Introduction and aim: National diabetes registry has been an important aim of Croatian diabetology for a long time. Currently, efforts are made to define a national registry based upon Cro Diab NET, a computer software designed as a patient record form and suitable for automatic extraction of data for the registry.
Methods: Diagnostic entities captured are defined by the International Classification of Diseases and Related Health Problems, 10th rev. (E10--E14, G63.2, and H36) and ICF (b540, b279, b298, b2108). Central CroDiabNET module is BIS (Basic Informatic Sheet), recognized by the international diabetology community as optimal data collection form for the follow-up and improvement of diabetes care. There are several ways of data collection: secondary and tertiary health centers with diabetes care within their scope of activities send BIS on their patients to the central base by a modem, other secondary and tertiary health centers as well as primary health care will send data on their diabetic patients either via the Internet or by mail (interactive BIS will be available on the registry www page). The preferred method of data collection is the electronic one. The central base automatically links records of uniquely identified patients. Other records are checked separately, compared with the existing records and linked manually. Data confidentiality is ensured by double password protection of excess at the level of both users and registry. Unauthorised access is highly unlikely.
Results: So far, most of the secondary and tertiary health centers dealing with diabetology have been active in data collection. Inclusion of primary care (family physicians) is in progress. The last report covered data from 13 centers. Besides data on the number of patients, the reports contained analyses of BIS patient data availability. This is a tool for the analysis of the quality of registry and it can also provide basic data for planned actions aimed at quality of care improvement. Regular data collection from all levels of health care with concomitant connection of the registry to the national death registry started in 2004. With complete implementation of these processes a unique national diabetes database will be defined.
Conclusion: Results of the use of CroDiabNET so far confirm its potency as a valuable tool for population registry of diabetes as well as for improvement of diabetes health care. Regular periodic reports reveal an increasing number of centers involved. With continuation of this trend the registry will become a national, population-based database.