Reproducibility And Validation Of A Food Frequency Questionnaire (FFQ) To Assess Dietary Intake Among Residents In Xinjiang,China

Objective: To evaluate the reliability and validity of the self-designed semi-quantitative FFQ(food frequency questionnaire)of Xinjiang,and to provide a simple and reliable tool for studying the correlation between eating habits and health of Xinjiang residents.Methods: Based on the multi-ethnic natural population cohort construction and health follow-up study project in Xinjiang,FFQ suitable for Xinjiang region was developed,and 111 food lists of 14 categories were determined.Volunteers living in different areas of Xinjiang were recruited from January 2022 to December 2022.Reliability analysis was conducted on the results of the FFQ1 and the FFQ2 of 124 subjects,and validity analysis was conducted on the mean value of the two FFQS and the results of the 24HRs(3-day 24-hour dietary recall).Spearman’s correlation coefficient,ICC(Intraclass correlation coefficients),triplet consistency and Bland-Altman statistical methods were used to verify its reliability and validity from the perspective of food group intake and nutrient intake,respectively.Results: A total of 124 participants were included in the FFQ reliability study.Spearman correlation coefficient between energy and nutrients in the two FFQS ranged from 0.28 to 084(median was 0.51),and ICC value ranged from 0.21 to 0.70(median was 0.47.Spearman correlation coefficients for food intake ranged from 0.29-0.60(median 0.48),and ICC values ranged from 0.27-0.58(median 0.47).A consistency test was conducted after tripartite grouping of food groups and nutrient intakes from FFQ1 and FFQ2,and it was found that less than 10% of subjects were assigned to different groups.The weighted Kappa coefficient ranges from0.21 to 0.67.A total of 125 participants were included in the FFQ validity study.Spearman correlation coefficients between energy and nutrient intake of FFQ and 24 HRs ranged from 0.21 to 0.78(median,0.28)and from 0.27 to 0.58(median,0.43)for food groups.After tripartiting 24 HRs and FFQ food groups and nutrient intake,less than 15%of subjects were divided into different groups.Weighted kappa values and Bland-Altman plots showed good consistency between the semi-quantitative food frequency questionnaire and 24 HRs.Conclusions:The semi-quantitative food frequency questionnaire in this study showed acceptable reliability and reasonable validity in evaluating the food categories and nutrient intake of Xinjiang population.The FFQ is an ideal dietary nutrition measurement tool to analyze the dietary intake status of Xinjiang residents and its relationship with diseases.