Reproducibility of a 69-Item School Children's Food Frequency Questionnaire (69-Item SCh-FFQ) and Diet Quality Scores Among Polish Children Aged 7-9 Years.

Background/

Objectives

Reliable dietary assessment tools are essential for evaluating dietary habits and diet quality in children. The aim of this study was to assess the reproducibility of a 69-item School Children's Food Frequency Questionnaire (69-item SCh-FFQ) and the diet quality scores among Polish children aged 7-9 years.

Methods

The reproducibility study was conducted among 77 children. The 69-item SCh-FFQ was administered twice (test and retest). Data on food consumption frequency were used to evaluate two diet quality scores: the pro-Healthy Diet Index (pHDI) and the non-Healthy Diet Index (nHDI). The reproducibility of the 69-item SCh-FFQ and the diet quality scores was evaluated using comparisons of mean values, cross-classification analysis, Cohen's kappa statistic, Spearman's rank correlation coefficient, intraclass correlation coefficient (ICC), and Bland-Altman analysis.

Results

No significant differences were observed between test and retest of the FFQ for mean values of the pHDI (2.9 ± 1.8 vs. 2.8 ± 1.9 points; p = 0.124) or the nHDI (4.3 ± 1.5 vs. 4.2 ± 1.5 points; p = 0.611). Total agreement for the classification of children into categories of the diet quality scores reached 81.8% for pHDI and 75.3% for nHDI, with low gross misclassification rates (1.3% and 0.0%, respectively). Cohen's kappa statistic indicated moderate to good agreement (κ = 0.68 for pHDI and κ = 0.51 for nHDI). For all food items, total agreement ranged from 66.2% to 94.8%, while the kappa statistic ranged from 0.56 to 0.81. Strong correlations and good agreement were observed between repeated measurements for both pHDI (r = 0.748, ICC = 0.778; p < 0.001) and nHDI (r = 0.783, ICC = 0.752; p < 0.001). Bland-Altman analysis demonstrated good agreement and no evidence of proportional bias.

Conclusions

The 69-item SCh-FFQ demonstrated satisfactory reproducibility for assessing food consumption frequency and diet quality scores in children aged 7-9 years. Therefore, this questionnaire is a promising tool for future epidemiological studies among early school-aged children, although its utility should be confirmed in further validation studies using an appropriate reference method.