Effectiveness of mental health chatbots in depression and anxiety for adolescents and young adults: a meta-analysis of randomized controlled trials

BACKGROUND: The mental health chatbot is dedicated to providing assistance to individuals grappling with the complexities of depression and anxiety. OBJECTIVE: The study aimed to evaluate the effectiveness of the mental health chatbot in alleviating symptoms of depression and anxiety among adolescents and young adults. METHODS: A systematic review framework was employed with a protocol pre-registered on Prospero (CRD42023418877). Databases were systematically searched, including PubMed, ACM Digital Library, Embase, Cochrane and IEEE. Data synthesis was conducted narratively, and meta-analysis was performed by pooling data from the original studies. RESULTS: Ten randomized controlled trials focused on an acute population, mainly females and university students. Chatbots designed for daily conversations and mood monitoring, using cognitive behavioral therapy techniques, showed efficacy in treating depression (95% CI = -1.09 to -0.23; p = .003). However, it is essential to highlight that these interventions utilizing chatbots for mental health were not found to be efficacious in managing symptoms of anxiety (95% CI = -0.56 to 0.4; p = .74). CONCLUSIONS: Evidence supports the effectiveness of mental health chatbots in treating depression, but further exploration and refinement are needed to optimize their efficacy in managing anxiety.