Reassessing quality-of-life instruments in the evaluation of new drugs.

Investigators use 2 fundamental approaches to the measurement of health-related quality of life (HR-QOL). Generic instruments include health profiles that tap into the full range of HR-QOL issues and are widely applicable, but may lack responsiveness to small but important changes in HR-QOL. Utility measures summarise HR-QOL in a single number between 0 (death) and 1 (full health) and are useful for economic analysis, but may lack responsiveness. Accumulating data suggest the alternatives to generic measures, instruments that are specific to a function or a health problem, are more responsive than generic measures. While direct comparison of the validity and responsiveness of alternative approaches remains limited and should be extended, it is already clear that comprehensive assessment of HR-QOL requires more than 1 type of instrument. To be useful, HR-QOL instruments must be interpretable. Investigators are beginning to elucidate what constitutes trivial, small but important, or large changes in HR-QOL. Approaches include both within- and between-patient global ratings, observing HR-QOL scores in different patient populations, and observing the magnitude of change in HR-QOL with established interventions.