The literature on quality has often focused on process indicators. In this paper we outline a framework for describing and measuring the quality of health systems in terms of a set of desirable outcomes. We illustrate how it can be measured using data collected from a recent evaluation of health system performance conducted by the World Health Organization (WHO). We then explore the extent to which this framework can be used to measure quality for all components of the system; for example, regions, districts, hospitals, and providers. There are advantages and disadvantages to defining quality in terms of outcomes rather than process indicators. The advantage is that it focuses the attention of Policy makers on whether systems are achieving the desired goals. In fact, without the ability to measure outcomes it is not possible to be sure that process changes actually improve attainment of socially desired goals. The disadvantage is that measuring outcomes at all levels of the system poses some problems particularly related to the sample sizes necessary to measure outcomes. WHO is exploring this, initially in relation to hospitals. The paper discusses two major challenges. The first is the question of attribution, deciding what part of the outcome is due to the component of the system under discussion. The second is the question of tin-Ling, including all the effects of current health actions now and in the future.
