Sample size for a dose-response study.

Abstract

This paper deals with a method of sample size allocation for a dose-response study assuming a logistic model for the dose-response curve. The method is based on the precision with which one wishes to estimate the dose that would produce the efficacy resulting in a clinically important difference from a placebo. An example is given to illustrate the methodology. The main development of the paper is for a binary response and is suitably modified for a continuous variable.