JMIR Publications

ABSTRACT

Background:

Chronic diseases necessitate effective self-management strategies to improve health outcomes and reduce healthcare demand. Conversational agents (CAs) offer promising, scalable, personalized solutions, yet robust evidence of their efficacy remains limited.

Objective:

This systematic review examines CA characteristics, integration of behaviour change techniques, and evaluation methods to guide future research and intervention design.

Methods:

We conducted a systematic search in PubMed to identify studies that evaluated the efficacy of CAs for chronic diseases self-management published from 2018 onward. Data extraction was guided by conceptual frameworks to ensure comprehensive reporting of intervention and methodologies: the Behavioral Intervention Technology model; and the CONSORT eHealth checklist.

Results:

Twenty-four studies were included, primarily focusing on text-based, rule-based CAs delivered via mobile applications. The chronic diseases predominantly targeted were diabetes and cancer. Commonly identified clusters of behaviour change techniques were ‘Shaping knowledge,’ ‘Feedback and monitoring,’ ‘Natural consequences,’ and ‘Associations’. However, reporting of behaviour change techniques and their delivery was lacking, and intervention description was limited. Studies were mostly early phase, with a great variety in intervention description, study methods and outcome measures.

Conclusions:

Advancing the field of CA-based interventions requires transparent intervention description, rigorous methodologies, consistent use of validated scales, standardized taxonomy and reporting aligned with standardized frameworks. Enhanced integration of AI-driven personalization and a focus on implementation within healthcare settings are critical for future research. Clinical Trial: Our protocol was not registered.