Elsevier

Clinical Nutrition

Volume 32, Issue 4, August 2013, Pages 613-618
Clinical Nutrition

Original article
A comparison and validation of child versus parent reporting of children's energy intake using food frequency questionnaires versus food records: Who's an accurate reporter?

https://doi.org/10.1016/j.clnu.2012.11.006Get rights and content

Summary

Background & aims

The aim of this study was to (i) to compare the accuracy of reporting for child's total energy intake from a food frequency questionnaire (FFQ) completed independently by the mother, father and child in comparison to total energy expenditure (TEE) measured using doubly labeled water (DLW) (ii) compare the accuracy of the weighed food record (WFR) and DLW.

Methods

Healthy weight children (mean ± SD age 9.8 ± 1.3years, n = 6 girls/3 boys) and their parents independently completed an FFQ about children's intake. A 4-day WFR of child intake was recorded simultaneously. The accuracy of energy intakes reports were determined by the absolute and percentage differences between estimated energy intake and TEE measured by DLW.

Results

The mean difference (limits of agreement LOA, ±2SD) when compared to DLW was; child 130 (−1518, 1258) kcal or (113 ± 35% of TEE); father 398 (0,796) kcal or (121 ± 13%); mother 807 (−213, 1824) kcal or (144 ± 26%) and for the WFR −153 (1089, −1395) kcal or 95 ± 32%.

Conclusions

Children were the most accurate reporters when compared to their parents, with fathers more accurate than mothers. The 4-day WFR was approximately equal to the child report FFQ in estimating EI in children 8–11 years.

Introduction

It is by many agreed that there is no perfect method of assessing dietary intake1 and that dietary intake measurement is complex. Measuring dietary intake in children is particularly challenging as limited valid tools exist to assess dietary intake in pediatric populations.1 Food frequency questionnaires (FFQs) ask respondents to report their usual food intake frequency of a defined list of foods over a nominated period of time and are a potentially valuable tool for assessing dietary intake in children. FFQs have been recommended for use in large cross-sectional and cohort studies1 as they capture usual dietary intake over longer periods of time compared to other methods and can be collected from a large number of people in a relatively short time frame.2 In addition FFQs have a relatively low respondent and analytic burden3, 4 and can be presented using a range of options including self, proxy, interviewer or web administration. However, FFQs have been criticized on the grounds of overestimation of intake,5 limited food lists, portion sizes, and recall bias.5

There are specific additional concerns in using FFQs in children, which are related to child cognitive development and capacity to concentrate, and may influence their ability to both recall foods and estimate portion sizes.6 For these reasons child dietary intake is commonly reported by a proxy, usually parents,6 and most commonly by the mother7 who is regarded as the gatekeeper of food provision within the family environment. In a review of validation studies for the assessment of dietary intake in children, parents were proxies or parental assistance was used in more than 50% of studies.1 The ability of a child to self report food and drink consumed increases with age.6 It has been suggested that to complete an FFQ, children need to be ≥12 years old.8 However, uncertainty exists when a child is between the ages of eight and 12 years as to who (i.e. parent or child) should be asked to report child intake, due to factors including increasing child independence, cognitive abilities and increased consumption of food and drinks outside the home outside of parental control.

The Australian Child and Adolescent Eating Survey (ACAES) is a child-specific FFQ that has been developed and validated in youth aged 9–16 years for fruit and vegetable intake, nutrient and fat intake.9, 10, 11 While the ACAES has been shown to be valid for energy intakes in the younger age groups (8–12 years) for both child and maternal reports,12 it was not clear who was the more accurate reporter, mother or child and accuracy of father report is unknown.

The gold standard method for validation of energy intake is by comparison with total energy expenditure determined using the doubly labeled water (DLW) technique in weight stable individuals. Unfortunately, this technique cannot be used widely due to its high cost and the need for specialized laboratory facilities and staff expertise for analysis.7 As an alternative, weighed food records (WFRs) are commonly used as the reference method in dietary intake validation studies as the errors associated with WFRs and FFQs are considered to be largely independent.13

Therefore this study aimed to determine and compare the accuracy of the reporting of child daily energy intake determined by parent-completed 4-day weighed food records and the ACAES FFQ completed independently by the mother, father and child, in comparison to total energy expenditure (TEE) measured by the DLW method.

Section snippets

Participants

Children aged 8–11 y with two eligible caregivers (defined as residing predominantly with the child) were recruited from the Hunter region, NSW Australia during February 2009 predominantly through university emails and community notice boards. Children were eligible to participate if they were a healthy weight (defined by age and sex-specific Body Mass index (BMI) 18.35–20.74)14; had two caregivers who were willing to attend all required sessions; no known medical conditions affecting body

Results

Eleven child participants were recruited to the study. One child became ill during the study and one child did not adhere to the study protocol; therefore data suitable for analysis were obtained from nine children. These children were all weight stable (±1 kg) during the data collection period. The mean weight variation during the 10-day collection period was 0.15 ± 0.17 kg (range −0.2 to +0.3 kg). The children (n = 9) were mostly female (n = 6) and had a mean ± SD age 9.8 ± 1.3years, BMI

Discussion

This study evaluated the accuracy of child total energy intake reporting by comparing the ACAES FFQs completed independently by the mother, father and child in comparison to total energy expenditure (TEE) measured using DLW. The child-reported FFQ evaluation was the closest to gold standard measure of DLW, indicating that the children were the most accurate reporters (113 ± 35%) using the ACAES FFQ, followed by fathers (121 ± 13%), with mothers (144 ± 26%) the least accurate.

The 4-day WFR

Conclusion

The accuracy of FFQs in estimating child EI compared to the DLW method appears consistent with other published literature. Children were the most accurate reporters when compared to their parents, with fathers more accurate than mothers. The child FFQ report was approximately equal to the 4-day weighed food record in estimating EI in children aged 8–12 years. The ACAES FFQ may be used in studies to estimate TEE in 8–11 year old children but consideration should be given to who reports child

Funding

This study was funded by a new staff grant for T Burrows, University of Newcastle.

Conflict of interest

None.

Acknowledgments

Clare Collins is a NHMRCCDA fellow. TB and, CC were major contributors in: study design, data collection, analysis, and preparation of manuscript. RC, PJ, HT, PD all provided input to study design and preparation of manuscript. All authors have approved the final article. None of the authors had any financial support or relationships that may pose conflict of interest.

References (30)

  • M.B. Livingstone et al.

    Issues in dietary assessment in children and adolescents

    Br J Nutr

    (2004)
  • A. Field et al.

    Reproducibility and validity of a food frequency questionnaire among fourth to seventh inner city school children; implications of age and day to day variation in dietary intake

    Public Health Nutr

    (1999)
  • J. Watson et al.

    Reproducibility and comparative validity of a food frequency questionnaire for Australian children and adolescents

    Int J Behav Nutr Phys

    (2009)
  • T. Burrows et al.

    Validation of food frequency questionnaire using red blood cell membrane fatty acids

    Eur J Clin Nutr

    (2012)
  • T.L. Burrows et al.

    Validation of overweight children's fruit and vegetable intake using plasma carotenoids

    Obesity

    (2008)
  • Cited by (81)

    • Relationship of parental feeding practices and diet with children's diet among South Asians in Canada

      2022, Appetite
      Citation Excerpt :

      These age groups were chosen to represent social environments and psycho-behavioural patterns primarily influenced by parents versus peers and self (Davison & Jago, 2009). For practical purposes, the 7–8 year old group was also considered to be the youngest group that could reliably contribute information about their preferences and health behaviours (e.g., dietary reporting by children as young as 8 years is as accurate as parent reports (Burrows et al., 2013; Livingstone et al., 2004; Lytle et al., 1993)). Schools with high rates of South Asian enrollment were asked to deliver packages containing an invitation letter and consent forms to entire classrooms in grade 2, 3, 9 and 10.

    View all citing articles on Scopus
    View full text