A knowledge of the chemical composition of foods is the first essential in dietary treatment of disease or in any quantitative study of human nutrition.
(McCance and Widdowson, 1940)
This statement is as true now as it was in 1940, when it formed the first sentence in the introduction to the book that has now evolved into the United Kingdom National Nutritional Database (Food Standards Agency, 2002a). The source of information on the composition of foods was, traditionally, printed food composition tables; these are now being replaced by computerized compositional databases from which the printed versions are usually produced. The information is widely used in the health, agriculture and trade sectors.
The data are used in research studies of the effects of diets on health, reproduction, growth and development. Data are also used for devising diets with specific nutrient composition in clinical practice, in the formulation of ration scales and in the devising of emergency food supplies. Nationally and internationally, compositional data are used in the assessment of the nutritional value of the food consumed by individuals and populations.
The recognition of the involvement of diet in the development of many diseases (McGovern, 1977) has led to an expansion in the number and range of studies of the relationship between diet and health and disease, which has led to a greater focus on nutrient data. Willett (1998) has drawn attention to this and to the need for databases to be reviewed regularly: “Diets of human populations are extremely complex ... Maximal insight into the relation between diet and disease will usually be obtained by examining diets both as constituents and as foods. Calculations of intakes of nutrients and other constituents require a food composition database that is complete and current.”
The evidence that has emerged from these epidemiological studies has led to a growth in the production of national and international guidance on choosing a healthy diet. Composition data provide the foundations for the development of education programmes on choosing healthy diets. As part of this guidance to consumers, many governments have implemented the nutrition labelling of foods. Some countries require the producers of food products to provide their own analytical data on the composition of their products.
However, in appropriate cases, most regulations allow the use of compositional data taken from an authoritative compilation, such as a national food composition database, as a substitute for direct analysis. This development has added a quasi-regulatory role to food composition databases and strengthens the need for maintenance of data quality in terms of both the representativeness of the samples and the quality of the analytical data.
Establishing the composition of foods often has advantages for the trade in foods because importing countries with nutrition-labelling regulations prefer (and may require) that imported foods conform to the standards expected of locally produced foods.
Computerized databases have substantial advantages over printed food composition tables: they can contain a greater volume of information and the data can be used in calculations much more easily. The information can also be reformulated in different ways relatively easily to accommodate the needs of different users.
These advantages of calculation from computerized databases are especially important for nutritional epidemiologists, who frequently have to work with very large numbers of subjects and a large number and variety of food consumption records.
The power of epidemiological studies can be greatly enhanced when they are implemented at the international scale. For this to be effective requires, first, compatible records of food consumption and, second, national databases that are compatible. Compatible in this context implies “capable of being used together”.
Achieving a worldwide system of compatible food composition databases lies at the heart of the INFOODS programme. INFOODS – the International Network of Food Data Systems– was established in 1984 on the basis of the recommendations of an international group, and it operates under the auspices of the Food and Agriculture Organization of the United Nations (FAO) and the United Nations University (UNU) (Scrimshaw, 1994). Its goal is to stimulate and coordinate efforts to improve the quality and availability of food analysis data worldwide and to ensure that anyone, anywhere, would be able to obtain adequate and reliable food composition data. It has established a framework for the development of standards and guidelines for the collection, compilation and reporting of food component data.
This book is a continuation of the INFOODS effort, building upon earlier books (Klensin et al., 1989; Rand et al., 1991; Klensin, 1992; Greenfield and Southgate, 1992). The principles and guidelines contained in this book are intended to aid individuals and organizations concerned with the construction of food composition databases. The primary objective is to show how to obtain information that will meet the requirements of a database system that is compatible with systems that have already been, or are being, developed worldwide.
The book focuses on the areas of information-gathering that are critical in determining data quality and must therefore be closely controlled.
It is important to recognize that the term “guidelines” is not used in a prescriptive sense but in the sense of the “principles” of preparing databases. These principles draw on and are a result of experience gained in the preparation of databases over many years and in different countries. The guidelines do not set out detailed sampling or analytical protocols but provide examples of approaches that have been used successfully. In many countries, the protocols that should be followed are set out within a legal framework that must, of course, be followed. However, by discussing and setting out the available options the guidelines may suggest where established programmes might be revised.
The nutritional and analytical sciences are developing continuously and these developments may indicate better approaches than those set out in these guidelines. It is expected that these principles will serve as a framework for the future development of food composition data programmes.
The structure of the book follows the stages in an idealized programme of work in preparing a food composition database. Chapter 1 describes the variety of uses of a food composition database that the compilers (those with executive responsibility for collecting and assessing the data to be used in the database and their presentation) have to meet. Chapter 2 describes the overall design of programmes for creating, or revising, a food composition database. Subsequent chapters deal with the selection of foods for inclusion (Chapter 3) and the selection of nutrients (Chapter 4). Chapter 5 describes the principles of sampling foods and Chapter 6 deals with the selection of analytical methods and their evaluation. Chapter 7 presents a review of the methods available for the nutrients, focusing on methods that have been shown to be compatible internationally. Chapter 8 describes the principles of assessing the quality of analytical data. Chapter 9 describes the presentation of data and the modes of expression that are central to producing compatible data. Chapter 10 discusses the compilation of data for inclusion in the computerized database. The processes and design of computerized systems for compositional databases lie beyond the scope of this book. Chapter 11 deals with the intrinsic limitations of nutrition databases that constrain their use. The chapter also provides guidance on the proper use of the food data. Finally, Chapter 12 discusses the future needs in the area of food composition.