A Protocol-Driven, Bedside Digital Conversational Agent to Support Nurse Teams and Mitigate Risks of Hospitalization...

Area of Science:

• Geriatric Medicine
• Human-Computer Interaction
• Digital Health

Background:

• Hospitalized older adults face risks of isolation, disorientation, depression, delirium, and falls.
• Embodied Conversational Agents (ECAs) are interactive technologies that can support socioemotional relationships.
• A novel relational ECA, Care Coach, uses an animated avatar monitored by live health advocates to implement clinical protocols.

Purpose of the Study:

• To evaluate the impact of a human-in-the-loop, protocol-driven relational ECA on loneliness, depression, delirium, and falls in hospitalized older adults.
• To assess the effectiveness of an animated avatar intervention for improving the well-being of elderly inpatients.

Main Methods:

• A clinical trial involving 95 adults over 65 hospitalized at an inner-city community hospital.
• Intervention group received a Care Coach avatar; control group received a daily 15-min visit from a nursing student.
• Measures included the UCLA Loneliness Scale, Geriatric Depression Scale, and Confusion Assessment Method, administered at enrollment and before discharge.

Main Results:

• Participants using the avatar showed a significantly lower frequency of delirium at discharge (P<.001).
• Avatar users reported fewer symptoms of loneliness (P=.01) and experienced fewer falls compared to the control group.
• No significant differences were observed in self-reported depressive symptoms between the groups.

Conclusions:

• Human-in-the-loop, relational ECAs are validated as effective tools for hospitalized older adults.
• The Care Coach avatar intervention demonstrates potential for enhancing care quality and patient outcomes in geriatric inpatient settings.
• Further research may explore optimizing ECA protocols to address depressive symptoms in this population.