The European Prospective Investigation into Cancer and Nutrition (EPIC) is a prospective multicentre study that has been implemented to further the understanding of the association between diet and chronic diseases with emphasis on cancer. In Heidelberg from June 1994 until October 1998 about 25,500 subjects aged 35 to 65 years in women and 40 to 65 years in men were recruited. Apart from extensive questions about food intake, the participants were also asked to provide detailed information about their smoking habits, physical activity, subjective well-being, medical history and use of medications. As well as completing the questionnaire and a personal interview, the participants also gave blood samples and anthropometiric measures and the blood pressure were taken in standardized manner. The analyses of the EPIC study will depend on achieving a comprehensive record of all new cancer cases, and all deaths, together with the corresponding cause of death, within the study population. To date all self-reported incident cancer cases are verified by comparing them with pathology reports and hospital records. They are then coded according to the coding list for the International Classification of Diseases (ICD-O2) issued by the World Health Organisation (WHO). Since at begin of the investigation in the study region no cancer registration existed, the participants are followed -up by interval questioning ('active follow-up'). In order to integrate increasingly the data of the Cancer registry Baden-Württemberg (EKR-BW) attempts were made to explore record linkage systems. For this purpose, in the years 2000, 2002, 2003 record linkages between EPIC-Heidelberg and EKR-BW were performed. Procedures were evaluated for performing an anonymous linkage of the EPIC data with the data of the EKR-BW. After a pilot project on the feasibility of the linkage the program was evaluated on the EPIC data, record linkages are performed regularly. Different coding systems were applied. Simultaneously, the EPIC data about on cancer cases among the Heidelberg study participants are passed on to the Cancer Registry, thus contributing to improve completeness of the registry. So far the active follow-up can not be replaced by passive follow-up through record linkage with the cancer registry, but in the long-term it may be possible. Since the technical requirements are complied with, attempts should be made to improve the completeness of the epidemiological cancer registry Baden- Württemberg.