FOOD, COMPOSITION OF

FOOD, COMPOSITION OF. Food composition activities include data generation in an analytical laboratory, data compilation in a database management system, data dissemination through print and electronic media, and data use by various professional and lay users.

Historically, food composition activities were limited to data on nutrients. Increasingly, food composition work deals with data for any component found in food: nutrients, bioactive nonnutrients, antinutrients, pesticide residues, other contaminants, additives, and more. A single food composition database, with proper documentation, can accommodate data on all these types of components.

Sectoral Elements

Food composition data are useful to many professions and sectors. Health, agriculture, environment, and trade are the sectors most fundamentally involved in food composition activities. Over time, agriculture has been the dominant sector involved in food composition research and service. This is demonstrated most clearly in international organizations such as the Food and Agriculture Organization of the United Nations (FAO), with a history of food composition work dating back to its inception in the 1940s; and in countries having the longest history of formal food composition activities such as the United Kingdom, where the Ministry of Agriculture, Fisheries and Food has had most of the responsibility, and the United States, where responsibility lies with the United States Department of Agriculture (USDA), and has since the late 1800s. Nevertheless, more than half of the participants in food composition conferences, and the researchers publishing food composition papers and books, are health sector professionals. In many countries it is the health sector that provides a high percentage of the funding for the work, and constitutes the highest percentage of the users of the information. Involvement of the environment sector is becoming increasingly important as it relates to the composition of indigenous or protected plant and animal species used as foods, and the content of environmental chemicals in the food supply. Trade has also gained more dominance in recent years. Nutrient information panels on processed foods have become regulatory requirements in many countries. And analytical data on both nutrient and contaminant content are necessary documentation for global food trade.

Organizational Elements

The international level. INFOODS, the International Network of Food Data Systems, was established in 1983 by United Nations University (UNU), with an organizational framework and international management structure that includes a global secretariat and regional data centers. Its mandate is "to improve data on the nutrient composition of foods from all parts of the world, with the goal of ensuring that eventually adequate and reliable data can be obtained and interpreted properly worldwide." In the mid-1990s, the Food and Agriculture Organization joined UNU in partnership for INFOODS. The main activities of INFOODS at the international level include development of technical food composition standards, assistance to Regional Data Centers and individual countries in developing their food composition activities, and publication of the Journal of Food Composition and Analysis.

The regional level. There are seventeen Regional Data Centers in operation. Some were created in the mid to late 1980s and have well-established and effective coordination (for example, LATINFOODS, ASEANFOODS, OCEANIAFOODS, and EUROFOODS); some are relatively new, yet making progress; and a few have been newly created and are trying to establish their regional or national priorities and capabilities. Regional food composition tables have been prepared, both electronically and in printed form (for example, Pacific Islands, ASEANFOODS, LATINFOODS), and many regions have regular food composition coordination activities and technical task forces involving all the individual countries in the region.

The national level. Most countries have food composition activities of one form or another. A national food composition program is usually the result of the combination and coordination of activities, within some defined administrative framework, related to food composition data generation, compilation, dissemination, and use. A steering committee is a useful structure, functioning well in many countries. This steering, or advisory, committee is ideally composed of individuals directly involved in food composition work, that is, the data generators, data compilers, and data disseminators. Crucial to the effectiveness of a steering committee is the involvement of data users. The users can be selected among dietitians, nutritionists, food industry personnel, and consumer group representatives.

Often a single organization holds the overall responsibility for managing a national food composition program, yet it is rare that a single organization accomplishes all the activities itself. Regardless of their affiliations, the laboratory-based data generators must interact closely with the data compilers, and the compilers must interact closely with the data users. In most countries there are other agencies with activities that have direct or indirect relationships with food composition data, but operate in concert with the national program. In addition to the desirability of a coordinated national approach for accomplishing essential activities, it is productive and important for a national food composition program to operate in conjunction with its Regional Data Center, and with ongoing international activities.

Technical Elements

Data generation is the process whereby foods are sampled, prepared for analysis, and analyzed in the laboratory. Data compilation is the process whereby the data from the laboratory are examined, manipulated, and incorporated into a food composition database. Data dissemination refers to the preparation and publication of books and electronic data products, which are made available to users in the various sectors. Data use also includes the application of these data to tasks, projects, and programs in the various professional sectors.

Data generation. Sampling, the process and procedures for obtaining foods that are representative of those available and consumed, is fundamental to any food composition activity. Preparation of a sampling plan often requires involvement of all the major contributors to a food composition program. Data generators must be involved in the sample collection, or at least the scheduling of sample collections, so that samples may be immediately and properly prepared for analysis. Data compilers must be involved because information on the sampling plan and details such as when and where sampling took place are important parts of a food composition database's metadata. Data users must be involved because they have the best appreciation of the foods that need to be analyzed, and often the location from which the samples should be collected. The services of a statistician are useful for developing a sampling plan, because representativeness is dictated by the number of food units collected—and analyzed—to achieve the goal. The goal might be to compare compositional differences between cultivars, or to achieve year-round, nationwide mean values for a food composition database. The overall quality of food composition data is determined largely by the sampling plan.

The collected samples must be properly handled so that they arrive at the laboratory without changes that might affect their composition. The key component, crucial to the correct determination of almost all other food components and most easily affected by improper handling and storage, is water (moisture). Once samples are delivered and documented, they are prepared for analysis. Preparation may involve separation of edible from inedible portion (for example, removal of bones from fish, or skins and seeds from pumpkins); kitchen-type preparation (for example, boiling rice); or combining of many samples into fewer samples (for example, combining five brands of similar biscuits into one representative composite sample). After this type of preparation, samples will be stored, or immediately analyzed. As with sample collection and sample handling, proper documentation of all aspects of sample preparation is essential.

Analyses. Most laboratories undertake a limited range of analyses for food composition purposes. This includes a set of core components and then additional components of interest, for example, laboratory research dealing with diet-related health problems. Core nutrients usually include the complete range of proximate components (water, nitrogen for the protein calculation, fat, glycemic carbohydrate, dietary fiber, ash, alcohol where relevant, and an energy value using factors applied to the energy-yielding proximates), some vitamins, and some nutrient elements. Additional components of interest often include cholesterol, individual fatty acids and aggregations of fatty acids (for example, total saturated fatty acids), carotenoids (both provitamin A carotenoids and antioxidant carotenoids with no provitamin A activity), other bioactive nonnutrients, heavy metals, and some so-called antinutrients (for example, phytates). Proper laboratory practices must be strictly adhered to, as well as laboratory quality assurance and quality control procedures, and details of analytical methodologies must be properly documented.

Data compilation. Data compilation requires a relational database management system, and adherence to international food composition standards where they exist. The database should accommodate numeric data, text, and graphics. Ideally, all the raw analytical data, and their attendant documentation, should be captured. The system should then be able to manipulate these data in many different ways. The same data system should provide an exhaustive reference database and any number of abridged user databases to satisfy the broad range of user requirements for food composition data. Many compilers only capture mean values, a practice that will satisfy many users. Other compilers provide more information, and therefore higher-quality databases, by including the number of samples and some expression of their variability. Other compilers are able to capture all the analytical data and prepare user databases with ranges (that is, high and low values), medians, and many different statistical expressions of the data, satisfying a broader spectrum of users and ensuring the highest quality database.

Some compilers prepare their databases with aggregations, excluding the baseline data (for example, a calculated value for vitamin A in retinol equivalents [RE] without individual values for retinol and each provitamin A carotenoid), whereas other compilers provide the analytical data for the individual components, in addition to the aggregations. This latter practice should be encouraged, since conventions for calculating aggregates based on biological activity change, and many of these individual components have other functions in addition to their roles as provitamins.

In data compilation, all food composition data can be included in the database. Complete information for all components in all foods is not necessary. Ideally, a database with one thousand foods should have complete information for core nutrients, but should also be able to accommodate sporadic data for other components in the foods included.

The early work of INFOODS included the development of standards and guidelines for compiling food composition databases for national and regional use (Rand et al., 1991), standards for unambiguously identifying food components (Klensin et al., 1989), and standards for ensuring international comparability and interchange of food composition data (Klensin, 1992). These standards are being maintained and further developed by INFOODS expert committees and consultative groups.

Data dissemination. With appropriate data compilation, food composition data can be disseminated in many different forms to satisfy all user requirements. Table 1 shows examples of some of the common forms in which food composition data are disseminated. Data disseminated as a set of relational files offers users with very specific needs, or those with customized software, the opportunity to use the data as they wish. Other common dissemination formats provide the types of information most often required by users.

Different countries have different approaches for charging, or not charging, for their data and data products. The United States Department of Agriculture prepares the largest single body of food composition data in the world and disseminates it freely via the World Wide Web, as both a downloadable set of relation files and a searchable reference volume.

Data Use

Food composition data are the basic, most fundamental information resource for most nutrition activities. Some of the specific uses of food composition data, along with examples of their uses, are listed below by sector.

Health sector. Food composition data are used in health protection activities in most countries in the world. "Food control" laboratories monitor mostly harmful components of foods. Other health protection activities include food composition activities involving total diet surveys or "market basket surveys" designed to determine the risk to populations from intakes of selected nutrients, antinutrients, and contaminants. The sampling, sample preparation, sample handling, analyses, and reporting requirements are virtually identical to the requirements of other food composition activities.

Health promotion. Health promotion activities include campaigns aimed at reducing or increasing the intake of certain nutrients in certain populations. Examples include healthy heart campaigns, typically using energy, fat, fatty acid, and cholesterol compositional data to educate the public about diet-related cardiac morbidity and mortality. In many developing countries, health promotion focuses on micronutrient data, including the necessity for including iodine in salt and provitamin A carotenoids in fruits and vegetables.

Clinical research. Food composition data are central to many clinical research trials. Examples include studies focusing on amino acid digestibility in ileostomy patients, vitamin A intake in breast-fed infants, and serum cholesterol levels in vegetarians. Knowledge of the composition of the test and control food(s) and/or diet(s) is fundamental to these studies.

Clinical care. Clinical dietitians must know the composition of foods in order to provide effective and therapeutic meals in a clinical setting. Special diets for patients are often based on individual nutrients in the foods: low-sodium diets for hypertensive patients, diets low in saturated fats for heart disease patients, diets containing proper ratios of protein and fat, and those containing the proper amount of carbohydrate for diabetics, high-protein diets for burn patients, diets containing low phenylalanine for phenylketonuric patients, and so forth.

Epidemiological research and diet studies. Epidemiological and diet studies take many forms. Some studies address food intakes and relate them to nutrient content of the diet and the incidence of diseases. Interpretations of the findings of these studies often focus on individual nutrients. Recent examples include Dutch and Finnish studies of dietary antioxidants and lung cancer, and studies of vitamin E and colon cancer among Iowa women.

Public health policies. Many public health policies relating to noncommunicable disease focus on food composition. Such policies set forth nutrition goals and guidelines and include recommended dietary intakes (RDIs). An example of such goals and guidelines is "choose a diet low in fat, saturated fat, and cholesterol"; and an example of an RDI is "females between the ages of fourteen and eighteen should get 15 mg of iron daily." In order for such recommendations to be useful, both health professionals and the public must have access to data on the nutrient composition of foods.

Nutrition intervention policies. Nutrition supplementation typically takes the form of fortification of the food supply or supplementation of the population. Examples of food fortification include the addition of iodine to salt (most countries), of vitamin A to sugar (for example, in Guatemala), and of minerals and B vitamins to refined cereal products (United States, United Kingdom). Such interventions are only carried out after the nutrients in the food and water supply of a country have been studied, and a baseline position has been established and carefully monitored over a period of time.

Household food security. Although food security is an issue that spans the health, agriculture, environment, and trade sectors, household food security is usually considered to be a health sector issue. Knowledge of the nutrient content of the foods consumed by household members is a precondition for assessing household food security.

Agriculture sector. The intensive livestock industries require accurate nutrient composition data on the feeds used. These data are generally far more extensive than those required for human foods, and include many micronutrients and individual amino acids. "Performance" in these animals usually refers to weight at time of slaughter; muscle tissue to fat tissue ratios; and in the case of milk-producing animals, an accurate profile of the proximate composition (protein, fat, lactose, water, and ash).

Food security. National and global food security is generally considered an agriculture sector issue related to food production, rural development, irrigation, fertilizer and pesticide use, crop yields, and so on. A common tool used to assess national and global food security is the FAO food balance sheets that examine, at the commodity level, the amount of food available to a country. The amount of food is then converted into individual components and reported as the amount of protein, fat, and energy available per person per day from the domestic food supply. Food composition data assigned to the commodity data are the basis for many food security assessments, including FAO's yearly report on the number of undernourished people in the world.

Export food industries. The agriculture sector is responsible for ensuring that food exports meet the regulatory requirements of the intended market. Food composition data are important, as product specifications (for example, the fat content of butter) and as nutrition label panels.

Domestic food production. Agriculturalists have long professed that malnutrition is not just a health problem, but also an agriculture problem. Increased consumption of imported food commodities has brought about changes in food patterns and diets that have contributed to the increase in diet-related health problems previously unheard of in certain parts of the world. Agricultural extension workers are combating the incidences of diet-related diseases in some developing Pacific Island countries by using nutrient composition data in family food production, helping families in designing home garden projects to supply nutrients that would otherwise be consumed in insufficient quantities.

Molecular and traditional food plant and animal breeding. Breeding has been done to modify certain nutrients in foods. Familiar examples include corn bred for higher lysine and cattle bred for lower fat content of the carcass.

Environment sector. Knowledge of the nutrient composition of the native diet of endangered animal species is an important requirement for protecting them. In New Zealand, scientists have undertaken studies to determine the nutrient composition of the original diets of birds in their native habitat, to ensure that the same nutrients in the same quantities and proportions were being supplied in their human-made offshore island sanctuaries and other protected, artificial habitats.

Climate change also affects food composition. Ozone depletion affects both food production and the composition of crops and agricultural products. Like ozone depletion, global warming affects agriculture in terms of production implications. Its other major effect, now and in the future, is the creation of conditions that will permit certain food products to be cultivated where temperature conditions did not permit their cultivation previously. This will alter the food supply, and along with it the nutrient composition of certain foods, in certain countries. Food composition data have been used as markers in modeling and predicting environmental change, for example, monitoring the changes in fatty acid composition of fish to chart the climatic phenomenon of El Niño.

Trade sector. Trade has emerged in recent years as one of the more important and demanding of the sectors involved in food composition activities. Food composition in various forms features in the World Trade Organization agreements, the Codex Alimentarius Commission and several of its committees, multilateral and bilateral trade agreements, and national food regulations and standards. More than other sectors, trade has illustrated most poignantly the need for standards and harmonization in technical food composition activities. Many trade-related court cases have involved food composition data, both in the charges filed and in evidence presented, and many of the food product detentions and rejections at U.S. borders are due to the absence of the Nutrition Facts panel of nutrient content data.

See also Agricultural Research ; Climate and Food ; Codex Alimentarius ; Ecology and Food ; FAO (Food and Agriculture Organization) ; Food Consumption Surveys ; Food Safety ; Food Security ; Food Trade Associations ; Government Agencies ; Green Revolution ; International Agencies ; Nutrition Transition: Worldwide Diet Change ; Toxins, Unnatural, and Food Safety ; Water: Safety of Water .

BIBLIOGRAPHY

AOAC. Official Methods of Analysis for AOAC International, 17th ed. Arlington, Va.: AOAC International, 2002.

Burlingame, B., ed. "Special Issue: 3rd International Food Data Conference." Journal of Food Composition and Analysis. Volume 13, 4. London: Academic Press, 2000.

FAO. Available at http://www.fao.org/. Provides links to Codex Alimentarius, FAO Statistical Databases (includes food balance sheets).

Finglas, P. M., ed. "Special Issue: The 2nd International Food Data Base Conference." Food Chemistry. Volume 57, 1. New York: Elsevier, 1996.

Greenfield, H., ed. Proceedings of the 1st International Food Data Conference. Quality and Accessibility of Food Related Data (1st ed., vol. 1). Arlington, Va.: AOAC International, 1995.

Greenfield, H., and D. A. T. Southgate. Food Composition Data: Production, Management, and Use, 2nd ed. Rome: FAO, 2002.

INFOODS. Available at http://www.fao.org/infoods. Provides comprehensive information and links to all food composition resources, including databases, printed food composition tables, standards and expert committees, regional data centers, reference and textbooks, software products, conferences, training courses, and so on.

Klensin, J. C. INFOODS Food Composition Data Interchange Handbook. Tokyo: United Nations University Press, 1992.

Klensin, J. C., D. Feskanich, V. Lin, A. S. Truswell, and D. A. T. Southgate., Identification of Food Components for INFOODS Data Interchange. Tokyo: United Nations University Press, 1989.

Rand, W. M., J. A. T. Pennington, S. P. Murphy, and J. C. Klensin. Compiling Data for Food Composition Databases. Tokyo: United Nations University Press, 1991.

Barbara A. Burlingame