With most healthcare organisations now investing in quality improvement, it is important to ask not only did it work but how did it work? Evaluation in healthcare is often considered to be synonymous with audit, but where audit looks at compliance with standards, evaluation looks more widely at the relationship between actions and outcomes and a major task in evaluation is that of differentiating between association and contribution. In the complex arena of healthcare services, it is not sufficient to select a simple pre–post (or before and after) design and to use any differences in data measured to draw conclusions about the impact of the initiative. Unlike the experimental laboratory from which a pre–post design is drawn, in ‘real life’ situations there are many other potential causes for the observed change. Whilst a single quality improvement intervention can be expected to contribute to some of the change, it is rarely expected to account for all of it. The challenge, then, is to determine both the extent of change and the different factors that might account for it.