VALIDITY AND REPRODUCTIBILITY OF A FOOD FREQUENCY QUESTIONNAIRE DESIGNED TO CHARACTERIZE THE DIET OF WOMEN OF CHILDBEARING AGE

Abstract

Objectives: To test the validity and reproducibility of a food frequency questionnaire (FFQ) administered by interview, consisting of 127 items, developed to characterize the food intake of women of fertile age, in comparison with the method of the 24 hours diet recall (24 HDR). Design: The FFQ assessment study was conducted on 47 women aged between 19 and 43 years old. The food frequency questionnaire was administered twice to the same subject at a period of three months. To test the validity, the diet was assessed by three 24 HDR, two applied at the same time with the FFQ and a third one a month after the first test. The reproducibility was tested by comparing the average nutrient intake and the food consumption in grams of the food divided into food groups of the two FFQ. The validity was assessed by comparing the average nutrient intake and the food consumption in grams of the food divided into food groups of the two food frequency questionnaires, with the average of the nutrient input of the three 24 HDR ones. Results: The reproducibility was assessed using the Pearson correlation, the correlation coefficients range between 0.615 for cholesterol and 0.863 for the energy application. The strongest connection was recorded for fruit (0.960) and the lowest for soups (0.637). The Pearson correlation coefficients to test the validity ranged between 0,051 for cholesterol and 0.829 for the energy application. In testing the validity by comparing the food groups the lowest correlation is recorded for vegetables (0.35) and the highest for beverages (0.72). Conclusions: The results indicate a favourable reproducibility and validity of the FFQ developed.