Objective: The objective of the study was to validate the use of 24-hour recalls assisted by food records as a dietary assessment tool for use with third-grade children.
Design: Trained staff observed children during mealtime at school, and parents observed and recorded what children ate in their presence. The following day children participated in a 24-hour recall interview. Children's ability to recall what they consumed during a 24-hour period was compared with observational data collected during the same period.
Setting: All data were collected in elementary school settings at four sites involved in the Child and Adolescent Trial for Cardiovascular Health.
Subjects: The sample of 49 children was self-selected, based on parents' willingness to observe and record their child's food intake.
Main outcome measures: Recalled and observed data for energy and nutrient levels were compared using mean energy and nutrient analysis and quartile classification. In addition, recalled and observed foods were compared by meal type and estimation of portion size.
Statistical analysis performed: Paired t tests, Pearson and Spearman correlations, and classification analysis were used to compare recalled and observed data.
Results: Comparison of observed and recalled food intakes showed no significant differences in percentage of energy from total fat, saturated fat, monounsaturated fat, and polyunsaturated fat or in the amount of sodium consumed, although there were differences in energy intakes. Spearman rank order correlations between recalled and observed nutrients ranged from .45 to .79. A 77.9% agreement was found across all meals in the food items children recalled having consumed compared with those adults actually observed them consuming.
Conclusions: We conclude that the 24-hour recall assisted by food records is a valid method for assessing the dietary intake of children as young as 8 years old for the purpose of group comparison.