OBJECTIVE: A precondition for the rational use of obstetric databases in biomedical research is detailed knowledge on how data are being generated. We identified the diagnostic procedures and criteria for pre-eclampsia (PE) and assessed the level of obstetric training of the personnel responsible for the records submitted to the patient registry at the Danish National Board of Health. STUDY DESIGN: A structured questionnaire, including three case stories, was sent to the chief consultant of the department. RESULTS: Thirty-three out of the 34 Danish departments (97%) returned the questionnaire. Reporters of pregnancy diagnoses to the National Patient Registry differed widely in training. For complicated pregnancies, departments ranged from having only specialists reporting all cases to secretaries reporting up to 50%. Cut off limits of blood pressure (BP) and protein loss used to diagnose pre-eclampsia showed large differences across departments. The diagnoses given to three case stories showed little correlation to the criteria the departments reported using. CONCLUSION: Even in a small country like Denmark with 34 obstetrical departments, there was little consensus on the diagnostic criteria for pre-eclampsia. The findings emphasize the need for standardizing diagnostic criteria and reporting practice and may have implications on how to interpret data regarding pre-eclampsia.