Planning analyses of quality-of-life studies: a case example with migraine prophylaxis

Assessment of health-related quality of life (QOL) has become increasingly common in clinical trials. Because QOL has multiple dimensions and is assessed repeatedly over time, multiple comparisons and missing data must be addressed in the design and analysis of these trials. While the development of an analysis plan will be a continuous process, described here are three critical phases in the ideal development of a statistical analysis plan for the QOL component of a clinical trial: concept, initial design, and final analysis plans. A clinical trial designed to evaluate a new therapy for the prevention of migraines was used to illustrate the critical decisions that are made during the design and analysis phases of a trial that incorporates QOL in the assessment procedure. Specifically addressed are clearly defining goals and objectives, endpoints, multiple comparisons, analysis of longitudinal data with dropout, and estimation of power.