Validity and reliability of a simple, low-cost measure to quantify children's dietary intake in afterschool settings.

BACKGROUND: Interest in evaluating and improving children's diets in afterschool settings has grown, necessitating the development of feasible yet valid measures for capturing children's intake in such settings.
OBJECTIVE: The purpose of this study was to test the criterion validity and cost of three unobtrusive visual estimation methods compared with a plate-weighing method: direct onsite observation using a 4-category rating scale and offsite rating of digital photographs taken onsite using 4- and 10-category scales.
DESIGN: Researchers observed and photographed 174 total snack meals consumed across 2 days at each program. PARTICIPANTS/
SETTING: Participants were 111 children in first through sixth grades attending four afterschool programs in Boston, MA, during December 2011. STATISTICAL ANALYSIS: Visual estimates of consumption were compared to weighed estimates (the criterion measure) using intraclass correlations.
RESULTS: All three methods were highly correlated with the criterion measure, ranging from 0.92 to 0.94 for total calories consumed, 0.86 to 0.94 for consumption of prepackaged beverages, 0.90 to 0.93 for consumption of fruits/vegetables, and 0.92 to 0.96 for consumption of grains. For water, which was not preportioned, coefficients ranged from 0.47 to 0.52. The photographic methods also demonstrated excellent interrater reliability: 0.84 to 0.92 for the 4-point and 0.92 to 0.95 for the 10-point scale. The costs of the methods for estimating intake ranged from $0.62 per observation for the onsite direct visual method to $0.95 per observation for the criterion measure.
CONCLUSIONS: Feasible, inexpensive methods can validly and reliably measure children's dietary intake in afterschool settings. Improving precision in measures of children's dietary intake can reduce the likelihood of spurious or null findings in future studies.