BACKGROUND
Regular physical activity is crucial to wellbeing, but healthy habits are difficult to create and maintain. Interventions delivered via conversational agents (eg, chatbots or virtual agents) are a novel and potentially accessible way to promote physical activity. Thus, it is important to understand the evolving landscape of research utilizing conversational agents.
OBJECTIVE
This mixed-method systematic review aimed to 1) summarize the usability and effectiveness of conversational agents in promoting physical activity, 2) describe common theories and intervention components utilized, 3) identify areas for further development, and 4) make recommendations for conversational agents targeting health behavior change.
METHODS
We conducted a mixed-method systematic review. We searched 7 electronic databases (PsycINFO, PubMed, Embase, CINAHL, ACM Digital Library, Scopus, and Web of Science) for quantitative, qualitative, and mixed methods studies that conveyed primary research on automated conversational agents designed to increase physical activity. Two reviewers independently screened studies and assessed methodological quality using the Mixed Methods Appraisal Tool (MMAT). Data on intervention impact and effectiveness, treatment characteristics, and challenges were extracted and analyzed using parallel-results convergent synthesis and narrative summary.
RESULTS
In total, 255 studies were identified, 20 of which met our inclusion criteria. Overall, conversational agents had moderate usability and feasibility and were effective in promoting physical activity. However, quality of evidence varied. Common challenges facing interventions were repetitive program content, high attrition, technical issues, safety, and privacy.
CONCLUSIONS
Conversational agents hold promise for physical activity interventions. However, there is a lack of rigorous research on long-term intervention effectiveness and patient safety. Future interventions should be based in evidence-informed theories and treatment approaches, and they should address users’ desires for program variety, safety and privacy measures, natural language processing, and delivery via mobile devices.
CLINICALTRIAL
The protocol for this systematic review was registered in the Open Science Framework (OSF) registries (osf.io/p4v6y).