Effectiveness of AI-Driven Conversational Agents in Improving Mental Health Among Young People: Systematic Review and Meta-Analysis

Background: The increasing prevalence of mental health issues among adolescents and young adults, coupled with barriers to accessing traditional therapy, has led to growing interest in artificial intelligence (AI)-driven conversational agents (CAs) as a novel digital mental health intervention. Despite accumulating evidence suggesting the effectiveness of AI-driven CAs for mental health, there is still limited evidence on their effectiveness for different mental health conditions in adolescents and young adults. Objective: This study aims to examine the effectiveness of AI-driven CAs for mental health among young people, and explore the potential moderators of efficacy. Methods: Five databases (PubMed, PsycINFO, Embase, Cochrane Library, and Web of Science) were searched from inception to August 6, 2024. Randomized controlled trials comparing AI-driven CAs with any control condition in young people aged 12–25 years were included, assessing depressive symptoms, generalized anxiety symptoms, stress, mental well-being, and positive/negative affect. Quality was assessed using the Cochrane Risk of Bias tool. Pooled effect sizes (Hedges g) were calculated using random-effects models. Results: Fourteen articles (15 trials) involving 1974 participants were included. After adjustment for publication bias, AI-driven CAs had a moderate-to-large effect on depressive symptoms (Hedges g=0.61, 95% CI 0.35–0.86). Effects adjusting for publication bias on generalized anxiety (g=0.06, 95% CI −0.21 to 0.32), stress (g=0.002, 95% CI −0.19 to 0.20), positive affect (g=0.01, 95% CI −0.24 to 0.27), negative affect (g=0.07, 95% CI −0.13 to 0.27), and mental well-being (g=0.04, 95% CI −0.21 to 0.29) were nonsignificant. Subgroup analyses showed particular effectiveness for depressive symptoms in subclinical populations (Hedges g=0.74, 95% CI 0.50–0.98). Conclusions: Findings highlight the potential of AI-driven CAs for early intervention in depression among young people and underscore the need to enhance efficacy across broader mental health outcomes. Key limitations include heterogeneity in CA therapeutic orientations and lack of follow-up measures. Future research should examine long-term effects of AI-driven CAs on mental health outcomes.

Yi Feng, Yaming Hang, Wenzhi Wu, Xiaohang Song, Xiyao Xiao, Fangbai Dong, Zhihong Qiao