A comparison of two screening methods to determine the validity of 24-h food and drink records in children and adolescents

Abstract

Background/Objectives: Dietary assessment in humans is hampered by the phenomena of under and overreporting of energy intake, when food records are used to evaluate habitual dietary intake. Different methods to evaluate mis-reporting have been proposed using cut-offs derived from estimates of reported energy intake and basal metabolic rate, or, from predictions of total energy expenditure. This study compares the effect of using two different cut-off approaches to screen food records for validity, completed by a large cohort of Australian children (n=2460), from Grades 1, 5 and 10 (aged 5-17 years). Subjects/Methods: Energy intake was calculated from 24-h food and drink records for each child. These data were screened using the Goldberg and McCrory cut-offs. The effect of using these two cut-offs on the collected dataset was explored by considering the mean and standard deviation of energy intake in each year level before and after the cut-offs were applied. Results: The use of the Goldberg cut-off resulted in 9% of the total cohort being classified as underreporters, with 60% of these subjects being in Grade 10. The McCrory cut-offs revealed that overall, 22% of the total cohort underreported EI. 33.3% of Grade 1 children were classified as overreporters with this value falling to about 20% of Grade 10 children, while 10-15% of Grade 1 children underreported, with this figure rising to about 30% in Grade 10. Conclusions: Both the Goldberg and McCrory approaches have their advantages and disadvantages, and we suggest that consideration should be given to the reason for screening data before a particular approach is used, with recognition that these methods do differ in their aims and outcomes. The McCrory method consistently classified a greater number of children as underreporters. © 2011 Macmillan Publishers Limited All rights reserved.