To assess nutritional status among individuals and communities different methods of dietary assessment are suggested. Compared to other methods of dietary assessment food frequency questionnaires (FFQs) are more feasible in community-based studies. Therefore, we validated a
dish-based semi-quantitative
FFQ to assess the intake of energy, macro and micro nutrients, food groups and dietary diversity in a population of Iranian people.Materials and methods: A total of 284 healthy people who lived in Tehran, aged between 9 to 65 years old (138 men), did not take vitamin supplements and were non-smokers volunteered to take part in the study. A semi-quantitative
FFQ containing 142 food items and mixed dishes was developed to assess food and nutrients intake among the population. The questionnaire was completed two times, the first time in the cold season (winter 2016) and the second time in the hot season (summer 2016). As a reference method a 24-hour recall per month (a total of six 24-hour recalls) were obtained for each participant. Meanwhile two blood and urine samples were collected in time of the second and the fifth interviews. The criterion
validity was measured by calculating the Pearson and Spearman Correlation Coefficient between the estimated energy, macro and micronutrients, food groups and dietary diversity evaluated by the FFQ, 24-recalls, and plasma levels of retinol and beta-carotene as biomarkers. The
reproducibility of the questionnaire was determined by calculating Pearson or Spearman correlation for energy, macro and micronutrients, food groups and dietary diversity in two FFQs. Results: Mean correlation coefficient for energy and nutrients and food groups between two FFQs were 0.5 and 0.41, respectively. Mean Spearman’s correlation between the
FFQ and recalls was 0.33 for food groups and 0.32 for energy and nutrients, respectively. The mean correlation coefficient between the
FFQ and plasma level of retinol and beta-carotene was 0.26 and 0.16, respectively. The mean percentage of energy and nutrients for people classified into the same or adjacent and opposite quartiles was 75% and 5%, respectively.Conclusion: The developed
FFQ in this study has a relatively good
validity and reproducibility, and can be used in community-based studies for evaluating and
screening of food consumption.