Contents - Previous - Next

This is the old United Nations University website. Visit the new site at

Urban malnutrition

Urban nutrition in the tropics: a call for increased attention to metropolitan population
in the developing world

Food aid as communal meals for the urban poor: the Comedor programme in Peru


Urban nutrition in the tropics: a call for increased attention to metropolitan population
in the developing world

Noel W. Solomons
INCAP, Guatemala City, Guatemala

Rainer Gross
Federal University of Rio de Janeiro, Brazil

The origins of the priorities expressed in world-wide nutritional activities are diverse. In the area of nutritional investigation, they can be set by tradition, by the emergence of new and interesting technology, or by trends and fashions in biological concepts within the professional community. For programmes of public health (diagnostic screening and intervention) in nutrition, policy-priority decisions are based on perceived relevance, political expediency for the national government, the global strategies of international development agencies, or simply accumulated habit. It is our perception that nutritional activities have traditionally been invested, and continue to be invested, in rural problems and rural populations. It is our premise, based on the obvious demographic shifts and the realities of current population distribution in developing countries, that an imbalance in the focus of efforts and interests has developed. This is reflected in the persistence of an inordinate concentration on and concern with rural nutritional issues and the relative neglect of urban problems.

It is not necessary to document here the pattern of emphasis on rural nutrition in developing countries since the end of World War II and the founding of the United Nations. Nor is it our purpose to deny the importance of studying and reaching the agrarian peasant, pastoral nomad, or tribal groups who truly constitute the world's "poorest of the poor." Rather, we focus on some new facts that should be factored into the nutrition equation for developing countries. In 1950, 47 per cent of the developed and 83 per cent of the developing world's people lived in rural areas; at that time there were no cities anywhere with more than 10 million inhabitants, and the grand capitals of industrialized nations-Tokyo, London, New York, Paris -were the great metropolises of the day. By the year 2000, it is projected that more than half the world's population will be living in urban or pert-urban (metropolitan) areas. In Latin America at the turn of the millennium, 80 per cent of the population will be metropolitan. Today Mexico City and Sao Paulo are the giant cities, and by the year 2000 these and nine others-aft in the developing world-will have more than 20 million residents each.

Implicit in this demographic reality is the potential for problems of under-nutrition, over-nutrition, and nutrient imbalance. The leading edge of this phenomenon has already been documented in the context of the rapid change in food habits of Mexican children and adolescents migrating from rural pueblos to the Distrito Federal [1]. The contribution - of soft drinks and convenience (junk food) snacks to daily intake increased dramatically. A calorie-dense, micronutrient-poor diet was the immediate result.

Ironically, two computerized software packets recently prepared for nutritional survey work in developing countries classify the grades of under nutrition from anthropometric data but do not have a parallel capacity to classify obesity

The world-wide nutrition community is showing increasing interest in urban nutrition issues. For example, the Dutch Foundation for the Advancement of the Knowledge of the Nutrition of Mother and Child in Developing Countries announced as the theme for its grant-in-aid programme for 1986-1987 "Nutrition of preschool children (O to 3 years) in urban communities." The Food and Agriculture Organization of the United Nations has established a programme concerned with the food supply to large cities, the first project of which was established in Sao Paulo, Brazil, in 1985. A recent book [2] cites 115 publications over the past two decades-review articles and original investigations-related to urban malnutrition in developing countries. In August 1985 at the thirteenth International Congress on Nutrition, held in Brighton, England, a workshop on tropical urban nutrition received outstanding interest and participation. In October of the same year, the first seminar on research on urban nutrition in Brazil brought participants from Sao Paula, Brasilia, and Rio de Janeiro together in Rio to share experiences and plan future collaborations. Noteworthy was the positive attitude manifested by officials of the Brazilian government.

What is now needed is not an abandoning of attention to rural malnutrition but a changing balance to reflect the growing relative importance of urban nutrition in the tropics.

The need to enlarge the circle of concerned nutritionists from both the developed and the developing world has motivated these comments. Input from the disciplines of both experimental and applied nutrition will determine the movement toward greater recognition of and attention to the nutrition of metropolitan populations in developing nations.

The biases against working in urban areas must be addressed. it is certainly not tenable to argue that the issue of urban nutrition lacks relevance. Fear of hostility and aggression from city slum dwellers and frustration with the demographic instability of the populations studied must be confronted and examined. The inclination of many nutritionists to be preoccupied with the most economically disadvantaged and under-fed populations in developing countries should be moderated: as urbanization generates a middle class, diet-related aspects of their life-style deserve attention. Nutritionists must also concern themselves with the ways of cities and urban populations: collaboration with professionals traditionally concerned with cities, sociologists and urban planners, should be sought. National and international meetings on nutrition and health should include issues of urban nutrition in colloquia, seminars, and other forums.

Finally, support must be sought and provided for research on public-health interventions and scientific research in urban areas of developing countries. The 115 citations found by Schurch and Favre 12] represent an average of slightly fewer than six publications per year over two decades. This production needs to be increased dramatically during the next five to ten years.


1. M.T. Cerqueira, "Effects of Urbanization and Acculturation on Food Habits: Studies in Mexico," in P.L. White and N. Selvey, eds., Malnutrition: Determinants and Consequences (Alan R. Liss, New York, 1984).

2. B. Schürch and A.-M. Favre, Urbanization and Nutrition in the Cord World (Nestle Foundation for the Study of Problems of Nutrition in the World, Lausanne, Switzerland, 1985.

Food aid as communal meals for the urban poor: the comedor programme in Peru

Judit Katona-Apte

Food-aid commodities have been traditionally used for the benefit of urban populations through various kinds of programmes such as food-for-work, maternal-and-child-health, vulnerable-group-feeding, institutional feeding, and child-feeding projects. Some of these are specifically targeted at recipients in urban areas and provide cooked meals to be consumed on site.

There are many disadvantages of such food-aid programmes from the perspective of both the community and the household. Child-feeding projects, for example, are beneficial only to those children targeted and only if the food provided is nutritionally superior to whatever is displaced from their usual diet [1]. An institutional feeding project is available only to those residing in the specific institution (e.g., hospital, orphanage, or boarding school) covered, and the food-aid commodities provide only a small portion of the needed nutrition.

In addition to meals received through food-aid programmes, a considerable proportion of the cooked food ingested by working persons, usually males, in urban areas may be purchased from shops and roadside stands. Such meals also have negative aspects; for example, only these adults receive improved quantities of energy and nutrients, and not the rest of the household. Consuming meals outside the home also keeps the family from eating together, and meal times may be the only occasions for parents to interact with their young children.

This paper was prepared for the Symposium on Urbanization and Nutrition, thirteenth session of the ACC Sub-committee on Nutrition, held 2-6 March 1987 in Washington, D.C.


Communal meals and sharing the task of meal preparation have been in existence since the beginnings of humanity. Hunter/gatherers shared the spoils of the hunt [2-4]; feasts still are often community-wide events and include sharing the consumption and often also the preparation of meals [5, 6].

An interesting institutionalized food-aid programme in which cooked meals are shared is the comedores project in Lima, Peru. The food-aid commodities (from USAID and the EEC) are given to a group of women who make up the comedor, or "dining room," which is a co-operative venture. Participating women take turns at preparing morning and midday meals. Standardized portions are then-sold at a set price to participants, who bring their own containers.

Each participant is entitled to receive a certain number of servings, presumably equal to or less than the size of her household. The money collected is used to purchase additional foods and ingredients.

The contents of the meals differ depending on the amount charged; comedores adjust their food purchases and preparations to the income level of the population they serve. Very poor members of the locality or community receive all their meals free; other participating households pay for some meals and not for others. There appears to be a limit to the number of free meals each comedor feels it can support.

From the food-aid perspective this is a closed monetized system [7] because the commodities are sold, in cooked form, to the recipients at a subsidized price. The benefits are targeted at the household rather than at any specific groups of members within the household. The most obvious advantage of a Comedor is economic: it is less expensive for each household to obtain its meals in this manner than to prepare them individually at home. Another advantage is that such meals are likely to be nutritionally superior to the usual intake, partly because the food aid commodities are often more nutritious than the foods they replace, and partly because more and better-quality foods can be purchased by a group than by an individual household due to the advantages of bulk buying, consumer clout, pooled knowledge, and so forth.

Ideally, the total amount of time spent in food preparation per week should be less for participating than non-participating households even if the meals brought home have to be warmed up and even if the women have to put time into working at the comedor. Other potential advantages should also not be overlooked: women have an opportunity to interact and socialize during meal preparation and can feel positive about the fact that they are helping to provide for members of the community who cannot afford to pay for meals.

There can be disadvantages as well. Meals prepared for a group may not necessarily taste the way a household would like them. Preferences for certain staples and other items cannot be met. A very important factor is that control over what foods to eat and how they should be prepared may be removed from the hands of the person in charge of food preparation.


In 1985 there was a unique opportunity to obtain in-depth data on eight households participating in a Lima food-for-work project sponsored by the United States Agency for International Development [8]. The data compiled included 240 sets of 24-hour household food records covering four households participating in comedores and four non-participating households, each for 30 days. Information was obtained regarding a number of other food-related behaviours, such as acquisition, expenditure, dispensation, and preparation techniques. The comedores received bulgur, non-fat dry milk, oats, vegetable oil, and wheat flour. From an analysis of the data we may draw the following conclusions.

TABLE 1. Average number of different food items consumed per day by members of comedor-participating and non-participating households classified by age and sex

  Food items per day
Comedor households Non-comedor
Adult males(over 16 years) 7.0 3.5
Adult females(over 16 years) 8.0 7.7
11-16 years 9.0 7.5
5-10 years 9.0 8.0
1-4 years 8.7 8.2

Households participating in the comedores programme had more elaborate meals: they consumed a greater number of food items per day (table 1), ate more total food items (table 2), and ate more frequently (table 3) than non-participating house

TABLE 2. Average total number of food items consumed in 30 days by members of comedor-participating households classified by age and sex

  Total food items
Comedor households Non-comedor
Adult males(over 16 years) 665 265
Adult females(over 16 years) 721 534
11-16 years 679 578
5-10 years 865 507
1-4 years 853 674
Nursing women 741 510

TABLE 3. Average total number of food eventsa in 30 days for members of comedor-participating and non-participating households classified by age and sex

  Food events
Adult males(over 16 years) 80 58
Adult females(over 16 years) 102 100
11-16 years 99 112
5-10 years 123 100
1-4 years 126 121
Nursing women 129 99

a. A food event is defined as the consumption of any food or foods at any time by one or more persons. The term is to be distinguished from meal which has connotations of specific types of foods or specific times of day.

TABLE 4. Cost of meals obtained from comedores

Household Cost per meal (solesa) % of meals received free Total value of free meals (soles)
Meals paid for All mealsb
1 2,792 532 81 857,144  
2 3,713 1,107 70 1,507,478  
3 1,500 90 94 394,800  
4 2,466 1,349 45 488,268  

a. US$1 = 13,943 soles on 31 Dec. 1985.
b. Average cost for all meals, including free ones. holds. This can be interpreted to suggest improved nutritional intake.

No significant income differences between participating and non-participating households can be discerned in this small sample. Incomes ranged from less than US$300 to US$1,780.

Participating households did not necessarily spend less total money or a smaller proportion of their effective income on food than non-participating households. One participating household, however, spent only 36 per cent of its income on food and received the greatest benefit from the comedor.

The monetary value of the free meals to those who received them from a comedor was considerable; each household received more than its total income if the value of the free meals is calculated on the basis of the meals that were paid for (see table 4). This is not to be considered displaced income, because none of the households would have spent this amount to prepare meals at home.

TABLE 5. Number of servings and meals received from comedors and their costs

Household Servings Mealsa Cost (soles)
1 380 68.5 33,500
2 580 130.3 115,100
3 280 43.4 71,500
4 440 96.7 7,500

a. To adjust for household size and composition, the following weightings (based on INCAP requirements) were used to arrive at average: adult males = 1; ages 11-16 years = 1; lactating women = 0.95; pregnant women = 0.9; adult females = 0.8; ages 5-10 years = 0.65; ages 1-4 years = 0.45; nursing children 1-4 years = 0.35; age under 1 year = 0.30; nursing children under 1 year = 0.2. This provided the following household weightings: household 1 = 5.55, household 2 = 4.45, household 3 = 6.45, household 4 = 4.55.

Meals received from comedores were used for more than one food event and thus had to be warmed up. They were often also used as part of a meal the rest of which was prepared at home.

Both participating and non-participating households purchased some of their meals from outside sources. It appears that this is a common practice in this society.

Adult males in participating households ate at home more frequently than those in non-participating ones (see table 2). This probably provided opportunities for interaction among family members.


Organized kitchens such as those in the comedores programme in Peru appear to have a number of advantages. First, they appear to be a cost-effective way to improve the dietary intake of the total household [9]. Cost effectiveness applies to both the food donors and the food recipients. (It is measured by V/K, where V=value to the recipient and K=cost. The value to the recipient can be calculated both as potential monetary savings to the household and as additional nutrients received.)

Second, the nutritional status of participating households is likely to be improved because the variety of foods ingested is greater. Quantity of food did not appear to be a problem in Lima, only the nutrient content of the diet.

Third, participating households realized no monetary savings. The sample size was limited, but it appears that food expenditures did not necessarily diminish as a result of participation, although the monetary value of the meals received was substantial. This can be interpreted to mean that people tend to pay whatever they can to the comedor in order to maintain their self-esteem.

Finally, it is difficult to ascertain if participation in such a programme saves time for the household. In the four participating households 186 hours were spent preparing food, compared with 331 hours in the four non-participating households. However, at least one participating household spent more time preparing food at home-including time spent warming up food from the comedor-than did two non-participating ones. Household preparation time also does not correspond directly to the weighted average number of meals per household (table 5). In addition, the participants provided labour in the comedor.

The opportunity cost of participating in this programme has generally not been considered, and no information about it is available. (Opportunity cost in this case refers to the cost to the individual of having to forgo other activities that might have been undertaken instead in the same time.) Yet these women are employed in one or more income-earning activities; in fact the US food-aid commodities are distributed in the context of a food-for-work project for women.


On the basis of information available from consumption surveys in several developing countries, it can be concluded that low-income urban households buy a significant amount of prepared food from outside commercial sources [101. As the comedor programme provides a number of advantages, it should be instituted on an experimental basis among the urban poor in all areas where food-aid programmes exist and where it is feasible. The nutritional advantages are obvious.

The major consumers of purchased ready-cooked foods at present are working males. A comedor-type programme provides nutritional advantages to the important vulnerable group of women and children as well, and communal meals are a cost-effective way to use food-aid commodities to improve the diet.

In addition to the nutritional benefits, such programmes have psychological advantages. Not the least of these is that they provide opportunities for women to organize and take charge or a leading role in those aspects of life they consider important, such as feeding their households. Also, receiving cooked food for the total household promotes consuming meals together at home. Eating together in the home provides psychological advantages that are important for the growth and development of children.


I wish to thank the ACC/SCN, and more specifically Drs. A. Horwitz and J. Mason; I also thank Dr. B. Popkin for encouragement and Dr. M. Apte for helpful comments.


1. Food and Nutrition Board, National Academy of Sciences, Nutritional Evaluation of PL 480 Title II Commodities (National Academy of Sciences, Washington, D.C., 1982).

2. L. Marshall, "Sharing, Talking, and Giving, Relief of Social Tensions among the !Kung," in R.B. Lee and 1. Devore, eds., Kalahari Hunter-Gatherers (Harvard University Press, Cambridge, Mass., USA, 1976.

3. J. Woodburn, "An Introduction to Hadza Ecology," in R.B. Lee and 1. Devore, eds., Man the Hunter (Aldine Publishing Co., Chicago, 1968).

4. A. Balikci, "The Natsilik Eskimos: Adaptive Processes," in R.B. Lee and 1. DeVore, eds., Man the Hunter (AIdine Publishing Co., Chicago, 1968).

5. A. Graham, "Let's Eat! Commitment and Communion in Cooperative Households," in M.O. Jones, B. Giuliano, and R. Krell, eds., Foodways and Eating Habits: Directions for Research (California Folklore Society, Los Angeles, 1983).

6. L.K. Brown and K. Mussell, eds., Foodways in the United States (University of Tennessee Press, Knoxville, Tenn., USA, 1984).

7. J. Katona-Apte, A Study of Commodity Appropriateness and Pricing in a Closed Monetized System: WFP Project India 2303 (World Food Programme, Rome, 19851.

8. J. Katona-Apte, Peru Household Consumption and Related Effects (US Agency for International Development, Washington, D.C., 1986).

9. S. Reutlinger and J. Katona-Apte, "The Nutritional Impact of Food Aid: Criteria for the Selection of Cost-Effective Foods," Food and Nutrition Bulletin, 6 (4): 1-10 (1984).

10. B.M. Popkin and E. Bisgrove, "Urbanization and Nutrition in Low-Income Countries,'' Food and Nutrition Bulletin, forthcoming.

Contents - Previous - Next