JMIR Publications

ABSTRACT

Background:

Cognitive Behavioral Therapy (CBT) is the most examined type of psychotherapy for depression and anxiety; however, traditional CBT delivery faces significant barriers such as limited access, cost, and time constraints. CBT-oriented psychological chatbots offer a promising means of addressing these challenges. Yet, their overall efficacy, user engagement, and acceptability have not been systematically synthesized.

Objective:

This systematic review and meta-analysis aim to fill this critical gap by evaluating current evidence on chatbot-based CBT interventions for adults with depressive and/or anxiety symptoms.

Methods:

A systematic search was conducted across nine databases on April 1, 2025. Two reviewers independently selected studies, extracted data, and assessed the risk of bias. Random-effects meta-analyses were performed to estimate pooled effect sizes (Hedge’s g) for depressive and anxiety symptom outcomes at post-intervention and follow-up. Subgroup and meta-regression analyses examined potential moderators. Data on user engagement and acceptability were extracted and synthesized using narrative and quantitative methods where available.

Results:

Out of 13783 records, 23 eligible randomized controlled trials (RCTs) were included for systematic review and meta-analysis. CBT-oriented psychological chatbots produced a moderate reduction in depressive symptoms at post-intervention (g = -0.60, 95% CI [-0.77, -0.42]) and a small but significant effect at follow-up (g = -0.30, 95% CI [-0.56, -0.03]). For anxiety symptoms, a small effect was found post-intervention (g = -0.27, 95%CI [-0.39, -0.15]), whereas no significant effect emerged at follow-up (g = -0.06, 95% CI [-0.23, 0.10]). Subgroup analyses indicated that participants symptom profiles, control conditions, intervention duration, platform type, and chatbot design might influence the short-term efficacy on depressive and anxiety symptoms. Notably, the efficacy of CBT-oriented chatbots appears to be symptom-specific, with depression benefiting most from longer interventions (8-16 weeks) while anxiety peaks at 5-8 weeks. The CBT-oriented chatbots received an adequate level of engagement that complied with digital intervention standards. Although user satisfaction ratings were generally favorable, technical limitations and repetitive interaction patterns await to be addressed to enhance overall acceptability.

Conclusions:

This study presents the first comprehensive synthesis of evidence on the efficacy, user engagement, and acceptability of CBT-oriented chatbots, demonstrating their potential to reduce depressive and anxiety symptoms among adults, particularly in the short term. The findings underscore the innovative promise and enhanced accessibility that CBT-oriented chatbots bring to digital mental health interventions. Future research should prioritize enhancing long-term effectiveness, standardizing engagement metrics, and exploring design features that strengthen therapeutic alliances and personalized user experiences. Clinical Trial: PROSPERO CRD42024615506; https://www.crd.york.ac.uk/PROSPERO/view/CRD42024615506