Evaluation of the Children's Eating Habits Questionnaire used in the IDEFICS study by relating urinary calcium and potassium to milk consumption frequencies among European children

Background: Measuring dietary intake in children is notoriously difficult. Therefore, it is crucial to evaluate the performance of dietary intake assessment methods in children. Given the important contribution of milk consumption to calcium (Ca) and potassium (K) intakes, urinary calcium (UCa) and potassium (UK) excretions in spot urine samples could be used for estimating correlations with milk consumption frequencies. Objective: The aim of this study was to evaluate the assessment of milk consumption frequencies derived from the Food Frequency Questionnaire section of the Children's Eating Habits Questionnaire (CEHQ-FFQ) used in the IDEFICS (Identification and prevention of dietary-and lifestyle induced health effects in children and infants) study by comparing with UCa and UK excretions in spot urine samples. Design: This study was conducted as a setting-based community-oriented intervention study and results from the first cross-sectional survey have been included in the analysis. Subjects: A total of 10 309 children aged 2-10 years from eight European countries are included in this analysis. Methods: UCa and UK excretions were measured in morning spot urine samples. Calcium and potassium urine concentrations were standardised for urinary creatinine (Cr) excretion. Ratios of UCa/Cr and UK/Cr were used for multivariate regression analyses after logarithmic transformation to obtain normal distributions of data. Milk consumption frequencies were obtained from the CEHQ-FFQ. Multivariate regression analyses were used to investigate the effect of milk consumption frequencies on UCa and UK concentrations, adjusting for age, gender, study centre, soft drink consumption and frequency of main meals consumed at home. Results: A significant positive correlation was found between milk consumption frequencies and ratios of UK/Cr and a weaker but still significant positive correlation with ratios of UCa/Cr, when using crude or partial Spearman's correlations. Multivariate regression analyses showed that milk consumption frequencies were predictive of UCa/Cr and UK/Cr ratios, when adjusted for age, gender, study centre, soft drink consumption and frequency of main meals consumed at home. Mean ratios of UK/Cr for increasing milk consumption frequency tertiles showed a progressive increase in UK/Cr. Children consuming at least two milk servings per day had significantly higher mean UCa/Cr and UK/Cr ratios than children who did not. Large differences in correlations between milk consumption frequencies and ratios of UCa/Cr and UK/Cr were found between the different study centres. Conclusion: Higher milk consumption frequencies resulted in a progressive increase in UK/Cr and UCa/Cr ratios, reflecting the higher Ca and K intakes that coincide with increasing milk consumption, which constitutes a major K and Ca source in children's diet. International Journal of Obesity (2011) 35, S69-S78; doi:10.1038/ijo.2011.37