Evaluation of the Relative Validity and Test-Retest Reliability of a 15-Item Beverage Intake Questionnaire in Children and Adolescents

Abstract

Background: Added sugar intake, in the form of sugar-sweetened beverages (SSBs), may contribute to weight gain and obesity development in children and adolescents. A valid and reliable brief beverage intake assessment tool for children and adolescents could facilitate research in this area.

Objective: The purpose of this investigation was to evaluate the relative validity and test-retest reliability of a 15-item beverage intake questionnaire (BEVQ) for assessing usual beverage intake in children and adolescents.

Design: This cross-sectional investigation included four study visits within a 2- to 3-week time period.

Participants/setting: Participants (333 enrolled; 98% completion rate) were children aged 6 to 11 years and adolescents aged 12 to18 years recruited from the New River Valley, VA, region from January 2014 to September 2015. Study visits included assessment of height/weight, health history, and four 24-hour dietary recalls (24HRs). The BEVQ was completed at two visits (BEVQ 1, BEVQ 2).

Main outcome measures: To evaluate relative validity, BEVQ 1 was compared with habitual beverage intake determined by the averaged 24HR. To evaluate test-retest reliability, BEVQ 1 was compared with BEVQ 2.

Statistical analyses performed: Analyses included descriptive statistics, independent sample t tests, χ² tests, one-way analysis of variance, paired sample t tests, and correlational analyses.

Results: In the full sample, self-reported water and total SSB intake were not different between BEVQ 1 and 24HR (mean differences 0±1 fl oz and 0±1 fl oz, respectively; both P values >0.05). Reported intake across all beverage categories was significantly correlated between BEVQ 1 and BEVQ 2 (P<0.001). In children (n=126), reported intake of milk and energy (in kilocalories) for total beverages was not different (all P values >0.05) between BEVQ 1 and 24HR (mean differences: whole milk=3±4 kcal, reduced-fat milk=9±5 kcal, and fat-free milk=7±6 kcal, which is 7±15 total beverage kilocalories). In adolescents (n=200), water and SSB kilocalories were not different (both P values >0.05) between BEVQ 1 and 24HR (mean differences: -1±1 fl oz and 12±9 kcal, respectively).

Conclusions: A 15-item BEVQ provides results that are similar relative to multiple 24HRs for determining habitual milk and total beverage intake in children, and water and SSB intake in adolescents. The 15-item BEVQ is a reliable indicator of habitual beverage intake in both children and adolescents. Future studies could explore whether adjustments to BEVQ beverage categories, portion size, and format could improve the tool's ability to measure beverage intake in young populations.

Keywords: Adolescent; Children; Dietary assessment; Sweetened beverages; Validation.