Methods and Applications of Statistics in Clinical Trials, Volume 1: Concepts, Principles, Trials, and Designs

Chapter 67

Prognostic Variables in Clinical Trials

Vance W. Berger and A. J. Sankoh

67.1 Introduction

A prognostic factor or variable is a patient characteristic that can predict that patient’s eventual response to an intervention. Prognostic variables include any and all known factors that could potentially impact the patient’s response outcome, including baseline demographic, concomitant illness, and medications factors that are likely to modify any treatment benefit, and those that may predict adverse reactions. These factors are also sometimes referred to as confounders because the presence of an imbalance at baseline between the treatment groups due to such factors may result in an apparent treatment effect when in fact none exists, or mask an effect that does in fact exist. Prognostic factors also include those potential confounders that have been used as stratification factors for randomization. Stratified randomization is often used when a baseline characteristic, such as tumor stage in cancer trials or baseline expanded disability status scale in multiple sclerosis trials, is known to affect the outcome or response. In such cases, the baseline characteristic is used as a stratification factor in the randomization scheme to minimize imbalance between treatment groups.

Similarly, when the study population contains subgroups of particular interest, the characteristics defining these sub-groups should be considered as potential progonostic factors. For example, in a long-term ...