Voice-Based Remote Care Program for Vulnerable Older Adults in a Rural Community: Single-Arm Pilot Clinical Study

Abstract

Background: Voice-based digital health technologies are highly feasible and acceptable tools for supporting older adults. However, their development has rarely focused on caregiving needs, and it is often poorly integrated with existing care services, thereby limiting their sustained effect. Objective: This study aimed to evaluate the feasibility and effectiveness of a comprehensive voice-based remote care program developed in partnership with a local public health center. Methods: A single-center, single-arm clinical study involving community-dwelling, socioeconomically vulnerable older adults was conducted using a Clinical Frailty Scale of 4‐5. Participants received a 6-month voice-based care program comprising smart speaker daily check-ins, an emergency response system, and artificial intelligence–driven well-being check calls. These components were integrated with the public health center for continuous monitoring. The primary outcome was caregiver burden, assessed using the Korean version of the Zarit Burden Interview. Secondary outcomes include depression (Patient Health Questionnaire-9), anxiety (Generalized Anxiety Disorder-7), and quality of life (Korean version of the Control, Autonomy, Self-realization, and Pleasure scale). Results: Among 100 enrolled participants, 96 (96%) completed the program. The caregiver burden slightly decreased from 17.1‐16.2 points (mean difference −1, 95% CI −2.17 to 0.24; P=.12). However, caregivers reported a significant reduction in their perception of being the sole support provider (P=.003). Among older adults, significant improvements were observed in depression (Patient Health Questionnaire-9; P