A taxonomy of health status instruments.

Clinicians and policymakers are recognizing the importance of measuring health related quality of life (HRQL) to inform patient management and policy decisions. Self or interviewer administered questionnaires can be used to measure cross sectional differences in quality of life between patients at a point in time (discriminative instruments) or longitudinal changes in HRQL within patients over time (evaluative instruments). Both discriminative and evaluative instruments must be valid (really measuring what they are supposed to measure) and have a high ratio of signal to noise (reliability and responsiveness, respectively). Reliable discriminative instruments are able to reproducibly differentiate between persons. Responsive evaluative measures are able to detect important changes in HRQL over time, even if those changes are small. HRQL should be interpretable--that is, clinicians and policymakers must be able to identify differences in scores that correspond to trivial, small, moderate, and large differences. Two basic approaches to quality of life measurement are available: generic instruments that attempt to provide a summary of health related quality of life; and specific instruments that focus on problems associated with individual disease states, patient groups, or areas of function. Generic instruments include health profiles and instruments that generate health utilities. The approaches are not mutually exclusive. Each approach has its strengths and weaknesses and may be suitable under different circumstances. Investigations in HRQL have led to instruments suitable for detecting minimally important effects in clinical trials, measuring the health of populations, and for providing information for policy decisions.