Development of a Novel Dietary Assessment Method Using Gamification Concepts: Exploratory and Application Study

Childhood and adolescent malnutrition, encompassing undernutrition and overnutrition, poses significant global health challenges, necessitating comprehensive dietary assessment tools. Existing dietary assessment methods, such as 24-hour dietary recalls (24HR), often fail to capture eating behaviors and food preferences.

This study aims to compare a newly developed gamified dietary assessment tool (GDA) with a traditional 24HR and to explore its applicability in assessing dietary behavior tendencies among children and adolescents.

A 2-phase study was designed, including an exploratory and an application study. The exploratory study included 30 school-aged participants, comparing the GDA with the 3-day 24HR. Nutrient and food intakes were analyzed using Pearson or Spearman correlation coefficients and Bland-Altman plots. The application study, conducted among 1541 adolescents (11-18 y), assessed dietary intake and eating behavior tendencies. Differences in dietary intake across age, gender, socioeconomic status, and weight status were analyzed using the Kruskal-Wallis rank sum test. Multiple linear regression models were used to examine the associations of dietary intakes with emotional eating and with dining environment tendencies, respectively.

In the exploratory study (n=30), the GDA demonstrated moderate agreement with 24HR for energy (r=0.46, P=.01) and carbohydrates (r=0.50, P=.005). Bland-Altman plots indicated good agreement for energy and carbohydrate intake between methods (mean differences around 0). For fat intake, although the mean difference was close to 0, the correlation was not statistically significant. In contrast, the GDA substantially overestimated protein intake (mean difference around 25 g). In the application study (n=1541), higher emotional eating scores were associated with higher snack consumption (β=0.438, 95% CI 0.035-0.840), and with lower protein (β=-0.159, 95% CI -0.267 to -0.052), fruit (β=-0.464, 95% CI -0.854 to -0.073), and nut consumption (β=-0.183, 95% CI -0.304 to -0.062). Participants who chose solitary screen eating consumed significantly more carbohydrates than those who selected "eat with peers" (β=4.2, 95% CI 1.2-7.1).

This study demonstrates that the GDA effectively captures both dietary intake and contextual eating behaviors in young populations, providing data distinct from traditional methods, such as 24HR. As a complementary assessment approach, it offers valuable insights into food preferences and eating patterns through its interactive design.