Interface Redesign of an Electronic Medical Record Review System Using User-Centered Design

Abstract

The healthcare industry in recent years has seen a rise in the adoption of Electronic Medical Records (EMRs). These EMRs have replaced the traditional paper-based records at hospitals, clinics, and nursing facilities. This transition has brought with it, numerous advantages of digitization such as improved patient care, timely reminders for checkups, and better health data tracking over time. But the EMR adoption has also come with its own set of challenges. The EMR systems are maintained by the medical coders/nurses at the hospitals. The coders are expected to gather information from different sources such as patient history logs, test results from different labs, etc. followed by entry into the EMR system. Due to the unstructured nature of the task, data entry in EMRs is susceptible to errors which lead to the poor data quality of patient records. Diagnostic decisions taken by the medical practitioners based on erroneous data can adversely affect the patient and at times, even prove to be fatal. To help address this issue of poor data quality of System X, an EMR, employs a unique data review process which allows reviewers (domain experts) to check patient records and communicate back the data entry errors to the coders for required changes to ensure high data quality. In this research, the user-centered design methodology was applied to improve the review process, with the aim of facilitating easier and quicker workflow. The usability issues faced by the reviewers were identified through heuristic evaluations, video walkthroughs, and user interviews methods. To address the issues identified, a new interface design was developed by employing low fidelity and high-fidelity prototyping techniques. Involvement of the reviewers throughout the research ensured that the design proposed was continually assessed and improved qualitatively until they were satisfied. Lastly, the Keystroke Level Model (KLM) was used to quantitatively assess the performance improvement gained from the new design. The final interface design was able to reduce the task-execution time of the patient record review process by 28.51%. This resulted in saving a significant amount of the reviewer’s time, thereby reducing their workload while improving data quality.