Previous - Next
This is the old United Nations University website. Visit the new site at http://unu.edu
INFOODS: Background and current status
Data: the user context
The INFOODS system
Introduction
Growing need for and availability of information
on components of foods
Generation and recording of food component data
INFOODS - an international network of food data
systems: a framework for discussion
Summary and conclusions
References
VERNON R. YOUNG
Laboratory of Human Nutrition, Department of Applied
Biological Sciences,
Massachusetts Institute of Technology, Cambridge, Massachusetts,
USA
Nutrient and food component data constitute a key and significant proportion of the body of knowledge required to understand the role played by the nutritional environment in human health and welfare. These data - their quality, quantity, and accessibility - constitute an essential resource for nutrition and food scientists and the various health professionals concerned with food supply and its implications for the well-being of individuals and populations at large. The data are generated and utilized by the international community of food and nutrition scientists, and thus it is essential that the continued development of this cumulative resource proceed in a way that promotes its maximum and effective exploitation for the resolution of nutrition and food-related problems both in this hemisphere and throughout the globe.
Information on the nutrient and non-nutrient composition of foods has been and continues to be used for a variety of purposes: assessment of the intake of and need for nutrient and nonnutrient constituents of foods by populations; meal planning and calculation of therapeutic diets; food production and nutrition policies; and the evaluation of food as a vehicle for transfer of toxic chemicals from the environment to man. To date, this information, particularly in relation to energy-yielding substrates and essential nutrients, has been obtained largely from food composition tables, a historical development of which has been presented by Hertzler and Hoover [5]. However, the increasing interest in and concern for the relationships between diet and food habits and degenerative diseases (including coronary heart disease, diabetes, hypertension, stroke, and cancers) has stimulated the further gathering of composition data on foods. Simultaneously, this has called attention to major limitations in the available data to support a variety of research activities in this area. These are especially evident in epidemiologic studies, where attempts are being made to characterize dietary differences between various populations exhibiting markedly different incidences of the diseases thought to be associated with diet and nutritional factors. Such studies have characterized interand intra-national differences in diet by broad categorizations, but suffer from the lack of detailed information on the nutrient and non-nutrient content of the enormous variety of foodstuffs consumed by different populations and subgroups within specific countries or geographic areas. Additionally, the introduction of nutrient labelling in the US, for example, has promoted the accumulation of data by the food industry on the composition of foods and their products. These data offer a further opportunity to expand the food data base.
Most of the information on the components of foods is still disseminated in the conventional, printed-page format; this is likely to continue as an important mode of communication for many users. However, there can be little doubt that the use of modern computer techniques will expand in the future, for a number of reasons:
1. Certain functions can be served more effectively through organization of a computerbased data bank than by conventional methods. Thus, one advantage of data dissemination in computer-readable form is the relative ease of updating; another is the possibility of carrying out sophisticated search strategies for material that has some specified set of desired properties, a problem that can be tedious or perhaps impractical when the information is tabulated in the conventional, hardcopy form.
2. The distribution of data in computer-readable form offers considerable advantages for its subsequent manipulation and usage.
3. The proliferation of small machines or of personal computers and interest by consumers in diet and food habits will inevitably result in an increased demand for computer-aided food composition data systems.
The number of food and nutrient data bases is increasing rapidly and many new entrepreneurs are becoming active. This is made evident, in part, by the growing number of meetings being devoted specifically to various problems associated with the development and uses of nutrient data bases [4]. Extensive nutrient data banks, such as the one established at USDA's Consumer Nutrition Center [6,10], and at other locations throughout the world, illustrate a clear trend in converting existing food component data to a computer format.
With increased use of computer techniques, both for archiving food component data and for application of these data to various purposes, it becomes absolutely critical to examine carefully the ways in which food composition data are handled and used. Although computerization can be very efficient, it does not guarantee optimum or effective service to the final user of the data. A study by Danford [2] is illustrative of this problem. It was found in this study that when a diet based on common, unambiguous foods and simple serving sizes was submitted for analysis using 11 commonly used computerized data bases, considerable variation existed between the output for various nutrients. This is a particularly disturbing finding when it is recognized that all of the data bases had evolved from a common data source, namely USDA Handbook No. 8. A similar, but more detailed, comparative study of a dietary record was considered by Hoover [7], and she also found that the values for nutrients varied considerably between the various computerized nutrient analysis systems included in the study, even after elimination of coding judgements relative to portion size and selection of food items. Dwyer and Suitor [3] have also pointed out the uncertainty about the reliability and validity of the data in many computer-aided diet analysis systems.
Because of the problem of the reliability of the available data in nutrient data banks and the disservice to the scientific and technical community that can easily arise from non-standardized, non-evaluated data storage and dissemination, it may not be unreasonable to suggest that consideration be given to the registration of food component or nutrient data banks by an appropriate authoritative body. This is especially important, as has been emphasized by Lide [8], when attempts are made to provide data directly to the user, because both quality assurance and the opportunity for users to assess the suitability of the data for meeting a particular need are key factors.
In addition to the ways in which information on food components is or might be handled, issues related to the availability and validity of the original, analytical data are critical as we further consider the ultimate nature of INFOODS. It is probably unnecessary to point out that the generation and recording of food composition data represents a difficult problem in comparison with, for example, data in physics and chemistry, where information about atomic structure and bond energies is determined under highly standardized and reproduceable conditions. In the context of foods, analyses are made on living or deteriorating material, and there is also variability between and within foods. Although this is not surprising, since biological variability is one of the major prerequisites of life, it does add to the complex task of achieving an orderly accumulation of composition data on foods consumed by people. Furthermore, the enormous variety of foods eaten represents a crucial problem with respect to the identification of foods and their systematic classification.
The precision found in chemistry and physics rests on the ability to repeat experiments under highly defined conditions in different laboratories. Comparable precision in the determination of food composition is not achievable, but further efforts must be made in this area to improve the quality of food component data. Attempts must be made to develop suitable reference, certified food matrices to permit the establishment of rigorous quality-control procedures for the analysis of foods both within and among laboratories.
In any event, data should be generated and presented according to accepted criteria or guidelines, with adequate descriptions of the procedures utilized, including concern for sampling and coding, as well as analytical chemistry. In order for progress to be made, careful consideration needs to be given to developing recommendations and criteria for reporting experimental data on food components. In this way, some of the experience of the evaluators of food component data could be transferred to those who generate new information, leading potentially to an upgrading of measurement practices.
To enhance the process of organizing food composition data into an international and consistent ensemble, it is essential that the quality of the archival data be indicated by the inclusion of information on methodology and sampling procedures. As Touloukian [12] points out, if this is done the task facing the critical analyst will be made more feasible and the quality of the data available to the user will be increased.
Some General Issues
Assuming, for the moment, that high-quality information on the nutrient and non nutrient profile of foods is available, a data system might be depicted simply, for discussion, as follows:
Inputs à Processing à Outputs
With this scheme in mind it is necessary to consider the following tasks, as discussed in greater detail by Vickery [13]:
1. Definition of the objectives of the food data system.
2. Examination of the present status of collection, analysis, and interpretation of food component data.
3. Assessment of the general design of a system, including technical, economic, and operational feasibility.
4. Practical implementation of the design and evaluation of the system in terms of objectives.
These, as well as other tasks, can be listed and, indeed, discussed separately. However, the building of a functional international food data system requires an iterative procedure.
As indicated above, the first question is: "What should be designed?" The answer involves exploring with users their aims and needs. Thus, the nature of the information required must be carefully identified, and consideration given to the functions that this information is meant to perform. It is fundamental to know early in our programme of work what uses are intended for the output of food component data, for then it is possible to assess the usefulness of the current information and the format(s) in which it is obtained. After this, one can begin considering alternative and perhaps more effective ways of meeting the needs of users of food composition data.
It is worth emphasizing further that immediate output of any data system is a set of products, intended to provide the community with information believed to be of interest to them. Clearly, a prime aim in evaluating, analysing, designing, or perhaps improving systems of food data information is to determine what specific information is of interest and whether the existing or intended products do or will provide what is required. Again it is critical to know the users and the specific uses that are made or might be made of food composition information.
Desirability of an International Effort
The establishment of national and international standardizing organizations and data bases has done much to improve the intelligibility, accessibility, and transferability of scientific and technical data. Thus, the initiation of an international, co-operative effort in food component data systems is worthy of particular consideration, especially when reference is made to the importance of food in international trade, in national and international aid programmes, and in the broad area of international health.
The precedent for an international food component data system has already been set by the many instances of international, non-government organizations working in the field of scientific and technological data; examples are the programmes of CODATA (Committee on Data for Science and Technology), the International Council of Scientific Unions, international organizations concerned with problems of handling data, such as the Rome-based Intergovernmental Bureau for Informatics (IBI), and the United Nations and its specialized agencies.
National food and nutrient data services are being organized in increasing numbers. It would be a major advantage to the community of nutritionists, food scientists, and health professionals if these services were compatible and, indeed, readily accessible. There is thus a need to consider the co-ordination of regional, national, and international food data services. This might range from a centralized file with on-line access, at one end, to a looser co-operation based on common guidelines and standards for information gathering and handling, at the other end. These possibilities are at least technically achievable, since there are numerous computer systems capable of accessing large centralized data banks from remote terminals via telecommunications, as well as systems based on decentralized data banks, updated in interchange of defined format tapes.
The task of placing existing food composition data in machine-retrievable storage is substantial but can be accomplished [6,10]. In terms of the human food supply, it is likely that this task will be done in piecemeal fashion even if a major commitment is made by the scientific community. As Abelson [1] has pointed out, if this problem is left entirely to individual initiative there will inevitably be enormous duplication of effort, and other problems will be neglected. Coordinating such an effort will require a high degree of international co-operation. However, international co-ordination is an attractive idea and we should strive to induce governments, academia, and food industries to provide support and sponsorship of these activities.
We must appreciate that private industries, governments, and non-profit institutions within and between countries will have their own distinct interests and responsibilities, and so a pluralistic approach to the generation, dissemination, and use of food component information is inevitable. It is, however, certain that, in view of the cost and effort involved, there are strong incentives for collaboration in building food component data bases, even if a multiplicity of information delivery systems turns out to be desirable. Furthermore, some central co-ordination is desirable, if only to encourage a degree of standardization and compatibility that will benefit all sources.
INFOODS at Present
Overall Construct
In consideration of the arguments presented above, and with the further assessments and recommendations of a group of experts convened to address the status of the field of food composition data [9], it was proposed to establish INFOODS, defined as:
An organization to promote international participation and co-operation in the acquisition and interchange of data on the nutrient composition of foods, beverages and their ingredients in forms appropriate to meet the needs of government agencies; nutrition scientists; health and agriculture professionals; policy-makers and planners; food producers, processors, and retailers; and consumers
A cornerstone of INFOODS' work plan is the idea that a majority of people involved with food composition data can be categorized as carrying out one of three major activities: generating the data in the laboratory; collecting and organizing the data of others for use by others; or using the data in a number of ways. Of course, these three groups may overlap in that many individuals are involved with two or even all three of the activities. However, they are logically distinct, and it is helpful to consider them as categories for purposes of further discussion in this paper.
Thus, apart from the users and uses of food composition data, there are many other activities involved, which concern the generation and preparation of the data. These would include the analysis of food, the development of analyses, the compilation of specialized data bases, and the design of diet analysis programs. These activities are all ongoing now, but a problem that must be faced is that they are often being done independently. Furthermore they can be, and often are, done in different ways.
This problem becomes obvious when one considers the flow of food composition data: food ~ analysis - data base - user. Just as there are different users and uses, there are different specific paths one might take: ways of selecting foods, performing the analyses, combining replicate analyses, and organizing data bases - each might differ. These paths can produce different numbers; food composition tables or data bases therefore differ, and the numbers in the various tables may mean different things.
The independence of all these individuals and activities relating to food composition data have contributed to the fact that the state of food composition data today is simply not satisfactory. Problems do exist, and many of them can be organized around four specific issues: (a) data do not exist; (b) data cannot be accessed; (c) data are not complete; and (d) data are not accurate.
It was to attempt to resolve these problems that INFOODS was organized essentially as two networks: a network of food composition data, and a network of people interested in food composition data. In addition, a secretariat, with Dr W. M. Rand of MIT as executive secretary, has been organized to co-ordinate the various activities, with major financial support for this purpose coming from the US National Cancer Institute, and additional sponsorship from the National Heart, Lung and Blood Institute, the US Food and Drug Administration, and the US Department of Agriculture. Administrative support is received from the United Nations University to facilitate the international aspects of the programme.
The scope and precise nature of the networks referred to above continue to evolve but the conceptual outline of the system is shown in figure 1. Thus, it is anticipated that local centres or individual users obtain their data and assistance from regional centres which communicate with each other, with institutions, and with the INFOODS secretariat. These regional centres could be individual countries or groups of countries. Thus, if a user needed data about the nutrient consumption of an immigrant population, some local data might be available or a request made for additional data from a regional centre. It might be that a regional centre would not have all of the information needed, but would, in turn, request such from another region, perhaps via consultation with the INFOODS secretariat to learn whether other data did exist and where they might be found. Most important, this schematic outline emphasizes that the key to this endeavour is communication, often between quite different entities.
Fig. 1. An overview of the International Network of Food Data Systems (INFOODS).
The Approach Being Taken
To achieve a working network, linking together a number of disparate entities, there must be careful and detailed planning. In this context, figure 2 summarizes the major components of INFOODS. In order to achieve our desired results, three major working committees have been established to produce standards and guidelines.
The first of these is chaired by Professor A. S. Truswell of the University of Sydney, Australia. This is the terminology/nomenclature committee, and is concerned with the major problem that foods are not well specified, and have different names in various parts of the world. Often the same name is given to foods that may differ in origin and form. Any interchange of information needs an unambiguous definition of a food. It is of paramount importance that this work be carried out well and carefully. The tasks of the committee include: a review of existing systems, evaluating these systems in terms of their suitability for international exchange of food information, and recommendation of universal, international, standardized documentation language.
The quality of the food composition data themselves are the subject of the second working committee. This is chaired by Dr D. A. T. Southgate of the Food Research Institute in the UK. The overall responsibility of this committee is to develop guidelines and standards for the data used in the construction of food composition tables and electronic data banks. Its long-term goal is to improve the quality of data by standardizing methodologies, and, in the interim, to develop measures of quality and conventions for data reporting. An initial product of this committee will be an updated and considerably expanded version of an earlier guideline for establishing food composition data tables [11].
Fig. 2. Schematic of approach being taken by INFOODS.
The third committee, under the direction of Dr J. Klensin of MIT, is concerned with the information systems component of INFOODS. Several of the more important topics that this committee is undertaking are: (a) data interchange and communication standards; (b) ideal data set content; (c) data manipulation; (d) regional data centre design; and (e) local data centre design, including personal computers and small machines for individual use in office and field settings.
An essential activity that underlies the work of these three groups is a careful survey of the users and specific uses of food composition data. Information to be gathered includes: (a) the magnitude of usage and types of users; (b) the data most frequently needed; (c) the unmet needs and why they are not met; and (d) the users' desires and expectations. What is being sought is not just a catalogue of uses, but a careful and comprehensive examination of how they constrain the INFOODS network itself. For example, how quickly do most users need their data, what precision do they require, and what tools do they have to use the data? With detailed information of this kind it will be possible to arrive at a design for the forms in which the data will be delivered. It is these types of questions that are essential to the design of INFOODS.
Among other activities shown in figure 2 is the promotion of regional INFOODS groups. This is important to the network of people that is a fundamental component of regional INFOODS and essential to its overall success; the data network operates between people, and it is operated by and for people. One of the regional organizations, EUROFOODS (Western Europe), began independently at about the time that INFOODS was established. Other regions include NORFOODS (Scandinavia), NOAFOODS (North America), LATINFOODS (Latin America), and ASIAFOODS (Asia). Early plans are under way to promote regional groups covering the African continent; Australia, New Zealand, and the Pacific Islands; and also the Soviet and Eastern European nations. These regional groups are at different stages of development but are being organized with the intent of bringing together the people within a specific geographic area so that they can work efficiently on their own problems as well as interact more effectively within their region. Some of the activities each regional group might be or are involved with include: (a) regional data centres/clearing-houses; (b) identification of regional food composition data problems; (c) co-ordination of regional resources and activities; (d) contribution to development and review of implementation standards and guidelines; and (e) communication with other regional groups and with INFOODS.
In order to further the international contribution that our collective efforts might make toward resolving the many issues related to food composition data, the various activities of INFOODS are also being conducted with formal representation on the working committees by IUNS (International Union of Nutrition Scientists), IUFoST (International Union of Food Science and Technology), and IUPAC (International Union of Pure and Applied Chemists). Additionally, whenever possible the World Health Organization (WHO) and the Food and Agriculture Organization (FAO) are represented at the meetings in view of the interest in and concern for food composition data by these UN agencies.
Information on the components of foods embodies the heritage of our understanding of the role of nutrition in human health. It serves as a medium for the education of future scientists and as a principal reservoir of concepts and data to be applied to the resolution of nutrition and health problems. Some reasons have been given indicating that the need for reliable food component data will become more pressing. In parallel, the use of computer-based methods for the storage, processing, and retrieval of such data will accelerate. It is essential that, with such developments, a co-ordination of effort be undertaken at an international level to standardize the gathering, synthesis, and application of this information.
To meet this challenge we must assess, in detail, the users and uses of food composition data. From this assessment a series of design specifications for INFOODS will emerge. In this way it is hoped that the goal of effectively and efficiently increasing the quality, quantity, and accessibility of food composition data on a worldwide basis will be significantly closer to becoming a reality.
1. P. H. Abelson, "Production and Use of Data in the Pure and Applied Sciences," in P. S. Glaser, ea., Data for Science and Technology (Pergamon Press, New York, 1981), pp. 5-9.
2. D. E. Danford, "Computer Applications to Medical Problems," J. Paren. Enteral Nutr., 5: 441-446 (1981).
3. J. Dwyer and C. W. Suitor, "Caveat Emptor: Assessing Needs, Evaluating Computer Options," J. Amer. Diet Assoc., 84: 302-312 (1984).
4. H. Greenfield and R. B. H. Wills, "Tables of Food Composition: An Australian Perspective," Food Tech. Australia, 33(3): 101-130 (1981).
5. A. A. Hertzler and L. W. Hoover, "Development of Food Tables and Use with Computers," J. Amer. Diet. Assoc., 70: 20-31 (1977).
6. F. N. Hepburn, "The USDA National Nutrient Data Bank,"A.J.C.N.,35: 1297-1301 (1982).
7. L. W. Hoover, "Computerized Nutrient Data Bases: I. Comparison of Nutrient Analysis Systems," J. Amer. Diet. Assoc., 82: 501-505 (1983).
8. D. R. Lide, "Critical Data for Critical Needs," Science, 212: 1343-1349 (1981).
9. W. M. Rand and V. R. Young, "Report of a Planning Conference Concerning an International Network of Food Data Systems (INFOODS)," A.J. C. N., 39: 144-151 (1984).
10. R. L. Rizek, B. P. Perloff, and L. P. Posati, "USDA's Nutrient Data Bank," Food Tech. Australia, 33: 112114 (1981).
11. D. A. T. Southgate, Guidelines for the Preparation of National Tables of Food Composition (Karger, Basel, 1974).
12. Y. S. Touloukian, "Reference Data on Thermophysics," in B. Dreyfus, ea., Generation, Compilation, Evaluation and Dissemination of Data for Science and Technology (Pergamon Press, New York, 1975), pp. 138-146.
13. B. C. Vickery, Information Systems (Butterworths, London, 1973), p. 238.
Introduction
The link between the user and the data
The variability of the data
WILLIAM M. RAND
INFOODS Secretariat, Massachusetts Institute of Technology,
Cambridge, Massachusetts, USA
It is the purpose of this conference to explore the linkage between food composition data and the user of these data. INFOODS has been set up to work towards "good" and "available" food composition data, and an essential first task is to establish just what is meant by "good" and "available." It is obvious that these terms only have meaning in the context of how the data are to be used, and thus a natural starting-point is to discuss and detail their users and uses. As a prelude to this, this paper first examines the context of food composition data usage, and then shows that the variability of the data is a key concept in unifying the various uses.
Most users of good composition data have to balance three major, different types of information - answers to three different questions: what is in foods?; what do people eat?; and what do people need?
While these can be written as distinct questions they are closely interrelated. Discussion of any one of them without consideration of the others would be possible; however, it would be, for the most part, academic. Since this conference is directed towards the relationship between the user and the single question of what is in foods, it is important to realize that there must be at least implicit realization of the importance of the other two aspects - what people eat and what they need. The strict inseparability of these questions becomes more apparent if we examine more closely the relationships between foods and users, following just the two questions of what is in foods and what people eat. Figure 1 shows the four major components of this linkage: food, data, the individual, and the user. These are connected by four types of activity: analysis, data usage, consumption, and diet evaluation. It is obvious that, for the user to be able to do anything meaningful, the two paths from food to user must be both compatible and consistent.
The linking of data and user, and how this link can be improved, is of special interest. It must be appreciated, however, just how constrained this link is by its following the analysis of the foods - for example, often we cannot get all the data we may think we need. Moreover, while we often feel that we really want to know precisely what is in foods, that information is really only useful if it can be correlated with what we can find out about what people really are eating.
Two additional points need to be made in conjunction with considering the data-user link in the context of our three questions. These are complementary and follow from the above. The first is that good, readily available food composition data, however we define good and readily available, will not solve all the users' problems. This leads to the second point that data are not of interest in and of themselves; they are of value only if they permit some generalizations to be made, some predictions or decisions.
