Instructor: Ajit C. Tamhane (Northwestern University)
AUGUST 18, 8:30—Noon
ROOM: LInC 368
Modern phase III confirmatory clinical trials often involve multidimensional study objectives which require simultaneous testing of multiple hypotheses with logical relationships among them. Examples of such study objectives include investigation of multiple doses or regimens of a new treatment, multiple endpoints, subgroup analyses, non-inferiority and superiority tests, or any combination of these. This short course will provide practical guidance on how to construct multiple testing procedures (MTPs) for such hypotheses while taking into account the logical relationships among them and controlling the appropriate Type I error rate.
Course Outline
- Introduction
- Error rates
- Methods for constructing multiple test procedures
- Common p-value based multiple test procedures
- Multiple test procedures for ordered hypotheses
- Gatekeeping procedures
- Primary and secondary endpoint testing in group sequential designs
- Graphical procedures
- Take-home lessons