Two approaches for the analysis of a titration study with a placebo control

Abstract

To analyze a titration study with a placebo control, we propose in this paper two approaches that address the issue of placebo effects. The two approaches differ primarily in the way that the response rates of the active drug are evaluated. Both approaches assume logistic linear models. However, the logistic linear models assumed describe quantities with very different interpretation. Iterated re-weighted least squares is used to estimate the model parameters under both approaches. The proposed procedures are applied to a double-blind, placebo controlled clinical trial that investigated the efficacy of an anti-hypertensive compound.