Annelotte van Bommel

23 The NABON Breast Cancer Audit of the 32 quality indicators were calculated for all 92 institutions and changes over the 4-year-period were evaluated using a χ 2 trend test. Since information regarding adjuvant treatments requires a longer period (~9 months) to be completed, information of the quality indicators involving adjuvant treatments was not available for 2014. A P-value <0.05 is considered statistically significant. Comparisons of indicator results between the individual hospitals are visualized by funnel plots and presented in relation to the mean or norm (if applicable) using funnels to represent 95% confidence intervals. Boxplots with median hospital performance and interquartile ranges were used to analyze changes over the years. All analyses were performed using SPSS 20 (IBM-SPSS, Inc., Chicago, IL). RESULTS Hospitals In the Netherlands, breast cancer care is provided in 92 hospitals. Full participation of all hospitals was realized in 2012, the second year of registration. One hospital stopped treating breast cancer patients in 2012, and one new hospital was founded in 2013. About one third of the hospitals registered the data themselves and a high rate of case ascertainment was found after comparing data with those registered by the NCR for these hospitals (data not shown). Patient and tumor characteristics After 4 years of auditing the NBCA database contained data of 56,927 patients: 7,649 patients with DCIS, 49,073 with invasive cancer ( Table 1 ). In 205 patients (0.4%of all patients), DCIS or invasive cancer was not specified. Most patients were aged between 50 and 65 years (57% for DCIS and 43% for invasive breast cancer, respectively) at the time of diagnosis. Patients diagnosed with invasive breast cancer most frequently had relatively small tumors (pT1, 63%) and the majority had no axillary lymph node metastases (pN0, 64%). Quality indicators An overview of the overall results for all NBCA-indicators per year is displayed in Table 2 . 2