Group sequential design in clinical studies

Presentation:

Sequential design and interim analyses in clinical studies. Seminarrom NSU1, Nevrosenteret, Wednesday 17 February 2010 at 1215-1400

Should we look at the results during a clinical trial? Should the trial be terminated before planned if the effect turns out to be larger than anticipated? Should we include more patients after the planned end of study if the p-value is 6%?

Many researchers are not aware that interim analyses must be specified in the study design. If not, the results will be misleading. Principles and practice for sequential designs will be presented. The presentation is open for all who are interested. This lecture is also part of the course KLMED8006 “Anvendt medisinsk statistikk”.

Lecturer: Stian Lydersen

Power point slides

Recommended literature:

Proschan, M. A., Lan, K. K., Wittes, J. T. (2006): “Statistical Monitoring of Clinical Trials: A Unified Approach” Springer *)

Jennison C and Turnbull, B W (2000): “Group Sequential Methods: Applications to Clinical Trials” Chapman & hall *)

Mazumdar M and Bang H (2008): “Sequential and group Sequential Designs in Clinical Trials: Guidelies for Practitioners”. Chapter 16 (pages 491-512) in Rao , Miller and Rao: “Handbook of Statistics Vol 27: Epidemiology and Medical Statistics”

Armitage P, Berry, G, Matthews, J N S (2002): “Statistical methods in medical research”. 4th ed. Section 18.7 Data Monitoring (page 613-623)

*) Available as e-book at UBIT