Maintenance Notice

Due to necessary scheduled maintenance, the JMIR Publications website will be unavailable from Wednesday, July 01, 2020 at 8:00 PM to 10:00 PM EST. We apologize in advance for any inconvenience this may cause you.

Who will be affected?

Accepted for/Published in: Journal of Medical Internet Research

Date Submitted: Dec 1, 2025
Date Accepted: May 19, 2026

The final, peer-reviewed published version of this preprint can be found here:

Therapeutic Interaction Features of AI Chatbots in Depression Interventions: Systematic Review and Meta-Analysis

Huang T, Li S, Wang Y, Liu W

Therapeutic Interaction Features of AI Chatbots in Depression Interventions: Systematic Review and Meta-Analysis

J Med Internet Res 2026;28:e88697

DOI: 10.2196/88697

PMID: 42379144

PMCID: 13318397

Therapeutic Interaction Features of AI Chatbots in Depression Interventions: A Systematic Review and Meta-analysis

  • Ting Huang; 
  • Shuangyu Li; 
  • Yanzhong Wang; 
  • Wei Liu

ABSTRACT

Background:

Depression is a prevalent mental health disorder and a leading cause of disability worldwide, creating substantial personal and societal burdens. Digital mental health interventions (DMHIs) have emerged as accessible and scalable solutions, with AI-driven chatbots increasingly applied to deliver therapeutic content, monitor symptoms, and provide personalised support. However, limited evidence exists on how chatbot interaction features influence treatment adherence and clinical outcomes in depression.

Objective:

To investigate the relationship between AI-driven chatbots, clinical outcomes, and user adherence in depression interventions.

Methods:

A systematic review and meta-analysis were conducted following PRISMA guidelines, searching six databases for randomised controlled trials (RCTs) published before June 2025. Eligible studies involved AI-driven chatbots for depression treatment and reported adherence rates or clinical outcomes. Data extraction included chatbot type, interaction features, adherence, and standardised mean differences (SMD) for symptom change. This systematic review and meta-analysis was preregistered on the Open Science Framework (OSF) prior to data synthesis.

Results:

A total of 11 RCTs involving 1,237 participants were included. AI-driven chatbots demonstrated a moderate and statistically significant improvement in depressive symptoms compared with control conditions (SMD = -0.45, 95% CI [-0.76, -0.14], p = 0.004). Subgroup analyses indicated notable variation in user adherence across chatbot types, with NLP/ML-based systems showing the highest retention, followed by rule-based and LLM-based systems. Further exploratory analyses of interaction features revealed that emotional responsiveness, structured feedback strategies, and interaction frequency were positively associated with adherence, whereas dialogue depth, self-disclosure encouragement, and user agency level showed weaker or inconsistent effects.

Conclusions:

AI-driven chatbots can improve depression outcomes, with adherence influenced by specific interaction features. These findings highlight the importance of targeted interaction design to enhance engagement and therapeutic effectiveness in digital mental health interventions.


 Citation

Please cite as:

Huang T, Li S, Wang Y, Liu W

Therapeutic Interaction Features of AI Chatbots in Depression Interventions: Systematic Review and Meta-Analysis

J Med Internet Res 2026;28:e88697

DOI: 10.2196/88697

PMID: 42379144

PMCID: 13318397

Download PDF


Request queued. Please wait while the file is being generated. It may take some time.

© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.