Validity and reliability of a food frequency questionnaire to estimate dietary intake among Lebanese children

Nutrition Journal

Dec 2015

DOI: 10.1186/s12937-015-0121-1

Citations:86

Influential Citations:7

Observational Studies (Human)

Enhanced Details

Methods

111 healthy Lebanese children aged 5–10 years (53 boys, 58 girls) from public and private schools in Greater Beirut, recruited by random cluster sampling. Mothers completed two FFQs (FFQ-1 at baseline and FFQ-2 after four weeks) and four 24-hour recalls were collected weekly over the 4-week period. Anthropometric measurements were taken. Study design: validation and reproducibility assessment of a culturally tailored FFQ, using four 24-hour recalls as reference and calibration analyses by linear regression; reproducibility assessed via ICC and weighted kappa.

Results

Spearman correlations between FFQ-1 and mean 24-HRs were significant (energy 0.54; p<0.001; MUFA 0.26). After calibration with sex-specific coefficients, mean intakes from FFQ-1 aligned with 24-HRs for energy and selected nutrients. Reproducibility (FFQ-1 vs FFQ-2) showed ICCs ranging from 0.31 (trans-fatty acids) to 0.73 (calcium); energy ICC = 0.71; over 80% were classified in the same or adjacent quartile for energy and nutrients. The calibrated FFQ is a reliable and valid tool for assessing past-year dietary intake in Lebanese children 5–10 years old and can support dietary monitoring and investigations of diet–disease relationships in this population.

Limitations

Limitations include proxy reporting by mothers which may introduce uncertainty and social desirability bias; sample size of 111 may limit generalizability; use of a USDA food composition database rather than a Lebanon-specific database; seasonal variation; potential self-selection bias with overweight/obese participants; results may not generalize beyond Beirut-area children.

Abstract

BackgroundNutritional status during childhood is critical given its effect on growth and development as well as its association with disease risk later in life. The Middle East and North Africa (MENA) region is experiencing alarming rates of childhood malnutrition, both over- and under-nutrition. Hence, there is a need for valid tools to assess dietary intake for children in this region. To date, there are no validated dietary assessment tools for children in any country of the MENA region. The main objective of this study was to examine the validity and reliability of a Food Frequency Questionnaire (FFQ) for the assessment of dietary intake among Lebanese children.MethodsChildren, aged 5 to 10 years (n = 111), were recruited from public and private schools of Beirut, Lebanon. Mothers (proxies to report their children’s dietary intake) completed two FFQs, four weeks apart. Four 24-hour recalls (24-HRs) were collected weekly during the duration of the study. Spearman correlations and Bland-Altman plots were used to assess validity. Linear regression models were used to derive calibration factors for boys and girls. Reproducibility statistics included Intraclass Correlation Coefficient (ICC) and percent agreement.ResultsCorrelation coefficients between dietary intake estimates derived from FFQ and 24-HRs were significant at p < 0.001 with the highest correlation observed for energy (0.54) and the lowest for monounsaturated fatty acids (0.26). The majority of data points in the Bland-Altman plots lied between the limits of agreement, closer to the middle horizontal line. After applying the calibration factors for boys and girls, the mean energy and nutrient intakes estimated by the FFQ were similar to those obtained by the mean 24-HRs. As for reproducibility, ICC ranged between 0.31 for trans-fatty acids and 0.73 for calcium intakes. Over 80 % of study participants were classified in the same or adjacent quartile of energy and nutrients intake.ConclusionsFindings of this study showed that the developed FFQ is reliable and is also valid, when used with calibration factors. This FFQ is a useful tool in dietary assessment and evaluation of diet-disease relationship in this age group. show less