Development and Validation of a Semi-quantitative Food Frequency Questionnaire to Measure Macro-Micro Nutrients Intake for Saudi Population in the Western Region of Saudi Arabia

Thumbnail Image

Publication or External Link





Background: Among numerous dietary assessment methods, the food frequency questionnaire (FFQ) is the most common dietary assessment in large-scale epidemiological studies. The aim of this research was to develop a semi-quantitative food frequency questionnaire (SFFQ), which is a form of food frequency questionnaire (FFQ) for the western part of Saudi Arabia, as well as to examine the validation of this SFFQ for the measurement of macro-micro nutrients intake. Additionally, this project was aimed at testing the efficacy of an electronic intervention in a randomized controlled trial among college populations. The specific aim was to determine whether the electronic interventions achieved increases in the consumption of fruits and vegetables and decreases in the consumption of saturated fats, and added sugars in the intervention group relative to the control group. Methods: A list of foods consumed by 150 healthy Saudi participants, between 18 to 75 years old was obtained from 24-hour recalls. This list was combined with foods listed in a nationally representative survey. The food list for the new SFFQ was selected by stepwise multiple regression. Twenty-one models were generated, first one for energy and 20 models for each nutrient. To assess the new SFFQ validity, energy and macro-micro nutrient intake estimated from the SFFQ were compared with those calculated from 24-hour recall. Spearman's rank correlation coefficient, cross-classification method and bland-Altman plots for 100 adults (aged 18 to 75 years) were used to analyze the validity. The electronic intervention study took place over ten weeks. The intervention delivered selected healthy diet goals using two methods (text messages and e-mail) to 200 college students weekly. Food intake was evaluated by three-day food records, at the beginning of the study (baseline), at the middle (week 5) and at the end (week 10). Also, weight, height and BMI were measured at these different points of time during the study. Outcomes were reported as nutrient intake in addition to average daily servings of food intake. Results: To propagate the SFFQ, 230 different food items were recorded. After performing stepwise multiple regressions, final SFFQ included 152 food items. The food list was transformed to a new SFFQ form. The correlation coefficient between the new SFFQ designed to be used in Saudi Araba and 24-hour recall ranged from 0.2 for vitamin D to 0.7 for energy. Moreover, correlation coefficients were statistically significant for energy, vitamin D, fat, polyunsaturated fat, and monounsaturated fat. Also, in the overweight group, the correlation coefficients were statistically significant for energy, vitamin D, fat, polyunsaturated fat, and monounsaturated fat, calcium and phosphorus. The SFFQ intake data were significantly higher than the recall data. Cross-classification analysis revealed that the average proportions of participants classified into the same or adjacent quartiles, one quartile apart, and those misclassified were 72.2 %, 20.4%, and 7.4 %, respectively. There is no systematic bias between the administration of the two methods, the new SFFQ and 24-hour recall, according to Bland-Altman plots analysis. The intervention study found a significantly higher fruit and vegetable consumption in the intervention group (4.68 servings a day) compared to the control group (3.55 servings a day) at week 10, (P ≤ 0.05). Conclusion: The new Arabic SFFQ has been developed and this study indicates the need for continued development and validation of this new Arabic SFFQ. Also, the intervention conducted in Saudi college students reported significantly improved fruit and vegetable consumption. However, further research is needed to evaluate the intervention efficacy on long-term health behavior changes in this population.