Factorial design. An approach to the assessment of therapeutic drug interactions in clinical trials
Factorial design is a useful experimental model for evaluating therapeutic drug interaction in clinical trials. Such a design was used to study interactions of pinacidil, a vasodilator, and hydrochlorothiazide, a diuretic, in hypertension. This paper describes the design of this study and statistical analyses that can be used to evaluate such data. Regression models that utilize response-surface methodology are very useful with factorial designs, especially in the presence of interaction between the factors. These models can be employed to test for linear (and more complex) effects of each drug and interaction, including linear-by-linear. Three-dimensional graphs are useful displays for visualizing such data and understanding the nature of interaction identified in the statistical analysis. Changes in postdose standing diastolic blood pressure are presented to illustrate the methods used. Response surface analysis of this data permitted simultaneous visualization of dose-response curves for pinacidil and hydrochlorothiazide and an interaction between the two drugs.