Discussion of "interventions" should distinguish between projects and movements. Ideally, projects are of limited duration-they have a starting point and a termination date. Projects are planned in advance and may or may not be open to adjustment while under way. Projects are usually dependent on resources from outside the project area-material as well as human resources. Problems as well as solutions are usually defined by well-meaning outsiders, not by those with the problems.

Ideally, movements are first and foremost characterized by control over actions by those who have the problem. The ideals) behind a movement may well come from "outside" the movement itself, but the ideas take time to grow and mature and become adjusted to the local reality. Those with the problem formulate their views about problems, priorities, and solutions. There is no fixed plan available a long time in advance, although good movements need both planning and analysis; movement leaders first have to use intuition to feel their way forward. Intuition is an important tool for understanding the exceedingly complex situation of which problems taken up by a movement are usually a part. This requires an intimate knowledge of both the problems and their causes. Movement leaders need to understand not only what is said and expressed openly but also what is behind the words and under the surface.

Movements take their own time, and have no defined time frame. Their starting point may be difficult to perceive, and so may their end. Either they die out, or they are suppressed (as is often the case), or they grow and lead to social change. The latter may be the most important aspect of movements as against projects: the much greater potential of movements for social change.

Systematic consideration of health and nutrition in agricultural and rural development programmes and projects

Fred T. Sai
Inter-regional Co-ordinator for Africa and Europe, World Hunger Programme, The United Nations University

One major dilemma in social and economic development is how to overcome the gulf in wealth between the developed and the developing countries and the in some ways even greater disparity among groups within the developing nations. In some areas poverty is so extreme as to make it impossible to conceive how those affected can have any higher aspirations than day-to-day existence against terrible odds. Such abject poverty is accompanied by many ills that help to maintain the status of poverty or even worsen it. Among the more glaring ills are poor health, poor nutrition, poor housing, poor environmental sanitation, and a generally low degree of security in life. The intractable problem of individual and community poverty, particularly as found in many rural areas of Africa, Asia, and Latin America, is both an impediment to the development of those areas and also a challenge to organized development efforts.

In the past, development projects and programmes meant to benefit nations and groups have mainly addressed economic issues. Many have been income-generating, while others have been geared to increased production; specific attention to what was happening to the rural poor was often lacking, and there were no explicit social goals.

Even where programmes and projects have been addressed to the rural areas, they have not confronted health and nutrition in a systematic way. Even specific health and nutrition projects may be so narrowly based and so ill-suited to the social and cultural realities as to have no lasting benefit for the rural poor. Many health projects have omitted any provison for nutrition. Food aid projects, ostensibly addressing an aspect of the nutritional problem, may aggravate it by their effect on production and the fragile rural economy. An over-enthusiastic concern for one nutrient (for example, protein or calories) may so distort the patterns of food production and consumption as to lead to a lowering in nutritional status rather than an improvement. It is necessary therefore to take a comprehensive approach to the analysis of programmes and projects meant to assist the rural poor.

From the point of view of the individual and for the rural poor as a whole, the objective of development as stated in the 1970 Development Strategy may still be our guide:

The ultimate objective of development must be to bring about sustained improvement in the well-being of the individual and bestow benefits on all. If undue privileges, extremes of wealth and social injustice persist, then development fails in its essential purpose. It is essential to bring about a more editable distribution of income and wealth for promoting both social justice and efficiency of production, to raise substantially the level of employment, to achieve a greater degree of income security, to expand and improve facilities for education, health, nutrition, housing and social welfare, and to safeguard the environment. Thus qualitative and structural changes in the society must go hand in with rapid economic growth, and existing disparities- regional, sectoral and social -should be substantially reduced. These objectives are both determining factors and end results of development; they should therefore be viewed as integrated parts of the same dynamic process and would require a unified approach.*

Even small-scale development projects and programmes meant to benefit the poor should be examined in relation to a set of objectives that take the above into account. It would be anti-development if a programme were to help a section of the poor to obtain a hold on the means of production and rapidly become wealthy at the expense of their neighbours. It is in this light, too, that one should examine such programmes and projects for their likely influence on, and contribution to, health and nutrition.

Nutrition is the basis of good health, and good health also contributes to the maintenance of good nutrition. It is therefore necessary to look at health and nutritional issues as an interrelated whole. For a complete and comprehensive analysis, the two must be taken jointly. The following are some of the social, health, and nutrition considerations that should be included in agricultural projects.

DEMOGRAPHIC SITUATION

The health and nutritional status of any population is related intimately to some basic demographic variables. Thus it is essential to examine what influence a programme will have on major demographic factors. What will the programme and project do to the growth rate and the spatial distribution of the population, and how will these affect demand and supply of foods and of the health services generally? A programme that creates centres of industrialization and urbanization may draw in the more active of the rural poor and depress agricultural output of the areas it is meant to help. On the other hand, with prior consideration for requirements, steps could be taken to form an interdependent series of projects within the programme to ensure increased production of food for the poor rural sector. It is important to realize that many organized agricultural programmes and projects concentrate on the male and alienate females from their traditional roles in agriculture. Resettlement schemes have been known to be so ill-conceived that they result in women losing their livelihood. This can also happen in efforts to make food crop production an organized, profitable business. Capitalization and financial constraints may transfer production to males exclusively.

SOCIAL AND CULTURAL STRUCTURES AND VALUES

In rural areas, and especially among agrarian societies, work and general styles of living are intricately mixed. Social structures are woven around work and other activities. Family types and relationships, the distribution of foods, eating patterns and habits-including consumption of prepared foods within and among families-are all dictated by complex cultural and social practices that may have existed for generations. These same structures control health practices and health decision-making. Some of these practices help group cohesiveness and may ensure reasonable management of food resources as well as the general health of the group as a whole.

The influence of any development programme on social and cultural structures, particularly those that relate to health, food, and nutrition must be carefully assessed. If women are given work outside the home, who tends to the household garden, prepares the family meals, and cares for the young? In particular, how will this affect the pregnant women, lactation, and breast-feeding, without which infant survival among the poor cannot be guaranteed A programme that incorporates women should therefore make adequate provision to minimise conflict between economics and their social and biological roles.

Even a project helping men to improve their income and providing food for them at their place of work may affect family food consumption in a negative way. Well-fed fathers may partially neglect their family responsibilities. In many rural areas of Africa, the consumption of alcoholic liquor is largely social. However, with modernization come beer bars. Even in rural areas money-earning heads of households have been known to spend disproportionate amounts on alcohol to the detriment of their families.

INCOME

In general, those earning good incomes are better able to look after their health and obtain health services than are the poor. Peasant populations increasingly need cash income. The feeling that peasant populations fed reasonably well on their own produce can have little use for money as far as food supply is concerned is of only limited validity. Income-however slight the modernization that has taken place among the poor-is of considerable importance to food intake as indeed it is in tiding them over shortages caused by adverse climatic conditions. Some studies undertaken in Ghana indicate that rural populations may need to purchase between 25 and 50 per cent of their foods. The foods usually purchased are the more expensive and nutritious components in the diet.

Even the rural poor now have some need for basic cash expenditures that become increasingly burdensome- education and clothing of children, fuel and transportation. Any increase in income will therefore play a major role in the nutritional status of the rural poor. Projects that help to generate income need to be accompanied by continuous education in household budgeting and in the requirements for good nutrition. Below a certain income level, food may enjoy high priority, but, when income reaches a point at which certain symbols of modernity can be acquired, these may be purchased at the expense of the food budget.

FOOD PRODUCTION AND DISTRIBUTION

Peasant societies generally mix their crops in such a fashion as to make nutritional and even economic sense, especially if enough land is available. Such problems as exist relate to land tenure, to the increasing partitioning of land under population pressures, or to the adherence to customs that were valuable in days long past.

A rough assessment of nutritional adequacy or inadequacy of a peasant group can be made on the basis of a study of the cropping patterns, the types and characteristics of foods produced, and land distribution.

Any project or programme to benefit the rural poor should be looked at from the point of view of its impact on cropping patterns, on food production and distribution, and, therefore, on nutrition. For example, insistence on monoculture has been detrimental in some situations, or cereal varieties have been introduced that populations would not use because the taste or other characteristics of these varieties made them ill-suited to traditional preparations.

Recently the introduction of cassava to help fill the calorie gap and to provide a surplus for export has been advocated. This may make economic sense, but there are many caveats. How will the introduction of this crop, which is relatively easy to grow and to store, affect the production of the more nutritious cereals? How and in what proportions will it enter the diet-in particular the diet of infants and pre-school children? Finally, will there be enough education about its use to avoid both acute and chronic toxicity? How will the cultivation of the "lazy farmer's crop" affect social organizations?

Another area of major concern should be the nutritional role of food items such as fruits and vegetables that are gathered in the wild. Will the project or programme destroy these to any great extent? If it does, then provision will have to be made for deliberate cultivation of substitutes and the population must be taught how to use them.

FOOD CONSUMPTION AND NUTRIENT INTAKE

A programme or project may affect the type, quality, and quantity of foods entering the home and thereby affect the composition of diets and the nutrient intake of the population. The classic example of a rise in beriberi with the introduction of dry milling of rice in East Asia is a constant reminder of possible pitfalls. For some reason, with increasing modernization, green vegetables disappear from many diets. If other sources of vitamin A are expensive or unobtainable, vitamin A deficiency may increase among a population that, on the surface, should be doing well. Changes in the method and frequency of feeding children may aggravate protein/calorie malnutrition. Many rural communities, especially poor ones, are generally on the border level with regard to intake of iron, and deficiencies can therefore be created relatively easily. Iron deficiency due to poor nutrition is fairly frequent among pregnant and nursing women of such communities as well as in very young children, and it is desirable to include monitoring of this particular deficiency in any nutritional surveillance. In parts of Africa, folic acid deficiency is fairly common and is aggravated by the presence of malaria. Here again, pregnant women are particularly vulnerable. Therefore, when changes take place in the supply and availability of food, it will be necessary to ensure that foods rich in folic acid are made available to especially vulnerable groups. Among the B group of vitamins, thiamine and riboflavin deficiency may create similar problems. Particular attention needs to be paid to the thiamine intake of rice-eating communities. Among many rural poor, the intake of riboflavin is marginal, and under any kind of stress, such as pregnancy and lactation, deficiencies may occur. Signs and symptoms of these deficiencies can also be observed relatively simply.

THE NUTRITIONAL STATUS OF THE POPULATION

It would perhaps be unnecessary to require that a comprehensive study of the nutritional status of the population be undertaken before mounting a project or programme or reorienting agriculture generally. Instead, a nutritional profile based on knowledge of some of the above issues and a few major criteria discussed below should provide a reasonable basis for assessing the present status of the population to be affected by the project and the likely impact of the project on it, as well as potential nutritional problems to be encountered within a programme. The profile would also provide a basis for monitoring any changes that take place during and subsequent to the implementation of the project. The assessment of nutritional status is a complex undertaking if comprehensively done; it is time-consuming and expensive. However, among many rural poor populations, a nutritional profile can be developed using a few indicators-with little extra investment from the programme or project-that will provide the basis for assessment of the project's nutritional impact and potential problems.

OTHER HEALTH CONSIDERATIONS

For many rural communities, agricultural practices are intimately related to health and disease patterns. The housing of agriculture labour forces and the location of villages in relation to water supplies and resources are issues of importance when planning agricultural programmes. Water management in relation to environmental sanitation and the role of health services should be carefully considered.

Water and Environmental Sanitation

Water management is a very important consideration for agriculture as well as for general health and nutrition in any agricultural programmes, therefore, the effects on the sources of water available to the population should be carefully studied. Apart from possible decreases in quantity, the even more important factor of biological and chemical pollution of water should be kept in view. Irrigation projects have been notorious for aggravating health hazards such as helminthic infestations (e.g., bilharzia and hookworm).

By changing the ecological balance, dams and irrigation projects may also increase the breeding of some disease vectors-the black fly (Simulium) and some species of mosquitoes are a case in point. The Volta Lake in Ghana has become a major source of schistosomiasis and of simulium, both of them nuisances and sources of ill health.

Health Services

The provision of facilities and services for health care in the community is an important consideration in any agricultural or rural development programme. There is no need to set up complex services that the people will not use. It is more important to help them create their own participatory health care system. The People's Republic of China has perhaps the best integrated health care system for agricultural communities in its Commune Health Care programmes. No one is more than a few minutes away from first-level care. Immunizations, antenatal care, and all basic public health measures are readily accessible at minimal cost.

It is important to make people responsible for their own health and nutrition education and service programmes. In this respect, the community's own structures, adapted if necessary, assume major responsibility in the programmes.

ASSESSMENT OF HEALTH IMPACT

Programmes meant to benefit rural populations can be assessed through a few indicators that have been proved to be good pointers to any changes in social health and nutritional status.

Infant Mortality Rates general health status of a population. A good rural development or agricultural programme that takes care of people will lead to a lowering of these rates.

Selected Disease Morbidity and Mortality

To monitor the impact of an agricultural programme on the population, it may be useful to identify those diseases whose incidence and prevalence can be influenced by alterations in agricultural practices and methods of settlement. These include infectious and communicable diseases and helminthic infestations.

Specific impact on women may be monitored through their pregnancy performance and the problems that arise, including what happens to their infants.

ASSESSMENT OF IMPACT ON NUTRITION

To assess nutritional impact, some or all of the following indicators can be recommended:

Age-Specific Mortality Rates

Studies of death rates of infants and young children give a relatively useful indication of the nutritional problems within a community. In their work on mortality among infants and children in Latin America, Puffer and Serano found that 5 per cent of all deaths in children under five years of age were due to malnutrition. In 55 per cent malnutrition was a major associated cause of death.* If malnutrition has some contribution to make to two thirds of deaths in these age groups, then the rates and changes in the rates over a period of time can be used as an indicator for changes in nutritional status.

Low Birth Weight and Prematurity

It has now been shown that the proportion of "low-birthweight babies" as well as prematurely is high among malnourished populations. A serial analysis of these indicators over time should give some idea of the impact of a project on the nutrition of at least women in the fertile age groups.

Mortality Registration from Specific Nutritional Diseases

Most countries have mortality registration. However, only in a very few developing countries can one place any confidence in the returns. The people who make the diagnoses are often not well trained, and efforts at analysing these returns are rather perfunctory. A great deal of information, however, can be obtained from such registration if training and supervision are improved and registration of major symptoms is encouraged. In the case of nutritional diseases, registration by both actual disease and deficiency symptoms can be performed, and these will give indirect evidence of the extent of the nutrition problem. This evidence has to be considered indirect because of the restrictions in the basis for diagnosis and the training of personnel. Where properly trained personnel are available to collect evidence of mortality from nutritional disease, then such evidence should be taken as direct evidence. Changes in these death rates with time are a significant indicator of what is happening within the population.

Direct Indicators of Nutritional Status

Incidence and prevalence of signs and symptoms of nutritional deficiency and their change over time should be noted. The incidence of marasmus and kwashiorkor, the most severe forms of protein/calorie malnutrition, should always be noted, because changes in the incidence of these diseases can be indicative either of deterioration in the health and nutritional status of the population as a whole or of changes in feeding practices of children. The incidence and prevalence of keratomalacia and xerophthalmia are used to assess the adequacy or inadequacy of vitamin A, Cheilosis, angular stomatitis, and changes in the tongue signify a deficiency of a member of the vitamin B group. These are seen most often in pregnant and nursing women and in young children and school children.

Growth and Development Studies

Growth studies of children from birth to five years old as well as annual cross-sectional height, weight, and arm circumference measurements in school children are good indicators of their nutritional status as well as of what is happening in the community as a whole.

CONCLUSION

In attempting to present an approach to a systematic consideration of health and nutrition in development programmes and projects aimed at the poor, one is acutely aware of the complexity of the subject. There is also the fact that-with some very few outstanding exceptions-no such work has been undertaken in any systematic way. Guidelines are badly needed to help both the ordinary programme planner and his health and nutrition advisors, if there are any. Considerations of nutrition must enter into programmes and projects at the pre-planning stage and be made evident in all subsequent phases.

If the guideline for food and nutrition prepared for the Food Policy Division of the FAO (unpublished internal FAO document) is found satisfactory in field trials, it will be a major step forward in efforts to analyse projects in terms of an important aspect of human development- adequate nutrition. A similar, even more comprehensive, exercise needs to be undertaken for the health field generally.

continue

Implementation of a conceptual scheme for improving the nutritional status of the rural poor in Thailand

Sakorn Dhanamitta, M.D., D.Sc., Suwanee Virojailee, M.S., and Aree Valyasevi, M.D., D.Sc.
Institute of Nutrition, Mahidol University, Bangkok, Thailand

INTRODUCTION

Thailand is located in South-East Asia, and has a population of approximately 47 million (1). The population growth is 2.3 per cent a year, which ranks among the highest in the world. About 85 per cent of the population live in the rural areas, are in poor health, and have a low nutritional status, especially infants and pre-school children and pregnant and lactating mothers.

During the years covered by the past three national development plans, nutrition has not been emphasized and was left to the health sector, which showed minimal involvement in agricultural, educational, and community development. Therefore, nutrition programmes for the past twenty years have been rather discouraging. In our Fourth National Development Plan, however, priority has been given to improving health and nutrition, with programme implementation carried out through the health care network at the peripheral or village level. Nevertheless, an effective model for integration of nutrition into the health care service has not yet been established. At the same time, it has also been recognized that, in order to improve the nutritional status of the villagers, it is essential to integrate health care, agriculture, and income generation into the nutrition package.

In an attempt to improve the nutritional status of the rural poor in Thailand, the Institute of Nutrition, Mahidol University (INMU), implemented a pilot project during the years 19761980 in the village of Nong-Hai. This village is part of Ubon Province and is located in the northeast of Thailand. The project initially followed a conceptual framework, which was later developed into a conceptual model. Infant supplementary food formulas, which were formulated by the institute, played an essential role in the implementation of the project. (The formulation of the supplementary foods is further discussed in the article by Tontisirin et al. on p. 00 of this issue of the Food and Nutrition Bulletin. )

The objectives of the project were:

a. to establish an integrated nutrition improvement programme utilizing a community approach model;
b. to integrate the model into the existing health care network;
c. to look for a suitable model which can be replicated for nation-wide use.

CONCEPTUAL FRAMEWORK

The project was implemented on the basis of the concept that good health and good nutrition among the people of a community will help boost agriculture and food production; that proper vocational training will promote income generation; and that these together will lead to an improved economy and result in a more advanced level of development in the community as a whole. The schematic flow of this concept is shown in figure 1.

PROJECT PLANNING AND IMPLEMENTATION

The basic principle of the project lies in obtaining maximum community participation with existing available resources. Since this project integrated nutrition, health, and community development, co-ordination with the district health officer, agricultural officer, and the community development worker was necessary in order to plan and implement the intervention programme. The principal personnel required for this project were a field researcher and a field implementor. The field researcher performed such duties as data collection, constant evaluation of project outcome, and motivation and support of the villagers on programme activities. The field implementor, who carried out the major activities of the project, was a midwife or nurse already stationed in the village by the government. An established, good rapport with the village people was indispensable before any behavioural changes could be made. A village centre was built in the village for purposes of programme implementation and as a gathering place for nutrition activities.

Baseline data on socio-economic status, nutritional status, and health status were obtained from 236 households with a total population of 1,655. The average family size was six. Specific data were collected on income, land ownership, water sources, housing, hygiene in the home, latrine availability, and illnesses. Cases of malnutrition were identified on the basis of the modified Gomez classification (as described in the footnote to table 1).

Health Care

The community development projects previously conducted by the Institute of Nutrition had shown the need for medical care as the first priority area among services for the villagers. Hence, curative health care was used as the initial entry point in programme implementation. This involved first-aid services, simple medical care, and a referral system. The second step, preventive health care, involved immunization of children at regular intervals, maternal and child health, and development of a school health programme. The third step in health care involved promotion of family planning, sanitation and hygiene, and health education.

In the area of sanitation and hygiene, attention was paid to instruction in the use of latrines and proper waste disposals and construction of a number of wells and water pumps for adequate and clean water supply.

Nutrition

In implementing nutrition as a major component of the programme, areas of concentration involved were nutrition surveillance, food production, processing, and distribution, food supplementation, nutrition education, and the promotion of breast-feeding (3, p. 527).

Before the programme was implemented, nutrition surveillance was done on a representative sample of 202 pre-school children. Surveillance data showed that 55 per cent of the children had protein-energy malnutrition (PEM). Twenty-four hour dietary recalls from the village mothers showed that the energy intake of pre-school children was only about 60 to 80 per cent of the requirement, with carbohydrate intake of about 80 to 85 per cent, fat 4 to 8 per cent, and protein 8 to 12 per cent of the total energy intake. The diet was bulky and very low in fat content. About two-thirds of the protein intake was derived from rice and vegetable sources. An important step toward improving the diet was to increase the intake of fat and to improve the quality of protein intake.

The institute formulated seven supplementary food mixtures, based on the Thai Standard for Infant Food. Each formula was proportionally prepared from a mixture of locally available carbohydrate, fat, and protein food sources. Preparation of these food mixtures was simple.

Recommended action for nutrition implementation on PEM children was given as follows:
- first degree: nutrition education;
-second degree: nutrition education plus supplementary feedings;
-third degree: nutrition education plus supplementary feeding plus therapy if required.

Mothers of second- and third-degree PEM children brought their children to be fed twice a week at the village centre. The food was prepared by these mothers, who took turns to cook the simple and nutritious food under the supervision of the field implementor. After feeding, the infant food packages were distributed to the second- and third degree PEM children, so the mothers could prepare the food and feed their children on other days of the week. Nutrition education was offered regularly at the village centre to everyone in the community. Nutrition education was also given on special accessions such as temple fairs and other village activities.

Aside from food processing and distribution, the village centre was also sometimes used for food demonstrations. Soybean milk, soybean chips, and other products were prepared and introduced to the community. Breast-feeding was promoted and advice on the subject given.

Agricultural and Food Production

The villagers were soon convinced of the beneficial effects on their children of the nutritious supplementary foods. A need was therefore seen for producing the raw ingredients of these foods. Using appropriate techniques in line with advice given by agricultural extension officers and project personnel, agricultural and food production in the village was greatly improved. Increased production of legumes and groundnuts was observed, and home gardening, poultry raising, and the use of fish ponds was initiated.

Vocational Training

Vocational training was another area included in the project. In an effort to promote home industries, short training sessions were offered to villagers to enable them to pick up some knowledge of those special skills in which they were interested. Silkworms were raised for the production of Thai silk. The villagers were shown how to use appropriate weaving equipment, and exhibitions were held to demonstrate and sell the finished products. As a result, the village's income-generating capacity was increased. Offering these training sessions also helped to encourage villagers to come to the village centre for nutrition education.

Increased agricultural production and the generation of more income through home industries ultimately resulted in an improvement in the local economy and, together with improved health and nutrition, led to the all-round development of the community.

PROJECT EVALUATION

The effectiveness of this project depended largely upon the strategies used in the integration of the four individual project components. The four components-health care, nutrition, agricultural and food production, and vocational training - played sequential roles in promoting the optimum nutritional status of the community

It was through the initial entry point of curative health care implementation that people began to gain trust in the project personnel. Certain dietary treatments on anaemic and beriberi patients had enabled the villagers to realize the importance of nutrition in relation to health and disease. People became more aware and concerned of their dietary habits, and this was when nutrition education could be implemented. The promotion of food supply through increased agricultural production, home gardening, and poultry raising provided the community with a steady supply of food. The introduction of appropriate technologies through vocational training in cloth weaving and other crafts enabled the community to generate more income. Finally, by processing supplementary infant foods from locally available sources at the village level, distribution and home delivery of the infant food packages reached second- and third-degree PEM children easily, hence improving their overall diet.

TABLE 1. Nutritional Status of Children before and after Food Supplementation in the Integrated Programme of Nong-Hai Village (1979-1980)

* Weight for age, using modified Gomez classification for protein-energy malnutrition (PEM) on the basis of body weight as a percentage of mean reference body weight (Harvard Standard), as follows: first-degree PEM, 75-85% of mean reference body weight; second-degree PEM, 60-70% of mean reference body weight; third-degree PEM, below 60% of mean reference body weight.
** Chi-square test, p < 0.01.
*'* Chi-square test, p < 0.001.

An evaluation of the project shows that following the initiation of curative health care, full community participation was achieved within six months of programme implementation. Nutrition education was found to be most effective when related to health and disease.

Supplementary food processing and distribution, the last aspect of the programme to be implemented, was undertaken during its final eighteen months. An evaluation of the effectiveness of the integrated programme over this period is shown in table 1. The nutritional status of the 202 infants and pre-school children, as measured by the modified Gomez classification, improved significantly. The proportion of normal healthy children increased from 45 to 79 per cent. The incidence of first-degree PEM decreased from 38 to 16 per cent. Second-degree PEM fell from 16 to 6 per cent, while third-degree PEM, which had an incidence of 1 per cent, disappeared altogether in eight months.

It can be hypothesized from the results that food assistance was required for the improvement of second- and third-degree malnutrition cases. The nutritional improvements in normal and first-degree PEM children were due largely to nutrition education and the other model components. Nutrition education played an important role in increasing the mothers' awareness of proper infant-feeding practices. Income generation alone did not necessarily improve the nutritional status of the rural poor, but when implemented with other components of the model, showed long term effectiveness.

DEVELOPMENT OF THE MODEL AND ITS SIGNIFICANCE

From our experience with this integrated nutrition, health, and rural development project, we have learned that, by utilizing curative health care as the initial entry point of implementation along with preventive and promotive health care and continuous nutrition education, the nutritional status of the community can be improved. We have also found that other components to do with the availability of local food supplies and income generation play equally important roles in promoting the outcome.

This project has led us to develop our original conceptual framework into a model (figure 2), which consists of four major components elaborated as follows:
a. health care-curative, promotive, and preventive;
b. food consumption-promoting the quality and quantity of food consumed by modifying people's dietary habits through nutrition education;
c. food supply-promoting agricultural production which leads to effective supplementary food processing and distribution at the community level;
d. income generation -the promotion of appropriate technology and occupations.

The sequential integration of these four components has endowed our project in Nong-Hai with popularity. Training sessions and workshops have been given to provincial and district health care personnel, agricultural extension workers, and community development workers in an attempt to replicate the model in other parts of Thailand.

CONCLUSION

An integrated health, nutrition, and rural development project was implemented in the village of Nong-Hai in an attempt to improve the nutritional status of the rural poor. The project followed a conceptual framework involving the implementation of four components: health care, nutrition, agricultural and food production, and vocational training. Curative health care was used as the initial entry point of programme implementation.

Seven supplementary food mixtures were formulated and tested for safety, digestibility, and acceptability by the institute. These food mixtures were introduced, fed, and distributed to second- and third-degree PEM children. Nutrition education and other related activities were initiated. Agricultural production and income generation were promoted, which ultimately resulted in an improvement of the local economy and the all-round development of the community.

An evaluation of the programme showed significant improvements in the nutritional status of normal children and first-degree PEM children. Nutrition education played an important role in increasing the mothers' awareness of dietary habits. Food assistance was needed for the improvement of second- and third-degree PEM children.

The original conceptual framework of programme implementation has now been developed into a model with four major components: health care, food consumption, food supply, and income generation. The proper sequential integration of these components is essential if the project is to have a successful outcome. As a result of the Nong-Hai project, it is hoped that this programme model will become our national model and that our experience can be shared with the people of other developing countries.

REFERENCES

1. The Population Reference Bureau, Inc., 1980 World Population Data Sheet (Washington D.C., 1980).
2. F. Gomez, G.R. Ramos, J. Cravioto, and S. Frank, "Malnutrition in Infancy and Childhoods with Special Reference to Kwashiorkor," in S.Z. Levine, ed., Advance in Pediatrics, 7:131 (Year Book Publishers, Chicago, 1955).
3. A. Valyasevi, "Public Health Program to Promote Nutrition in Rural Areas-Thailand Experience," in W. Santos, N. Lopes, J. Barbosa, D. Chaves, and J. Valente, eds., Nutrition and Food Science (Plenum Press, New York, 1980).

Perspectives on infant feeding: decision-making and ecology

Women's Employment and the Feeding of Infants

At the outset it is important to note that "employment" is often far from being a simple variable in its effect on women's lives and roles. We tend to think of people in our society as either employed or "unemployed" in relation to "jobs" that earn wages or salaries. In much of the world (including major sectors of North American society) the gradations of "jobs" and "employment" (in and outside the home) can be complex indeed.

One of the few studies that directly addressed the interrelations of women's employment and infant feeding is the work of Popkin and Solon (38). Their study is focused on the Philippines, where they note first of all that women are more actively involved in trade, service, professional, industrial and other work which draws her [sic] outside the home for employment.... Nevertheless market related activities which take place in the home such as embroidery, basket weaving and sari-sari store management play a major role in her economic activities. [38, p. 197]

In their sample of women on the island of Cebu, they found that "26% of the urban and 31% of the rural women were engaged in some form of market employment." They note that some studies in the Philippines report infants nursed 7 to 8 times per day, with each feeding lasting 15 minutes to half an hour. This schedule of nursing could require from 3 to 4 hours, in addition to which one should allow for "travel time" if the mother is working somewhere and returns to her infant for breast-feeding. In any case, the scheduling of breast-feeding requires a very considerable time investment from the mother.

Dugdale (39) examined data on breast-feeding patterns for 2,009 infants in Kuala Lumpur, Malaysia, for the years 1960, 1962, and 1965. The information included family income (divided into high and low), number of children in the family (presumably of that particular mother), and ethnicity (Chinese, Malay, Indian). The data are interesting because they are roughly comparable to situations in many other developing nations, especially in Asia. Focusing on the data for families of two to four children (the modal category) in 1965, according to the different ethnicities, we find that the income variable is important in each of the ethnic groups, with the higher income (higher socio-economic status) women breast-feeding less often land for shorter duration) than the lower status mothers (table 1). Dugdale feels that there are two possible reasons for the differentials based on income:

The first is that the social and advertising pressures are all away from breastfeeding. The mother who has a higher income is probably more educated and is more able to afford artificial feeding. The second reason is that many families have a higher income because both parents work outside the home. Under these circumstances, breast-feeding is not possible. [39, P. 233]

TABLE 1. Percentage of Mothers Breast-feeding at 28 Weeks in Kuala Lumpur (1965) according to Income and Ethnicity

*n in parentheses. Source: Dugdale (39).

Also, as the author comments, "the extended family and the availability of domestic help both make possible the care of infants while the mother works." Further, among the Malays the mothers with children go out to work much less than do the Chinese, hence an overall higher incidence of breast-feeding among the Malay mothers (in the three years surveyed).

A comparison of the data for 1960 with the 1965 data shows a consistent decrease in breast-feeding in the Chinese sample, (n is 1020) and an increase in breast-feeding among the Malay women. The author does not comment on this difference in the trends but notes that the increase in breast-feeding is mainly among low income mothers with one child.

Marchione (40) has developed an economic model to explain infant-feeding patterns in the Caribbean. He emphasizes the relationship of women's work activities with the type of economy in which those activities take place as a significant factor in infant-feeding patterns. In a plantation economy-first under slavery, then as low-pay resident workers-women in the Caribbean have for centuries been under pressure to devote long hours to work and short hours to breast-feeding. Marchione's study suggests that, wherever we find the combination of a past or present plantation agricultural system and the presence of a variety of breast milk substitutes such as bush teas, starchy gruels, or paps, we can hypothesize on the existence of direct pressures on women to limit breast-feeding.

Another type of economic model also concentrates on structures outside the household, specifically on the economic motivations of the commercial interests that stand to gain from a large-scale shift to bottle-feeding- the infant food manufacturers, the advertising media, and, to a limited extent, health-care professionals. The research strategy that has been pursued to develop this model has focused primarily on describing the economic structures of the formula-manufacturing companies and the promotional strategies of corporation and advertising agencies (41, 42). There has been little linking of these structures and activities to maternal selection of feeding mode, perhaps largely because it presents serious methodological and ethical problems for investigators.

Bio-cultural Models of Infant Feeding

Another approach to explaining infant-feeding patterns examines issues from a biological perspective, seeking to understand the extent to which problems with lactation, as a socially mediated biological process, may be involved in current trends. It is important to note that this approach is more limited in that it applies only to one segment of the women who make up the statistics-those who attempt breast-feeding but shift rapidly to bottle-feeding. In any examination of infant-feeding practices it is necessary to separate exclusive bottle-feeders from those women who begin breast-feeding, even if they only breast-feed for a few days. The decision to bottle-feed an infant-if the decision is made before the infant is born-is conceptually a different kind of decision from the decision to try breast-feeding. Bio-cultural explanations which seek to understand "lactation failure" or "premature weaning" apply, of course, only to those women who attempt breast-feeding.

While a biological failure of the capacity to initiate lactation is rare, post partum problems in lactation are not (43). It has been hypothesized that separation of infants from their mothers at birth, which has been a common practice in western medicine, affects lactation success. Klaus and Kennell (44) have studied the relationship of early separation to problems in mother-infant bonding, which, in turn, appears to be implicated in lactation problems. Newton and Newton (45) and others have suggested that separation, which delays the onset and frequency of breast-feeding in the immediate post partum period, may be related to breast over-engorgement and sucking problems for the infant that lead to a "vicious cycle," resulting in early termination of breast-feeding.

Raphael has also proposed a big-cultural explanation of lactation failure, an explanation based on the of the "let down reflex" in the absence of supportive social networks for the newly delivered mother (46). The breakdown of social networks and social support is postulated as an urban phenomenon which, together with other stresses of urbanization and modernization, leads to psychophysiological stress inhibiting milk production (47, 48).

Recently Gussler and Briesemeister (43) have presented a big-cultural explanation for what they refer to as the "insufficient milk syndrome." Their discussion begins with a review of a significant finding in many studies, namely, that the most common reason mothers give for premature weaning or early termination of breast-feeding is "insufficient milk." From many different cultural and social settings in both industrialized and developing countries, mothers report that their milk was insufficient or inadequate to satisfy their infant's needs:

Anecdotal and research data suggest that, quite naturally, mothers become concerned about presumably unhappy or uncomfortable infants, and respond to their behavioral cues with an explanation of breast milk insufficiency, even though the explanation is not confirmed by poor weight gain.

These mothers are both right and wrong. They are probably correct in interpreting the behavior of their fussy infants as caused by hunger. They are incorrect that this necessarily indicates that something is inherently "wrong" with their milk or ability to lactate. [43, p. 6]

From this base the authors develop their theory. The primary source of the problem is what they call "non-biologic breast-feeding," the pattern of "scheduled, widely spaced feedings of breast milk." Scheduled, infrequent feeding leads, in turn, to a crying baby, to sucking difficulties, to maternal anxiety, to insufficient sucking and emptying of the breast, and hence through several neural and cognitive pathways to the interpretation of "insufficient milk." In many cases the interpretation becomes a biological self-fulfilling prophecy. The authors conclude their paper with a call for research on this important aspect of contemporary breast-feeding experiences.

Each of these approaches-socio-cultural (modernization), economic, and psycho-physiological-has much to offer to our understanding of contemporary infant-feeding practices. Of course, the significance of particular factors will vary in different contexts. To date, none of the theoretical approaches have been fully explored through systematic research, nor have efforts been made to integrate them. While the bio-cultural approach of Gussler and Briesemeister does present a theory that integrates social behaviour (frequency of feeding) with biological variables, it does not provide a full framework since it is focused specifically on milk insufficiency. The development of a full framework is probably best achieved through successive approximations based on feedback from empirical research. The remainder of this paper is devoted to outlining some of the features of a "first approximation," noting several key components of what can be regarded as a "cultural-ecology" approach.

THE CULTURAL-ECOLOGY FRAME OF REFERENCE

An ecological perspective directs attention to the specific features of the physical and social environments within which behaviour takes place. The question of how individuals and groups meet essential needs (for food, shelter, and so on) remains a central concern. Although people are seen as acting within a system of environmental constraints and opportunities land having available resources and technologies), they are considered to be making conscious choices about behaviour. The ecological approach thus emphasizes people's rational decision-making behaviour and calls attention to intra-community, intra-population variations in the specific choices, as related to situational (environmental) differences (34).

Anthropologist John Bennett has described the theoretical perspective of cultural ecology clearly:

A second meaning of the term ecology emphasizes adaptation or adaptive behavior. Here we refer to coping mechanisms or ways of dealing with people and resources in order to attain goals and solve problems. Our emphasis here is not on relationships between institutions, groups or aggregates of data, but on patterns of behavior: problem-solving, decision-making, consuming or not consuming, inventing, innovating, migrating, staying. [49, p. 11]

In the nutrition and public health field, the issue of infant feeding is often approached from a narrow viewpoint in which the infant is seen as a passive receiver of food and the mother as the giver of food. Problems arise when the mother fails to fulfil adequately her role as food-giver. However, the woman's behaviour with respect to feeding her infant is rarely examined in the total context of her life. While a new infant may seem to be a major focus of a mother's energy and attention, the fact is that a new mother can be said to have acquired an additional role- that of mother of her new-born-but she does not necessarily lose her other roles. In most circumstances she may continue to have the responsibilities of being a wife, a daughter-in-law, a mother to her other children, a farmer, a housekeeper, a lawyer, and so on.

While the extension of biologically oriented theory places primary emphasis on women's roles as child-bearers and child-rearers, an ecological approach would tend to focus on women's responsibilities in the economic subsistence system, as well as in maternal tasks. The contrast between the two approaches is evidenced in this statement from the study by Nerlove:

In her discussion of the division of labor by sex, Brown (1970) stated that the degree to which women participate in subsistence activities depends upon the compatibility of the latter with simultaneous child care responsibilities. The present study qualified this statement by showing that child care responsibilities may be adjusted to accomodate the mother's subsistence activities. [6, p. 212]

The ecological framework is not one of economic determinism, but a search for all the relevant factors (economic and others) affecting women's roles. Aside from direct food production practices and ways of making a living, many other factors must be taken into account in an ecological perspective, including such features as shelter, modes of travel, and environmental characteristics. Political organization and religious beliefs and other ideological components are also examined. Perhaps the major feature of the cultural-ecology approach is the attention to a multiplicity of factors affecting every aspect of human behaviour.

ECOLOGICAL THEORY AND WOMEN'S ROLES

The growing body of literature on women's situations in various cultures and diverse environmental contexts is beginning to provide a base for generalizations about women's roles. These studies point to some of the major variables and parameters that are necessary for a well-developed model of women's roles in the care and feeding of offspring. It would appear that throughout history women have played a very significant and primary role in child care, not least because of the nature of their biological relationship with their offspring. It is also clear that in many societies women have major responsibilities for food production and that, in these societies, their child-care roles are adjusted to these responsibilities.

The introduction of new crops into traditional societies can have various and diverse effects on women's roles. The economic position of Afikpo Ibo women of Nigeria is an interesting illustration. According to LeVine, when the new crop cassava, was introduced in the region, it was considered to be of very low prestige and value. Women were allowed (by the men) to grow the cassava as a marginal activity:

This despised crop eliminated the annual famine before the yam harvest and attained a high and stable market value. The Afikpo women became capable of supporting themselves and their children without aid from their husbands.... Afikpo husbands have found it increasingly difficult to keep their wives at home in their formerly subordinate position. [50, p. 178]

A different sort of example is provided by the Azande people in East Africa. Until recent times the Azande engaged in frequent warfare, and women were the main source of labour and symbols of wealth. Traditionally, women were "treated much like slaves and . . . barred from the main source of power." Increasing contacts with Europeans and the introduction of cash crops brought large-scale changes. "Women became emancipated, crops acquired cash value, war disappeared, hunting was reduced, and men were compelled to invest most of their efforts in agriculture" (51, p. 201). In contrast to the Afikpo situation, the Azande men placed high value on the major cash crops and controlled the access of women to this new sphere of activity. Although the women gained a new measure of freedom, they did not become anything like as economically powerful as the Afikpo women.

The growing number of studies of women's roles in different types of societies have underscored two major themes: a. Women's overall status and prestige and the degree they contribute to the basic economic system exhibit a very wide range of variations in non-modern societies. b. Degrees of participation in food production and other economic activities have strong effects on the patterning of child care activities.

Some of the other variables that have important effects on women's roles include the patterning and degree of warfare (52), kinship structure and household composition (53), and religious system (54). This is hardly an exhaustive list of the variables that may be important in affecting women's roles and infant-feeding practices, but it provides a starting point for an examination of individual cases and contexts.

THE SHAPING OF WOMEN'S ROLES

From an ecological perspective, women's roles in a social system can be seen as composed of a number of crucial dimensions that can take on a range of values. Following role theory we can conceptualize that an individual has a cluster of roles. Many of the roles an individual assumes are of long standing and taken to be stable over the generations. For example, the central roles of wife and mother are thought of as relatively unchanging in basic content from one generation to the next. Also, in a stable society the supposed rules of behaviour are sufficiently well established and pragmatic that regular expectations about conduct in the role of "wife" do not seriously conflict with the requirements of the role set to "mother." On the other hand, everyone is familiar with examples in which different aspects of role expectations come into sharp conflict. In such situations individuals have had to make agonizing choices: "Shall I give up my family ties to marry this man?" "Shall I protect my husband from this unwarranted demand by our son?"

The factors that shape the multiple-role behaviour of any given women in any particular society are variable and specific to the situation. At the same time, it can be argued that these factors have regular, patterned effects, so that generalizations can be made about the nature of role behaviour:

1. Although people do not always act in their own best interests, in general individuals weigh alternatives and make choices based on their perceptions of the relative merits and shortcomings of particular actions. All persons seek to be optimizers in their daily behaviour.
2. "Micro differences" in characteristics such as household composition, distance from resources, material wealth, etc., create differential behaviour within even the smallest and most traditional populations.
3. Therefore women within practically all communities will exhibit a range of variation in their tendencies toward particular types of role behaviour, including breast-feeding. Intra-community variation is a pervasive fact of life.
4. Individual choice behaviour is, of course, always constrained by knowledge of what other persons-kin and non-kin-expect as appropriate, right conduct.
5. For most categories of behaviour and activities there are alternatives that people can choose. Life is not a rigid prescription.
6. Recent decades have brout new dimensions of choice in practically all parts of the world. People have many more options to consider, even though poverty and lack of technical skills may in practice limit their selections from the theoretically available alternatives.
7. Although new options and alternatives have spread rapidly in recent decades, change and development occurred in most societies and communities in past centuries as well. Most people are used to the prospect of change.

INFANT FEEDING AND DECISION MAKING

The statistics on trends in both breast-feeding and bottle" feeding can be regarded as a reflection of multiple decisions by numerous women and their families. The concept of decision implies, of course, a choice among alternatives. With the advent of large-scale availability of bottles, it is apparent that many women perceive themselves as having an alternative. Seen against the whole of human history the decision to use a bottle appears to represent the exercise of choice, comparable to the choice involved in using contraceptives to regulate fertility.

However, just as with the use or non-use of contraceptives, we must be careful not to assume that choices are made in a vacuum, without constraints. It is apparent from the available literature that many factors influence and constrain a woman's decision to breast-feed or bottle-feed her infant. Among the factors that influence this decision are the views of significant other persons, particularly spouses, as the following illustrations make clear.

In an Egyptian village, anthropologist Soheir El Sukkary interviewed women concerning their views and decisions about infant feeding. One woman, the mother of seven, told her:

In our village all women have to breast-feed. I told my husband that I was tired of breast-feeding after seven children and I want to use cow's milk. You know what he said? He told me even if I break my neck I still have to breast-feed. We are no city girls. [El Sukkary, personal communication]

Contrast the constraints on this woman with those on a young Puerto Rican woman interviewed in Miami by anthropologist Carol Bryant:

The doctor said to breast-feed them, especially the boy. But I can't explain it, maybe it has a lot to do with my husband. He's sort of shy about that, too . . . You know, I think he's jealous. When I told him I wanted to breast-feed, he said, "You're going to breast-feed? Well, I don't know about that." And stuff like that. But if I had to do it again, I'd try it. [37]

In another cultural context, it is the in-laws who effectively control the decision-making concerning infant feeding. Johnston (55) has reported about the situation in West Indian households in Trinidad, where the low-status young wife is subordinate to her mother-in-law in an extended family household. According to Johnston, the young mother is expected to "entrust the daytime care and feeding of the infant to her in-laws and devote her attention to chores for the benefit of the whole family." The grandparents therefore give bottles, filled with a variety of substances from skim milk powder to sugar water and tea, while the mother works at other types of activities for the household. This Trinidad study is perhaps the clearest example Yet reported of bottle-feeding motivated by the work demands of a woman's extended household. The bottle-feeding regimen does not, however, eliminate breast-feeding. Moreover, Johnston reports that the ratio of breast-feeders to bottle-feeders has not changed substantially during the past decade.

In this Trinidad example, the significance of in-laws in the infant-feeding process lies in the fact that they are in the same economically integrated household. Clearly the variable of household size and structure may in this case be more significant than simply the "cultural attitudes" of in-laws per se. There are a few other cases, however, in which the in-laws appear to have some influence. In a small-scale study in New England, for example, the opinions of the subjects' in-laws appeared to be the only major anti-breastfeeding influence.

MEDICAL ADVICE

The doctor Avicenna's advice against any substitutions for natural (mother's) breast-feeding (1) reminds us that health professionals may be among the most important sources of influence on women's infant-feeding practices. Jelliffe and Jelliffe (48, 56), Cole (57), and others have noted the significance of health professionals in affecting infant feeding. Halpern et al., for example, reported that among 1,700 infants in Dallas, Texas, there was a strong correlation between breast-feeding and the paediatrician's attitudes (58). In developing countries, the role of western medical care may significantly influence women to adopt bottle-feeding (42), and lack of encouragement and assistance from medical and nursing staff has been identified as a factor in lactation failure among Asians recently migrated to Great Britain (59).

MULTIPLE SOURCES OF ADVICE

Many of the studies of infant feeding in both industrialized and developing societies have provided information on the effects of social ties, significant others, opinions of relatives, and other influences on women's decision-making. While these are undoubtedly important factors, they appear to assume significance mainly in relation to other variables.

In practically all situations where the opinions of significant others-kin and non skid-appear to be important, there is likely to be a considerable mixture of sources of information and opinion. Karkal reports that in a survey conducted in Bombay the following were identified as sources of advice regarding supplemental foods: tradition, 11 per cent; elders, 19.2 per cent; self, 17.1 per cent; doctor or nurse, 13 per cent; radio or reading, 4.7 per cent; and "cannot say," 34 per cent (60). An effective explanatory model must make sense of this mixture.

BIOLOGICAL INFLUENCES

In the earlier discussion on big-cultural approaches, the role of birth practices, feeding schedule, and the absence of a supporting social network were mentioned as factors that affect infant-feeding practices. Additional features of maternal health and pregnancy experience, fatigue, maternal intake of fluids and calories, and other dietary variables must also be carefully considered. In pursuing a cultural-ecology approach, the problem will be to identify the ways in which biological variables enter into the complex calculus of decision-making of individuals.

CONCLUSION

The main features of a cultural-ecology framework as a guide to research and application of research for programme planning and development in infant feeding include the following:

1. The aggregate pattern of infant-feeding modes for any particular population or identifiable social groups is a composite result of a series of individual decisions
2. To understand these decisions, attention must be focused both on the individual decision-makers (mothers and their households) and on the characteristics of the environment in which the decisions are made.
3. In choosing among alternative strategies for feeding their infants, mothers may be influenced by a series of factors: economic conditions; health characteristics and concerns (their own and their baby's); requirements and desires related to allocation of their own time; the presence of alternative caretakers; beliefs and values related to the social acceptability of the choices; and advice from other people and media sources.
4. Selection from theoretically available alternatives is strongly conditioned by external [environmental) and internal (individual) constraints. Thus, such features as the availability of work, characteristics of the economic system, characteristics of the health care system, family and community structure, and organization may all exert powerful constraining influences on an individual mother's decisions about how to feed her infant.
5. Decisions on infant feeding, as with most other activities, are not made simply at a single point in time but should more properly be regarded as a process, in which the key actors (mothers, infants, families, and other significant participants) monitor and adjust their behaviour over time.

These points suggest that the reasons for particular patterns of feeding will differ from one social setting to another over time. It is fallacious to conclude that a common pattern is necessarily the result of common causes. The utilization of a cultural-ecology approach does not imply a simple eclecticism in which "all variables are equal." As conditions change, the "power" or significance of particular factors in the decision-making process may change.

Following a cultural-ecology framework, the next steps to improving our understanding of contemporary infant feeding patterns is to develop research that is holistic and multidisciplinary. Attention must be directed to include mother-infant pairs, the households in which they reside, the communities and environments in which the households exist, and the larger socio-political, cultural, and economic systems from which the new choices (e.g., bottles and formulas) and new constraints (e.g., wage employment) are ultimately derived.

There is no evidence, at present, that justifies eliminating any of the major domains of causal influence from investigation. The extent to which economic factors, household characteristics, the individual's beliefs, goals, desires, and values, and psycho-biological characteristics affect decisions about infant feeding is amenable to empirical investigation. Since the relative importance of these characteristics is likely to be different in different environments, future research must be cross-cultural and must be carried out in different regions and cultural conditions. This does not mean that research on infant feeding has to be undertaken in every community on the face of the earth before we can have a better understanding of the contemporary situation. But it does mean that we must be careful not to assume that the picture is the same in all urban centers of the industrialized world or that we can easily extrapolate findings from a rural Asian community to a plantation in Guatemala.

Direct interventions by policy-makers can bring about dramatic changes in the environment, changes which presumably will have powerful effects on individuals' decisions. It would seem logical to conclude, for example, that banning bottles from the market-place would create an environmental constraint of overwhelming proportions. However, alternative routes to utilization of artificial feeding may develop readily if the other factors that are influencing mothers to select bottle-feeding remain unchanged. We return to the basic premise: decision-making about infant care (including infant feeding) is a rational process in which the outcome reflects the "best choice" or the "best compromise" among competing demands, conditions, and values. When the "best choice" conflicts with "best" psycho-physiological and health outcomes (as judged from a big-medical perspective), health care professionals and other people are appropriately distressed. The problems, however, cannot begin to be resolved unless infant-feeding patterns are examined in the complex context in which they occur. Thus, research that is sensitive to differences in micro-ecological variations (for example, in the WHO research and action projects) is essential for progress in our understanding of these issues (61). A cultural-ecology approach contains the guidelines for beginning that process.

REFERENCES

1. B.M. Duncum, "Some Notes on the History of Lactation," Brit. Med. Bull., 5:1141 (1947).

2, W. Butz, "Economic Aspects of Breast Feeding," in W.H. Mosely. ea., Nutrition and Human Reproduction (Plenum, New York, 1977).

3. A.E. Dugdale and D.P. Doessel, "Family Well-being and Infant Feeding" (unpublished paper, Department of Child Health, University of Queensland, Brisbane, Australia, n.d).

4. J.W.M. Whiting and l.L. Child, Child Training and Personality: a Cross Cultural Study (Yale University Press, New Haven and London, 1953).

5. L. McInturn and W.W. Lambert, Mothers of Six Cultures: Antecedents of Child Rearing, (John Wiley and Sons, New York, 1964).

6. S.B. Nerlove, "Women's Workload and Infant Feeding Practices: A Relationship with Demographic Implications," Ethnology, 13:125-214 (1974).

7. H. Barry, l.L. Child, and M,K. Bacon, "Relation of Child Training to Subsistence Economy," American Anthropology, 61: 51-63 (1959),

8. R. LeVine and B.B. LeVine, Nyansongo: A Gusii Community in Kenya, (John Wiley and Sons, New York, 1966).

9. B.B. Whiting, ed., Six Cultures: Studies of Child Rearing (John Wiley and Co., New York, 1963).

10. I.G. Wickes, "A History of Infant Feeding," Arch. Dis. Childhood, 28:151-158, 232-240, 495-502 (1953).

11. P. Aries, Centuries of Childhood: A Social History of Family Life, (Random House, Vintage Books, New York, 1962).

12. J. Prince, "Infant Feeding through the Ages," Midwives Chronicle and Nursing Notes, Dec. 1976, pp. 283-285.

13. A.L. Wood, "The History of Artificial Feeding of Infants," Journal Amer. Dieter Assoc., 31: 474-182 (1955).

14. A. F.C, Wallace, Rockdale: The Growth of on American Village in the Early Industrial Revolution (Alfred A. Knopf, New York, 1972).

15. D.A. Sehring, "Infant Feeding Trends in an Industrialized Culture," Mod. Prob. Paediat., 15 (1975): 231-237.

16. H. Meyer, "Breastfeeding in the United States," Clin. Pediat., 7:708-715 (1968).

17. B.A. Underwood, H. Van Arsdell, E. Blumenstiel, and N.S. Scrimshaw, "Implications of Available Information on Breast-feeding Worldwide" (paper presented at the International Symposium on Infant and Child Feeding, Michigan State University, East Lansing, Mich,, USA, 1978).

18. G.E. Hondershot, "Trends in Breastfeeding," Advanced Data, Vital and Health Statistics of the National Center for Health Statistics, PHS No. 59 (Washington, D.C., 1980).

19. E. Helsing, "Feeding Practices in Europe-Beliefs and Motivations and Possibilities for Change" (paper presented at the International Symposium on Infant and Child Feeding, Michigan State University, East Lansing, Mich., USA, 1978).

20. WHO/UNICEF, part one of the working paper of the Meeting on Infant and Young Child Feeding (organized by WHO and UNICEF, Geneva, 1979).

21. A. Ferris, Ph.D. dissertation (Department of Food Science and Nutrition, University of Massachusetts, Amherst, Mass., USA, 1979).

22. N.W. Jerome, "Do Families and Women of the Third World Have the Right to Establish Their Own Priorities? " (paper presented at the Conference on Infant Formula in the Third World: Does It Serve a Useful Role? New York, 1979).

23. T. Phayer, Boke for Children (1510-1560).

24. L. Tiger and R. Fox, The Imperial Animal (Holt, Rinehart and Winston, New York, 1971 ).

25. L.P. Gerlach, "Milk, Movements and Multinationals: Complex Interactions and Social Responsibilities," in Responsibilities of Multinational Corporations to Society, vol. III, proceedings of the Fifth Panel Discussion, Council of Better Business Bureaus, Washington, D.C.

26. L.P. Gerlach, "The Flea and the Elephant: Protest, Response and Consequences in the Infant Formula Controversy, Transaction/Society special issue on food, agriculture, and development (1979).

27. A. Berg, The Nutrition Factor: Its Role in National Development (The Brookings Institution, Washington, D.C., 1973).

28. C. Hirschmann and J.A. Sweet, "Social Background and Breastfeeding Among American Mothers, Social Biology, 21 :3957.

29. W.O. Robertson, "Breastfeeding Practices: Some Implications of Regional Variations,"Amer. J. Publ. Health, 51:1035-1042 (1961).

30. E.J. Salber and M, Feinleib, "Breastfeeding in Boston," Pediatrics, 37:299-303 (1966).

31. A. Yankauer, A. Boek, E.D. Lawson, and F.A.J. Ianni, "Social Stratification and Health Practices in Child-bearing and Child Rearing," Amer. J. Publ. Health, 48:732-741 (1958).

32. M. Harris, The Rise of Anthropological Theory, (Crowell, New York, 1968).

33. C.M. Woods and T.D. Graves, The Process of Medical Change in a Highland Guatemalan Town, Latin American Studies, 21 (University of California Press, Los Angeles, 1973).

34. P J. Pelto and G.H. Pelto "Intracultural Diversity: Some Theoretical Issues," Amer. Ethnologist, 2:1-18 (1975).

35. J. Mohrer, "Breast and Bottle Feeding in an Inner-City Community: an Assessment of Perceptions and Practices," Med. Anthro, 3:125-145 (1979)

36. R.A.K. Jones and E.M. Belsey "Breastfeeding in an Inner London Borough-A Study of Cultural Factors," Soc. Sci. Med., 11:175-179 (1977).

37. C. Bryan, "The Impact of Kin, Friend and Neighbor Networks on Infantile Obesity" (Ph.D dissertation, University of Kentucky, Lexington, Ky., USA, 1978).

38. B. Popkin and F. Solon, "Income, Time, the Working Mother and Child Nutriture," J. Trop. Pediat, 22:156-165 (1976).

39 . A. E. Dugdale, " Breastfeeding in a South East Asian City ," Far East Med. J., 6:230-234 (1970).

40. T. Marchione, "A History of Breast-feeding Practices in the English-speaking Caribbean of the Twentieth Century," Food and Nutr. Bull., 2 (2): 9-18 (1980).

41. T. Greiner, The Promotion of Bottle Feeding by Multinational Corporations, Cornell international Nutrition Monograph Series, No. 2 (Cornet/ University, Ithaca, N.Y., 1975).

42. T. Greiner, Regulation and Education.: Strategies for Solving the Bottle Feeding Problem, Cornell International Nutrition Monograph Series, No. 4 (Cornell University, Ithaca, N.Y., 1977).

43. J.D. Gussler and Briesemeister, "The Insufficient Milk Syndrome: a Biocultural Explanation," Med. Anthro., 4:1-24 (1980).

44. M, Klaus and J. Kennell, Maternal-Infant Bonding: The Impact of Early Separation or Loss on Family Development (C.V. Mosby, St. Louis, Mo., USA, 1976).

45. M. Newton and N. Newton, "The Normal Course and Management of Lactation," Child and Family, vol. 9 (1) (1970).

46. D. Raphael, The Tender Gift: Breast Feeding (Schocken Books, New York, 1976).

47. N.R, Newton and M. Newton, "Psychological Aspects of Lactation," New Engl J. Med., 277 :1179-1188 (1967).

48. D.B. Jelliffe and E.F.P. Jelliffe, Human Milk in the Modern World, (Oxford University Press, London, 1978).

49. J.W. Bennett, Northern Plainsmen: Adaptive Strategy and Agrarian Life, (Aldine/Atherton, Chicago, 1969).

50. R. LeVine, "Sex Roles and Economic Change in Africa," in J. Middleton, ed., Black Africa (Macmillan Co., New York, 1970), pp. 174-180.

51. P.R. Sanday, "Female Status in the Public Domain," in M.Z. Rosaldo and L. Lamphere, eds, Women, Culture and Society (Stanford University Press, Stanford, Calif., USA, 1974), pp. 189206.

52 M. Ember and C.R. Ember, "The Conditions Favoring Matrilocal Versus Patrilocal Residence," American Anthropologist, 73:571-594 (1971).

53. A. Schlegel, Male Dominance and Female Autonomy (HRAF Press, New Haven, 1972).

54. M.Z Rosaldo, "Women, Culture and Society: A Theoretical Overview," in M.Z. Rosaldo, and L. Lamphere, eds., Woman, Culture and Society (Stanford University Press, Stanford, Calif.,USA, 1974), pp. 17-42.

55. J. Johnston, "The Household Context of Infant Feeding Practices in South Trinidad" (paper presented at the annual meeting of the American Anthropological Association, Houston, Texas, 1977).

56. D.B. Jelliffe and E.F.P. Jelliffe, "The Uniqueness of Human Milk,"Am. J. Clin. Nutr., 24:968-1024 (1971).

57. J. Cole, "Breastfeeding in the Boston Suburbs in Relation to Personal-Social Factors," Clin. Pediat., 16:352-356 (1977).

58. S.R. Halpern, W.A. Sellars, R.B. Johnson, D.W. Anderson, S. Saperstein, and S. Shannon, Jr., "Factors Influencing Breastfeeding: Notes on Observations in Dallas, Texas," South. Med. J., 65:100-102 (1972).

59. N Evans, I.R. Walpole, M.U. Qureshi, M.H. Memon, and H.W.E Jones, "Lack of Breastfeeding and Early Weaning in Infants of Asian Immigrants to Wolverhampton," Arch. of Dis. in Childhood, 51: 608-612 (1976).

60. M. Karkal "Sociocultural and Economic Aspects of Infant Feeding," Indian Pediatrics, 12:13-19 (1975).

61. M. Carvallo, "World Health Organization Programs in Breast-feeding," in D. Rafael, ea., Breastfeeding and Food Policy in a Hungry World (Academic Press, New York, 1979), pp. 245-252.