Dietary intake data Food matching

1 downloads 4 Views 4MB Size Report
1 International Life Science Institute (ILSI) – Argentina; 2 Instituto Pensi - Hospital Infantil Sabara – Brazil; 3 Departamento de Pediatría – UNIFESP – Brazil; ...





* E-mail:[email protected] [email protected]

Irina Kovalskys1*, Mauro Fisberg2,3, Georgina Gómez Salas4, Attilio Rigotti5, Lilia Yadira Cortés Sanabria6, Martha Cecilia Yepez Garcia7, Rossina Gabriela Pareja Torres8, Marianella Herrera9, Viviana Guajardo1, Ioná Z. Zimberg10, Natasha Aparecida Grande de França11, Maria Paz Amigo1, Guadalupe Echeverria5 on behalf of the ELANS Study

1 International Life Science Institute (ILSI) – Argentina; 2 Instituto Pensi - Hospital Infantil Sabara – Brazil; 3 Departamento de Pediatría – UNIFESP – Brazil; 4 Escuela de Medicina – Universidad de Costa Rica; 5 Centro de Nutrición Molecular y Enfermedades Crónicas – Pontificia Universidad Católica – Chile; 6 Departamento de Nutrición y Bioquímica – Pontificia Universidad Javeriana – Colombia; 7 Universidad San Francisco de Quito – Ecuador; 8 Instituto de Investigación Nutricional – Peru; 9 Centro de Estudos do Desenvolvimento da Universidade Central da Venezuela – Venezuela; 10 Departamento de Psicobiología – UNIFESP – Brazil; 11 Departamento de Nutrição – USP – São Paulo, Brazil


Dietary intake data

Between-country comparisons are particularly prone to error when different food composition tables are used to estimate dietary

intake. In this sense, the objective of this study is to describe our

procedures and rationale for the selection and adaptation of food

composition from a single database during pilot study of ELANS to make cross-country nutritional intake comparisons.

A standard study protocol was design to investigate nutritional intake in all countries enrolled:

Two 24-hour recalls using Multiple Pass Method

Data from the 24-HR were entered into the Nutrition Data System

for Research program (NDS-R, version 2013) after a harmonization process between local foods and NDS-R database.

Food matching


Venezuela Colombia

Costa Rica Ecuador



A standardized procedure involving a nutritional equivalency of local food items reported by the study subjects of each country to foods

available in NDS-R database was strictly conducted by professional nutritionists in each the country.

Identification of foods commonly consumed in each country ELANS is a household-based multicenter cross-sectional

According to national food consumption data available and pilot study

study of a representative

sample of 8 Latin American

Chile Argentina


countries (Argentina, Brazil,

Chile, Colombia, Costa Rica,

Ecuador, Peru and Venezuela). Figure 1. ELANS countries

A standard study protocol was designed to investigate dietary intake of 9,000 subjects enrolled aged 15.0-65.0 years, stratified by geographi-

cal location (only urban areas), gender, age and socioeconomic status.

Each site-specific protocol was approved by the ethical review boards at the participating institutions.

A total of 4,730 standardizations, considering food items and recipes, were made in ELANS.

A mean of 407 food items and 185 recipes were standardized in each country.

Table 1. Standardization of Food Composition Database in pilot study

Country Argentina Brazil Chile Costa Rica Colombia Ecuador Peru Venezuela

Number of standardizations Food Recipes Total 638 195 833 666 495 1161 130 31 161 512 235 747 145 65 210 220 130 350 652 281 933 291 44 335

Food matching

Development of recipe files Broken down into ingredients and entered into the software

Concordance rate (%) 80-120 80-120 90-110 90-110 85-105 68-105 85-115 80-120


Nutritional equivalency of local food items (food, beverages) to foods available in NDS-R database; concordance rate of 80-120%

Standardization at the food and nutrient levels will likely

minimize systematic and random errors in nutrient intake estimations and allow comparisons between these Latin American countries.

Documentation of food matching

This is an important initiative for harmonization of dietary

According to a common format in excel

manner in different populations and could therefore gen-

The ELANS protocol was approved by the Western IRB

(#20140605) and registered at Clinical Trials (#NCT 02226627).

A pre-test was conducted with 400 subjects (50 cases in each country).

assessment that could be applied in a standardized

erate comparable dietary data in multicenter epidemioFigure 2. Food matching standardized procedure in ELANS


logical studies.

This research is funded by The Coca-Cola Company

Suggest Documents