ABSTRACT
The term “quality of life” (QOL) is used in a variety of ways. Given that most cancer clinical trials are designed to identify new drugs or combinations that improve survival, the role of QOL measures is often questioned. Even in the evaluation of interventions likely to change survival, there are lessons to be learned from QOL outcomes. Sometimes clinical investigators assume that a change in a biomedical outcome will also improve the patient’s QOL. Developing composite measures from multiple scales is more controversial as it leads to endpoints that are a combination of different aspects of QOL that may or may not be impacted in the same direction. Summary statistics across time are a very useful way of minimizing the impact of multiple comparisons procedures as well as facilitating the interpretation of results. The critical trial decision is whether to control the type-I error solely at the level of the primary endpoints with no adjustments for secondary endpoints.
