Annelotte van Bommel

33 The NABON Breast Cancer Audit national audit providing benchmark information to participating hospitals catalyzes quality improvement and insurance on various levels in the healthcare system. 13 An example of improvement on hospital level is a hospital that recognized itself as an outlier on the indicator “frequency of HER2-positive tumors”. Having observed a significant higher frequency they evaluated their pathology processes and found out that their laboratory used a different method of tracking HER2 positivity. This was subsequently adjusted. Another hospital observed low rates of patients discussed in a pre-operative MDT meeting, identifying that this was associated with a lack of meetings during holiday periods and they changed their clinic days to make sure every patient is discussed in an MDT meeting. On another level, regional cancer centers have organized network meetings reflecting on observed differences between the institutions within the network. Apart from the actions of the individual hospitals that were triggered by benchmarking their results, the comprehensive audit outcomes have led on a national level to in depth research into hospital variation in breast MRI use and immediate breast reconstruction facilitated by research grants of the Dutch Cancer Society. As such, the NBCA serves as a monitor to identify variation as well as a database that identifies factors explaining variation and eventually ought to catalyze guideline adjustments. The absence of consistency between indicator sets used by different other audits internationally is a limitation of individual audits as only uniform definitions of quality indicators can enable international benchmarking. 7–9,14,15 Nevertheless, guidelines may well differ between countries and therefore differences in quality parameters will remain, as the main goal of an audit is the quality assurance in a particular area. Process indicators A number of trends were observed since the introduction of the audit in 2011. For most quality indicators with a predefined quality norm, the mean value of all hospitals improved and the variation between the hospitals decreased, as was observed at an earlier moment in the Netherlands. 16 Significant changes were 2