User Testing of a Diagnostic Decision Support System with Machine-assisted Chart Review to Facilitate Clinical Genomic Diagnosis

Abstract

Objective: There is a need in clinical genomics for systems that assist in clinical diagnosis, analysis of genomic information and periodic re-analysis of results, and can utilize information from in the electronic health record to do so. Such systems should be designed using the concepts of human-centered design, fit within clinical workflows, and provide solutions to priority problems. Materials and methods: We adapted a commercially available diagnostic decision support system (DDSS) to use extracted findings from a patient record and combine them with genomic variant information in the DDSS interface. Three simulated patient cases were created, and a structured interview guide was used to assess technical issues, workflow fit, and acceptability. Three testers conducting evaluation of genetic conditions were selected for user testing; 3 additional testers were included to evaluate fit and system implementation potential. Qualitative analysis of test session transcripts was done. Results: Tester responses were positive and noted good fit within real-world clinical genetics workflow. Technical issues related to interface, interaction, and design were minor and fixable. Testers suggested solving issues related to terminology and usability through training and added infobuttons. Time savings was estimated at 30-50% and other uses for in-house clinical variant analysis were noted to increase fit with workflow and address priority problems. Discussion and Conclusion: We provide preliminary evidence that a system including DDSS and automated chart review may help improve efficiency and consistency of genetic diagnosis and addresses perceived needs for clinical end-users. Further development could improve functionality and implementation potential.