How do automated reasoning features impact the usability of a clinical task management system? Development and usability testing of a prototype.

BACKGROUND: Electronic clinical task management systems (ECTMSs) have been developed and adopted by care providers to improve care coordination. Some systems utilised automated reasoning (AR) to enable more intelligent task management functionalities, such as automated task allocation. Yet, the impact of such features on usability remains unclear. Poor usability of health information systems has been described to cause frustration and contribute to patient safety incidents. AIM: To design AR features for an ECTMS and to evaluate their impact on usability. METHODS: In this mixed methods study, four ECTMS feature prototypes were co-designed with two clinicians. For each prototype, one AR variant and one non-AR variant with equivalent functionalities were developed. A moderated usability testing was conducted with seven clinicians to obtain ease-of-use ratings of prototypes and measure task durations. Parameters related to demographics and attitudes of participants were obtained via a questionnaire. A framework analysis was performed to summarise qualitative feedback. To determine statistical relationships of study variables, Spearmans rank coefficients were calculated and presented as a correlation matrix. RESULTS: Three out of four prototypes received higher median ease-of-use ratings for AR variants and were associated with shorter average task durations. Multiple clinical use cases suitable for AR were identified. Preference for AR was found to moderately correlate with digital proficiency and prior experience with ECTMSs. Insufficient trust in automation, alert fatigue, and system customisation were identified as challenges in the adoption of AR features. CONCLUSIONS: This study provides evidence for the potential of AR to enhance usability in ECTMSs. Consideration of psychological and organisational context of users in the feature design was found to be decisive for usability. Future research should explore implications for operational and clinical outcomes.

Trajectories of Anxiety Symptoms for COVID-19 Patients Using Multimodal Data Collected from Telemedicine.

We designed and developed Remote Patient Monitoring (RPM) system specific for coronavirus (COVID-19) patients, and collected multimodal data. Using the collected data, we explored the trajectories of anxiety symptoms for 199 COVID-19 patients quarantined at home. Two classes were identified using latent class linear mixed model. Thirty-six patients showed an exacerbation of anxiety. Presence of initial psychological symptoms, pain on the start day of quarantine, and abdominal discomfort at one month after finishing the quarantine were associated with exacerbation of anxiety.