Monitoring of Clinical Trials

Authors:
Editors:
  • BRBernard Ravina
    Bernard Ravina
  • JCJeffrey Cummings
    Jeffrey Cummings
  • MMMichael McDermott
    Michael McDermott
  • MPMichael Poole
    Michael Poole

Abstract
This chapter highlights some important aspects of the design and analysis of clinical trials, and sketches a number of relevant statistical concepts. A controlled clinical trial of a medical intervention should have at least one primary hypothesis that drives its design. Well-designed and well-executed trials include an unambiguous protocol approved by the Institutional Review Boards (IRBs) or Ethics Committees of the participating clinics, laboratories, and data centers. The chapter also describes the basic frequentist statistical testing paradigm used by the typical randomized clinical trial with particular reference to ideas necessary in selecting sample size. Most clinical trials study more than one outcome of interest. Many neurological clinical trials compare therapies with respect to time to occurrence of the primary outcome. In the past, few clinical trials were performed in the Bayesian framework, but Bayesian methods have become more widely used recently.

This publication has 9 references indexed in Scilit: