Comunicaciones Orales

CO279. HARMONIZATION OF THE SELECTION OF FOODS FOR PHOTO QUANTIFICATION OF PORTION SIZES IN DIETARY SURVEYS IN LATIN-AMERICA: LADIETA PROJECT REPORT

Sandra Patricia Crispim1, Regina Mara Fisberg2, Genevieve Nicolas3, Viktoria Knaze3, Silvia Bel-Serrat3, Tania Sanchez4, Dirce Maria Lobo Marchioni2, Lilia Susana Pedraza Zamora4, Josiane Steluti2, Aline Mendes Peralta2, Claudia Choma Bettega Almeida1, Marina Araujo5, Edna Yokoo6, Rosangela Alves Pereira7, Juan Rivera4, Nadia Slimani3.

1 Universidade Federal do Paraná, Curitiba, Brazil; 2 University of São Paulo, São Paulo, Brazil; 3 International Agency for Research On Cancer, Lyon, France; 4 National Institute of Public Health of Mexico, Cuernavaca, Mexico; 5 Fundação Oswaldo Cruz, Rio De Janeiro, Brazil; 6 Fluminense Federal University, Rio De Janeiro, Brazil; 7 Federal University of Rio de Janeiro, Rio De Janeiro, Brazil

The LA-DIETA Project is the Latin American (LA) branch of a broader international initiative (GloboDiet) aiming to pilot implementation of a standardised 24-hour recall methodology worldwide (GloboDiet software). For that purpose, there is the need to adapt its related picture book for estimating food portion amounts in LA. The objective is to report on the work conducted in parallel in Brazil (BR) and Mexico (MX), as pilot countries. The results of the selection procedure to deliver a list of foods that require photos for estimation of portion quantification in dietary surveys in both countries, having in consideration harmonized results in LA and worldwide. Data from adults participating in the latest national surveys were used to identify the most consumed foods (n=1090 BR; n=1239 MX). The number of foods selected for photos was based on variability analysis using stepwise regression to detect betweenperson variation for selected nutrients by region and sex. Foods with over 90% accumulated r-square for the different nutrients were selected. In addition, a top-list with 50 most often consumed foods was generated. Next, each one of the most consumed foods (n=272 BR; n=675 MX) were evaluated as for their inclusion in the album, following IARC guidelines (i.e. recommendations for quantifying a food with photos or other types of quantification). Very rarely reported foods (by <0.01% of the population) were excluded. Selected foods were then searched in the existing GloboDiet album and divided into two groups (existent and new photos to be created). For the existing photos, intake ranges were compared between national survey and GloboDiet photos. As main conclusion, 34 new photos will be developed in BR and 45 in Mexico. BR and MX will have 7 new photos in common and the final albums will comprise 91 and 66 photos, respectively.