Abstract
The Danish Hereditary Non-Polyposis Colorectal Cancer Register (HNPCC-R) has formed the basis for many epidemiologic studies supporting risk stratification and both national and international guidelines on surveillance. However, the HNPCC-R is based on voluntary reporting and the completeness of registration is unknown. Hence, we aimed to assess the validity and completeness of the colorectal cancer (CRC) registrations in the HNPCC-R. We combined the registrations in the HNPCC-R with the registrations in the national Danish Cancer Registry (DCR), which is validated and based on mandatory reporting, and matched the cases from each register using a 13-step algorithm based on unique, national Central Population Registration number, date of diagnosis, location, and morphology thus identifying 9160 verified CRCs in 49,799 individuals. The overall agreement between the registers was 85%, and the completeness of the HNPCC-R was 95%. The DCR had the highest number of registrations before 1975, and the HNPCC-R had the highest number of registrations after 1985-especially more synchronous and metachronous cases. In conclusion, data from the HNPCC-R on CRC are valid and should be preferred for studies on CRC in families with a heritable increased risk of colorectal cancer-especially in Lynch syndrome, which is known for multiple CRCs, though a combination of both registers would secure the most optimal dataset.