Usability of digital health technologies

Background Motivated by the mixed outcomes of previous digital health initiatives, such as the National Programme for IT and the adoption of US-based electronic health record (EHR) systems, this thesis investigates how the usability of NHS digital systems could be improved through the development of modernised user interface (UI) guidelines and how quantitative usability modelling could be used to evaluate and improve the design of digital health technologies. Method The Design Science Research framework was used, spanning six stages: problem identification, objective definition, artefact development, demonstration, evaluation, and communication. EHR usability problems were identified by analysing the Certified Health IT Product List (CHPL) database of usability tests. The objective was defined through a document review of the existing NHS usability guidelines, which informed the development of novel UI design artefacts demonstrating how existing usability guidelines could be feasibly modernised. Human-computer interaction (HCI) modelling was used to evaluate the artefacts, and the results have been communicated through peer-reviewed journal articles, conference presentations, and this thesis. Results The CHPL analysis showed significant variability in the usability performance of EHR systems, establishing the need for improved usability guidelines. The document analysis of the NHS Common User Interface (CUI) and Design System (DS) guidelines provided the information needed to create updated UI design artefacts by integrating the CUI guidelines with the modern technology stack and design principles from the NHS DS. HCI modelling showed how different designs performed and provided a foundation for scientifically validated usability improvements without reliance on human-subject experiments. Conclusion This thesis contributes evidence for variability in the usability of existing EHR systems, proposes technically feasible updated design artefacts for two safety-critical components of EHR systems, the Patient Banner and Patient Name Input, and demonstrates how HCI modelling can be integrated into the Human-Centred Design process of designing evidence-based digital health technologies.