Comparing intake estimations based on food composition data with chemical analysis in Malian women

• Published in: Public health nutrition Vol. 20; no. 8; pp. 1351 - 1361
• Main Authors: Koréissi-Dembélé, Yara, Doets, Esmee L, Fanou-Fogny, Nadia, Hulshof, Paul JM, Moretti, Diego, Brouwer, Inge D
• Format: Journal Article
• Language: English
• Published: Cambridge, UK Cambridge University Press 01.06.2017
• Subjects: Adolescent; Adult; Analytical chemistry; Assessment and Methodology; Black People; Body Mass Index; Carbohydrates; Carotenoids; Chemical analysis; Chemical analysis of duplicate portion; Correlation analysis; Correlation coefficient; Correlation coefficients; Data processing; Diet; Diet Records; Dietary Carbohydrates - administration & dosage; Dietary Fats - administration & dosage; Dietary fiber; Dietary Fiber - administration & dosage; Dietary Proteins - administration & dosage; Digitaria exilis; Documentation; Estimation; Female; Food; Food Analysis; Food composition; Food composition database; Food consumption; Food intake; Food weighed record; Health care; Humans; Iron; Laboratories; Mali; Micronutrients - administration & dosage; Middle Aged; Nutrient intake of Malian women; Nutrients; Nutrition; Nutrition Assessment; Nutrition Surveys; Pregnancy; Quality; Recipes; Reproduction (copying); Research Papers; Validation; Vitamin A; Vitamin A - administration & dosage; Young Adult; Mali; Validation; Food weighed record; Nutrient intake of Malian women; Food composition database; Chemical analysis of duplicate portion
• ISSN: 1368-9800; 1475-2727
• DOI: 10.1017/S1368980017000015

Food composition databases are essential for estimating nutrient intakes in food consumption surveys. The present study aimed to evaluate the Mali food composition database (TACAM) for assessing intakes of energy and selected nutrients at population level. Weighed food records and duplicate portions of all foods consumed during one day were collected. Intakes of energy, protein, fat, available carbohydrates, dietary fibre, Ca, Fe, Zn and vitamin A were assessed by: (i) estimating the nutrient intake from weighed food records based on an adjusted TACAM (a-TACAM); and (ii) chemical analysis of the duplicate portions. Agreement between the two methods was determined using the Wilcoxon signed-rank test and Bland-Altman plots. Bamako, Mali. Apparently healthy non-pregnant, non-lactating women (n 36) aged 15-36 years. Correlation coefficients between estimated and analysed values ranged from 0·38 to 0·61. At population level, mean estimated and analysed nutrient intakes differed significantly for carbohydrates (203·0 v. 243·5 g/d), Fe (9·9 v. 22·8 mg/d) and vitamin A (356 v. 246 µg retinol activity equivalents). At individual level, all estimated and analysed nutrient intakes differed significantly; the differences tended to increase with higher intakes. The a-TACAM is sufficiently acceptable for measuring average intakes of macronutrients, Ca and Zn at population level in low-intake populations, but not for carbohydrate, vitamin A and Fe intakes, and nutrient densities.