A Matrix-Based Method for Optimizing the User Interface Design of Electronic Medical Record Systems Open Access
Downloadable ContentDownload PDF
Usability issues present in current Electronic Medical Record (EMR) systems diminish healthcare provider satisfaction, lower productivity, and introduce medication-related errors, all major factors that impact improvements in the quality-of-care and lead to unrealized savings estimated at $12 billion over the next ten years. The impact of usability on user productivity and quality-of-care improvements has contributed to the low adoption rate of EMR systems. Such systems have complex user interfaces (UIs) caused by the precise details in medical practice. A tempting UI design solution is to present as much information on the screen as possible, so that all functions and options are available to the user. This often results in interfaces that cause user errors and create risks to patient safety. If the number of UI elements on the screen is not optimized for simplicity, users will be inefficient in completing their tasks and are more susceptible to accepting erroneous data.This dissertation describes a systematic method of UI inspection and design by combining the Design Structure Matrix (DSM) modeling technique and interface design metrics from the fields of Systems Engineering and Human-Computer Interaction. The proposed method captures and analyzes the interactions between UI elements of an EMR system, and organizes these elements on the screen based on information exchange, spatial proximity, and similarity to improve screen density, layout simplicity, and task completion time. The method was tested in an academic hospital setting, where the UI design of an EMR medication dosage calculator was examined. Medication dose adjustment task time was estimated using a cognitive simulation model that predicts user performance. Medication dosage calculators, used mainly in pediatrics to determine the medication dose for young patients, require simplicity in UI design to prevent user errors that can adversely affect a patient's well-being. If the UI design is not optimized for simplicity, users will be less efficient in adjusting medication dose due to an increase in task time.The study shows that improvements to the medication dosage calculator design lower screen density and layout complexity by an average of 17% and 124%, respectively. The predicted user performance model based on cognitive simulations shows that medication dose adjustment task time was reduced by an average of 21% as a result of design improvements, which are estimated to have economic value to the hospital by enabling physicians to collectively see an additional 75 patients over the course of a year. The study shows that the application of the DSM modeling approach to UI inspection and design can improve the usability of the EMR system and user productivity. Moreover, the presented method can be extended to inspect and improve the UI design of medical devices and other non-medical software systems.