Retrieval of information in electronic patient records; experience with use of manual semantic tagging

摘要

Objective: To assess the value of manually supplied semantic tags on retrieving information from narrative patient records. Design: Retrospective evaluation of narrative electronic admission notes with respect to consistency of semantic tags supplied by means of flexible document templates. Assessment of effectiveness of retrieving information about clinical neurological examination using methods based on semantic tagging and string matching. Setting: Dep. of Neurology in a Norwegian university hospital using a document based electronic patient record system offering template assisted semantic indexing of textual record notes. Results: 75% of information regarding clinical neurological examination of the patient was consistently tagged. Retrieval models based on simple string matching performed better than models based on semantic indexing alone with respect to maximum obtainable mean recall and mean precision at high recall-levels. Models combining both string matching and the presence of semantic tags performed better than separate models with respect to maximum obtainable mean recall (0.95) and mean precision at high levels of recall. Conclusion: Although template assisted semantic indexing of narrative electronic patient record notes showed suboptimal consistency it had the potential of improving retrieval effectiveness with respect to maximum obtainable mean recall and precision at any recall level.