LabVis: usability testing of a prototype tool for integrating timeline graphs and clinical notes
摘要
Effective patient care requires clinicians to develop a comprehensive understanding of a patient’s medical history, a task made challenging by the complexity of electronic health records (EHR). Timeline-based overviews have the potential to simplify this process, but proposed solutions often lack seamless integration between graphical summaries and free-text notes. The purpose of this study was to evaluate a prototype (LabVis) that uses temporal relationships to help clinicians quickly contextualize notable lab graph findings by enabling fast access to clinical notes from the same time period.
MethodsLabVis was developed based on input from experienced clinicians and existing literature. It features interactive timelines and semantic zooming to manage data density, along with tools for connecting findings in graphs to free-text notes. The prototype was tested with 10 physicians. Each session included task-based interaction with LabVis, direct observation, a System Usability Scale (SUS) questionnaire, and a semi-structured interview. Observational data were systematically reviewed to identify and rate usability issues, and interview transcripts were analyzed using Systematic Text Condensation.
ResultsParticipants reported high usability, with an average SUS score of 83. Timelines were said to help quickly identify key events, while integrated document navigation enabled efficient access to detailed notes. While participants emphasized the need for greater customization, there were no uniform suggestions for how to achieve this without compromising usability.
ConclusionsLabVis demonstrates efficient strategies for linking timeline-based overviews with free-text notes. The study underscores the importance of intuitive navigation, prioritization of key information, and customizable views for clinical workflows. These insights can be leveraged to design EHR systems that better support clinical decision-making. Further research is needed to explore how best to ensure customizability without compromising usability in a real-life setting.