Contents - Previous - Next
This is the old United Nations University website. Visit the new site at http://unu.edu
N. Snehalatha Reddy, S. Y. Waghmare, and Vijaya Pande
One way to prevent protein-energy malnutrition is to provide nutritious weaning foods to infants. In India, several commercial mixes are available, but they are often too expensive for persons of low socio-economic status. To find a more affordable product, four weaning mixes were formulated using local foods and traditional processing techniques. They were evaluated for acceptability by a sample target population of mothers and infants, and for protein, iron, and energy content using established laboratory methods. All the mixes were inexpensive and nutritious, and remained acceptable even after being stored for one month.
The principle of low-cost vegetable-protein-based weaning foods was demonstrated to be practical in the late 1950s by the Institute of Nutrition of Central America and Panama. Through experimental, clinical, and field studies a mixture of approximately one-third oilseed meal and two-thirds cereal suitably fortified with vitamins and minerals was determined to be fully equal nutritionally to animal-protein-based mixtures and much less expensive (see "A Look at the Incaparina Experience in Guatemala," Food and Nutrition Bulletin, 2 (2):1-2 [Apr. 1980]). Legumes could replace oilseeds at a slightly higher proportion. This work became the basis for the various formulas for Bal-Ahar, which was first produced in India in 1962 and is still important there. These mixtures are still being produced and used on a large scale in India and Guatemala, and the principle has been widely applied elsewhere.
The present paper expands the search for locally available components for such weaning foods and explores processing alternatives. There is a continuing need for the development of vegetable-based weaning foods of this type in other developing countries.
Protein-energy malnutrition is an important nutritional deficiency condition that often occurs during the critical transitional phase of weaning in infants, crippling their physical and mental growth. This condition can be prevented to a large extent by introducing weaning foods of quality and quantity at the right time in the right proportions.
Several commercial weaning mixes are being marketed in India, but they are too expensive for the population of low socio-economic status, especially those in the rural areas. It is therefore imperative to formulate inexpensive weaning mixes from locally available resources, that can easily be prepared at home.
This paper discusses the evaluation of several experimental mixes made from a variety of locally available foods by traditional processing techniques.
Materials and methods
The foods used to formulate the experimental weaning mixes were jowar (Sorghum vulgare), bajra (pearl millet, Pennisetum typhoideum), wheat (Triticum aestivum), rice (Oryza saliva), Italian millet (Setaria italica), green gram (Phaseolus aureus Roxb), Bengal gram (Cicer arietinum), moth beans (Phaseolus aconitifolius Jacq), rajkeera seeds (Amaranthus panicultus), potatoes (Solanum tuberosum), and spinach leaves (Spinacia oleracea).
The four processing techniques used were roasting, malting (germination), puffing, and fermentation. Figure 1 outlines the steps that were followed for each of these techniques.
FIG. 1. Steps followed in the four processing techniques used in making the weaning mixes
A different combination of foods was used for each of the four techniques:
Each combination was tried in three different sets of proportions (table 1), which were evaluated using a four-point scale, and the proportions found most acceptable for each combination were selected for further investigation.
The organoleptic qualities, nutritive value, cost, keeping quality, and acceptability to members of the target group of each of the formulations were evaluated.
The organoleptic characteristics were assessed under controlled conditions in the laboratory. Nutritive value was assessed by determining the protein and iron content by standard methods  and by calculating the energy value using data on the nutritive value of Indian foods . The cost of the mixes was calculated on the basis of the current local prices of the raw foods and processing costs.
TABLE 1. Composition of formulated weaning mixes
|JRGP||roasting||jowar, rice, green gram, potatoes||10:5:4:1a||5:10:3:3||3:3:2:2|
|GW||malting||wheat, green gram, moth beans||3:2:0a||3:5:2||3:2:5|
|JR||puffing||jowar, rajkeera, bajra||2:1:1||1:1:1a||3:5:2|
|JIBS||fermentation||jowar, Italian millet, Bengal gram, spinach||10:4:5:1a||5:10:3:2||3:3:2:2|
The keeping quality of the mixes was ascertained by storing a set of freshly prepared mixes at room temperature and another set at 0 °C in air-tight tin containers and testing their quality at weekly intervals for four weeks.
The weaning mixes were assessed for acceptability in 25 infants from families of low socio-economic status, with each formulation being tested for one week. The mixes were given to the mothers of the infants in amounts of 100 grams, enough to feed an infant three servings a day after blending with either water or milk. The quantity consumed by each child each day, the opinions of the mothers concerning the acceptability of the mixes, and data on any side effects observed in the infants were recorded.
The results were analysed statistically by analysis-of-variance and critical-difference tests .
Results and discussion
The mean scores of the organoleptic evaluations of the four weaning mixes are presented in table 2. The highest scores for appearance and colour were given to the JRGP mix and the lowest to the JR mix. With respect to taste and flavour, however, the JR mix received the highest scores and the JIBS mix the lowest. No significant differences were found in the scores for texture. The JRGP mix may be considered superior with regard to the five organoleptic characteristics, as it received high scores in each category. Next in order was the JR mix. The GW and JIBS mixes were considered the least acceptable.
The crude protein and iron content and the energy value of the mixes are shown in table 3. Wide differences were found in crude protein (10.0-17.4 g) and in iron (5.3-12.0 mg). The differences in energy content were not as noticeable. Regardless of the differences, however, all four mixes were adequate in protein content and in energy density to meet the guidelines set by the ICMR for weaning foods .
TABLE 2. Organoleptic evaluations of the four selected weaning mixes (mean scores on a four-point scale)
a. Standard error.
b. Critical difference.
TABLE 3. Nutritive value of the four weaning mixes per 100 g
|Protein (g)||Total iron (mg)||Energy (kcal)a|
a. Calculated value
TABLE 4. Keeping quality of the four weaning mixes stored for four weeks
|Mix||Standard error||Critical difference (0. 5)|
The total cost of each mix was about 60 paise per 100 g, which is significantly higher than the recommended 25 paise for home-made weaning mixes . This higher cost is due in part to the rise in the cost of foods with inflation. However, it is still much lower than the prevailing price of commercial weaning mixes.
The mean quality scores of the mixes stored at room temperature in tin containers for four weeks did not vary significantly from those of the control samples (table 4).
The average amount of the JRGP, GW, and JR mixes consumed by infants per day was significantly more than that of the JIBS mix. This indicates that JIBS is inferior in acceptability to the other three mixes. All 25 mothers agreed that the JRGP mix was the most difficult to prepare and feed. The mixes all blended well with milk or with water, although the JR mix was the least satisfactory. No adverse side effects were noticed in the infants due to the intake of any of the weaning mixes. Thus they all were satisfactory, with JIBS being the least acceptable.
John C. Klensin
At the 14th International Congress of Nutrition in Seoul, Korea, in August 1989, the development of the International Network of Food Data Systems (INFOODS) and its regional groups, organized under the sponsorship of the United Nations University, was presented and discussed. This paper by Dr. Klensin discusses the activities of the central INFOODS secretariat, and the succeeding papers deal with the INFOODS regional associations in Africa, Asia, Europe, Latin America, the Middle East and North Africa, and Oceania.
Introduction and history
Almost from the beginning, INFOODS assumed that, while international work on and exchange of food composition data was needed and desirable, most of it should be done in regions of the world with similar food cultures, rather than centrally as a single international activity. During the last six years, regional groups have been established to cover most of the world. As earlier reports indicated, some of them, including EUROFOODS and NOAFOODS, had organizations and activities in operation before INFOODS itself was fully organized; others, such as AFROFOODS and MENAFOODS, are still in the process of holding organizational meetings. The project's initial goals and thinking were discussed in more detail in the report of the planning conference .
During its initial four years of operation, INFOODS also initiated several scientific and standardization activities to address gaps in the literature, to help to focus its own efforts, and to respond to its particular mandate to provide the technical structure for efficient and unambiguous data interchange among regions. The results of that work appear in several publications and in plans for additional work.
Publications and new results
INFOODS sponsored a conference in March 1985 to provide a better understanding of the ways in which food composition data were used and of the problems involved in storing, organizing, and processing them. This conference in many respects followed up on the organizational meeting, focusing on these specific topics. Its results were summarized in a book published two years ago .
For the scientist in search of data values from outside his or her own country or region, one of the most important tools is a good directory of what tables are available. The FAO used to publish such volumes  but stopped doing so. More recently, in-depth directories have been produced in a few regions, notably in Europe  and in North America . INFOODS has supplemented these, and provided listings for regions without their own more detailed listings, with an international directory. That directory has been published in two editions, the more recent one in September 1988 , and additional editions will be produced when the quantity of new tables and corrections requires them. Resources are being sought for a more ambitious and more computer-intensive project that would provide an international index of food composition data and databases organized by foods and nutrients, not only by geography.
Somewhat unexpectedly, differences between countries about which nutrients and other food components are included in tables and how these are analysed raise major challenges to comparisons based on these data. For example, some countries report only carotene for vitamin A, others report carotene and retinol, while still others report retinol equivalents, and so forth . Misunderstandings occur when these values are compared as if they were all the same. INFOODS first surveyed this problem to determine its extent and to understand how precise various food tables were about identifying the methods used and the chemicals being reported. The problem is quite extensive. and it is not surprising that tables differ as significantly in their degree of precision and detail about nutrient and method identification as they do about other elements of data quality. A system of identifying specific foods components was then developed based on a great deal of international participation and comment . While that system was designed primarily for purposes of data interchange, it also may be useful for organizing tables and reporting methods of analysis and calculation in printed tables. Several countries have begun to use it in these ways.
Much of the complexity and detail of the system to identify food components stems from historical differences in procedures and approaches. If we could start over today rather than using older tables and data, and could agree on a single set of standards and conventions, many of the problems would be eliminated. Unfortunately, starting over is unrealistic and will be for some time; and there are a great many very useful older data. For new tables and databases, however, and for extensive revisions of existing tables, determining how to do things well and in a generally agreed-upon fashion is more desirable than simply describing what has been done in the past without making normative judgements about those practices. INFOODS has initiated two efforts to provide guidelines for the developers of new tables and databases. The first of these, concerned primarily with methods of food selection, sampling, and analysis, is currently being prepared by Heather Greenfield and David Southgate. The second concentrates more on the organization of food composition tables and databases for specific tasks and on ways of estimating values that have not been directly measured.
The data-interchange system itself is designed to provide a standard format for transferring food composition data between regions, and also perhaps within regions and even within countries. The system is very precise and permits storing and transferring any data or descriptors of data that are available, without need for the separate introductions, usage notes, and ancillary files that have characterized the formats and organization of databases in the past.
Preparing a file for interchange involves converting it to a special structured format in which all data and all descriptions of them are specifically and precisely identified with special names or "tags." Unlike historically more common formats, this permits whatever information is associated with the file to be represented and exchanged, without requiring extra space, or data fields, for information-whether nutrients or description-that is not available for the file. The specific details of the system have been explained at conferences and are covered in a series of working papers that have been circulated widely. Those working papers and some new results have been consolidated into a lengthy technical manuscript that includes rules for syntax as well as all of the tags that are not associated with identifying food components. We expect to publish this manuscript when all the ideas can be tested together and in context, in conjunction with the implementation of the first regional data centre.
Finally, the first issue of the INFOODS-sponsored Journal of Food Composition and Analysis was published in late 1987, and subsequent issues have come out at regular intervals since then. It is now operating to a great extent independently of the secretariat. Subscriptions are available through Academic Press (Duluth, Minnesota, USA).
The next steps
From one perspective, INFOODS has had three goals. The first is to improve communication among scientists interested in food composition data and to organize effective regional groups. Partially because the time was right for regional development, and in a few cases without significant input from INFOODS staff, this process has been very successful. A periodic INFOODS newsletter and the Journal of Food Composition and Analysis also contribute to improved communications.
The second goal is to identify the scientific barriers to effective inter-country and inter-regional use of food composition data, and to eliminate or lower them. Significant progress has been made, especially in the areas of guidelines, data identification, and facilities for meaningful data interchange. In other areas, much work still remains to be done to either solve problems or at least increase the general understanding of them. The matter of food coding and nomenclature is especially important: many systems have been proposed as meeting all needs and being applicable for broad international use, and the usual assumption is that precise food identification is a problem that must be solved. This assumption has led both to good systems and serious fallacies, discussed in detail below.
Finally, we hope to provide for actual international data availability and exchange. Many of the problems in this area are now solved, as mentioned above. An up-to-date directory of tables has been completed, and several regional groups, including EUROFOODS, have developed comprehensive surveys of data available. An interchange system has been designed and tested with data from several tables from various parts of the world. The next step, and the one on which INFOODS is concentrating its efforts, involves the development and construction of operational regional centres that can maintain and exchange data and, where appropriate, act as focuses for specifically data-related activities within their regions.
Status of the secretariat
With the completion of the initial specific INFOODS core tasks and expiration of the initial core funding, the INFOODS secretariat has entered a period of consolidation and reorganization, focusing its efforts on obtaining the resources to construct regional centres in a few parts of the world. The proposals and discussions so far assume joint efforts, with the secretariat providing assistance and technical expertise but with most of the actual development work being done within the target regions by regional personnel. Fund raising efforts so far have concentrated on regions in developing areas, partially on the assumption that the more developed areas will eventually take care of themselves; however, we are looking for independent or partnership arrangements in the latter as well.
Partially as a result of this shift in emphasis, visible secretariat activities have been considerably reduced in scale. The newsletter will be published less frequently and less regularly, at least until we have more to report, and incoming letters and requests are taking longer to be answered. We apologize for any inconvenience, but these delays should not be taken as an indication that we have disappeared; we have not.
Food terminology and classification: The key question or an interesting distraction?
How foods should be named or coded in an international system has been the most controversial of the questions INFOODS has had to deal with, in part because many people seem to hold strong opinions about it. The INFOODS conclusions are perhaps somewhat radical.
The matter of food names, coding, and description is key to many ideas for comparing values among different tables and databases, as well as to food coding for dietary studies. The idea of developing a single international nomenclature system is quite seductive. Unfortunately, it does not appear to be practical, and regional systems that focus on particular goals and objectives seem to be more reasonable. Systems such as the EUROFOODS-developed EUROCODE 2  have precisely the correct properties. They reflect the foods of a relatively small number of countries with fairly similar foods, and the goals for use are well defined and limited.
As one moves away from this set of properties, systems inevitably become either more complex and less problem-specific, or very general and imprecise. They may even become unworkable. This is not a defect in any particular system that is subject to improvement by adding or changing a few terms; it is inherent in the nature of the underlying foods and varying societal assumptions about them. This is perhaps best illustrated by some examples.
For dietary assessment purposes, a food-coding system should reflect only those distinctions between products that the consumer can make in the marketplace. Additional distinctions and the attempt to elicit them in a survey are, at best, a waste of time and classification effort and, at worst, sources of scientific error. By contrast, a regulatory database may require considerable additional classification information and coding; organizing principles based only on consumer distinctions might not be useful. Adequate computer programming would permit use of the less precise parts of a very detailed system for dietary assessment without other effects, but only if the detailed system were designed using appropriate criteria for dietary assessment. No one has yet demonstrated that this can be done optimally for a general-purpose food composition database.
A multilingual thesaurus is a useful supplement to a food-coding system, and such thesauri have been constructed, primarily for agricultural products generally  and animal foods  but also for human ones. Once a system is adapted to regions with widely differing food cultures, however, a simple thesaurus that translates words of one language into words of another is no longer sufficient: it is necessary to describe the foods in great detail, comparing them to relevant local products where these exist. In the terminology of thesaurus systems , this implies carefully translated scope notes, not merely terms.
Some of the food similarities that would be required in order for it to be possible to create multilingual (and multicultural) scope notes simply do not exist; that is, there are no corresponding foods in different countries. Similarly, marketplace distinctions in one country may be different from those in others-more precise in one place, less so in others. One could easily invent a system that would permit local variations to compensate for this, but then one would have a common method for building coding systems with similar features, not a single international system.
An example from another area may help to illustrate the underlying problems. Since computer systems began to become important for processing things besides numbers, concern has been raised about eliminating the tyranny of character sets based on United States English, in which most other languages cannot be expressed correctly. After a slow start, considerable development has occurred in the registration of national graphic character sets and devices to handle them. If the registration sequences are used, one still may not be able to use a French character set on a US-developed device, but at least no assumption will be made that it is "pure" ASCII. At the same time, the hardware and software technology now permits more convenient use of multiple or extended character sets: for example. INFOODS has gradually succeeded in spelling the names of people and organizations correctly, something that just a few years ago we could do only by marking in with a pen characters that do not occur in US English.
In some cases, where the differences between marketplace distinctions in two countries involve only a higher degree of distinction in one than the other (for example, it is well known that many more distinct cheeses are identified in Denmark and France than in the United States), it appears possible to build an international system by adding hierarchic levels for those countries that will use them and allowing others to ignore them. This is useful, practical, and important, as long as its implications are understood: comparisons between the foods can be carried out only at the minimum of the level of distinction made in either country. There is no way that a coding system will invent information that does not exist, and it is improbable that the use of such a system will of itself change what manufacturers sell in the marketplace.
More serious problems arise when one group of countries has different, non-overlapping systems for naming or selling foods. The most commonly cited example is cuts of meat, but there are others.
Of course, while no coding system can eliminate simple errors, such as the identification of a highly processed cheese product as "cheddar" (J. A. Campbell, unpublished data, 1983), more complex and precise systems may be more prone to errors than simpler ones in single countries where terminology is more obvious. Descriptions of foods in prose can act as a useful check on complex classifications and can be used to resolve or approximate differences among them. If the "cheddar" were also described as '`a soft, processed cheese product with several ingredients," the potential for serious scientific error when comparing values in several tables would be reduced. On the other hand, if that description were there, it is not clear that assigning a specific name to the product would add much information, at least outside the country that produced the food table in question.
In summary, these difficulties, especially the potential for errors of classification and the lack of precisely analogous foods that makes multilingual thesauri difficult, are strong arguments that the "general food code" problem should not be the primary focus of international efforts. Instead, work should concentrate on the development, where needed, of highly specific problem-oriented codes such as the EUROCODE, and the development of good food-description procedures and checklists for more general comparison and exchange of values. If that approach is taken, the unsolvable problem of a universal food nomenclature will become moot.
Claver R. Temalilwa
Concern over the quality and quantity of data on foods consumed in Africa caused many professionals in Africa and overseas to be interested in establishing a network of food data systems for that continent. In 1987 Dr. William Rand of INFOODS requested Dr. T. N. Maletnlema, Managing Director of the Tanzania Food and Nutrition Centre, to organize a workshop to establish the Africa Network of Food Data Systems (AFROFOODS). At the third Africa Food and Nutrition Congress, held in Harare, Zimbabwe, 3-5 September 1988, a workshop was convened to discuss the quality and quantity of data on the composition of foods consumed in Africa and the accessibility of such data to users.
Participants were drawn from Africa, Europe, and North America-with representatives from Nigeria, Uganda, Ghana, Zimbabwe, Lesotho, Kenya, Malawi, Senegal, Benin, Burkina Faso, Madagascar, Ethiopia, Cameroon, Tanzania, Zambia, South Africa, Algeria, Botswana, and Swaziland. INFOODS was represented by Roslyn M. Romberg from the INFOODS secretariat, and EUROFOODS was represented by Dr. Clive West of Wageningen, Netherlands.
Six papers were presented, which covered global and regional activities related to data on food composition tables and analysis. The presentations were followed by a general discussion, at the end of which it was unanimously accepted that both the quality and quantity of data on foods consumed in Africa were deficient, and that an urgent need existed to establish a body to deal with the many aspects of the issue. The main shortfalls were seen to be (1) the lack of an inventory of food tables in Africa, (2) unacceptable quality of data in the tables in some cases (e.g. missing and incorrect nutrient values), and (3) a lack of co-ordination at both regional and sub-regional levels, leading to a lack of forums for the exchange of information.
A committee was established with 10 representatives from the sub-regional groups that make up AFROFOODS-east central and southern Africa, anglophone West Africa, francophone central Africa, North Africa, and francophone East Africa. The committee was given a mandate to deal with the following areas: (1) to take an inventory of food tables available in Africa, thereby improving access to existing and new data on food composition, (2) to improve the quality and quantity of food composition data, (3) to organize the training necessary to reach the goals of AFROFOODS, (4) to establish and strengthen forums for the exchange of information, (5) to explore sources of funding for the various AFROFOODS activities, and (6) to establish contacts with established bodies such as INFOODS, EUROFOODS, LATINFOODS, and OCEANIAFOODS.
AFROFOODS was established as an independent body for Africa, but it works within the framework of INFOODS. Its initial task is to construct food tables and data bases for local use, rather than placing too much emphasis on access to international databases.
Because of a lack of funds, the committee was unable to meet as planned in early 1989 to follow up on the recommendations made at the 1988 workshop and to draw up a five-year programme for establishing AFROFOODS. In spite of this, however, activities are in progress.
The Composition of Foods Commonly Eaten in East Africa was published in 1988 and is now being revised . The new version will be printed in both French and English, and it will also be available in poster form. Data included in the composition table are also available on a floppy disk designed for use with the MicroNap database-access software produced by Dr. Gustaaf P. Sevenhuysen of the University of Manitoba, Winnipeg, Canada. Our hope is that other sub-regional groups will make an effort to produce food tables for use in their respective regions.
Work is in progress on the inventory of food composition tables available in Africa. AFROFOODS has started participating in international forums.
Almost all the countries in Africa have established laboratory facilities and personnel to carry out various food analyses. Too often, however, laboratory work is hindered by malfunctioning equipment. Equipment repair and maintenance are normally big problems. According to a survey of 35 institutions, conducted by the International Foundation for Science (IFS) in the developing countries, 340 researchers had a total of only 600 pieces of crucial research equipment, 35% of which were defective. In response to the requests of these scientists, the IFS recently initiated a pilot project in the Southern Africa Development Coordination Conference (SADCC) countries (Angola, Botswana, Lesotho, Malawi, Mozambique, Swaziland, Tanzania, Zambia, and Zimbabwe) which approaches the equipment problem from the viewpoint of researchers and technicians, with the primary goal of assisting a selected group of researchers to make maximum use of their equipment by providing technical documentation (technical sheets, users' manuals), spare parts, supplementary equipment (e.g. to improve water quality), repair services for key equipment, and training of local scientific and technical staff. The accumulated experience will be useful to the rest of the continent as well as to other countries in the developing world.
A lack of funds to carry out the various activities is another serious constraint. The AFROFOODS development programme needs money to hold forums, publish a newsletter, compile food composition tables, and so on. As of now, no monies are available for any of the planned activities. We are, therefore, appealing to international organizations and governments to support AFROFOODS in any way possible.
Contents - Previous - Next