Comparison of national food supply, household food availability and individual food consumption data in Portugal

Abstract

The present study aimed to compare Portuguese food and nutrient data retrieved from three distinct sources of dietary data: national food supply, household food availability and individual food consumption. In particular it was intended to validate the use of the dietary data derived from the Portuguese household budget surveys (HBS). The data sources used were food balance sheets (FBS) (1990, 1995 and 2000) generated by the Food and Agriculture Organization (FAO), HBS (1990HBS ( , 1995HBS ( and 2000 collected by the National Statistics Institute (INE) and individual consumption obtained through a food frequency questionnaire (FFQ) applied in the context of the EpiPorto study circa 2000. Thirteen comparable food groups were established for analysis. Total energy, macronutrients and other selected dietary components were considered. The average daily values of different dietary sources were compared in absolute amounts (g/person per day) and in relative energy contribution (%E). To evaluate association and agreement between sources, Spearman's correlation coefficients and distance matrices based on Euclidian distances were computed. The obtained average daily per capita amounts reflect the different points of data collection in the food chain; HBS fall in the expected position, below FBS and closer to FFQ. Agreement between FBS, HBS and FFQ was higher when data were expressed in relative energy contribution than when absolute amounts were considered. In general, the observed discrepancies between the three data sources could be attributed to known differences in the methodological approaches. This study supports the finding that despite their constraints the HBS-derived data are a valid option to evaluate dietary habits in Portugal.

Are the eating habits of university students different to the rest of the Spanish population? Food availability, consumption and cost

“…Our results are similar to those of previous studies by Rodríguez et al (1996), SerraMajem et al (2003), Rodrigues et al (2007) and Naska et al (2009), for all of the food groups with the exception of beverages (nondairy). For the category of beverages (nondairy), for both the HBS and the university students in Castilla-La Mancha, consumption levels are higher than availability, where the difference is more significant for the group of young university students.…”

“…In any case, knowledge of the eating habits of the population is the starting point for designing public health policies to intervene and monitor diets can help to explain these differences like structure according to sex, the geographic area where the surveys were carried out (let us keep in mind that the UCLMS was carried out on students from Albacete city), and educational level. These variables are just some of those mentioned by Rodrigues et al (2007). It should also be remembered that consumer preferences and habits influence food consumption (Grunert, 2005;Honkanen et al, 2005;Verbeke et al, 2010).…”

“…Studies based on food surveys show the eating models of a group of countries (Lagiou & Trichopoulou, 2001;Serra-Majem et al, 2003;Naska et al, 2009), those of a particular country (Lagiou & Trichopoulou, 2001;Rodrigues et al, 2007) or the worldwide tendencies of food consumption (Kearney, 2010).…”

“…grains); fruits; legumes, vegetables and greens; meat; fish and seafood; oils and fats; dairy products; eggs; pastries, sugar and honey; and beverages (nondairy). These groups were based on the mostused food groups reported in previous nutritional literature (Rodrigues et al, 2007;Cervera et al, 2014;García-Meseguer et al, 2014). The characteristics of the sources of data as well as the particularities of each food group, depending on the source of information used, are explained in detail below.…”

The seafood market in Portugal: Driving forces and consequences

“…When interpreting consumption data, different data sources have different methodological limitations that need to be considered. Data from both FBS and individual surveys are not directly comparable but can complement each other [30]. There are limitations related to the statistical sources and it is difficult to correctly assess values of per capita consumption by type of product [31].…”