Acceptability, Usability, and Quality of a Personalized Daily Meal Plan Recommender System: The Case of Virtual Dietitian

Abstract

Nutrition research is now entering the subfield of personalized nutrition, where dietetics professionals are using it as an approach to support individuals in formulating unique dietary interventions and guidelines. Despite a large number of meal recommender systems that endeavors to incorporate the concept of personalized nutrition, the existing artifacts remain preliminary in the nutritional health context largely due to lack of integrated nutrition knowledge. Hence, a nutrition system called Virtual Dietitian (VD) was developed and grounded on the Nutrition Care Process and Model. Unfortunately, the beta evaluation (Phase 1) revealed some vital modifications that are needed to accomplish as per the feedback from experts. Hence, another sprint of development was achieved to comply with the requirements set forth by experts. This study reports the alpha evaluation (Phase 2) of 397 non-expert users on the revised VD on three factors: acceptability, usability, and quality. Using the scores from these factors, statistical analyses were performed to determine if there were significant differences between these scores and variables linked to users’ profile. Results show that VD passed on all factors, and there were significant differences between the scores and users’ profile (living condition, current physical activity, nutritional status, monthly household income, and average daily meals). Several recommendations were still offered on how to move beyond the existing features of VD and with considerations to relevant modern technologies.