Random research

Advances in computing have combined with the rapid dissemination of treatme nt discoveries for diseases of public health importance to create pressure for accelerated promulgation of promising research results to the medical c ommunity. The 2 recent examples of the US Carvedilol Heart Failure program and the Evaluation of Losartan In the Elderly (ELITE) study demonstrate the importance of the prospective nature of research design, as well as the co nsequences of its abandonment. This article explains in nonmathematical ter ms the rationale for the tenet "first say what you will do, then do what yo u said" in sample-based research.