Systematic review and meta analysis of chatbots in the management of depressive and anxiety symptoms

Abstract

Mental health chatbots have proliferated rapidly, yet their effectiveness remains unclear. This systematic review and meta-analysis included randomized controlled trials comparing chatbots with any control condition for depressive and/or anxiety outcomes. PubMed, Embase, PsycINFO, Scopus and Web of Science were searched from January 2017 to October 2025. Risk of bias was assessed using the revised Cochrane tool. Pooled effect sizes (Hedges' g) were calculated using random-effects models. Of the 39 eligible studies, 38 (n = 7,401) were analyzed for depression and 34 (n = 7,621) for anxiety. Chatbots produced statistically significant reductions in depressive (g = 0.31, 95% CI [0.17, 0.46]) and anxiety symptoms (g = 0.28, 95% CI [0.05, 0.51]) compared with controls. Subgroup analyses for depressive symptoms showed larger effects in clinical and subclinical than in nonclinical samples (p = 0.001). Contemporary chatbots thus appear to alleviate depressive and anxiety symptoms, especially in individuals with greater depressive severity. (PROSPERO registration: CRD42024598761).