The over-all strategy given by the joint FAO/WHO consultation 1995 was used as the starting point, i.e. target foods or food patterns for public health nutrition programmes should be identified from an analysis of prevailing food and nutrient intakes. Prevailing data for food and nutrient intakes in 14 EU countries were examined and different principles and options for the derivation of FBDG were explored. Methodological issues and their influence on the interpretation of data for the development of FBDG were also examined.

The process from nutrients to foods can be briefly: 1) identification of major food sources of the nutrient of interest, 2) identification of foods contributing substantially to population intakes, 3) identification of foods or food patterns compatible with desirable nutrient intakes or explaining variations in nutrient intakes, 4) formulation of FBDG into foods, portion sizes, frequency of intake, meal composition taking attainability and acceptability as well as compatibility of co-existing guidelines into account.

The level of complexity that can be applied in the analytical approaches depends on the characteristics of available intake data. A detailed analysis requires data on an individual level for nutrients, foods, food patterns, eating and meal habits etc. When individual data are available different analytical approaches (examination of distribution of intakes, correlation analysis between foods and nutrients, examination of food intakes in compliers/non-compliers to nutrient goals, discriminant analysis, cluster/factor analysis) can be used to identify key foods or food patterns fulfilling nutrient goals.

The examination of prevailing food and nutrient intake data in the EU revealed:

– a number of methodological differences in approaches to dietary surveys exist in the EU countries e.g. regarding mediods used, selection of population, classification of foods, which have to be kept in mind in pan-EU comparisons at present there is a substantial gap between actual intakes and present nutrient goals suggesting that major changes of dietary habits are needed

– while some food patterns were consistently related to intake of specific nutrients in most EU countries, other patterns showed large variations between countries

– methodological issues, such as survey duration, survey techniques, under- or over-reporting, could have substantial influence on the identification of target foods or food patterns.

A science-based analysis of nutrient and food intakes allows development of FBDG, which, if implemented, are likely to result in mean population intakes closer to nutrient goals. Acknowledging the social and cultural differences within the EU as well as the need to focus on the most relevant public health problem in die population, FBDG should first be developed widiin member states. Harmonisation of survey methods within the EU would facilitate development of regional and EU FBDG.