluate the methodological properties of the re-randomisation design. Chapter 2 defines a set of treatment estimands that can be used for rerandomisation trials, and chapters 3 and 4 evaluate the use of independence estimators and mixed-effects models for these estimands. I find that independence estimators are generally unbiased, though can be biased for certain estimands in specific situations. Mixed-effects models are generally biased, except under very strong assumptions. In Chapter 5 I compare re-randomisation with cluster, crossover, and parallel group designs. I find that re-randomisation compares favourably with the other designs, though depending on the specific research question (i.e. estimand of interest), other designs may be more appropriate in certain settings. In chapter 6 I evaluate a set of trials of granulocyte colonystimulating factors for patients with febrile neutropenia which include both parallel group and re-randomisation designs. I found that using re-randomisation led to an increase in recruitment and provided similar results to parallel group trials. In conclusion, the re-randomisation design is a valid design option, and should be used more often.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available