Data and safety monitoring boards of randomized trials: evolving principles and practical suggestions

Author(s): Charles H Hennekens,David DeMets4

Randomized trials designed a priori to test particular hypotheses provide the most reliable evidence concerning the most plausible small-tomoderate effects of drug therapies or interventions and are a necessary component of a totality of evidence upon which to make rational clinical decisions for individual patients and policy decisions for the health of the general public. As the methodology for the design, conduct and analysis of randomized trials continues to evolve, so do the principles and practical suggestions for their Data and Safety Monitoring Boards (DSMB)s. The implementation of these principles and practical suggestions should enhance the functioning of DSMBs, trial investigators and sponsors, protect the safety of the randomized subjects as well as the independence and integrity of the randomized trials.