Preliminary Assessment of the Healthy Beverage Index for US Children and Adolescents: A Tool to Quantify the Overall Beverage Intake Quality of 2- to 19-Year Olds.

Abstract Background Improving beverage patterns of children and adolescents is recommended for combatting obesity and reducing disease risk. Therefore, it is important to assess beverage intake quality in this population. For adults, the Healthy Beverage Index (HBI) was created to assess beverage intake quality, but a similar tool did not exist for children and adolescents. Objective The objective was to develop an HBI for US Children and Adolescents (HBI-CA), and then assess the validity and reliability of this tool. Design Modeled after the adult HBI, age-specific, evidence-based beverage recommendations were compiled. Ten components were included to assess beverage intake quality. Validity and reliability were assessed using cross-sectional data and methods similar to those used for the evaluation of the Healthy Eating Index. Participants The 2015-2016 National Health and Nutrition Examination Survey provided 24-hour dietary recall data for 2,874 children and adolescents aged 2 to 19 years. Main outcome measures HBI-CA scores were the main outcome measure. Statistical analyses performed To assess validity, independent t tests were used to determine differences in HBI-CA component and total scores among groups, and principal component analysis was completed to examine multidimensionality of the HBI-CA. Pearson bivariate correlations were used to assess reliability. Results The HBI-CA produced a (mean ± standard error) total score of 69.2 ± 0.8, which is similar to the adult HBI mean total score of 63. Principal component analysis identified six factors, indicating the multidimensionality of the HBI-CA, with more than one combination of components contributing to variation in total scores. Most HBI-CA components were significantly correlated to the total score, with met fluid requirements, total beverage energy, sugar-sweetened beverage, and water components demonstrating the strongest correlations (r range = 0.335-0.735; P ≤ 0.01). Conclusions The results provide preliminary evidence to support the validity and reliability of the HBI-CA. If future research establishes the predictive validity and sensitivity of the HBI-CA, this tool could be useful to quantify the beverage intake quality of children and adolescents.