JMIR Publications

ABSTRACT

Background:

Chronic diseases are the leading global cause of death, largely driven by Western lifestyles characterized by poor diets and physical inactivity. Digital interventions, including apps and chatbots, offer promising tools to promote health behavior change. In particular, interactive conversational agents (CAs) provide real-time, personalized support for meal planning, dietary advice, and healthy eating habits. Their ability to engage users makes them valuable in addressing diet-related chronic diseases. Despite their potential, evidence of their effectiveness remains limited. Systematic evaluations of their impact, features, and user acceptability are needed to determine their role in public health strategies for improving dietary behaviors and preventing chronic diseases.

Objective:

The main aim of this mixed systematic review is to evaluate the effectiveness of CAs designed to improve dietary behaviors, to list their basic features, functions, and conversational capabilities, and to assess the impact of interactive conversational agents on nutrition knowledge as well as their usability, acceptability, user experience, and engagement.

Methods:

Five electronic bibliographic databases were used: MEDLINE (Ovid), CINAHL (EBSCO), Embase (Embase.com), Web of Science (Core Collection), and PsycINFO (Ovid). Reference lists of relevant articles were reviewed to ensure that all eligible studies were included. Studies published since 2013 in English, French, or Spanish that examined the use of conversational agents for dietary modification were included. Two independent reviewers assessed the studies, and a third reviewer resolved any disagreements. The quality of the studies was assessed using the Mixed Methods Appraisal Tool (MMAT).

Results:

Studies reported improvements in fruit and vegetable intake, adherence to Mediterranean diet, and overall nutritional knowledge following CA interventions. Challenges in maintaining recommended servings of whole grains and other dietary components were observed. While some studies highlighted caffeine and snack consumption reductions, these changes were not statistically significant. Overall, dietary behavior improvements varied across studies and populations. Although the selected studies had dietary behaviors change as their primary outcome, other outcomes were also addressed. For example, the benefits of CA interventions on physical activity and stress management were also noted, while CA intervention engagement and long-term adherence presented challenges. Usability and user experience were generally positive, with features like goal setting and feedback enhancing satisfaction. However, the limited effect on social support highlights the need for further refinement, and the variability in study design makes comparisons between studies difficult.

Conclusions:

CAs demonstrate promising potential for improving dietary behaviors, with generally positive usability and user experience. However, variability in outcomes, engagement, and long-term adherence challenges, as well as limited impact on social support, highlight areas requiring further investigation. Systematic evaluations and refined designs are essential to optimize their effectiveness and establish their role in public health strategies aimed at preventing diet-related chronic diseases.