ABSTRACT
Identification of the components of HRQoL that are relevant to the disease and its treatment, development of specific priori objectives and an analytic plan are essential to both good trial design and subsequent scientific review. This chapter will focus on design and protocol development. Later chapters will deal with strategies for analysis of longitudinal studies (Chapters 3 and 4), definition of endpoints that summarize information over time or domains (Chapter 13), handling multiple endpoints (Chapter 14), development of a statistical analysis plan (Chapter 16), and sample size calculations for studies with dropout (Chapter 16). Figure 2.1 is a checklist for protocol development. All principles of good clinical trial methodology are applicable [Friedman et al, 1985, Meinert, 1986, Fleiss, 1986, Spilker, 1991, Piantadosi, 1997], but there are additional requirements specific to HRQoL. These include clear identification of the construct(s) that are relevant for the disease and treatment, selection of an appropriate measure of that construct and the conduct of the assessment to minimize any bias. The historical tendency to add a HRQoL assessment to a trial often at the last minute without careful thought about all the components of its necessity, implementation and analysis should be just that ... historical! The HRQoL outcomes (using the broad definition) should be a well integrated component of the protocol design.
