A Mobile Application-Based Relational Agent as a Health Professional for COVID-19 Patients: Design, Approach, and...

Area of Science:

• Human-Computer Interaction
• Digital Health
• Artificial Intelligence in Healthcare

Background:

• Relational Agents (RAs) offer potential benefits for supporting patients with chronic conditions, including COVID-19.
• The unique socio-emotional and informational needs of COVID-19 patients necessitate tailored digital health interventions.

Purpose of the Study:

• To identify design principles for a Relational Agent (RA) to function as a healthcare professional for COVID-19 patients.
• To explore the tasks and characteristics of an RA suitable for supporting individuals recovering from COVID-19.

Main Methods:

• Qualitative thematic analysis of interviews with 33 COVID-19 survivors to identify RA tasks.
• Development and qualitative evaluation of four hypothetical RA conversation scenarios by 43 healthcare professionals.

Main Results:

• Identified essential RA functions: modeling clinical protocols, delivering evidence-based interventions, patient education, and reminders.
• Highlighted the importance of RAs building trusting relationships and addressing patients' socio-emotional needs.
• Healthcare professionals confirmed the suitability of proposed RA characteristics.

Conclusions:

• Design principles for RAs in COVID-19 care emphasize clinical accuracy, patient support, and relationship building.
• Findings provide a framework for developing RAs in diverse healthcare contexts beyond the pandemic.
• Future RAs can enhance patient care by integrating technological capabilities with empathetic interaction.