Consultation of medical narratives in the electronic medical record

This article presents an overview of a research project concerning the consultation of medical narratives in the electronic medical record (EMR). It describes an analysis of user needs, the design and implementation of a prototype EMR system, and the evaluation of the ease of consultation of medical narratives when using this system. In a questionnaire survey, 85 hospital physicians judged the quality of their paper-based medical record with respect to data entry, information retrieval and some other aspects. Participants were more positive a bout the paper medical record than the literature suggests. They wished to maintain the flexibility of data entry but indicated the need to improve the retrieval of information. A prototype EMR system was developed to facilitate the consultation of medical narratives. These parts were divided into labeled segments that could be arranged source-oriented and problem-oriented. This system was used to evaluate the ease of information retrieval of 24 internists and 12 residents at a teaching hospital when using free-text medical narratives divided at different levels of detail. They solved, without time pressure, some predefined problems concerning three voluminous, inpatient case records. The participants were randomly allocated to a sequence that was balanced by patient case and learning effect. The division of medical narratives affected speed, but not completeness of information retrieval. Progress notes divided into problem-related segments could be consulted 22% faster than when undivided. Medical history and physical examination divided into segments at organ-system level could be consulted 13% faster than when divided into separate questions and observations. These differences were statistically significant. The fastest divisions were also appreciated as the best combination of easy searching and best insight in the patient case. The results of our evaluation study suggest a trade-off between searching and reading: too much detailed segments will delay the consultation of medical narratives. Validation of the results in daily practice is recommended.