Sample size in clinical protocols

Abstract

Sample size calculation in clinical protocols submitted to the review board of the Instituto Nacional de Ciencias Médicas y Nutrición Salvador Zubirán frequently has three types of problems: insufficient data for sample size recalculation, use of obsolete equations that do not take the beta error into account, and use of equations that do not correspond to the study design. In this document, we discuss the four data that should be included in the clinical protocols of randomized controlled trials that compare drugs versus placebo in superiority trials, and versus active controls in non-inferiority studies.