Contents - Previous - Next

This is the old United Nations University website. Visit the new site at



Aree Valyncevi


ASIAFOODS was established in 1984. The first conference was held in Bangkok, with participants representing 12 Asian countries. The current status of food composition tables in the various countries was presented, and future activities were identified. Technical and financial resources both within Asia and from the outside were identified as necessary to carry out those tasks.

Asia is the world's largest continent, with a population of more than two thousand million. Geographically, it can be divided into three subregions- southern, south-eastern, and eastern-with a great diversity in socio-economic development, culture, and life-style, including categories of food consumption and habits. Therefore, it would be desirable for each subregion to develop its own activities as long as information is shared between them. It is expected that the Federation of Asian Nutrition Societies (FANS) will help to facilitate this information transfer.


The ASEAN Food Data Network

Through the ASEAN Sub-committee on Protein, the Australian government has supported food-habits research and development since 1985. Because it was recognized that data on food composition in ASEAN countries are vital to the success to this research, the ASEAN Food Data Network was established by the Sub-committee on Protein in 1986. Its goals and objectives are in line with those of ASIAFOODS and INFOODS. The plan is to explore the problems and needs for improving the status of food composition data, and to develop and upgrade data production and availability within and outside the ASEAN region. Particular attention will be given to nutrients related to health.

The first ASEAN workshop on food composition, held in Indonesia in October 1986, identified problems and specific needs as well as the current status of food composition tables in ASEAN countries. The problems identified were as follows:

  1. Proper sampling procedures are lacking.
  2. Analytical methods are being used without collaborative national and regional precision testing.
  3. No quality-assurance programme has as yet been established.
  4. National food composition tables are inadequate in terms of food items, nutrient constitutents, appropriate format, and background information.

There is a great need for collaborative studies within and among the ASEAN countries to establish a scientific base for the standardization of food sampling procedures and analytical methods. In addition, quality-assurance programmes must be developed at the national and regional levels. Finally, we must establish a systematic method for compiling and using food composition data. The ultimate goal of the Food Data Network is to develop national and regional food composition tables of high quality, in adequate quantity, that will be readily accessible to users in the ASEAN region.


Recommendations for activities

At the national level, the current status of food composition tables must be documented. One method is to organize a national food data network. Another is to convene workshops or task forces to review existing methods for sampling and analyses, the adequacy of existing national food composition tables, and the format of existing tables in terms of appropriateness to specific users, adequacy of background information, and accuracy and precision of the data.

Current practices should be improved and guidelines developed to improve sampling procedures. For immediate needs, the INFOODS manual Guidelines for the Production, Management, and Use of Food Composition Data Systems, being prepared by Greenfield and Southgate, can be followed. For the future, studies should be carried out in each country on key food items within food groups to determine more precisely the sources and extent of variability in nutrient content. This should result in a model sampling scheme for the foods concerned.

Sampling procedures need to be documented. For example, to enhance the value of the data, it was recommended that all laboratories should document full sampling procedures for each food prior to analysis, using the format given in the INFOODS manual.

Each member country needs to be able to determine the accuracy of data that are gathered by all methods used in each laboratory. Quality-control practices in each laboratory must be documented. Finally, inter-laboratory tests should be conducted to compare proficiency between laboratories.

To improve food composition data, each member country should prepare an inventory of the individuals who compile the information, and review the methods they use. Seminars and workshops could be held to discuss and design guidelines and systems for compiling the data and to consider establishing national data-compilation centres.

At the regional level, it was recommended that the network co-ordinator should organize an inter laboratory test programme to assess existing methods and provisions for quality assurance. A continuing programme would monitor this through recovery data and control charts sent periodically to the coordinator, and outlying laboratories would be encouraged to improve their methods. In addition, food reference materials would be developed and used to further improve the quality of data.

A programme to promote the exchange of personal contacts and information between laboratories through a co-ordinating office, national and regional workshops, and/or publications was encouraged. Finally, it was suggested that the co-ordinator should assist in organizing training programmes in the areas of food-laboratory management, data generation and compilation, and the dissemination of food composition data in ASEAN countries, Australia, or other countries of specialists in the particular areas.


Activities in member countries

During 1986-1989, each member country carried out these tasks. The co-ordinator of the Food Data Network organized the regional collaborative precision testing of the existing methods of food analysis used among ASEAN countries as recommended. There are many activities still to be completed.


Second ASEAN workshop on food data systems

During the past three years limited activities at the national level and inter-laboratory testing between ASEAN countries have been carried out. A second workshop is needed, however, to strengthen cooperation and acquire information on current conditions within each member country, and plans are to hold it in Bangkok in October 1990. Specifically, it is intended to review the activities on food composition data systems, develop guidelines for systematic development of the data, and formulate a programme of collaboration for activities for the next three years.

Three participants from each member country who are actively involved in the food data system will attend the workshop. In addition, resource persons will be invited to address a specific relevant issue.


The contribution of EUROFOODS to the improvement and dissemination


Clive E. West


EUROFOODS was formed by a group of nutritionists and food scientists after a meeting held in Wageningen, Netherlands, in May 1983 [1]. It has two aims: to improve the quality of food composition data, and to increase the accessibility of nutrient data. More than 20 countries in Europe, both east and west, have been involved in its work.



The three major areas of achievement of the organization have been in improving the quality of food composition data, the dissemination of food composition data, and the process of networking and bringing people together.


The quality of food composition data

Food analyses

It was realized quite early that differences in procedures between laboratories, especially in different countries, may be an important source of discrepancies between nutrient values in food tables and nutrient databanks. Thus, the EUROFOODS inter laboratory trial was carried out in 1985, with 20 leading laboratories in Europe and the United States participating. Each laboratory received a well homogenized, dried sample of six foods and was requested to perform analyses of dried weight by a prescribed vacuum method, and of protein, fat, available carbohydrates, total dietary fibre, and ash by its own routine methods. The results were far from satisfactory [2].

The between-laboratory coefficient of variation (CV) for dietary fibre, for example, ranged from 23% for French beans to 84% for biscuits. This large range could be explained by the use of different methods. The CV for protein ranged from 2.8% for egg to 6.4% for wheat and rye, but recalculation of these values using uniform Kjeldahl factors reduced the figures to 2.7%, 4.7%, and 5.2% respectively. It was concluded that leading laboratories in different countries may produce widely different values for proximate constituents (macronutrients) in common foods.

To improve the quality of nutrient data, better standardization of methods is required. As an initial step, it was decided that reference materials of certified nutrient content should be produced and made widely available. The organization of this task has been taken over by the Community Bureau of Reference of the Commission of the European Communities.


Missing values

The usefulness of any food composition table or nutrient database depends on the quality of the information used to construct it. In general, original analytical data provide information of the highest quality. In most countries, however, it is not feasible to construct tables with only these. Priorities for analytical work have to be drawn up, and the greatest effort usually is directed toward analysing foods that make a major contribution to the national diet. For specific individuals or groups, other foods may be of greater importance. Therefore, in order to convert food intake to nutrient intake, it is necessary to estimate nutrient content, rather than have no data.

A project was set up with the following aims: (1) to classify the nutrient values in food composition tables and nutrient databanks in Europe, so as to identify which values are missing, and to obtain information on how the problem of missing data is being handled; (2) to bring about the exchange of data to assist in filling in the missing values; (3) to make proposals for priorities for the analysis of nutrients in foods; and (4) to establish international guidelines for estimating missing values where no data are available.

In the first phase a survey was carried out to examine missing values on the nutrient content of 50 foods commonly consumed in Europe. The results [3] gave a general impression of the quality of the data in the various European food composition tables examined. The quality, as expressed by the proportion of original analytical data, varied between tables and also within tables with respect to various nutrients. The second phase of the study, involving 50 of the most important foods in each country apart from the 50 basic items, provided further information on the quality of data in the various tables. It also enabled draft guidelines to be drawn up, and these were discussed at the third EUROFOODS meeting, in Warsaw in May 1987.


Nutrient losses and gains during food preparation

The nutrient content of foods per unit weight changes when the foods are prepared. Such losses and gains can be classified in two ways. The first is related to yield when the primary ingredients at the pre-cooking stage are compared with the prepared food at the cooking stage and as consumed at the post-cooking stage. The second is related to changes in the amount of specific nutrients when foods are prepared.

In a perfect world, original analytical data would be available for foods at all stages of preparation; in reality, however, it is often necessary to make estimates. In addition to assessing changes in nutrient content when foods are prepared, it is possible to calculate the nutrient content of foods prepared from raw ingredients. Data are lacking for many foods consumed in Europe, however, and a five-pronged project has been implemented to improve the situation.

The first subproject is a description and inventory of the methods used to calculate the nutrient composition of prepared dishes. The second involves a comparison of these results. The last three involve an inventory of research being carried out in Europe, a bibliography on the subject, and a database to include this information.


The dissemination of food composition data


To let people in Europe know what information is available, an inventory of food composition tables and nutrient databases has been established. This was published [4] and distributed at EUROFOODS meetings. The latest edition was distributed at the Uppsala meeting in draft form and will be published in the proceedings of the meeting.


Development of the EUROCODE

Since no two foods are identical and very few can be considered to be similar in most respects, a system is necessary by which they can be grouped together on the basis of their similarities. After meetings in Luxembourg in December 1984 and in Heidelberg in February 1985, a coding system based on 14 major food groups was agreed on by representatives from 17 countries. The 14 groups are divided into 2,500 subcategories, and the system developed has been revised five times. The result is EUROCODE 2, which is part of a food coding and descriptor system [5].


European nub/ant databases

In February 1985 the Commission of the European Communities (DG XIIIB) awarded a contract to enable EUROFOODS to study the feasibility and methodology of developing an easily accessible database on food consumption derived from tables and databases currently existing in various European countries. The construction of such a merged database was found to be technically feasible. The problems encountered were tedious but not difficult to overcome. They were mostly related to language and terminology, and proposals for resolving them included the use of the EUROCODE referred to above. There are also questions related to copyright, but these are not insurmountable.

On the basis of the experience obtained, a proposal was developed in conjunction with the Agrofoods programme of the Commission of the European Communities at a meeting held in Brussels in June 1988 for the establishment of a series of nutrient databases for specific purposes at the national, household, and individual levels.


National food balance sheets

To be able to convert information from national food balance sheets on food supply to nutrient supply, it is necessary to have an improved nutrient database, because the present data are inadequate. Thus it is proposed to prepare an appropriate, specifically tailored database on about 300 food items for use at this level.


Household food consumption

In 1987 a workshop was held in Athens to review methods of producing data on household food consumption in the European Community, and it was recommended that a compositional database should be constructed for use at this level. It is proposed to use existing national food composition data and food consumption statistics to construct a European database.


Food consumption at the individual lewd

A number of current and planned studies depend on the use of compatible databases across Europe. The present proposal is a development of the EUROFOODS study on the feasibility of merging databases and the development of compatibility and interchangeability between national databases. It is proposed to resolve the remaining issues of comparability of the different national data sets and to use the interchange proposals from INFOODS to establish compatibility between the national databases. This approach will provide a cost-effective means of using national data across Europe in epidemiologic studies, for example.



Perhaps the most significant achievement of EUROFOODS has been that it has increased the awareness of people in Europe and elsewhere of the problems of the inadequacy and inaccessibility of data on the nutrient composition of foods. This has been done in a number of ways, the most obvious through holding ten meetings between 1983 and 1989.

At these meetings, many people came together and planned activities related to the interests of EUROFOODS but outside its orbit of influence. It really does not matter who organizes the activities as long as the work is carried out. Many people within EUROFOODS have also participated in the activities of INFOODS, which has been responsible for drawing up a series of guidelines aimed at improving the quality and accessibility of nutrient data throughout the world.


The Uppsala meeting: Update

A successful meeting attended by about 100 people from more than 20 countries was held in Uppsala, Sweden, 31 May-3 June 1989. Highlights of some of the workshops held during this meeting are discussed below.


Experiences with EUROCODE 2 were discussed, as well as proposals for revising the code and the procedures to be adopted when making such changes.


Databases for non-nutrients

There was a general discussion on the use and problems involved in establishing databases for non nutrients. No specific suggestions for further action were made.


Reference standards and inter-laboratory control

A survey of current projects on reference standards, such as those being organized by the Community Bureau of Reference of the Commission of the European Communities in conjunction with the State Institute for Quality Control of Agricultural Products in the Netherlands, was presented. Problems associated with inter-laboratory control were discussed, and participation in inter-laboratory programmes was encouraged.


Software for portable computers

A distinction was made between programs developed for professional use and those for use by the general public, and the need to direct attention mainly to the former was recognized. More information is required on programs available and their capabilities. Better naming of foods in nutrient databases is necessary since this is a problem for the developers of software.


Dietary methods in epidemiologic research

Three papers were presented at the workshop. One was a review of epidemiologic studies on diet and colorectal cancer. The second suggested that errors in estimating the frequency with which foods are eaten is more important that those in estimating portion sizes. Experience with the five-day weighed record was the topic of the third paper. Sources of errors were associated with omitting foods, day-to-day variation, changes in diet, coding mistakes, and errors in converting foods to nutrients. It was concluded that simplified dietary methods are required for use in epidemiologic studies.


Meal classification in dietary surveys

An attempt was made in Uppsala to establish a system for classifying meals. The developers were faced with the problem of classifying meals of people who did not conform to normal dietary patterns, such as air hostesses and forestry workers working shifts. The system that was developed was based on the number of groups (protein-rich, carbohydrate-rich, and fruits/ vegetables) and the degree of preparation required by the person or family consuming the food. However, it was agreed that the system was too complicated and required further work.


Composition data for industrially manufactured products

It is difficult to obtain information from manufacturers: qualitative information on ingredients is available, but quantitative data are difficult to obtain. Industry has difficulty dealing with the variability in products, and this contributes to the reluctance to provide data. Also, such information may be used by regulators against the interests of companies. An attempt should be made to include industrial nutrient databases in the inventory of food composition tables and nutrient databanks prepared by EUROFOODS.


Future EUROFOODS activities

Major meetings

European meetings have been held every two years since 1983, and it is envisaged that this practice will continue in the future. The next one is planned for Athens in May 1991. Offers to hold the subsequent meetings have been received from Barcelona, Spain, and Bratislava, Czechoslovakia. These meetings make a major contribution to building up the EUROFOODS network, which has assisted in the development of the quality and accessibility of nutrient data in many countries in Europe.

Working groups

Since its inception, much of the work of EUROFOODS has been carried out by different groups: the computer group, the group on nutrient losses and gains in the preparation of foods and missing values in food composition tables, the group on the inventory of food composition tables, the food analysis! reference materials group, the tourist food tables group, and so on. Many have been successful, and it is proposed that their work should be completed and written up for publication before the end of 1989.

Major projects

Major projects must be funded by outside agencies. It has proved impossible to fund major projects through national governments and other country-based organizations. Thus EUROFOODS must depend on funds from bodies such as the Commission of the European Communities.

EUROFOODS was successful in obtaining an initial grant that enabled a European nutrient databank to be demonstrated. In fact, this small grant acted as a catalyst for many more activities than those financed by the Commission. Since that time, funds have been made available by DGVI (Agriculture) to make possible a workshop to discuss the establishment of a series of three European nutrient databases to convert food intake into nutrient intake, for national food disappearance data, for household budget surveys, and for food intake at the individual level. In the coming months, efforts will be continued to obtain a contract to facilitate this work.

It was agreed at Uppsala that EUROFOODS did not have to expend any effort on the reference materials for food analysis, as this work was being carried out by the Community Bureau of Reference of the Commission of the European Communities (CEC) in conjunction with the State Institute for Quality Control of Agricultural Products. Special projects have been selected for particular attention. The nutrient database project is described above, and it is envisaged that it will be funded by CEC.

It was agreed that insufficient information existed on software for accessing nutrient databases. Most people find it difficult to know what software is available and which would fulfil their specific requirements. It was decided to establish a committee to contact Ms. Sallie Bassham of the University of Salford, who has prepared an inventory of database-access software.

Nutrient databases are now reaching a high level of sophistication, whereas the development of databases for non-nutrients is still in its infancy. Little is known about the databases that exist, and the problems involved in establishing them are poorly understood. Thus it was proposed that a committee should be established to address these matters.



  1. West CE, ed. EUROFOODS: towards compatibility of nutrient data banks in Europe. Ann Nutr Metab 1985;29(suppl 1):1-72.
  2. Hollman PCH, Katan MB. Bias and error in the determination of common macronutrients in foods: an inter laboratory trial. J Am Diet Assoc 1988;88:556-63.
  3. Meyer B, Van Oosten-Van der Goes HJC, van Staveren WA, West CE. Missing values in European food composition tables and nutrient data bases: preliminary results of a survey. Food Sci Nutr 1988;42F:29-34.
  4. EUROFOODS. Review of food composition tables and nutrient data banks in Europe. Ann Nutr Metab 1985;29(suppl 1):11-45.
  5. Arab L, Wittler M, Schlettler G. European food composition tables Heidelberg: Springer-Verlag, 1987: viii.

Contents - Previous - Next