(Conference paper at the HealthInf conference, Madeira, January 28-31, 2008.) Abstract: Patient health data has a valuable potential for secondary use, such as decision support on a national level, reimbursement settlements, and research on public health or on the effects of various treatment methods. Unfortunately, extensive secondary use of data has disproportionate negative impact on the patients’ privacy. The Norwegian health data processing regulation prescribes four different ways of organizing health registers (anonymous, de-identified, pseudonymous or fully identified data subjects). Pseudonymity is the most innovative of these methods, and it has been available as a legitimate means to achieve extensive secondary use of accurate and detailed data since 2001. Up to now, two different national health registers have been organized this way. The evidence from these experiences should be encouraging: Pseudonymity works as intended. Yet, there is still discernible reluctance against extending the pseudonymity principle to encompass other national health registers as well.