ABSTRACT
Quality clinical research must be well planned, closely and carefully monitored and conducted, and appropriately analyzed and reported. Greater attentiveness to detail at the design stage argues for greater efficiency at the analysis and reporting stages. An aspect of good design of protocols for new drugs is determining the
number of patients required by the clinical investigation to adequately address the objective. Not only is this important on a per-protocol basis, adequate numbers of patients must also be studied within and across the phases of clinical development to support regulatory filing. Although this chapter could go directly to a presentation on computation
of sample sizes, similar to what one would present in a statistics class, a quick review of the phases of clinical trials and their objectives is presented first. Then the clinical development plan and its connection to labeling are discussed. Thereafter the statistical requirements for sample size determination are presented. Proceeding in this manner sets the stage for the recommendations that are made relative to the size of trials in each phase of clinical development pre-market approval. Some philosophical if not controversial issues then follow as well as concluding remarks.
