Classificação de alimentos para a estruturação de bases de dados alimentares no Brasil

Abstract

The standardization of food classification and consequent construction of harmonized food database helps to improve the characterization of the food profile, allowing the refinement of results from studies. This thesis aims to describe the procedures and elements for the structuring of a classification system of food items into groups in a harmonized way, using informations about the availability of food in the Brazilian households of the 2008-2009 Household Budget Survey collected between May 2008 and May 2009. In the secondary data corresponding to information of 55,790 households, 1716 food items were identified, obtained through the evaluation of the household food acquisition of seven consecutive days. The harmonization procedures were based on the Data Food Networking (DAFNE) classification, which consists in a joint initiative, including several European countries. For the construction of harmonized food groups and food allocation in groups, guidelines by the United Nations, the International and National Food Classification System (Codex Alimentarius and ANVISA), informations about the classification of foods according to the level of processing adopted by the Food Guide for the Brazilian Population and an experts group were also consulted. All these criteria were considered to preserve the classification of the food according to nutritional characteristics, a preponderant factor to study and understand the contribution of the food to the state of individuals health. Food items were allocated into 14 groups and 42 subgroups, according to the proposal of the harmonization study. The per capita household acquisition averages were presented in kg/per capita/year, according to the harmonized food groups. The proposed harmonization system allowed the allocation of food in groups, systematizing the identification of food items and the sustainable structuring of groups. Our study suggests that the use of harmonized food databases is an accessible and low cost methodology that facilitates the analysis of consumption profiles, allowing the correlation of these profiles with socioeconomic characteristics, as well as comparisons between different studies, including international level. These data can be very useful in linking food consumption with health or disease outcomes, and to assist in the development of health and nutrition policies