Abstract:
|
Blinding is a fundamental issue in randomized clinical trials. If blinding is violated, the behavior of the subject may be affected by knowledge of the intervention and a bias can be introduced. This can adversely affect the study by influencing the interpretation of clinical outcomes and subsequent clinical decision-making. Regulatory guidance suggests that the study design should include an evaluation of the integrity and effectiveness of the blinding (FDA Guidance, http://www.fda.gov, 2013). Several published papers (James et al, 1996; Bang et al, 2004) proposed some methods to assess the success of blinding in clinical trials but at the end of the study only. It is expected that the difference in efficacy between the active and control groups will lead to some unblinding gradually overtime. This paper therefore investigates a new approach to sequentially monitor blinding during the entire clinical trial. Based on the cumulative evaluation data, sequential blinding assessment tests are constructed at each stage and provide investigator a warning of unblinding. Results of extensive simulations are presented.
|
ASA Meetings Department
732 North Washington Street, Alexandria, VA 22314
(703) 684-1221 • meetings@amstat.org
Copyright © American Statistical Association.