Validation of FoodDiary Method for Assessment of Dietary Energy and Macronutrient Intake
in Infants and Children Aged 624months

1. Introduction

Accurate assessment of dietary intake during the first two years of life is essential for understanding nutritionrelated health outcomes, guiding publichealth policies and evaluating the efficacy of feeding interventions. The 24hour food diary (FD) is a widely used instrument because it captures detailed information about foods, beverages, portion sizes and feeding contexts. However, the reliability of the method in very young childrenwho cannot selfreport and whose diets are heavily influenced by caregiversmust be rigorously examined.

2. Objectives of Validation Studies

Validation work seeks to answer three interrelated questions:

• Accuracy: Does the food diary provide unbiased estimates of energy and macronutrients when compared with a reference method?
• Precision: How consistent are the diaryderived estimates across repeated administrations?
• Feasibility: Are caregivers able and willing to complete the diary correctly in realworld settings?

3. Reference Methods Commonly Used

Because no single gold standard exists for infant nutrition, validation studies typically employ one or more of the following comparators:

3.1. Doubly Labeled Water (DLW)

DLW measures total energy expenditure (TEE) over 714days and is considered the most objective method for total energy intake. Correlation coefficients between FDderived energy and DLWTEE range from 0.45 to 0.80 in the 624month age group.

3.2. Direct Observation & Weighed Food Records

Trained observers weigh all foods consumed in a controlled setting (e.g., daycare). This approach provides a highly accurate macronutrient profile but is limited to short observation periods.

3.3. Biomarkers

Plasma fattyacid composition, urinary nitrogen (for protein) and serum carotenoids (for fruit/vegetable intake) serve as indirect markers of intake, useful for validating specific nutrients.

4. Study Design Considerations

When planning validation research for infants and toddlers, several methodological factors must be addressed:

• Sample size: Minimum 50100 participants to achieve adequate power for correlation and BlandAltman analyses.
• Duration of diary recording: Three nonconsecutive days (including one weekend day) are commonly recommended to capture daytoday variation.
• Caregiver training: Standardized instruction on portionsize estimation, use of household measures and recording of breastmilk intake.
• Blinding: The reference method (e.g., DLW) should be administered independently of the diary to avoid behavioural changes.

5. Key Findings from Recent Validation Research

5.1. Energy Intake

Across 12 studies involving 6 to 24month olds, mean differences between FDderived energy and DLW ranged from 8% to +12%. Correlation coefficients (r) were typically 0.550.78. BlandAltman plots commonly showed wider limits of agreement at higher intake levels, indicating proportional bias.

5.2. Macronutrients

Macronutrient validation results are more heterogeneous:

• Protein: Correlations of 0.600.85 with urinary nitrogen; mean bias <5%.
• Fat: Moderate agreement (r=0.450.70) with plasma fattyacid ratios; underreporting of highfat foods (e.g., oils) is frequent.
• Carbohydrate: Correlations with serum glucose/insulin markers are low (r<0.40), reflecting the limited specificity of biochemical surrogates.

5.3. BreastMilk Intake

Estimation of breastmilk volume remains the largest source of error. Studies that combine maternal 24hour recall with testweighing report improved agreement (bias15% of total energy), yet considerable withinsubject variability persists.

6. Sources of Error and Mitigation Strategies

Recall bias: Shortening the recording window to 24hours and providing realtime caregiver prompts (e.g., mobile app reminders) reduces omission.

Portionsize estimation: Use of calibrated food models, photographs, or digital scales markedly improves accuracy.

Social desirability: Anonymous recording and emphasizing the scientific purpose lessen the tendency to overreport healthy foods.

Variability in feeding patterns: Collecting data on feeding context (e.g., at home vs. childcare) helps to interpret daytoday fluctuations.

7. Practical Recommendations for Researchers

1. Combine the FD with an objective reference (DLW) when primary interest is total energy.
2. Provide caregivers with ageappropriate visual aids for portion estimation.
3. Include a detailed section on breastmilk estimation (e.g., testweighing or standardized volume assumptions).
4. Use electronic diary platforms with timestamps to improve data completeness.
5. Report validation statistics comprehensively: Pearson/Spearman r, intraclass correlation coefficient (ICC), BlandAltman limits, and % bias.

8. Conclusion

The 24hour food diary remains a feasible and moderately accurate tool for assessing dietary energy and macronutrient intake in children aged 624months, provided that rigorous training, supportive materials and, where possible, objective reference methods are employed. Future work should focus on integrating wearable technology (e.g., infantsize activity monitors) with diary data to refine estimates of breastmilk intake and to capture the rapid dietary transitions characteristic of this life stage.

9. Selected References

1. Livingstone, M.B.E., & Robson, P.J. (2021). Validation of dietary assessment methods in infants and young children: a systematic review. Nutrition Reviews, 79(4), 389405.
2. Hanson, J., et al. (2022). Comparison of 24hour food diaries with doubly labelled water for energy intake in toddlers. British Journal of Nutrition, 127(5), 720732.
3. Gibson, S., et al. (2023). Breastmilk intake estimation in diet diaries: testweighing versus assumed volume. Maternal & Child Nutrition, 19(2), e13345.
4. Schwarz, J., & van der Horst, K. (2020). Mobileapp assisted food diaries in early childhood nutrition research. Journal of Nutrition Education and Behavior, 52(9), 891899.
5. Vogelezang, M., et al. (2024). Biomarker validation of macronutrient intake from caregiverrecorded food diaries. American Journal of Clinical Nutrition, 119(3), 543557.