The utilization of Artificial Intelligence-based conversational agents on mental health care for older adults: a scoping review

OBJECTIVES: Conversational Agents (CAs), commonly referred to as chatbots, are increasingly employed in mental health care, with applications ranging from assessments to patient communication and psychoeducation. Recent advancements in Artificial Intelligence (AI) have enabled CAs to deliver personalized and engaging interactions. However, most existing mental health CAs overlook the specific needs and preferences of older adults, potentially increasing user burden and limiting adoption. To bridge this gap, the present study reviewed AI-based CAs designed for mental health support in older adults. METHOD: A comprehensive literature search across 13 databases (e.g. PubMed, PsycINFO, Scopus) identified 20 peer-reviewed articles. RESULTS: Of these, 12 studies examined usability, with the majority of them reporting positive acceptance and perceived usefulness, while one study presented mixed findings. For studies that assessed the effectiveness, most demonstrated significant improvements, and one reported mixed effects. Challenges for older adults using CAs were also identified, including the need for more user-centered designs specifically tailored to older adults, the absence of objective measurements of usability and effectiveness, and the limited range of therapeutic techniques within these interventions. CONCLUSION: These findings offer valuable guidance for researchers and practitioners aiming to optimize AI-based CAs to better meet the unique mental health needs of the aging population.