Our paper presents the results of a research study concerning the accuracy of the diagnosis and/or therapy for a cardiovascular patient. From the electronic patient records we have built a special database. The structure of this database offers possibilities for performing research studies on: cardiovascular pathology, risk factors action, medical treatment action, efficiency programs for primary and secondary prevention of diseases. The patient's data can be introduced from all intranet workstations of the cardiology department. The database structure is established to permit: optimal allocation of disk space, robust statistical analyses.

The coding scheme can be explained as follows: if the information can have two or more values which exclude each other, the information is represented by a character; if the information can have more values which don't exclude each other, the field contains many characters representing possible values.

Our paper shows that this database structure makes it possible to analyze every data piece that is stored. This system makes it possible to obtain a comprehensive summary of the cardiovascular population in real time by using standard reports and graphics, drill‐down and roll‐up on hierarchical dimensions, and analysis of temporal indicators. The database now contains information on 4000 patients undergoing effort tests as well as 2500 patients in the recovery program, and the system is being tested for all registered patients. We hope that field tests can begin in the near future, in the framework of a national cardiovascular infrastructure program.