Abstract Objective: to evaluate the results of parallel use of both paper based and electronic patient records with respect to concordance of corresponding information in two continuously updated versions of the same records. Design: retrospective evaluation of patient records, comparing documentation in electronic and paper based patient records. Setting: Department of Neurology in a Norwegian university hospital using paper based and electronic patient records in parallel during migration towards completely electronic patient records. Material: electronic and paper based patient records of 90 randomly selected patients visiting the department between 1 November 1997 and 30 April 1999. Results: seven percent of the electronic documents were significantly different in some way from the corresponding paper documents. About 4–13% of the documents in the electronic record were missing; one percent were missing from the paper record. Conclusion: parallel use of electronic and paper based patient records has resulted in inconsistencies between the record systems in our setting. Documentation is missing in both the electronic and paperbased records. When implementing electronic record systems intended to operate in parallel with paperbased systems, focus should be on securing the validity of all versions of the record.