Measuring outcome can be an insensitive way to detect differences in the quality of health care. This paper captures the implications of this poor sensitivity for the interpretation of studies of outcome that compare provider performance, and considers in what circumstances monitoring outcome might be useful. When interpreting studies, it is important to consider the size of the effect that a difference in the quality of care might be expected to have on outcome and whether it is likely that important differences in quality might not have been detected. It is argued that outcome measures may be of value when how you do something is as important as what you do, when process measures are invalid or impractical, and when the overall effectiveness of an intervention is critically dependent upon its complication rate.