Design of Clinical Trials with the Desirability of Outcome Ranking Methodology

We discuss issues in the design of clinical trials comparing the benefit:risk of two interventions using the Desirability of Outcome Ranking (DOOR) methodology, particularly sample size determination. The DOOR is a paradigm for the design, analysis, interpretation and reporting of clinical trials and other research studies based on patient-centric benefit:risk evaluation. The DOOR methodology uses a DOOR outcome representing a global patient response, constructed on the basis of important clinical outcomes (e.g., benefits, harms, and possibly quality of life) to evaluate the benefit:risk of interventions. We describe power-based methods for assessing sample size in superiority trials when the DOOR outcome is used as the primary endpoint and the “DOOR probability” is used to compare the DOOR outcomes between intervention groups, where the DOOR probability represents the probability of a more desirable DOOR outcome for a trial participant assigned to one intervention relative to a trial participant assigned to the other. We illustrate the sample size assessments using these methods with an example. We also briefly discuss a power-based method for sample size determination in noninferiority trials, and precision-based methods for controlling the width of the confidence interval of the DOOR probability.