Cover Image
close this bookFood, Nutrition and Agriculture - 12 - Food Composition Data (FAO - FPND - FAO, 1994)
close this folderImproving food composition data through training
View the document(introduction...)
View the documentAmelioration des données relatives à la composition des aliments grace à la formation
View the documentUntitled


J. Castenmiller and C.E. West

Jacqueline Castenmiller is Research Associate and Clive E. West is Associate Professor at Wageningen Agricultural University, Department of Human Nutrition, Wageningen, the Netherlands.

Information on the composition of foods is essential for any nutrition programme, whether the diets concerned are for healthy individuals or for those suffering from illness. In addition, there is a growing demand for high-quality and compatible food composition data to meet the needs of research on the relationship between diet and disease, Because of the heightened interest in the nutritional quality of foods, exporting countries and the food industry find that data on the nutritional composition of foods are essential. Thus, food composition data must meet the needs of government agencies, nutrition scientists, health and agriculture professionals, policy-makers and planners, food producers, processors and retailers, and consumers.

In developed countries considerable effort has been expended in generating and making available data on the composition of foods. However, in developing countries the data are often outdated and inadequate.

Although the primary objective of a nutrient database or food composition table is to provide data on the nutritional value of foods, it is also important to have information on the content of constituents that influence the bioavailability of specific nutrients, Data are needed to estimate the edible portion of foods. In terms of prepared foods, the yield and retention factors should be measured in a uniform way. There is a paucity of data on some nutrients, such as vitamin K, Finally, there is a paucity of data on “non-nutrients”, such as flavonoids, which have antioxidant activity.


To meet the goals of the International Conference on Nutrition (held in Rome in December 1992), it is essential that a programme of action to produce and disseminate data on the composition of foods be initiated, especially for developing countries and eastern Europe, As is often the case, resources and knowledge are two main limitations to progress in the area of providing sufficient data of acceptable quality.

Perhaps the principal barrier to the production and compilation of food composition data is the lack of knowledge and interest among those who are responsible for policy and the allocation of resources, It is very important that these decision-makers be aware of the need for this information. The programme of action should include short sessions on the importance of high-quality food composition data, which could be organized during regional meetings of ministers of agriculture and of health.

Another aspect of the programme of action would be the establishment of a network of regional centres for analysing foods, These institutions could act as reference and training centres for the region. To support the regional centres, backup centres, which could assist with training, quality control and the development and testing of appropriate methods, should be established in industrialized countries, The regional centres could assist in the collection, evaluation and dissnation of food composition data.


Training is necessary for those generating, compiling or using data on food composition. A systematic approach to the collection of information required for the creation of a comprehensive database and the construction of new databases is necessary to ensure compatibility with other databases and systems, Furthermore, the production of sound nutritional data depends on an integrated series of activities involving the users of the data, the analysts who generate data and the compilers of the database.

The effective use of food composition tables and nutrient databases requires a level of training and expertise that depends on the sophistication of the database or tables concerned. Even simplified food tables designed for lay use require some background knowledge of weights and measures and of terms such as “kilojoules” and “energy”. More sophisticated databases require an understanding of modes of expression, food descriptors and concepts such as “edible portion”, A professional nutritionist or dietician must become familiar with the principles of sampling, analytical methodology and data management and must be aware of common mistakes that can arise in database usage. The professional user also requires training in the evaluation of databases for specialized uses such as research projects.

General training

Often food analysts do not understand how their data will be used and for what purpose, Compilers of food composition tables are not familiar with the technical problems involved in food analysis and cannot adequately assess the quality and reliability of data.

It is essential that those involved in the process of creating databases be trained to be aware of the needs of the users, as they are responsible for ensuring that the database meets users’ requirements. Users expect all data that they require for specific purposes to be available. In turn, they must be aware of the problems encountered by those generating and compiling food composition data.

Even the best food composition table or database is of little value if it is used incorrectly, The compilers must define for the user the limitations of the database so that the data can be used appropriately, Those who train the users, and the users themselves, are ultimately responsible for correct use of the data.

To meet this training need, a postgraduate course on the production and use of food composition data in nutrition is held in Wageningen every two years, The first of these courses was organized in October 1992 by the FLAIR Eurofoods-Enfant Project (see Box), The second course (FoodComp ‘94), held in October 1994, was organized by the Graduate School for Advanced Studies in Nutrition, Food Technology, Agrobiotechnology and Health Sciences (VLAG) of the Wageningen Agricultural University, in cooperation with the United Nations University, FAO and the International Union of Nutritional Sciences.

The course is based on guidelines developed initially under the auspices of INFOODS and later by the FLAIR Eurofoods-Enfant Project (Greenfield and Southgate, 1992), The course is the major international training activity on the subject and could be used for the training of facilitators for regional courses. In 1994, the course had 33 participants from 22 countries, most of whom were involved in updating existing food composition tables or preparing food composition tables where none existed previously, for example in Singapore and Lebanon.


Some 15 years ago, there was recognition in Europe that substantial benefits would flow from coordinating the production of food composition tables in various countries. Such cooperation would lead to improved quality, availability and compatibility of the European nutrient databases and the food composition values In these: databases, and this would be valuable for multicentre epidemiological studies in Europe.

An organization which was initially called Eurofoods was established. Later it was incorporated into the Food-Linked Agro-Industrial: Research (FLAIR) Programme of the Commission of the European Communities (now the Commission of the European Union) as the Eurofoods-Enfant Project, This project collaborates with and resembles the International Network of Food Data Systems (INFOODS). These organizations have been working towards the improvement of the quality and compatibility of data on food consumption and composition in Europe.

One of the main achievements of the FLAIR Eurofoods-Enfant Project has been the establishment of a network of persons working in the field of food consumption and food composition in Europe, Meetings have been organized at regular intervals in which: people could meet and exchange information, expertise ano) experience. The project has also supported the further development and evaluation of food coding systems in Europe.

During the 1994 training course the following issues were addressed:

· selection of foods to be included in food composition tables in order to cover foods commonly consumed by the population in general and foods consumed by specific groups;

· selection of nutrients and other components based on policy decisions or objectives of research;

· sampling procedures with a well-described and properly documented protocol in order to obtain representative food samples;

· choice of analytical methods for the production of food composition data that are appropriate and valid, with consideration also for the resources available;

· quality of analytical data and the need for full documentation;

· other quality considerations in the compilation of a food composition database;

· food coding, modes of expression and dealing with missing values;

· the use of food composition data in association with food balance sheets and household budget surveys and in individual dietary intake studies.

Training in analytical techniques

Training in analytical techniques for generating food composition data is also required. Those responsible for food analysis do not always understand the principles involved in the different facets of generating food composition data, such as developing sampling protocols, sample handling and sample preparation, Nutritional analysis is a specialized branch of food analysis, and specific training is required in this area, Ideally, nutritional analysis should be included in university courses on analytical chemistry. However, a range of postgraduate courses could be directed at graduates in food science or nutrition who work in the field or plan to do so.

Quality control is an important issue. All laboratories involved in food analysis should be involved in collaborative interlaboratory studies and should use standardized reference materials and materials for internal quality control. Collaborative interlaboratory studies should be carried out to evaluate laboratory performance and to stimulate its improvement.


While it is technically feasible to develop international databases, this work is impeded by the variable analytical quality, the incompatibilities and even the unknown origins of existing data. One of the most important requirements for compatibility is that the data be of high quality. Food composition data must be reliable and comparable. Two examples demonstrate the importance of reliable data.

In 1985, seven centres were given the task of calculating the mean intake of energy and nutrients from a two-day dietary protocol (Arab, 1985). As can be seen from Table 1, estimates of the intake of calcium and fibre varied considerably, with coefficients of variation of 44 and 42 percent, respectively. Differences between extreme values were greater than 50 percent for fibre, calcium, iron, vitamin A and the B vitamins. Because the quality of the nutrient composition data applied Is not known, it is not clear whether these differences are due to natural variability in the composition of the foods, method of preparation, processing techniques, sampling procedures or method of analysis of the foods.

In a study on the carotenoid content of foods, particularly In developing countries (West and Poortvliet, 1993), a comparison was made between provitamin A data found in the Indonesian food composition table and the best data on the provitamin A content of foods currently available (Table 2). As a result of this finding, the estimate for the provitamin A intake of pregnant women in Indonesia has had to be reduced by at least 50 percent.

TABLE 1 - Daily intake of energy and various nutrients calculated from a two-day dietary protocol by seven research centres

Research centre

Energy (kj)

Carbohydrate (g)

Fibre (g)



Vitamin A (µm)

Vitamin B1

































































Standard deviation








Coefficient of variation (%)








Maximum difference

1 708







Source: Arab, 1985.

TABLE 2 - Provitamin A content of foods: values in Indonesian food composition table compared with revised data

English name

Indonesian name

Scientific name

Provitamin A contenta

Indonesian data

Revised data



Daucus carota

1 800

1 000

Cassava, leaf

Daun singkong

Manihot utulissima

1 650

1 055

Sauropus, leaf, sweet shoot

Daun katuk

Sauropus androgynus

1 556




Amaranthus viridis



Sweet potato, yellow

Ubi jalar

Ipomoea batatas



Tomato, ripe

Tomat buah

Lycopersicum esculentum



a Measured as retinol equivalent per 100 g edible portion.
Source: West and Poortvliet, 1993 (with corrections by the authors).

Generally, training in a number of specialized techniques requires three to six months in a specialized laboratory, which could be followed by four to six weeks of on-site training and two- to four-week visits at annual intervals. On-site training offers greater applicability to local conditions and also allows simultaneous training of several analysts.

Methods of food analysis must be adapted to local foods and facilities. Techniques of evaluating data from such methods should also be developed.

Apart from the training, sufficient resources should be guaranteed to the institution where the trainee works to allow the planned analytical work to take place, Additional problems to be solved may include: the initial cost of purchasing equipment, the cost and availability of supplies, adverse climatic conditions and unreliable and fluctuating electric power supplies, In addition, the institution where the training has taken place should be able to provide backup in the form of continued quality control and the supply of chemicals and spare parts.

Standard reference materials

There is a need for the ready supply of standard reference materials. Both the United States National Institute of Science and Technology (NIST) and the Community Bureau of Reference of the Commission of the European Union are active in the area of standard reference materials. More standard reference materials should be established, maintained and made available at an affordable price to recognized institutions. In addition, training should be provided in the use of standard reference materials and of in-house materials with values that are linked to those of certified standard reference materials.


Effective use of food composition tables and nutrient databases as well as international harmonization of food composition data and data management require training, education and expertise. A training programme covering all aspects of production, management and use of food composition data should form a unit in any tertiary or professional course in nutrition. Thus, considerable responsibility rests with those who train users of food composition database systems.

Education and training programmes will develop a network of workers with common goals and standards who will contribute to the development of common approaches to the organization of food composition programmes, to food nomenclature and nutrient analysis and expression, and to food sampling and data quality assurance programmes, Data will become more compatible as their quality improves.

Food composition tables and nutrient databases are the basis for all national nutrition programmes and studies relating nutrition to health, It is absolutely essential that sufficient resources, including trained personnel, be devoted to their production.


Arab, L. 1985, Summary of survey of food composition tables and nutrient data banks in Europe, Ann. Nutr. Metab., 29 (Suppl. 1): 39-45.

Greenfield, H. & Southgate, D.A.T. 1992, Food composition data: production, management and use. London, Elsevier Applied Science/Chapman and Hall.

West, C.E. & Poortvliet, E.J. 1993. The carotenoid content of foods with special reference to developing countries. Arlington, Virginia, USA, International Science and Technology Institute, Vitamin A Field Support Project (VITAL).

Amelioration des données relatives à la composition des aliments grace à la formation

II est essentiel pour tout programme de nutrition humaine comme pour l’industrie alimentaire de disposer de donn sur la composition des aliments qui soient de bonne qualitt compatibles, Etablir un syst mondial de bases de donn compatibles sur la composition nutritionnelle des aliments est une te de grande ampleur qui nssite une approche systtique de la collecte des informations. Les ddeurs doivent e conscients de la nssit’avoir des donn de qualitur la composition des produits alimentaires, et des fonds adats doivent e alloues activit

La production de donn valables sur la nutrition dnd d’une se d’activitauxquelles doivent participer les utilisateurs des donn, les analystes des aliments qui fournissent les donn et le personnel charg’blir la base de donn, L’exigence de qualites donn doit e intdans le programme dle drt. Il est nssaire d’assurer la formation de ceux qui produisent les donn sur la composition des aliments comme de ceux qui les utilisent, afin que tous comprennent les contraintes de la production et de l’utilisation de ces informations.

La bonne utilisation des tables de composition des aliments et des bases de donn sur les ments nutritifs, de m que l’harmonisation internationale des donn sur la composition des aliments et de la gestion des donn nssitent formation, enseignement et expertise. L’enseignement professionnel en mati de nutrition devrait comprendre un programme de formation couvrant tous les aspects de la production, de la gestion et de l’utilisation des donn sur la composition des aliments. Un programme de formation international organisar l’Universitgricole de Wageningen, l’Universites Nations Unies, la FAO et l’Union internationale des sciences de la nutrition comprend des matis telles que la sction des aliments et des ments nutritifs, les procres d’antillonnage, les modes d’analyse et autres considtions de qualitla codification des aliments et l’utilisation des donn sur la composition des aliments en nutrition.

Il est essentiel pour l’lution de la science nutritionnelle que les activitgrales de recherche nutritionnelle comprennent l’blissement et l’entretien de bases de donn sur la composition des aliments, La formation devrait faire partie d’un plan d’action gral visant l’amoration de la qualitt de la compatibilites donn sur la composition des aliments; la qualites bases de donn ne pourra s’amorer ’avenir que si les utilisateurs sont correctement formet vigilants.

La disponibilidad de datos compatibles y de alta calidad sobre composicie alimentos es imprescindible tanto para cualquier programa de nutriciumana como para la industria alimentaria, El establecimiento de un sistema mundial de bases de datos compatibles sobre este tema es una tarea fundamental que spodrealizarse mediante un plan sistemco de recogida de informaciLos encargados de formular polcas han de ser conscientes de la necesidad de datos de alta calidad sobre composicie alimentos y de los requisitos financieros para estas actividades.

La elaboracie datos fiables sobre nutriciepende de una serie de esfuerzos integrados en los que participan los usuarios de los datos, los analistas de los alimentos que generan la informaci los compiladores de la base de datos. La buena calidad de los datos debe formar parte del programa desde su principio. Es necesario que quienes crean y utilizan datos sobre composicie alimentos reciban capacitacide modo que comprendan los factores que limitan la producci utilizacie esta informaci

El uso efectivo de cuadros de composicie alimentos y bases de datos sobre nutrientes y la armonizacinternacional de los datos sobre composicie alimentos y su gestion exigen capacitaciformaci competencia profesional. Los cursos especializados sobre nutricieber incluir un programa de capacitaciue abarcara todos los aspectos de la produccimanejo y utilizacie datos sobre composicie alimentos. Un programa internacional de capacitaciorganizado por la Universidad Agraria de Wageningen (Pas Bajos), la Universidad de las Naciones Unidas, la FAO y la Uninternacional de Ciencias de la Nutricicomprende temas como la seleccie alimentos y nutrientes, procedimientos de muestreo, mdos de ansis y otras consideraciones relacionadas con la calidad, codificacie alimentos y utilizacie datos sobre composicie alimentos en la nutrici

Para que progresen las ciencias de la nutricis imprescindible que se elaboren y mantengan bases de datos sobre composicie alimentos como parte de las actividades generales de investigacielacionadas con la nutriciLa capacitacia de formar parte de un plan de accieneral para mejorar la calidad y compatibilidad de los datos sobre composicie alimentos, La calidad de las bases de datos smejorarn el futuro si los usuarios estperfectamente capacitados y permanecen vigilantes.