Contents - Previous - Next

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

Social, economic, health, and environmental determinants of nutritional status

David Nabarro
Department of Human Nutrition, London School of Hygiene and Tropical Medicine, London, England


This paper examines primarily the factors that determine the nutritional status of children. The initial focus is on the relationships between nutrient consumption, body function, and physical size in the individual child. The child is dependent on an adequate supply of nutrients if he is to be capable of operating at his full potential. Many different influences-acting both inside and outside the child's body-may prevent him from having the nutrients needed for these functions available for use in his body tissues.

The child is subject to continuous change as he passes through life. He is growing mentally, physically, and emotionally. Initially he is nurtured entirely by his mother, but gradually he becomes independent of her and learns to feed and look after himself. He has very special nutritional needs in early childhood and is fully dependent on his mother to provide for them. At this time, too, he is particularly susceptible to infection with pathogens from the home environment and to those that cause the specific infectious diseases of childhood, described below.

The frequent observation that children in less developed countries are particularly prone to the dangerous combination of undernutrition and illness has prompted the establishment of health and nutrition programmes that focus on the needs of small children as a "vulnerable group." These programmes are designed to provide resources that will make it easier for mothers to support their children through infancy (e.g., immunization, medical care, and supplementary foods). They also provide health and nutrition education with the aim of improving mothers' nutritional and child-rearing practices.

However, the relationship between a mother and her children will be influenced, to a very large extent, by the viability of her family as an economic unit. Viability will primarily be determined by the resources available that lead to production and/or income for the family. It may also be affected during particular times of the year (e.g., in subsistence farming communities with a single main harvest period) or by general economic conditions (e.g., a drought or the creation of new employment opportunities).

We can therefore see situations in which the mother-child relationship may be directly influenced-even constrained -by the economic realities facing a family. In the subsistence farming community described above, this will be the times of the year in which food is in short supply. At the same time, the mother's activities will also be influenced by her status in the family, particularly with regard to decisions made about tackling food shortage and searching for income.

When problems arise, like children's illness or lack of food grains, economically constrained mothers will be faced with a limited number of options, and choosing the right one may be very difficult. Perhaps the most difficult will be choice between using time for economically productive work or for child care and other domestic activities. The choice will be especially difficult if the mother has more than one infant and is therefore facing a heavy demand on her domestic time. Decision-making will be eased if there are other family members or relatives who are able to look after the children. The consequences of the mother making the "wrong" decision can be extremely severe. Concentrating her time on the sick child could result in the reduction of a subsequent grain harvest-a heavy burden to face later. Alternatively, if the time is not made available for the child, a more immediate con sequence (his disability or death) may be the result. Poor families may not have the resources available to have another opportunity to help, or obtain help, for the child. Use of health facilities in particular may involve high real (or opportunity) cost.

It is not realistic to view an individual family in isolation. Each family will be linked to a greater or lesser degree to others in the community's social and cultural groups.

Normally, it is the strengths of kinship and ties among families inside the social groups that are most noticeable to the outsider visiting urban and rural communities. More important, at least from the standpoint of children's nutrition, may be the economic relationship between families, not only within but also between different social groups. In any community, these economic "systems," if unchecked, will be likely to work to the advantage of some groups and to the disadvantage of others. The assets at the disposal of most families in the favoured group will be on the increase; those of the disadvantaged group may show less of an increase or be static, and they may even be decreasing. Unless legislative systems can be introduced to prevent it, existing economic resources in a community will be both distributed and maintained by a small and definable group of families with power. They will be likely to have close links with similar groups in other rural communities, too. Unless there is an alternative resource allocation system in a community, attempts to change the social and economic conditions of any particular group of people through the provision of new resources will have to be implemented by (or under the influence of) the powerful groups. A corollary is that if there is a high incidence of malnutrition among children from a particular community group, this may signify the group's powerlessness.

Nutrient Consumption and Body Function

Energy needs for children per unit of body weight are surprisingly high-proportionately much higher than they are for adults. A diet that is low in energy will, therefore, be associated with reduced activity and slowed growth. If body maintance is impaired and nutrient stores are decreased, weight will fall and the child will become thin. Several recent longitudinal studies in India and Bangladesh (1-3) have confirmed that children who are thin by anthropometric criteria have an increased likelihood of death.

Growth rates and physical dimensions are, however, not solely influenced by nutrient intake. Illness may impair digestion, absorption, and metabolism, thus affecting the efficiency with which a child makes use of his food. The enormous impact of different illnesses on child growth has been studied in the Gambia. Children with diarrhoea experienced the largest weight losses (5). Briscoe (6) has pointed out that it is the effect of diarrhoea on the child's appetite, and the tendency for the mother of a child with diarrhoea to withhold foods, that is the largest contributor to the effect of illness on energy availability. Histories of clinically malnourished children frequently reveal that they have had infections as well and that their intake is low. For example, a recent study in Guatemala (7) confirms that diarrhoea and infectious diseases cause a marked reduction in energy intake.

The Mother's Influence on Child Nutrition

Nutrition and infection relationships in early childhood

Children are generally viewed as the group in any population who are most vulnerable to poor nutrition. If undernutrition is detected through anthropometric measurements (deficits in weight for age, weight for height, or arm circumference), its prevalence will be greatest in those of weaning age. Children have very particular requirements, not only for nutrient intake, but also for the type of food that they can eat. They are also especially liable to suffer from diarrhoeal illness and other infectious diseases. A child's nutrition starts from the moment of conception. Before birth it is dependent on transfer of nutrients from the mother across the plancenta. If a pregnant mother has a low intake of energy, the child may suffer, too. Iron requirements are also increased in pregnancy; the anaemic mother will not be able to transfer nutrients to her child efficiently. Maternal illness, e.g., malaria, will also influence intra-uterine growth. The result of all these factors is that the child has a low birth weight, and this will again be associated with an increased risk of death during the perinatal period.

Birth weight can be successfully increased through maternal iron supplementation, anti-malarial treatment, and dietary supplementation of the mother during pregnancy. The newborn child is also dependent on his mother for nutrients. For the first few months of his life, at least, he will be solely breast-fed. Lactating mothers will produce adequate amounts of breast-milk for growth during these months even if they are mildly undernourished, though a recent review of growth patterns of children solely breast-fed in the first months of life suggests that the growth rates may lag behind those of wellnourished communities from as early as three months (8). The seasonal variability in mothers' dietary intake has been shown in a rural community in the Gambia, and this may influence milk output (9). Certainly, most mothers are unable to produce sufficient milk to sustain adequate child growth after six months, and supplementation will often begin around, or before, that time. Delayed supplementation or extreme lack of breast-milk production will lead to slowed weight gain and put the child in danger of undernutrition and death; early supplementation may provide adequate nutrient intake but put the child in danger of diarrhoea This will be compounded if breast-feeding is stopped early because of economic or social pressures.

Two problems are therefore associated with supplementation. One is bacterial contamination of food supplements because of unhygienic preparation, increasing the likelihood that the child will be exposed to pathogens causing gastro-intestinal illness, especially if the supplement is left in a warm place after it has been prepared (10). The other is the range of foodstuffs that are available to be given to the child. Small children have limited appetites. Their energy needs should be supplied through small, frequent feedings. Generally, food supplements consist of staples boiled and then mixed with water to form a soft paste (on rare occasions they will be ground to flour and fed as porridge). To be palatable they need to be liquid, and to be liquid they need to be heated and/or diluted with large amounts of water. The energy density of these foods is often very low indeed; children must eat large volumes to satisfy their energy needs. Maintaining energy requirements will therefore cause difficulties even if the child is well unless high-energy foodstuffs are available to be added to the child's diet (fats, oils, sugars). The addition of fats increases energy density, reducing the volume of food that provides a fixed amount of energy. Fats also increase the palatability of food, so that it may not be necessary to heat it or to add water before the child eats it.

As high-energy, dense foods are seen as important in enabling the maintenance of energy intake by all small children, they are vital if the effect of illness on intake and nutritional status is to be minimized. The problem is compounded if the child is ill and appetite falls; illness is necessarily accompanied by weight loss when high-energy dense foods are not available. The weight that is lost by the ill child cannot easily be regained. "Catch-up" will be slow unless the child is given at least one and one-half times the normal maintenance energy requirement; doubling of energy intake will enable return to normal weight 20 times faster than the maintenance value diet (11). Illness also precipitates malnutrition through an effect on absorption and metabolism. As well as affecting appetite, diarrhoea and other diseases (parasites and tuberculosis) may influence the body's ability to absorb nutrients.

By the time the child has reached the age of one year, the process of food supplementation may give way to weaning. A positive attempt will be made to transfer him from the breast to a more adult staple diet (though the one-year-old will still need special feeding and more frequent meals than the rest of the family). He will be exploratory, crawling around the home, putting everything into his mouth, and therefore particularly liable to disease. It may not be easy for his mother to prepare everything for him that he needs, especially if she has also given birth to another child or has reached the end of her "maternity leave" and needs to return to economically productive activities. In fact, the danger period for small children lasts from six months until at least three years, when they are able to feed themselves and may be exposed to fewer pathogens.

The foregoing discussion identifies the particular nutritional and health needs of small children, but it also shows that it is fallacious to attempt to distinguish the pregnant woman from the small child, at least when attempting to assess the problems that communities face. Mothers and children are certainly an extreme group with particular needs, but in fact their difficulties are often experienced by other community groups (e.g., the old, disabled, and even sometimes the healthier children and other adults), though to a lesser degree.

Maternal practices: in need of changing?

Mothers and their children are often identified as a group who should be given priority help. Usually this takes the form of a "package" of inputs, including supplementary feeding; nutrients like vitamins A and D, folate, and iron; family planning advice; etc. Some mothers and children will be considered high-risk-there is a strong chance that the child might die-and monitored more frequently, using longitudinal body weight or arm circumference to help identify those who are failing to thrive and need health or nutritional intervention (e.g., feeding, medical care, and education).

Health and nutrition education is a central feature of most mother and child health programmes. This is undertaken with a view to changing mothers' practices that are seen as potentially harmful to their infants and supporting those likely to be beneficial. Educational strategies have concentrated on the provision of simple messages (often relayed to mothers by extension workers and supported by mass media like radio advertisements or posters) in the hope that, if mothers are provided with sufficient knowledge to understand the dangers of their existing practices, they will change their beliefs and attitudes and adopt new patterns of behaviour. This approach has been used to promote increased frequency of infant feeding, the use of locally made weaning foods, oral rehydration for diarrhoea, the addition of high-energy density foods to infants' diets (particularly after illness), and so on. Often, however, despite the fact that the message is communicated well and understood by the mother, change still does not occur. Failure of an education programme to change behaviour is often ascribed to "cultural blocks," and intensive efforts are made to break them down.

In this context, "culture" is not an easy entity to define. We can conceive of a community's culture as the pattern of consistently held beliefs that represent community members' perceptions of their relationship with their environment. In practice, however, the influence of a community's culture on its members' actions has to be considered in the context of the life-style of the community's families and its overall social structure.

Many culturally determined actions have a demonstrably rational basis for those who practice them. They may consider that the results obtained are to their advantage- for instance, the reduction in a child's stool volume when food and fluids are withheld during diarrhoea or a lower birth-weight infant associated with a limited food intake during the last trimester of pregnancy. The risks that we perceive to accrue from these practices (dehydration of the child or low-birth-weight infant) may not be large enough to outweigh the immediate advantages that are seen by the mother or other advisers. We could perhaps refer to these as "pragmatic" practices.

Other culturally determined practices do not appear to have any immediate advantage even when considered from the mother's and family's perspective. They appear more to be ritual behaviour. One example is the highly complex series of rules that determine the types of foods to be given to adults and children who are ill. This kind of distinction is found in many parts of the world (e.g., the "hot" and "cold" dichotomy). The observance of these kinds of rituals may be reinforced by the community's authority figures-as a necessary assertion of the individual's commitment to maintaining the integrity of the community as a whole. Many such rituals concern the use of food. They are less likely to be adhered to in communities whose social structure is fragmenting.

A third reason culturally determined practices may be slow to alter is that the changes advocated may require additional effort or expense that family members do not consider to be worth the returns they expect to obtain. For example, if a mother is to provide locally made weaning foods for her child, add in fats, and ensure satisfactory oral rehydration when diarrhoea intervenes, she may need to increase the amount of time she spends on child-minding and purchase special foods when the child is ill. If she lives in an area where the majority of agricultural activity has to be undertaken during a few months, her time at that period will be extremely valuable. The additional investment required for her to be looking after her child, while apparently small from the outsider's perspective, may seem to the family to be excessive because of its potential effect on the next harvest. Even if the outsider is convinced that the family is wrong, there is nothing he/she can do to persuade them to change until they see the gains of changing to be worth the costs as they perceive them. This may explain the limited capacity of community health and nutrition programmes to change behaviour unless, at the same time, additional material resources are provided for the family (these may not have to be great- again, it is the perceived value that matters in promoting change, not the "real" value, or the outsider's assessment). In this respect, the displeasure of leaders that might be incurred if community rituals are not properly observed may also be thought of as a cost to the family. This could have two components-an economic cost (the family, for example, not being able any more to borrow on interest free loans) and a social cost (the family losing social prestige in the village).

So, for successful change to take place as a result of health or nutrition education, it would appear to be important (a) that the family should understand the likely benefits of the change suggested and (b) that the costs of the change suggested should not appear to be too great-including costs expressed as time, food, or money, displeasure of community "elders," or loss of social position.

But there are at least two other factors that may influence this process:

1. If the costs are quite large, behaviour change might still take place if there is overall social and economic change taking place in the community anyway. This is particularly so if economic or political changes have led to alterations in the processes by which the social structure of the community is maintained. However, economic changes per se can themselves lead to behaviour changes, though they may not bring about changes in the practices that are important features of the regulation of community structure.

2. Even if the costs of change are perceived to be large, it may still take place if a firm and trusting relationship builds up between members of the target community and the agent of change. This may explain why many small, non-governmental programmes are sometimes able to achieve much more in the way of changes in behaviour than larger, government-run programmes even when the latter provide the same input in terms of manpower, resources, etc. It also suggests that the relationship between extension worker and target families is an all-important factor in determining whether change will occur. The extension worker must show the family that he/she truly understands the kinds of economic and social constraints under which the family is living and support the family through its difficulties. Outsiders find this difficult to do; however, the community member who is closely identified with the privileged group in the community (if there is one ) may also start with a disadvantage.

There are, therefore, a number of quite sound reasons why it is often hard to initiate a successful behaviour-change programme in a community. Important factors are not only the design of the programme but also the social and economic features of the life-style of the target group families. The same maternal and child health (MCH) programme may be successful in one place and not in another, despite little difference in personnel. This emphasizes the the social and economic constraints that limit families' potential for changing what they do-and the importance of identifying these constraints.


We will consider a situation in which the initial focus for improving children's nutrition was on modifying mothers' child-care and nutrition practices through an MCH clinic, a nutrition rehabilitation unit, and a home visiting programme. After the programme had been in operation for a year, evidence collected suggested that, while many mothers had understood the messages that were being conveyed, relatively little impact had been made on their child-rearing practices. Reports from community health workers suggested that there were major constraints inhibiting successful change-at least, using the techniques that were at their disposal.

Subsequently, much detailed information was collected from families in the vicinity of the MCH project to find out whether education could be more effectively implemented through a village women's group. At the same time, a large integrated rural development programme was being set up in the surrounding area, and a baseline survey to assist with its planning was under way.

We looked at information collected by the Nepal Children's Organisation (NCO) child-care programme in Dhankuta (supported by the Save the Children Fund, UK) and by the Kosi Hill Area Rural Development Programme (administered by the Ministry of Local Development, Government of Nepal, and supported by the British Government's Overseas Development Administration).

FIG. 1. Nutritional Status of Children in Nepal, 1975 (source: ref.12)

The National Nutrition Status Survey

The Government of Nepal, in conjunction with USAID, undertook a nation-wide nutritional survey in 1975 (13), which revealed that 52.1 per cent of Nepal's children were chronically undernourished, or stunted, by international criteria; 6.7 per cent were acutely malnourished, or wasted; 3.8 per cent were both acutely and chronically undernourished. The distribution of this malnutrition by ages is shown in figure 1. The proportion of children with signs of chronic malnutrition rose from 23 per cent in the 6-11 -month age group to 48 per cent (12-23 months), 52 per cent (24-35 months), and 57-65 per cent (36-71 months). The prevalence of acute malnutrition was highest in the 12-23-month age group, 15.3 per cent (with 8.9 per cent showing signs of both acute and chronic malnutrition), and remained high in the 2435-month group; these are the ages when children are transferred from the mother's breast to an adult diet. It is no surprise that the percentage of stunted children-children who have experienced acute or chronic malnutrition in the past-is much greater for the 12-23-month age group than for 0-11 months and is constant from about 36 months onwards.

The Nepal Children's Organisation ChildCare Programme, Dhankuta, Eastern Nepal

Stimulated by concern about child malnutrition, the Nepal Children's Organisation (non-governmental) joined forces with the Government of Nepal's Mother and Child Health/Famity Planning Project to establish, in July 1977, a children's clinic in Dhankuta Bazaar with the aim of providing simple curative and preventive health facilities and nutrition education to try to reduce the incidence of childhood malnutrition and under-five mortality. The clinic incorporated a small residential unit in which mothers with severely malnourished children could stay for one month and participate in nutritional rehabilitation.

Attendances at the clinic and the nutrition unit in the first two years are shown in figure 2. There is pronounced seasonal variation: clinic attendance was lowest between September and March (Nepal) months Ashwin to Falgun), rising to a peak in May-July (Jesth-Srawan). In the second year attendance rose dramatically in April (Chaitra), probably as a result of an outbreak of measles in the area at that time. In 1977/78 the numbers of children registered in the nutrition unit were highest in May through August (Jesth-Bhadra). A similar pattern was found in 1978/79, though attendance started to rise earlier, again, probably because of the measles outbreak.

The proportion of clinic attenders with diarrhoeal diseases was particularly high during the pre-monsoon and monsoon seasons. This is also the time of year when the majority of families are short of food. Similar seasonal patterns were found in attendance at other centres in the area.

FIG. 2. Monthly Attendance at the Children's Clinic and New Cases in the Nutrition Rehabilitation Unit, Dhankuta, Nepal, 1977/78 and 1978/79

Longitudinal Studies of Child Health and Nutrition

These seasonal variations in clinic attendance figures promoted a further study of malnutrition and morbidity patterns in a rural community. In mid-1977 the NCO initiated a pilot home-visiting health and nutrition education programme in Chuliban Panchayat, to the south of the bazaar (13). The weights, heights, and illnesses of approximately 200 children in the panchayat (now part of

Dhankuta Bazaar Panchayat) were continuously monitored. Chuliban is a community of dry-land farmers, many of whom have very small landholdings. It includes the village of Santang, which was the subject of more intensive investigation. The Chuliban information has been analysed to investigate seasonal changes in nutritional status, the prevalence of diarrhoea, and the prevalence of vitamin A deficiency.

Nutritional status (anthropometry)

Preliminary analysis of the Chuliban data was undertaken in the summer of 1979, concentrating on the Nepali months Jesth, Bhadra, Mangsir, and Falgun. The prevalence of wasting malnutrition in children under five showed considerable variation during 1978/79 (Nepal) years 2035/ 36). Results are shown in figure 3.

Wasting prevalence was highest at the end of the monsoon, during August/September (Bhadra), and lowest in November/December (Mangsir), three months after the maize harvest (differences statistically significant). Prevalence was higher three months later, in February/ March (Falgun).

The percentage of children whose weights were greater than the expected weights for their heights showed corresponding variation. It was lowest (16.7 per cent) at the end of the monsoon, in August/September (Bhadra), and highest (35.9 per cent) in November/December (Mangsir); the percentage then fell to 25.8 per cent in May/June (Jesth). The differences are again statistically significant.

It was thought that the likely explanation for this fluctuation was that food was in short supply in many Chuliban households during the pre-monsoon and monsoon months.

FIG. 3. Seasonal Variation in Children Who Are Wasted and Children Who Are of Greater Than the Expected Weight for Height-Chuliban Panchayat, Nepal

FIG. 4. Preliminary Analysis of Information on Weight Gains of Children in Chuliban Panchayat, 1979/ 80. Mean and standard error of weight gain of individual children aged 5-23 months over successive two-month periods (no observations were made in January 1980). A difference of 300 9 is significant.

Additional problems would be caused by diarrhoeal and other illnesses influencing children's appetites and, perhaps, by the lack of time available to mothers to care for their children during this period. The first maize harvest comes in early August (Schrawan); so it is likely that nutrition will only start to improve from August onward.

Recently attempts have been made to look in more detail at the weight and height changes of individual children. This kind of growth velocity information provides a sensitive measure of changes in a child's nutritional status.

Figure 4 shows the marked difference in growth velocity found when weight gains of a subsample of about 20 children from four wards in Chuliban Panchayat were examined (this is a preliminary analysis and needs confirmation). It was not possible to obtain consecutive weight change figures for all children for each interval, but means were taken of those available. February/March, March/ April, and May/June gains were all low. Weight gains were significantly greater in July/August (presumably related to the maize harvest in early August), but were greater still in September/October and November/December.

The expected weight gain of a child in the second year of life is between 2 and 2.5 kg. Assuming that the mean weight gain in January/February is 300 9, the annual weight gain of the children in this group is around 2.3 kg, well inside the expected range. However, as shown in table 1, there is immense variation in weight gain throughout the year.

TABLE 1. Variation in Mean Weight Gains of Individual Children in Chuliban Panchayat through the Year

Gain (g)
of Annual
Weight Gain
Mar.-Apr. 1979 11-1 70 3
May-June 1-3 210 9
July-Aug. 3-5 460 20
Sept.-Oct. 5-7 700 31
Nov.-Dec. 7-9 650 28
Jan.-Feb. 1980* 9-11 200 9
Total   2,290 100

* Estimated

FIG. 5. Seasonal Variation in Point Prevalence of Children with Diarrhoea and Measles (Measles Expressed as Percentage of Susceptible Children)

Diarrhoea and vitamin A deficiency

Figure 5 shows the pattern of diarrhoeal illness and measles among the children under five years old in Chuliban in 1978/79. The histograms indicate the percentage of children who, when visited between August 1977 and June 1979, were found to have abnormally frequent stools; this should represent the monthly point prevalence rate of diarrhoea. The prevalence of measles is expressed as the percentage of susceptible children (i.e., those with no known previous infection) who developed the disease.

In 1978 the prevalence of diarrhoea was greatest in May through July; in 1979 it was also high in April (perhaps related to the measles outbreak? ). Comparison with the Dhankuta district rainfall figures (from the meteorological station two miles north of Chuliban) shows that the month of peak prevalence of diarrhoea coincided with the start of increased rainfall-the early monsoon showers. This may be circumstantial evidence to suggest that the early rains cause a high level of environmental contamination by faeco-oral bacteria; bacteriological studies of environmental and watersupply contamination following early rains have yet to be undertaken in Nepal.

The findings reported in this section suggest that when food supplies are lowest, the likelihood of childhood diarrhoeal illness is highest. The peak prevalence of diarrhoea precedes the peak prevalence of wasting malnutrition by two to three months. It is therefore not at all surprising that children's nutrition deteriorates during the monsoon months. Children, especially those who are ill, may have reduced energy and nutrient intakes during this period. Parents will not easily be able to afford the hidden cost of taking a child to a hospital or health post. They certainly will not be able to manage the actual cost of extra ghee or sugar needed to provide a sick child with the energy it requires in a high-density, low-volume form. There was similar seasonal variation in the prevalence of mild signs of vitamin A deficiency. This coincided with the diarrhoeal peak, which is also the period in which green vegetables are in short supply.

Case Histories from Malnourished Children

The staff working in the Nepal Children's Organisation Nutrition Unit, under the supervision of Durga Pokhril, collected information from each mother with a malnourished child who registered in the unit. To date, over 500 case histories have been collected. These make sober reading. The choices that many mothers have to make when their children become ill and lose their appetites during the early monsoon months are extremely difficult, especially if the family faces a substantial food deficit. There seemed to be no doubt that economic constraints exert an influence on the decisions that families make about the kinds of food to be given to their children under normal circumstances, and especially when they are ill.

Nutrition and Family Economy

A recent report ( 14) reviewed some of the difficulties faced by communities in less-developed countries during different seasons, particularly communities of subsistence farmers, and presented the following scenario for regions with "unimodal" rainfall:

"Towards the end of the dry season water becomes scarce. There is a rise in the labour and energy requirements for fetching water and watering livestock. The poorer people, who may have no land or small plots and/or weak family labour, begin to suffer more than others. They have less food because they have been able to grow less, have fewer livestock and less money. They may eat less to save food for the crucial time of cultivation. Work is scarce and wages are low at this time of year.

"The rains bring the start of the 'hungry season.' For cultivators, future food supplies and cash income depend on timely agricultural activities during an often brief period for land preparation. Poorer farmers may have difficulty obtaining inputs like seeds, fertilisers or draught power. For all those with land, heavy and urgent demands have to be met-land preparation, transplanting and weeding-by both men and women. These demands come at a time when food supplies are generally short, high energy foods are a luxury. Labourers benefit from being able to get some work, though many subsistence farmers are in negative energy balance and lose weight. At the same time, food prices are high and transport difficulties may limit the scope for intervention in the food market by central authorities. The hard work may induce mothers to reduce frequency of breast feeding, breast milk output may fall as a direct result of reduced material dietary intake and frequency of feeding of weanling children may also be reduced. Food preparation is hurried, diets are constrained by family food availability and may be less nutritious. Less time may be spent on cooking, house-cleaning, water collection, fuel gathering and child care, and more of women's time is spent on agricultural operations.

"The rains are also the least healthy time of the year. Some of the more serious infections peak during or just after the rains including malaria and diarrhoea. This is also the period when Protein Energy malnutrition is likely to develop and increase susceptibility to illness by causing a reduction in the immune response. Infections increase the vulnerability of rural people by coinciding with peak labour demand; failure to cultivate, weed or harvest may critically affect future income or food supplies. Inability to work at this time has high costs.

"In such a situation poorer people are often driven to distress sales and borrowing during the wet season. They sell or mortgage land, livestock, jewelery, their future crop or their future labour. They get into debt and are forced to pay high interest rates-becoming dependent on creditors with whom they have to stay on good terms as further indebtedness is likely in subsequent seasons. This is the time of year when dependent relationships begin or are reinforced and deepened.

"When harvest comes, wages are high. Work is hard but people are still unwell or weak from food shortage and sickness in the lean season. Food prices are low, which is good for the landless who buy in stocks at the cheap rates to last them through to the next season, though this is a problem for the sharecroppers and cultivators who have to sell crops in order to raise cash to repay loans and for harvest festivals soon to follow.

"After harvest things improve. Food intake is better both in quality and quantity. Morbidity drops, as does mortality, and body weights rise. There are ceremonies, celebrations, marriages and, perhaps, a peak in rates of conception. But gradually the cycle starts all over again."

Several of the Nepal Children's Organisation's findings tied in with the nutritional outcomes predicted by the scenario. We needed to identify whether there was a similar interplay between family economy and child health/nutrition in eastern Nepal and, if so, to find out which families were affected the most.

Two opportunities emerged for a study of relationships between family economy and child nutrition of the hill area of eastern Nepal. One was a socio-economic survey of 2 per cent of the population in the Kosi Hill area (total population 500,000) undertaken in preparation for the beginning of an integrated rural development programme there (15). The other, mentioned above, was a detailed study of families around the NCO children's clinic in Dhankuta (most of the families came from Santang Village in Chuliban Panchayat, the population of the panchayat was 2,500). This preceded an educational/ development programme that was to be catalysed in that village by the Dhankuta office of the Women's Affairs Training Centre (16). The results of the two studies are summarized below.

The Socio-economic Survey

In 1978 (a good year) farmers cultivating less than 0.5 hectare of land ("small farmers") produced, on average, 54 per cent of their calculated cereal needs, while those cultivating more than 1 hectare ("large farmers") produced 113 per cent of their needs-a surplus. In 1977 (a bad year) the figures were 30 and 90 per cent respectively. The 0.5-1hectare group ("medium farmers") had productions between the extremes in both years. Approximately 43 per cent of the area's farmers have small holdings, 31 per cent medium, and 26 per cent large.

If the type of land cultivated is taken into account, the disparity is greater still. Farmers with up to 1 hectare of non-irrigated land averaged a production that was less than half of their requirements in 1975, and only about a quarter of their requirements in 1977. This marginal group constitutes 20 per cent of the total in the Kosi Hill area.

In the hill regions, the first crop to be harvested is maize (August, towards the end of the monsoon); millet and rice follow. Deficits are therefore most pronounced between January and July, coinciding with the periods of heavy agricultural activity in April, May, and June and the early monsoon peak of diarrhoea prevalence.

Families experiencing deficits expend much time and energy on obtaining sufficient food for the rest of the year. They manage their deficit by

Compared with farmers with surpluses, those likely to experience deficits tend to spend a higher proportion of their total annual cash expenditure on food-spending proportionately less on meats, fats, alcoholic drinks, vegetables, fruits, and spices but more on cereals. Purchasing power is influenced by seasonal variations in cereal prices.

They obtain more of their income off the farm and less from the sale of farm produce. Farm produce sold by small farmers consists primarily of animals and animal products; larger farmers sell cereals. Small farmers depend on off farm income; two major sources are local labour and porterage. In medium-size farm families, returns from migrants are also important.

Small farmers sell more small livestock and other assets (e.g., land) which, if they were not sold, would enable them to produce food for sale. They incur more debts for consumption and obtain fewer production loans: and they benefit less from institutional loans.

The results of deficit management for the small farmer are

One likely consequence of an inability to manage the deficit is permanent migration. This occurs especially in the years when the monsoons are poor. Families particularly likely to be affected are those who experience a crisis during the deficit months-e.g., an unexpected demand for cash to pay for medicines or funeral expenses.

A second likely consequence is seasonal undernutrition of family members. The Chuliban Longitudinal Survey revealed seasonal variation in the nutritional status of children. Adults-especially women-commented that during the monsoon months they become thinner and their clothes hang loosely on them. However, it has not yet been possible to relate seasonal nutritional information to the degree of family deficit in food production in Chuliban.

A more widespread cross-sectional nutrition survey was undertaken in the middle of the monsoon in a subsample of the families studied for the socio-economic survey. Anthropometric data were collected for 648 children (no height data on 30) from 307 households. Table 2 shows the percentage of children who were wasted (less than 80 per cent of expected weight for height) in three different age groups (12-35, 36--59, and 60-95 months), analysed by the amount of land cultivated by their family. Table 3 shows similar percentages for children who were stunted (less than 85 per cent of expected height for age). There was a higher prevalence of wasting amongst small farmers' children, particularly in the 36-59-month group; and a higher prevalence of stunting amongst small farmers' children in both the 12-35-month group and the 60-95month group.

TABLE 2. Percentage of Children Who Were Wasted, by Age Group and Area of Land Cultivated by Household-Kosi Hill Area Nutrition Survey, 1979


Percentage Wasted*

Area Cultivated 12-35 months 36-59 months 60-95 months 0-95 months
0.0-0.5 ha 22 (27) 11 (28) 2 (44) 9 (108)
0.51-1.0 ha 23 (43) 4 (55) 1 (74) 6.5 1198)
Over1.0 ha 2 (91) 1 (76) 1 1113) 2 (312)
Total 11 (161) 4 (159) 1 (231) 5 (618)

* Wasting defined as weight less than 80 per cent of expected weight for height, using WHO/NAS standards. Figures in parentheses indicate total number of children in each group.

TABLE 3. Percentage of Children Who Were Stunted, by Age Group and Area of Land Cultivated by Household-Kosi Hill Area Nutrition Survey, 1979 Percentage Stunted*


Area Cultivated

  12-35 months 36-59 months 60-95 months 0-95 months
0.0-0.5 ha 24 (25) 34 (29) 52 (40) 39 (94)
0.51-1.0 ha 19 (42) 37 (54) 45 (74) 36 (170)
Over 1.0 ha 10 (81) 32 (71) 27 (108) 23 (260)
Total 15 (148) 34 (154) 37 (222) 30 (524)

* Stunting defined as height less than 85 per cent of expected height for age, using WHO/NAS standards. Figures in parentheses indicate total number of children in each group.

FIG. 6. Summary of Factors Likely to Be Influeneced by Seasons and Family Economy Which Affect the Nutritional Status of Small Children

Taken together, these findings are strong evidence for a relationship between seasonal undernutrition and family wealth. We conclude too that, given the constraints faced by small farmers' families when managing deficits, the undernutrition in many cases will be virtually unavoidable. It will, however, be exacerbated by other influences such as birth interval, the mother's actions when children are ill, etc., and these will be responsible for the variability in children's nutritional status found within each economic group. These factors are summarized in figure 6.

A third possible consequence would be the untimely death of family members. Given that the undernourished child is at greater risk of death than one who is well nourished, we would anticipate that the rate of death of children under five years old would be greater in small farmers' families than in large farmers' families. The data collected to date have not enabled us either to support or to deny this hypothesis.

Family Profile Studies

A rather different-but complementary-approach was used to study relationships between family economy and beliefs and attitudes relating to children's nutrition in two villages near Dhankuta Bazaar, Santang (in Chuliban), and Nigale. The purpose of the study was to explore factors that influence the potential of families to change their behaviour. A major feature of the approach was the investigation of the way in which a family's previous experiences influenced its actions at present and its expectations for the future. The assessment of a family's economy, therefore, did not depend solely on its present status but was more concerned with what had happened to the family over the previous 20 years

Thirty families were studied in depth, with formal recall questionnaires covering assets, production, and farm management. Open-ended interviews were used to elicit beliefs and attitudes, often structured by asking family members to comment on how they might handle particular situations both now and in the future. A three-day period of observation by "participant obervers" provided some insight into behaviour patterns, though, because this was a pilot study, the general format of data collection was not rigorous enough to permit more than a very little quantitative analysis. Lessons learnt may, however, provide useful clues for using quantitative data-handling techniques in the future.

The results are being prepared as annotated and interlinked case studies. Four different groups of families have been defined in Santang, and their patterns of likely behaviour now or in the future have been generalized as far as possible. Since Santang is a highly unrepresentative village (though what village is representative? ), the four groups are identified and the patterns described in brief. A fifth group, found in Nigale but not in Santang, is also described.

Group 1: Farmers producing a grain surplus, employing labour, and investing in land

Members of the first group produced an annual grain surplus, which provided them with a regular income that was invested in purchasing more land. They employed labourers. They were also in a position to lend money and could easily afford to have their children educated.

Group 2: Farmers normally having a small deficit in grain production-desiring to invest and diversify and, if possible, to increase land holdings

Group 2 consisted of farmers who had small deficits in grain production which they were able to compensate for through sales of livestock and agricultural produce (including fruit and vegetables), sales of natural products (e.g., firewood), and wage labour. They have managed until recently to obtain sufficient cash in the deficit months to purchase food for consumption, though they faced severe and unexpected difficulties if their economy was put under stress by a poor harvest or a demand for unusual expenditures (e.g., medical costs or a funeral). Their diminishing land yields and increasing costs (as a result of inflation) have encouraged them to try to accumulate cash reserves, invest in livestock, and further diversify production into relatively risk-free, noncompeting produce (in this case pineapples).

Group 3: Farmers having substantial deficits, dependent on the labour market, and with regular crises associated with disinvestment

The farmers in group 3 faced substantial deficits each year. They too obtained cash through sales of fruit, vegetables, livestock, natural products, and wage labour. However, annual crises and shortages were for them the rule and not the exception. They were dependent on an uncertain labour market and were often heavily in debt-earning cash not only to purchase food but also to pay off interest charges. Their chances of buying land were extremely small; they were not easily able to raise production loans to diversify production. They were more likely to disinvest by selling animals and mortgaging land. They were already experiencing increasing poverty, and stresses such as poor harvests, funeral expenses, etc., exacerbated this process.

Group 4: Landless artisans meeting food needs through casual labour, with substantial debt and continuous crisis and no prospects of investment at all

The fourth group in Santang consisted of villagers who were virtually landless. They were mostly artisans in caste trades (blacksmiths, tailors, etc.) who were entirely dependent on wage labour (agricultural and non-agricultural) to meet their annual food needs. They were in severe debt, and their assets were mortgaged. They lived a handtomouth existence, struggling from crisis to crisis, saw no prospect of accumulating wealth, and had no sense of any ability to invest either now or in the future. If they perceived safe income-earning opportunities elsewhere in the district or outside, they would move; such a move might be precipitated by the arrival of a crisis they were unable to overcome.

Santang farmers in groups 3 and 4 have learnt to be successful manipulators-identifying low-risk methods of obtaining cash and avoiding new ideas and schemes that have yet to prove their worth. In many ways, it is the group 2 families who may be in greatest danger during an unforeseen crisis, especially in the deficit months; they may not be used to deficit management and the opportunistic responses it requires.

Group 5: Farmers with surpluses who invest in businesses, doing well as entrepreneurs

The farmers of group 1 with surpluses, like many other families in Dhankuta district, invested their profits in land. However, in other villages close to Dhankuta Bazaar, there was another group of entrepreneurs, group 5, who had made profits from sales of surplus grain and other crops and who invested in businesses-tea shops, construction, trade in animals, etc. Their profits, besides permitting increased consumption, also enabled them to repay production and other loans, to lend money, to educate their children, and, if appropriate, to expand their businesses. Members of group 5 were in an excellent position to benefit from many of the opportunities from development programmes in the district, especially if they lived close to the bazaar or main trails.

Though no village in Nepal can be said to be altogether typical, Santang has certain major features that make it particularly exceptional. It is inhabited by a close-knit group of Atapre Rais-a tribe concentrated in only a few discrete areas close to Dhankuta Bazaar. In this respect it is relatively introverted. Those Atapres who are able to buy land tend to do so in the village even though prices are rising and yields are failing. Proximity to the bazaar provides a market for fruits, vegetables, natural products (firewood and lime), and labour. Thus, group 4 families can survive, whereas if they lived further afield they would probably have had to migrate. It is possible, though, that some features of the group 3 families are similar to those of households with 0-0.5 ha of land in the Kosi Hills area socio-economic study, the group 2 families may share some characteristics with the households with 0.51-1.0 ha of land, and the group 1 families with the households having more than 1.0 ha.

However, other results of the Santang study show that we should be careful when making generalizations about the perceptions and actions of groups of farmers who appear to be of similar economic status. One of the major factors in Santang that has placed many families in difficulty is the tradition of multiple inheritance, whereby all male heirs inherit an equal share of the family's assets. The extreme scarcity of land in the Santang area has meant that in many cases landholdings that in the past easily supported the families of grandparents have been divided into small fragments by the time they are inherited by grandsons.

In some families impoverishment has been rapid; in others, it has progressed slowly. A few families may, in fact, be gradually accumulating small surpluses through careful economic management. The land holdings, staple-production potential, and animal holdings of families from groups 1, 2, and 3 may be found to be similar when assessed at the present time. However, each family's familiarity with poverty, perception of the options available, and expectations for the future will be different-in part, at least, conditioned by its experiences over the preceding few years. It is not at all surprising, therefore, that the ways in which families of similar economic standing react to stresses such as seasonal food shortage and superimposed crises such as illness, harvest failure, etc. may not be the same. The Santang studies suggested that to some extent these diversities can be explained by reference to the family's history and to the economic and social influences that have been encountered.

The Santang analysis emphasizes the importance of the mother's perception in determining the way in which her family-particularly small children-are fed. Her perception -influenced by priorities defined by authority figures in her community group and in the family, by her previous experience, and perhaps by knowledge-will in turn affect the actions she takes in response to different situations she perceives, the resources she has, and demands she faces (e.g., time). These include child hunger and illness.

Not all mothers will act in the same way in response to specific situations even if they come from the same economic group or have similar economic histories. For example, some will be keener than others to adopt new ideas. One purpose of the Santang studies was to explore the way in which economic constraints-as perceived both by the mother and by outsiders-appear to influence the potential for adopting new practices and for managing crises of the sort that may lead to the development of child malnutrition during food-deficit periods. There was a suggestion that those in groups 1 and 5 were most likely to innovate by choice, those in group 4 out of desperation. Families in group 4 would also be more likely to be used to crisis management, while families in group 3, in particular, might not be able to adapt quickly enough to cope with difficulties.

Using the Results of the Investigation to Devise Objectives for Development Programmes

We have described a series of studies seeking to explore relationships between family economy and social status, child health and nutrition, and practices that influence nutritional status. These have been examined in different seasons during the year. The results suggest that children's nutrition can be expected to improve if an economic development programme is effective and perhaps as the result of an intensive and well-administered child health and nutrition programme.

We might outline possible objectives for child health care and nutrition programmes and for integrated rural development activities, concentrating on improving the well-being of children and their families, as follows:

a. Intermediate outcomes

(i) child health-care end nutrition education

(ii) integrated rural development

b. OveraII outcomes

The material presented in the case study also shows that there are various interrelated processes working at the family level that may have implications for both programme design and the selection of approaches to evaluation. For example, there is a clear inverse relationship between a household's wealth and its children's nutritional state, especially if the child's age is taken into account. Studies of the socio-economy of families with varying wealth and of the demands on family members during food-deficit periods reveal the economic background of the association. The family profile studies suggest that the nutritionrelated behaviour and innovative potential of families of varying economic levels can be related to patterns in their history. As with all descriptions of reality, where attempts are made to go into increasiny detail about any aspect of life at progressively deeper levels, such as the individual child, the mother-child dyed, the family, or the community, increasingly complex processes are identified. The challenge is to distil them into representations that are comprehensible. We try to identify indicators of these community processes that can be used by policy makers and programme planners to help identify potential for successful action.

Our current view is that long-term improvements in the well-being of the poorest households (those cultivating less than 0.5 ha) will only result from increasing the productive resources at their disposal. Short-term improvements will only result from sympathetic supportive welfare in the form of consumption credit, curative as well as preventive health care, subsidized foods available throughout the year, and guaranteed off-season employment with controlled wages.

These options are both financially costly and difficult to implement successfully. However, the necessity for radical action to help the poorest in the community is accepted by many governments in South-East Asia, including that of Nepal. Certainly the prediction that food supplementation and health education will not have a sizeable long-term impact on malnutrition in the poorest families is useful for programme-planners.


This material has been obtained through the efforts of a large number of government officers in the ministries of Health, Agriculture, Education, and Local Development, staff in the respective programmes, village leaders, and visiting workers. Over the last four years, they have spent many months collecting, analysing, interpreting, and discussing the data and using them to plan and evaluate interventions with either a primary or a secondary objective of improving community nutritional status.


1. A. Kielmann and C. McCord, "Weight for Age as an Index of Risk of Death in Children," Lancet, i: 1247 (1978).

2. A. Sommer and M. S. Lowenstein, "Nutritional Status and Mortality: A Prospective Assessment of the QUAC Stick," Amer. J. Clin. Nutr, 28:209 (1975).

3. L. C. Chen, A. K. M. A. Chowdhury, and S. L. Huffman, "Anthropometric Assessment of Energy-Protein Malnutrition and Subsequent Risk of Mortality among Preschool-Aged Children,' Amer. J. Clin. Nutr., 33: 1836 (1980).

4. A. Tomkins, "Nutritional Status and Severity of Diarrhoea among Preschool Children in Rural Nigeria," Lancet, i: 860 (1981).

5. T.J. Cole and J. M. Parkin, "Infection and Its Effect on the Growth of Young Children: A Comparison of the Gambia and Uganda," Trans. Roy. Soc. Trop. Med. Hyg., 71 13): 196 (1977).

6. J. Briscoe, "The Quantitative Effect of Infection on the Use of Food by Young Children in Poor Countries," Amer. J. Clin. Nutr., 32:648 (1979).

7. R. Martorell, C. Yarbrough, S. Yarbrough, and R. E. Klein, "The Impact of Ordinary Illnesses on the Dietary Intakes of Malnourished Children." Amer. J. Clin. Nutr., 33: 345 (1980).

8. J. C. Waterlow, A. Ashworth, and M. Griffiths, "Faltering in Infant Growth in Less Developed Countries," Lancet, ii: 1176 (1980).

9. M, G. M. Rowland, A. Paul, A. M. Prentice, M. H. Muller, R. A. E. Barrell, and R. G. Whitehead, "Seasonal Aspects of Factors Relating to Infant Growth in a Rural Gambian Village," (paper presented at IDS/ROSS Conference: Seasonal Dimensions to Rural Poverty, IDS, Sussex, UK, 3-6 July 1978.

10. M. G. M. Rowland, R. A. E. Barrell, and R. G. Whitehead, "Bacterial Contamination in Traditional Gambian Weaning Foods," Lancet, i: 136 (1978).

11. A. Ashworth, "Progress in the Treatment of Protein-Energy Malnutrition," Proc. Nutr. Soc (UK), 38: 89 (1979).

12. E. W. Brink, 1. H. Khan, J. L. Splitter, N. W. Staehling. J. M. Lane, and M. Z. Nichaman, "Nutritional Status of Children in Nepal, 1975," Bull. WHO, 54: 311 (1976).

13. D. N. Nabarro, "Report for the Period March 1977 to March 1978," Joint NCO/SCF Nepal Project at Dhankuta, Eastern Nepal (1978).

14. R. Longhurst and P. Payne, "Seasonal Aspects of Nutrition" (paper presented at IDS/ROSS Conference: Seasonal Dimensions to Rural Poverty, IDS, Sussex, UK, 3-6 July 1978.

15. S. Conlin and A. Falk, Kosi Hill Area Rural Development Programme (KHARDEP), Nepal-A Study of the Socio-Economy of the Kosi Hill Area: Guidelines for Planning end Integrated Rural Development Programme, vols. 1 and 2, KHARDEP report no. 3 (Land Resources Development Centre, Surbiton, Surrey, UK, 1979).

16. Nepal Children's Organisation, First Report of Joint NCO/ Women's Affairs Training Centre Child Health Support Programme, Dhankuta, Nepal (1979).


Human nutrition is an area which allows the holistic study of individuals and communities. Although nutrition itself probably has a limited role in health, it is currently accepted that a person interested in nutrition may investigate clinical medicine, epidemiology, demography, economics, sociology, agriculture and politics, and other relevant disciplines. Nutrition, viewed as a focus of many human actions, has the capability of integration and synthesis that few of the specialized sciences can claim. In this sense nutrition combines the analytical power of the "hard" laboratory disciplines with the broad view of the social sciences, the satisfaction of critical bench experiment with the slow long-term goal of community change. It is therefore possible to use a wide range of scientific methods to investigate "nutrition" problems.

However, many of the problems are still poorly defined. The appropriate methodology is to determine that the problem exists through case reports, to identify the principal associated factors by epidemiological methods, and then to seek the underlying physical and social mechanisms. Although it is necessary for an academic to follow problems through, many of these have such urgent implications that remedial measures are needed before knowledge is complete. The design of intervention programmes in the face of uncertainty and their critical evaluation are valid areas of academic interest.

Contents - Previous - Next