JMIR Publications

ABSTRACT

Background:

The global surge in mental health challenges has placed unprecedented strain on healthcare systems, highlighting the need for scalable interventions to promote mental health self-care. Chatbots have emerged as promising tools to address these gaps by providing accessible, evidence-based support. However, most existing studies focus on the clinical population and symptom reduction, with limited exploration of preventive strategies targeting self-care and mental health literacy among people in the community.

Objective:

This study evaluates the effectiveness of a rule-based, topic-specific chatbot intervention in improving self-care efficacy, literacy, self-care intention and self-care behaviors as well as mental well-being immediately after 10 days of usage and after 1 month.

Methods:

A two-arm, assessor-blinded randomized controlled trial was conducted. 285 participants were randomly assigned to either the chatbot intervention group (n=140) or a waitlist control group (n=145). The chatbot intervention consisted of ten topic-specific sessions targeting stress management, emotion regulation, and value clarification, delivered over 10 days with a 7-day free-access period. Primary outcomes included self-care self-efficacy, behavioral intentions, self-care behaviors, and mental health literacy. Secondary outcomes included depressive symptoms, anxiety symptoms, and mental well-being. Assessments were conducted at baseline, post-intervention at 10 days, and 1-month follow-up. Primary data analysis was performed using linear mixed models with intention-to-treat approach.

Results:

Participants in the chatbot group demonstrated greater improvements in behavioral intentions (d=0.31) and mental health literacy (d=0.45) compared to the control group as revealed by the significant interaction effects (mental health literacy: F(2, 423.57)=4.27, P=.015; behavioral intentions: F(2, 379.74)=15.02, P<.001). The chatbots were also able to bring immediate improvement on self-care behaviors (d=0.36), mindfulness (d=0.37), mental health outcomes of depressive symptoms (d=-0.26) and overall well-being (d=0.22) and positive emotions (d=0.28). Yet, these improvements did not differ significantly at 1 month when compared to the waitlist control. Adherence was higher among participants who received push notifications.

Conclusions:

This study highlights the potential of rule-based chatbots in promoting mental health literacy and fostering the intention to take care of own mental health. Despite recent advancements in generative AI, rule-based systems remain valuable due to their content control and safety assurances. However, improvements in chatbot design, including greater personalization and interactive features, may be necessary to enhance self-efficacy and long-term mental health outcomes. These findings underscore the utility of chatbots as scalable interventions to bridge gaps in mental health service delivery. Future research should explore hybrid approaches that combine rule-based and generative AI systems to optimize intervention effectiveness. Clinical Trial: ClinicalTrials.gov NCT05694507