Child-Reported Food Insecurity Is Negatively Associated with Household Food Security, Socioeconomic Status, Diet Diversity, and School Performance among Children Attending UN Relief and Works Agency for Palestine Refugees Schools in Lebanon.

BACKGROUND

Recent studies have shown that children are able to accurately describe their own experiences of food insecurity; however, few studies have quantified this experience.

OBJECTIVES

We aimed to develop and validate a child food security scale to be used in Arabic-speaking countries.

METHODS

As part of the Healthy Kitchens Healthy Children study on food security and nutrition in schools, 14 questions were administered to 5-15-y-old children (n = 1433) attending 4 UN Relief and Works Agency for Palestine refugees schools. We used statistical methods based on Item Response Theory to assess the internal validity of a proposed food security scale using these items. Further validation of the scale was conducted by examining associations with sociodemographic, economic, household food security, and diet diversity variables through logistic regression models adjusted for age, sex, and school of the children.

RESULTS

A total of 1287 children (89.8%) provided complete responses to the questionnaire. Four items were dropped from the scale because of high infit statistics (>1.3) and high residual correlation with other items. The resulting 10-item scale had acceptable reliability (Cronbach α coefficient = 0.89). Children who reported food insecurity were more likely to be from food-insecure households (OR: 2.3; P < 0.001). Food-insecure children came from households with lower expenditures per capita (OR: 0.80; P = 0.007) and higher household crowding (OR: 1.70; P = 0.001), and had less educated mothers (OR: 0.89; P = 0.009), lower household diet diversity (OR: 0.85; P < 0.001), and lower household consumption of meat (OR: 0.89; P < 0.001) and fruit (OR: 0.95; P = 0.001). Child food insecurity was also associated with higher school absenteeism (OR: 1.64; P = 0.01) and lower academic grades in languages and science (P < 0.05).

CONCLUSIONS

Child food security was accurately measured through a 10-item scale. This tool is appropriate to assess child food insecurity experience directly with children, and can be used in school-based interventions, where obtaining data from parents may be cumbersome.