BACKGROUND: A self-administered semi-quantitative food frequency questionnaire with 65 food lists (FFQW65) was developed for the nutritional education of patients of diabetes mellitus. METHODS: Relative validity and reproducibility of nutrient intakes as assessed by FFQW65 were investigated in volunteers of 71 males. To assess the relative validity, a one-week diet record was used as a reference. The FFQW65 was administered twice at an interval of 10 month in order to assess the reproducibility. Log transformed data was used to obtain Pearson correlation coefficients. RESULTS: As for the validity, Pearson correlation coefficient of total energy intake of breakfast was 0.74 and the relative difference between predicted value and diet record was relatively small (the difference was -5 kcal and relative difference rate was 1.1%). Compared with breakfast, those for lunch, supper and one-day total energy intakes were relatively lower (0.52, 0.66 and 0.64, respectively) and the relative differences were around 10%, respectively. The one-day intakes of sixteen nutrients (except for potassium and niacin) ranged from 0.28 to 0.64. Energy adjusted correlation coefficients for protein, iron and cholesterol were low (0.16, 0.14 and 0.19, respectively). Those for the other thirteen nutrients ranged from 0.26 to 0.65. As for the reproducibility, the second survey using FFQW65 was administered to 64 males. Pearson correlation coefficient of one-day total energy intake was 0.76 and those for the other nutrients ranged from 0.54 (vitamin D) to 0.81 (calcium). CONCLUSION: The results suggest that FFQW65 can be used for the evaluation of energy intake by usual meals for nutrition education of patients of diabetes mellitus.
BACKGROUND: A self-administered semi-quantitative food frequency questionnaire with 65 food lists (FFQW65) was developed for the nutritional education of patients of diabetes mellitus. METHODS: Relative validity and reproducibility of nutrient intakes as assessed by FFQW65 were investigated in volunteers of 71 males. To assess the relative validity, a one-week diet record was used as a reference. The FFQW65 was administered twice at an interval of 10 month in order to assess the reproducibility. Log transformed data was used to obtain Pearson correlation coefficients. RESULTS: As for the validity, Pearson correlation coefficient of total energy intake of breakfast was 0.74 and the relative difference between predicted value and diet record was relatively small (the difference was -5 kcal and relative difference rate was 1.1%). Compared with breakfast, those for lunch, supper and one-day total energy intakes were relatively lower (0.52, 0.66 and 0.64, respectively) and the relative differences were around 10%, respectively. The one-day intakes of sixteen nutrients (except for potassium and niacin) ranged from 0.28 to 0.64. Energy adjusted correlation coefficients for protein, iron and cholesterol were low (0.16, 0.14 and 0.19, respectively). Those for the other thirteen nutrients ranged from 0.26 to 0.65. As for the reproducibility, the second survey using FFQW65 was administered to 64 males. Pearson correlation coefficient of one-day total energy intake was 0.76 and those for the other nutrients ranged from 0.54 (vitamin D) to 0.81 (calcium). CONCLUSION: The results suggest that FFQW65 can be used for the evaluation of energy intake by usual meals for nutrition education of patients of diabetes mellitus.
Authors: Roberta Aguiar Sarmento; Bárbara Pelicioli Riboldi; Ticiana da Costa Rodrigues; Mirela Jobim de Azevedo; Jussara Carnevale de Almeida Journal: BMC Public Health Date: 2013-08-09 Impact factor: 3.295