Health-related quality of life: an indicator of quality of care?

There is an increasing interest in the use of outcome indicators to monitor the quality of care. Traditionally, outcome indicators have been based mainly on biological indicators reflecting death or disease. Now that various instruments for health status measurement have become available, questions have been raised as to the potential application of health status scores in monitoring the quality of care. This paper identifies conditions which should be fulfilled before such applications can be recommended. Firstly, the relationship between care delivery processes and health status outcomes must be established. In order to achieve this, health status measures which are clearly able to detect health status variations between groups of patients (i.e. discriminative ability) and variations over time (i.e. sensitivity to change) are needed. Secondly, health status data should be available, preferably from established data collection registries (e.g. computerized hospital records or national registries) where data relating to the description of variations in health status (between physicians, hospitals, regions, etc.) are routinely collected. Thirdly, methods should be found to collect additional data, including 'case-mix' information and health status reference data, in order to enable the interpretation of variations in health status. Because most of these conditions are currently not being fulfilled, we conclude that the state-of-the-art of health status measurement has not yet matured sufficiently to allow for the use of health status as an indicator of quality of care. The present paper provides a framework for both future research and data collection that is needed to improve the applicability of health status measures as quality-of-care indicators.