A total of 315 children aged 9 to 12 years-old were recruited from the primary schools throughout the city of Mashhad, north-east of Iran. Data were collected from December 2018 to March 2019. Study subjects were chosen using a multi-stage random cluster sampling method. Children in the fourth, fifth and sixth grades who had not any chronic or acute diseases, were eligible to participate in the study.
Sociodemographic data including child’s age, birth order, parents' age, parental education was obtained through interviews with students and verified by their mothers or caregivers by experienced interviewers.
Anthropometric measurements were performed by a trained dietitian using the calibrated equipment. Body mass index (BMI) was calculated as weight in kilograms divided by the square of height in meter. The BMI Z-score for age and sex was calculated based on the World Health Organization Child Growth Standards software (AnthroPlus, World Health Organization, Geneva, Switzerland, 2007). The weight status of children was reported in four categories including underweight (z-score < 2 standard deviation (SD), normal (z-score ≥–2 SD and ≤ 1 SD), overweight (z-score > 1 SD and ≤ 2SD), and obese (z-score > 2 SD).
The Household Food Insecurity Access Scale (HFIAS) was used to assess HFI in the study sample. The validity and reliability of the Persian version of questionnaire in the Iranian population was confirmed by Salarkia et al (13). The questionnaire consists of 9 items investigating a wide range of food-related behaviors, experiences, and conditions due to the financial limitation over a recall period of past month. Based on the total score, households were categorized as food secure (0–1 point), mild (2–7 points), moderate (8–14 points), and severe food insecure (15–27 points).
The FNLIT was measured using a developed questionnaire. The questionnaire examining FNLIT in two distinctive domains with 7 subscales, including 1) cognition domain: understanding food and nutrition information and nutritional health knowledge; 2) skills domain: functional FNLIT, interactive FNLIT, food choice literacy, critical FNLIT and food label literacy, respectively.
We confirmed the validity and reliability of FNLIT in this population. Content Validity Ratio (CVR) and Content Validity Index (CVI) of the 40-item questionnaire were at acceptable levels of 0.87 and 0.99. The internal consistency and test-retest reliability were assessed using Cronbach α (subscale-specific, range: 0.68–0.8) and intra-class correlation coefficients (ICC: 0.97, CI: 0.94-98), respectively. FNLIT scores were ranked into 3 categories as low FNLIT (≤ 58), medium FNLIT (> 58-<81) and high FNLIT (≥ 81).
Statistical analysis was performed using SPSS version 25 (SPSS Inc., Chicago, Illinois, USA). Chi-square test was used to compare the demographic and anthropometric variables between the food-secure and food-insecure subjects. Crude and adjusted linear regression models were performed to determine the influence of HFI on changes in overall FNLIT score and its sub-categories. Also, to determine the odds of having low FNLIT score in food insecure subjects in comparison to the food secure one, the crude and adjusted multiple regression models were used. The covariates included in the adjusted analyses were sex, grade, BMI, birth order, as well as parental age and education. Categorical variables were presented as frequency and percentage, while the numerical data were reported as mean, standard deviation (SD), beta coefficient (β), odds ratio (OR), and 95% confidence interval (CI). Significance level was considered as a p-value less than 0.05.