Repeated reports of frequent and often preventable errors in medicine are leading to increasing public pressure for programmes to improve the quality of care. However, the results of such undertakings are disappointing. Consequently there is a need to identify the reasons for such a slow pace of improvement. Three interrelated questions must be systematically answered: (a) How do we measure the quality of care?; (b) If the quality of care proves to be deficient: what should be changed and how?; (c) How can we determine whether progress is being made? We are confronted with a myriad of issues and we must learn how to deal with the organisational and biological complexity of these. Therefore aggregated measures such as performance indicators are not suitable. But there is also another crucial question: who are we trying to satisfy? The increasing number of supervisors or our patients? The quintessence of quality of care is first and foremost putting responsibility for the care of our patients first and then accepting the consequences of this choice.