The main goals of hospital-based and population-based cancer registries are respectively to contribute to patient care by providing information on cancer patients and to produce statistics on the occurrence of cancer in a defined population in order to assess and monitor the impact of cancer in the community. To achieve these goals the use of complete quality data is essential. Therefore, cooperation between a hospital-based cancer registry (HBCR) and the Belgian Cancer Registry (BCR) was set up. A pilot study was conducted to exchange clinical data (date of death) between the HBCR and the BCR. Secondly, we wanted to analyse the completeness and the quality of data delivered by the HBCR to the BCR. For the pilot study, all new patients with a diagnosis of head and neck cancer in 2005 and 2006 were included. For the analysis of the completeness and quality of the data all invasive or in situ cancers with an incidence date of respectively 2005 and 2006 were included. The HBCR could be supplemented with 23 dates of death (42%). Overall, the completeness of the registration was near 100%. Except for the TNM-data of malignant melanoma the quality of the data delivered by the HBCR to the BCR showed a maximal rate of missing data of 1.7% (basis of diagnosis) and a maximal rate of conflicting data of 2.8% (basis of diagnosis combined with specific histology). Cooperation between the HBCR and the BCR gives an added value to both registries. The HBCR could be complemented with data from the BCR. The feedback report can increase the completeness and accuracy of the data of the HBCR because it provides a focus on the quality of the data.

(BELG J MED ONCOL 2013;7(3):74–79)