Technical Reports : Adaptive Designs
Technical Reports : Adaptive Designs
Issues in the use of adaptive clinical trial designs.
Summary: A discussion of a variety of issues that must be considered when using an adaptive clinical trial design.
Ref: Emerson SS, Statistics in Medicine 25: 3270-3296 (2006).
Adaptive Methods: Telling ‘The Rest of the Story’.
Summary: A commentary on the draft FDA Guidance on adaptive designs.
Ref: Emerson SS, Fleming TR, Journal of Biopharmaceutical Statistics 20: 1150-1165 (2010).
Exploring the benefits of adaptive sequential designs in time-to-event endpoint settings.
Summary: When using a time to event as a primary endpoint there are some features that might suggest an advantage to a more adaptive approach.
Ref: Emerson SC, Rudser KD, Emerson SS, Statistics in Medicine 30:1199-1217 (2011).
Summary: A commentary on a paper written by Mehta & Pocock (Statistics in Medicine)
Ref: Emerson SS, Levin GP, Emerson SC, Statistics in Medicine 30:3285-3301 (2011).
Summary: An exploration of efficient rules for adaptive modification of maximal sample size.
Ref: Levin GP, Emerson SC, Emerson SS, (BEPress) (2011)
Summary: G. Levin’s Ph.D. Dissertation exploring adaptive designs that allow modification of the sampling plan, as well as exploring inferential procedures following termination of the clinical trial.
Ref: Levin GP, University of Washington, Department of Biostatistics (2012)
Estimation Following Self-designing Clinical Trial
Summary: S Shi MS Thesis: A comparison of the MLE and self designing trial Z statistic for estimation following an adaptive RCT.
Ref: Shi S, University of Washington, Department of Biostatististics (2003)
Evaluation of Strategies for the Phase II to Phase III Progression in Treatment Discovery
Summary: Brittany J. Sanchez Thesis: An investigation of the importance of well-planned progression from Phase II to Phase III RCT in order to ensure a sufficiently low false positive rate of adopting ineffective treatments (a low type 1 error), a sufficiently high probability of adopting truly effective treatments (a high statistical power), and a sufficiently high probability that adopted treatments will be truly effective (a high positive predictive value). Also investigated in the bias associated with early termination of group sequential RCT.
Ref: Sanchez BJ (2014)