Previous - Next
This is the old United Nations University website. Visit the new site at http://unu.edu
INTRODUCTION
In response to a request from the Office of Nutrition of the US Agency for International Development (AID), the National Research Council established the Subcommittee on Maternal and Infant Nutrition in Developing Countries to provide guidance to the Office of Nutrition for its expanding programme in maternal and infant nutrition. The subcommittee serves under the Committee on International Nutrition Programs of the Food and Nutrition Board, Commission on Life Sciences. The subcommittee was asked to (a) review the state of knowledge of infant feeding practices and maternal nutrition in developing countries with a view toward programme implications, (b) evaluate the scientific and operational basis of programmes in breast-feeding, weaning, and maternal nutrition, (c) evaluate programme needs and guidelines, and (d) identify research priorities in the above areas.
The subcommittee determined that the approach to this assignment that would best utilize its expertise was to review and evaluate the scientific basis of maternal nutrition and infant and young child feeding practices, especially those aspects that may form the basis for interventions. The subcommittee felt it beyond its expertise and the time allotted for the task to systematically evaluate specific maternal and child health programmes in developing countries.
To accomplish its task, each subcommittee member prepared a paper on a specific topic in infant or maternal nutrition that then formed the basis for discussion of research needs and approaches to intervention. This report, which is a product of the entire subcommittee, summarizes the subcommittee's analysis of the scientific knowledge upon which interventions in maternal and infant nutrition can be based and identifies gaps in knowledge and recommendations for research. This report was transmitted to the funding agency in July 1983.
The subcommittee assumes, based upon discussions with AID Office of Nutrition staff, that the primary audiences for this report are AID field staff and Washington staff who have at least a general health background, but who may not have access to the literature cited here. Thus, the subcommittee chose to present a summary of existing knowledge rather than an exhaustive literature review.
Several additional comments are necessary to provide the context for the subcommittee review and recommendations and to understand the premises underlying the subcommittee's analysis:
1. Although the importance of nutrition and health during pregnancy as determinants of pregnancy outcome has been recognized for some time, a more comprehensive concern for the needs of the woman during other phases of her life cycle is only beginning to receive the attention it deserves. The subcommittee endorses, and the report reflects, the more comprehensive approach that attention should be paid to women not only during pregnancy and lactation but also before they reach reproductive age and in between pregnancies. Furthermore, the period of lactation is approached not solely from the perspective of the infant's welfare, as has often been the case in the past, but also from the perspective of its effect on maternal nutrition and health status.
2. A basic premise underlying this report is that nutrition and health must be considered as a unit. Given this premise, the subcommittee is convinced that actions can be taken within the broad context of health that will improve the nutrition and health of women and young children in developing countries.
3. The subcommittee also recognizes that nutrition and health are not an isolated unit and that in developing countries problems in growth, development, and attainment of full functional capacity have multiple causes, among which nutritional, socio-cultural, economic, and environmental conditions and health care play synergistic roles. Consequently, interrelated non-nutritional factors need to be considered in the design, implementation, and evaluation of nutrition interventions. The subcommittee is especially aware of the close association of socio-economic status and nutrition.
4. Severe malnutrition is clearly a serious problem in a number of developing countries. However, for this report it is considered a medical problem that is not within the scope of the subcommittee's charge. The primary concern in this report is prevention of mild and moderate (and hence, severe) malnutrition.
NUTRITION OF WOMEN AND CHILDREN- STATEMENT OF THE PROBLEM
Nutrition of Women of Reproductive Age
The paramount importance of nutrition of a woman during pregnancy and lactation for her own health and that of her offspring is widely recognized and has received increasing attention in recent years. In many Third World countries malnutrition is endemic, fertility rates are high, and women enter the reproductive stage at an early age and subsequently attain high parity. Maternal mortality rates are high, and the incidences of foetal wastage, low birth weight, premature birth, and perinatal death are several times those reported for industrialized societies (1).
Furthermore, from studies on women's roles in developing countries (2-7), it has become evident that for poor women pregnancy and lactation do not interrupt the flow of daily activity. In communities where women carry heavy economic responsibilities, often as the sole or primary supporter of their children, neither pregnancy nor lactation can be allowed to disrupt the pattern of work. Under such circumstances, when food availability is also constrained, the physiological burden of pregnancy and lactation is immense.
Clearly, there must be behavioural and physiological adjustments for the majority of women to survive. Although it is important to learn more about the nature of these adaptive processes, it is even more important to understand the cost-to women, their children, and society -of their adaptive adjustment. It is only by focusing on these, as well as on the many gaps in knowledge outlined in this report, that we can begin to comprehend the impact of nutritional deficits on maternal health and well-being and soundly plan our efforts for effective correction of undesirable consequences.
Nutrition in Preparation for Reproduction
Undernutrition of future mothers during their growth and development is one factor influencing the reproductive life of women. One mechanism through which undernutrition acts is by delaying the age at which menstruation starts (menarche). For example, there is evidence that "undernourished" girls are older when menstruation starts (810) but are of similar skeletal maturation compared with "well nourished" girls (10). Bongaarts (8) succinctly reviews the evidence for a nutritional effect on menarche.
Recent reports based on data from industrialized countries suggest that menarche is delayed or amenorrhoea may ensue when women, especially lean women, engage in heavy physical exercise (11-13). Whether strenuous physical activity among young women in marginally nourished agricultural populations potentiates a delay in menarche is still unknown.
Whether delayed menarche constitutes a relative advantage for an "undernourished" group, because it allows time to attain a greater pelvic size, as seems to be the case for young "well-nourished" women whose menarche occurs later (14), is still unknown. However, if chronic undernutrition delays pelvic growth as it delays menarche, any relative advantage of late maturation could be lost. Moreover, even if delayed maturation results in larger pelvic size at menarche (14), there still may be other negative effects of poor nutrition on reproductive performance and maternal nutritional status, especially if pregnancy follows menarche promptly.
Although undernutrition may increase the age at which the birth canal reaches adult size, there is no conclusive evidence that undernutrition specifically alters pelvic growth except in cases of severe rickets occurring during childhood and of osteomalacia. However, perinatal complications are more common among populations where nutrition problems exist, particularly if prenatal care is poor (15). The contribution of cephalopelvic disproportion to these complications is still unclear, as is the contribution of nutrition to pelvic growth. It is believed by a number of obstetricians practicing in developing countries that small pelvic size is a major cause of difficult labours and deliveries (Naeye, personal communication), but, again, conclusive evidence does not exist.
Chronic undernutrition resulting in maternal stunting, or acute or chronic undernutrition resulting in low prepregnancy weight, may also negatively affect reproductive performance as reflected in reduced newborn size and its attendant risks of mortality and morbidity. A positive association between maternal size (height and weight) and newborn size is widely recognized (16-22). The relationship between height and pregnancy outcome is confounded by the fact that maternal height reflects environmental, early nutritional, and genetic influences. The influence of height is also confounded because, although height appears to have an independent effect on birth weight (18, 23, 24), evidence from industrialized countries suggests that prepregnancy weight or pregnancy weight gain have a larger effect on birth weight than does height (18, 23). The interaction of these effects is discussed in the next section.
In contrast to the work reported from industrialized countries, where women were on average taller and had better childhood nutritional status than is commonly found in developing countries, multivariate analyses of data from a longitudinal study in Guatemala of the effect of supplementation on birth weight suggest that the independent contribution of height to birth weight is statistically significant (p < .05). This effect is as large as the effects of maternal weight and caloric intake from supplementation during pregnancy (19). Furthermore, because the women studied came from villages where childhood growth, especially in height, is affected by differences in nutrition (20), the authors assume that the adult height of the women studied was determined in large part by their childhood nutrition (19). Thus, these findings suggest that, at least in this population, the mother's nutritional status during childhood and immediately preceding pregnancy is as important a determinant of birth weight as is maternal nutrition during pregnancy.
Another issue of concern in preparation for pregnancy that has not been clearly defined is the capacity for restoration during puberty and adolescence of growth deficits incurred early in life. Can effects of chronic undernutrition on height and lean body mass persisting into adolescence be ameliorated by improved nutrition during adolescence or later in life?
Lastly, we still lack considerable knowledge regarding what constitutes an optimal nutritional status as a preparation for pregnancy. With the possible exception of iron, additional information about critical levels of specific nutrient stores and ways to measure them is required before such levels can be suggested. Given current knowledge, the goal should be a dietary intake and a state of health that allows the population of women of reproductive age to achieve an adequate muscle mass and at least 16-18 per cent of adipose tissue, 300-500 mg of iron reserves, and adequate levels of other vitamins and minerals.
Nutrition during Pregnancy
Data on the clearly negative effects of poor nutrition during pregnancy on maternal health and reproductive performance, including birth weight, are available. For example, it is generally recognized that, in populations where chronic undernutrition is common, risk of low birth weight has been associated with increased risk of severe undernutrition, poor growth and development, and even death in infancy. In addition, famine during any part of pregnancy long enough to deplete maternal nutritional reserves has been shown to reduce birth weight and increase perinatal complications and mortality of both mothers and infants. However, depletion of maternal stores even due to famine during the first two-thirds of pregnancy may be compensated by adequate nutrition during the third trimester (25, 26).
Weight gain during pregnancy is an acknowledged determinant of birth weight. However, the size and importance of its effect (i.e., the amount of variability in birth weight explained by pregnancy weight gain) differs according to the condition of the woman entering pregnancy and reflects an interaction of weight gain during pregnancy and prepregnancy size and body composition. In the Collaborative Perinatal Project in the United States, birth weight increased significantly and perinatal mortality decreased as weight gain of thin mothers increased. However, birth weight and perinatal mortality were less sensitive to weight gain among normal and particularly among overweight women (27). Winikoff and Debrovner (23) also reported that weight gain explained most of the variability in birth weight for thin women, whereas prepregnancy weight was more important for women of intermediate weight for height. In their study thin women were defined as women with weights less than the lower limit for medium frame women of the Metropolitan Life Insurance Table of Desirable Weights, and intermediate women were those with weights within the medium frame range. In heavier women (defined by weight above the upper limit for medium frame women) weight gain was not a significant predictor of birth weight. Thus it appears that maternal reserves buffer the influence of diet such that when reserves are depleted diet becomes critical.
An average, well-nourished healthy woman irrespective of geographic or ethnic origin is estimated to gain 12.5 kg during pregnancy. Approximately 4 kg of this weight is considered to be fat deposited in different depot organs. Surveys of pregnancy weight gain conducted in Guatemala (28), Ethiopia (29), the Gambia (30), and India (31) show that a large proportion of the women gained between 2.7 and 6.8 kg. Although not all trials supplementing pregnant women have shown a positive effect on pregnancy weight gain (32) dietary supplementation studies in indict (33, 34), Bogota (35, 36), Guatemala (37), and Mexico (38) were associated with increased pregnancy weight gain, suggesting that the low weight gain is at least partially due to insufficient food intake. Recent supplementation studies in the Gambia showed a seasonal effect of supplementation on birth weight and pregnancy weight gain (39). Thus, the authors suggest that mean monthly weight gain during pregnancy could be used to determine which communities would benefit most from prenatal supplementation (39).
Evidence published primarily in the last decade indicates that placental growth and function are compromised by severe (25) and moderate (40) maternal undernutrition (reviewed by Rosso [41]). Although studies in humans of the mechanisms by which malnutrition affects placental function have been limited, primarily for ethical reasons, a number of studies in rats show a reduced rate of placental transfer of nutrients in malnourished rats (42-44). Reduced availability of specific nutrients could conceivably account for some of this impairement in nutrient transfer (45). However, a more likely explanation is that inadequate maternal blood volume expansion (41) is the mechanism through which undernutrition affects placental nutrient transfer.
Maternal blood volume normally increases by more than one litre during pregnancy (46). The increase is regarded as an adjustment that allows the pregnant woman to expand blood flow to her uterus without reducing blood flow to other vital organs. It is postulated that reduced blood volume expansion in turn results in an insufficient increase in cardiac output, decreased placental blood flow, and reduced nutrient transfer and placental size (41, 45).
No blood volume measurements have been published for women from undernourished populations in developing countries. Thus, there is no direct information about the importance of low blood volume expansion as a contributor to perinatal mortality rates. However, preliminary indirect analyses from the Collaborative Perinatal Project in the United States provide some information. In these analyses the presence of hand/facial oedema was used as a proxy for adequate blood volume expansion. Hand/facial oedema is positively correlated with blood volume expansion (46). When pregnancy weight gains were low, perinatal mortality rates were lower if hand/facial oedema was present than if it was absent (Naeye, personal communication). This suggests that at least in thin women blood volume may be associated with perinatal mortality rates. These finding require validation using actual blood volume measurements in women from developing countries.
Maternal acetonuria (ketonuria) has been hypothesized to be responsible for some of the excessive perinatal mortality associated with low pregnancy weight gain in the Collaborative Perinatal Project in the United States (27). Increased plasma and urinary ketones are associated in pregnancy with the increased fat catabolism (increased use of fat as a fuel) that occurs during starvation (47) and/or after short-term fasting (12-18 hours) (48). The increase in ketones (along with a decrease in glucose and other biochemical changes) is thought to be the result of the mechanisms operating to rapidly adapt the mother to the metabolism of fat so that glucose and amino acids can be spared for the foetus (49). However, some of the lipid products may also cross the placenta and it is not known whether they are completely innocuous in the foetus (48). Although there are data suggesting that newborn animals including humans have a heightened ability to use ketones for energy (50-52), other mechanisms related to ketosis (e.g., acidosis, specific metabolic effects of acetone, widely fluctuating metabolic fuel substrates) may also have unexpected effects on the foetus. Currently, only indirect evidence links acetonuria with perinatal mortality and further elucidation of hypothesized mechanisms is required.
In most developing countries, the overall risk of exposure to various kinds of infections during gestation is considerable. Both overt and incipient intra-uterine infections may cause foetal growth retardation or other clinical syndromes that may affect the foetus (53). However, the relation of these infections to nutritional deficiencies (general or specific) has not been thoroughly investigated. Examples of questions to be addressed include: Why is chorioamnionitis more prevalent in marginally nourished or moderately malnourished pregnant women than in well-nourished women? What mechanisms are responsible for foetal growth retardation among pregnant women with urinary infections? Are they equally damaging among pregnant women of varying nutritional states? Are acute phase immune responses equally effective in pregnancy as when a woman is not pregnant? Are there nutritional interrelations? The well-known infection-induced defect in utilization of fat as a metabolic fuel may be particularly deleterious during pregnancy because of the greater dependency on fat as an internal source of fuel under those conditions (49).
Some evidence suggests that zinc deficiency is related to preterm labour and delivery (54). It has been hypothesized that zinc deficiency results in reduced antimicrobial activity and therefore amniotic fluid bacterial infections (54, 55). Further work is needed, however, to validate this hypothesis. If it is true, additional work will be needed to determine, given other nutritional and non-nutritional factors influencing preterm delivery, whether zinc deficiency as a contributor to preterm labour and delivery is a public health problem and amenable to intervention.
Degree of physical work during pregnancy may also affect the health of the mother and the outcome of pregnancy. Several investigations demonstrate that pregnancy weight gain and pregnancy outcome (birth weight) improve if a woman maintains energy balance by reducing physical work to counter-balance reduced dietary energy intake (56). However, more systematic data are required to determine if changes in physical activity have an effect on pregnancy outcome when dietary intake is adequate. In other words, the effect of physical activity on pregnancy weight gain and outcome throughout the range of dietary intakes is not understood.
Other factors to consider during pregnancy, at least in some developing countries, are a woman's habitual alcohol intake and smoking habits. Alcohol is a foetal toxic agent throughout pregnancy, producing different types of foetal anomalies, including foe/al-alcohol syndrome (57, 58). A deleterious effect of smoking on foetal growth is also well recognized (59-61). The possible role of adequate nutrition in correcting and possibly preventing the effects of smoking on foetal growth and development deserves greater attention. Whether a high concentration of smoke and carbon monoxide, observed in closed highland dwellings with a central open fire, as found in some developing countries, has an effect similar to smoking and whether it has nutritional impacts are still unclear.
In many developing countries anaemia during pregnancy is a recognized public health concern. Reported prevalence of anaemia (expressed as the percentage of women with haemoglobin values less than 11 g/dl) for women in tropical Asia in the third trimester of pregnancy ranges from 35 per cent in urban Nepal, to 66 per cent in Bangladesh, to 72 per cent in the Philippines, and to 88 per cent in India (62). A large collaborative study in Latin America reported a prevalence of 22 per cent for third trimester pregnant women (63).
The most common causes of anaemia are dietary intake deficiencies, malabsorption defects, blood losses, infections and infestations, and in certain areas, haemoglobinopathies. In general, iron deficiency is the commonest cause of anaemia (64). In pregnant women folate deficiency is also an important cause (62, 65).
Severe anaemia in pregnant women is associated with an increased risk of premature delivery and maternal morbidity and mortality. Mild to moderate anaemia (small reductions in haemoglobin) caused by severe iron deficiency has been shown to decrease performance in maximum or near maximum exercise in humans (66, 67) and alter mitrochondrial energy metabolism and hormonal balance in animals (68-71). There is also some evidence that mild to moderate anaemia may limit performance of tasks involved in daily occupations (72, 73). Lastly, there is some experimental evidence that iron deficiency anaemia may be related to reduced resistance to infection and impaired immunity (74, 75). However, current findings are not conclusive, and the subject requires further controlled study.
Iron and folate supplementation have been shown to effectively increase the haemoglobin concentration of pregnant women in Israel (76), Burma (77), and India (78). The Narangwal studies in the Punjab suggest that maternal iron and folate supplementation of pregnant woman contributed to a decrease in perinatal mortality (79).
Nutrition in Relation to Pregnancy during Adolescence
A special case of nutrition and growth related to reproductive efficiency is that of pregnancy during adolescence, because two components of growth (maternal and foetomaternal) take place simultaneously. A number of studies of pregnancy in adolescence carried out in developed and developing countries indicate that rates of low birth weight and prematurity are higher in the young, physiologically immature adolescent than in mature women (15, 80-84). There are exceptions to this evidence, however, such as that presented by Zuckerman et al. (85), which shows no such association for a group of pregnant adolescents in Boston. There is little, if any, information from developing countries about whether early pregnancy has a deleterious effect on maternal nutritional status.
Agreement is lacking on the severity and incidence of pregnancy and labour complications, partially because of methodological differences among studies. However, it appears that the risks for pregnant adolescents under 16 and their infants are high (84). For example, although not universal (85), there is evidence suggesting that the incidences of toxaemia and pre-eclampsia are significantly higher among young adolescents than any other group of reproductive age (84). Whether this is nutrition-related is not clear, however.
Although dietary practices contribute to the increased risks of early pregnancy, inadequate prenatal care (15, 86), low socioeconomic status (82), unmarried mother (87), and inadequate social support systems (84) have been associated with adolescent pregnancy and its risks. Thus age per se may not always be a risk factor (85). There is still some debate about the relative importance of chronologic age and gynaecologic age (years after menarche) for the outcome of pregnancy (14, 88). Although the importance of gynaecologic age in developing countries is not well understood, data from industrialized countries suggests that reproductive performance is poorer among adolescents who become pregnant within two years after the onset of menarche (88). Again, however, the data available suggest that, although low maternal age is a risk factor, environmental, social, health, and nutritional conditions surrounding young childbearing are likely even more important determinants of a successful pregnancy outcome.
Nutrition during Lactation
Factors affecting milk production and the effect of milk on infant growth have been the main focus of recent lactation research. Maternal nutritional status and other maternal characteristics are almost always examined in relation to their effects on milk quality and quantity. Very rarely is the direction reversed to inquire about effects of lactation on the mother. Historically, this has probably occurred for several reasons. First, high rates of infant mortality have led to a focus on nutrition as a determinant of infant mortality. This, in turn, led to a tendency to see the lactating mother as a vehicle through which nutrients are delivered to the infant, with a research emphasis on how milk quality and quantity affect infant health. Second, when nutrition researchers have studied maternal nutrition, they have been more likely to focus on pregnancy than on lactation because it generally has been perceived, although not tested, that foeto-maternal nutritional interactions that take place during pregnancy are more important than infant-maternal interactions occurring during lactation, and that pregnancy places a greater strain on maternal resources than does lactation.
Little information about direct effects of lactation on maternal nutritional status has been compiled in the literature. To help fill this gap the discussion of nutrition during lactation presented in this report concentrates on effects of lactation on the mother's nutritional status and includes review of three factors that may have particular effects on maternal nutritional status during lactation-culturally determined food restrictions, contraception, and infection. The effects of maternal nutritional status on milk yield and composition have been extensively reviewed (89, 90) and are summarized only briefly here.
Maternal Nutritional Status during Lactation
Knowledge about maternal nutrient requirements and standards for assessing maternal nutritional status during lactation are inadequate. For most nutrients, current published standards for lactating women have been derived by adjusting requirements for non-pregnant, non-lactating women upward to include the additional amounts of nutrients in human milk. For some nutrients, knowledge of requirements in the non-lactating woman is highly inadequate. That is, the reference base on which to add the additional requirements of lactation is not secure. In addition, a requirement may be based on assumptions not applicable to women living in developing countries. For example, the assumption of an 80 per cent efficiency in the conversion of food energy to milk may not always be correct. Similarly, the assumption of a highly effective transfer of nutrients from maternal plasma into milk has been studied very little in humans or experimental animal models.
From a review of available evidence we conclude that the current state of knowledge about the effect of lactation on nutrient requirements and nutritional status of women in developing countries is inadequate and that generalizing from such an inadequate data base must be done cautiously. A more detailed review of the specific effects of lactation on maternal energy balance and protein, vitamin, and mineral status is presented as an appendix.
There is a widespread belief in the health care, nutrition, and social science communities, that cultural food restrictions practiced by breast-feeding women have deleterious consequences for maternal and infant health. There is evidence from many parts of the world that beliefs concerning the importance of lactation food restrictions are very common (91). However, although the literature contains numerous discussions of types of lactation restrictions (92, 93), often with ethnographic examples, little solid research addresses the impact of cultural food restrictions or dietary changes in general on the health and nutritional status of lactating women.
The effects of hormonal contraceptives on the nutritional status of lactating women appear to be no different from those on non-lactating women (94). For example, Prema and colleagues (95) found no effects of a low-dose injectable progestational contraceptive on the nutritional status of low-income lactating Indian women as assessed by anthropometric techniques and clinical signs of nutritional deficiencies. There is, however, some evidence that vitamin B6 requirements may be changed in women who use oral contraceptives (96, 97). Roepke and Kirksey (98) demonstrated that long-term use of oral contraceptives prior to pregnancy significantly decreased the levels of vitamin B6 in maternal serum and in breast milk.
Recent evidence from the Gambia (99), Guatemala (100), India (101), and Zaire (102) suggests that improving the nutritional status of breast-feeding women in developing countries may upset the hormonal mechanisms prolonging postpartum amenorrhoea, shorten the period of postpartum amenorrhoea, and, in the absence of alternative contraceptive techniques, cause fecundity and fertility to rise, i.e., reduce the contraceptive effect of breast-feeding. However, analyses of other data suggest that the impact on fertility of improved maternal nutritional status is modest (103105). Another explanation for a shortened period of postpartum amenorrhoea is that supplementing an otherwise fully breast-fed infant reduces the intensity and/or frequency of sucking and therefore alters the hormonal balance maintaining the period of amenorrhoea (100, 106, 107). Both of these hypotheses require further investigation and may have important consequences for programme planners in nutrition, who may have to improve contraceptive services at the same time that they are improving the nutrition of lactating women or their nursing infants.
In the large literature on nutrition-infection interactions, we were unable to identify any empirical studies in developing countries that focused specifically on the effects of maternal infection on nutritional status during lactation. Does lactation create special conditions that influence the course and severity of infection? What is the prevalence of mastitis and its impact on the health status of poorly nourished women? These and related questions require further study.
Maternal Nutritional Status and Milk Production
Studies of human milk production and composition in both well and poorly nourished communities in different ecological circumstances have had different emphases and methodologies (90). As a result inferences and comparisons must be made cautiously. Studies carried out in different population groups seem to indicate that mean breast-milk output in mothers from privileged populations tends to range from 600 to 900 ml/day (89). Volumes of breast milk reported for women in countries with undernutrition and poor living conditions are about 400-700 ml/day in the first six months, 300-600 ml/day in the second six months (89), and 300-500 ml/day in the second year (90). Higher figures of 700-1,000 ml/day have been reported for non-privileged Ethiopian women (Gebre-Medhin, personal communication).
Available information seems to suggest that the quality of human milk is remarkably preserved even in the mildly to moderately malnourished mother except in advanced deficiency states (90, 108), although the evidence for an effect on quantity is not as good. For example, in famine milk output may decline and ultimately cease. Marasmus in the first six months of life may occur in babies of very poorly nourished women. Although the concentration and the content of individual nutrients in breast milk may be somewhat reduced in malnourished mothers compared with healthy controls, breast-feeding even by malnourished mothers seems capable of achieving adequate growth and nutrition in the early postnatal period. The determinants of milk production in humans are not well understood, but the stress associated with living under difficult socioeconomic conditions is considered an important cause of variation in milk yield. The effect of maternal disease burdens on milk composition and volume is also not understood well.
Although, as mentioned above, hormonal contraceptives may not affect maternal nutritional status directly, there is concern that oestrogen-containing contraceptives may decrease milk output (94, 109, 110) and could result in early termination of breast-feeding and/or infant growth retardation. For example, combined oestrogen progestogen oral contraceptives containing as little as 30 mg of oestrogen have been reported to significantly decrease milk volume (111). When used in the first few months of the postpartum period, these contraceptive agents can decrease milk volume by as much as 40 per cent within 3 to 6 weeks (94).
Until recently it was thought that hormonal contraceptives containing 50 ?gor less of oestrogen did not affect infant growth (109). However, preliminary results of at least one study suggest that this may not be the case (Zeitlin, personal communication). Given their widespread implications for infant health in developing countries, these findings require further clarification. Until results of such studies are available, current knowledge, as summarized by the recent WHO/NRC Joint Workshop on Breastfeeding and Fertility Regulation (94), suggests the following: (a) The appropriate time for introducing contraceptive techniques other than lactational amenorrhoea varies among social groups and therefore should be set by family planning and health officials on the basis of breastfeeding and lactational amenorrhoea patterns in their specific situations. (b) If contraceptive use is deemed necessary, non-hormonal contraceptives should be encouraged for lactating women. (c) For lactating women who wish hormonal contraceptive protection in the early months of lactation, progestogen-only hormonal contraceptives should be made available, and women should be warned about the effect of combined oral contraceptives on the quantity of breast milk (94).
Nutrition of Infants and Young Children
The presence and extent of poor childhood nutrition in a population can be assessed in various ways. Impaired physical growth (e.g., weight and length or height) is perhaps the most useful indicator. Although lacking specificity, growth is a very sensitive indicator of nutritional adequacy. It is possible that other functional alterations, such as those reflected in decreased physical activity and psychosocial interaction and mental wellbeing, occur during malnutrition prior to evident reductions in growth rates. However, anthropometric measurements of physical growth are still the simplest means of assessing adequate nutrition at the population level, while other functional impairments are difficult to quantify.
Available data on physical growth document that infants in developing countries almost universally begin to lag behind their counterparts in industrialized countries by the middle of the first year of life and continue to do so for the ensuing years (112). Although non-nutritional factors, including infection, undoubtedly play an important role in growth retardation, inadequate food intake is an important factor.
More dramatic evidence of inadequate nutrition is provided when frank malnutrition in its various clinical manifestations is observed. Severe malnutrition is often life threatening and presents serious problems of treatment and rehabilitation. Severe malnutrition also has profound negative effects on immunity and resistance to infection. Impaired immunity is responsible for much of the increased frequency and severity of infectious illnesses. In the case of infectious illnesses, especially diarrhoea, the illness itself may lead to more malnutrition, and set in motion the vicious malnutritioninfection cycle (113). Impaired bodily defence against infection contributes substantially to the high death rate among infants and young children in developing countries.
The occurrence of cases of overt malnutrition may provide a marker of poor nutriture existing in a community. Thus, even a few cases of severe childhood malnutrition in any given community suggest the presence of significant nutritional problems, perhaps involving large segments of the susceptible population.
Mortality statistics also often provide evidence of inadequate nutriture. Although infection is often the primary cause of death, malnutrition is associated with a large proportion of deaths occurring in developing countries. Kielmann and McCord (114) and Chen et al. (115) have demonstrated an inverse relationship between weight for age and mortality. Much of the weight deficit is due, directly or indirectly, to inadequate nutrition. Inadequate nutrition and its consequences are important contributing factors to the increased mortality in infancy and childhood.
Reduced physical activity and impaired psycho-social functioning are commonly observed in poorly nourished children. These factors may in turn cause impaired interaction with other family members and result in impaired cognitive and psycho-social development. The behavioural effects of malnutrition are difficult to separate from the effects of early malnutrition on the development of the central nervous system. Whichever the more important of the two mechanisms may be, there is no doubt that inadequate nutrition and poor psycho-social environment during early childhood can have adverse effects on intellectual and psycho-social development.
Limited food intake associated with restricted cultural or
physical availability of foods or with insufficient purchasing
power is an important cause of malnutrition, in conjunction with
poor sanitary conditions and insufficient knowledge about
nutritional needs. However, even when general causes of
malnutrition are understood, specific local conditions need to be
explored to identify combinations of causative factors that may
be amenable to intervention. For example:
- Poor maternal nutritional status plus infections during
pregnancy have been shown to be determinants of low birth weight.
Poor nutrient intake during lactation, combined with heavy
physical labour, may lead to diminished milk production. These
factors often contribute substantially to early faltering of
growth, which begins before diarrhoeal disease begins to play a
role (116).
- In some cultural settings late introduction of complementary
foods is common practice. Often these foods are of less than
ideal composition, i.e., caloric density is low, protein content
is low. Furthermore, if the complementary foods are prepared and
fed under poor sanitary conditions, they become major sources of
pathogens contributing to weanling diarrhoea.
- Frequent infectious illnesses, especially diarrhoea, contribute
to insufficient food intake in children and result in anorexia,
lethargy, and diminished nutrient absorption. Each of these
causes of malnutrition of infants and children is susceptible to
intervention. Clearly the chances of success are greatest if
intervention simultaneously aims at all causative factors
operating in a given population.
A discussion of nutrition in infancy would be incomplete without comment on breast-feeding, complementary foods, and weanling diarrhoea. Thus, the remainder of this section will highlight these interrelated topics.
Breast-feeding and Complementary Foods
In many traditional societies breast-feeding is a common practice. Over the past several decades, however, the incidence and duration of breast-feeding has declined in many countries, particularly in urbanized areas (109, 117). This has aroused concern among the health and nutrition communities (118).
Where breast-feeding is practiced, breast milk usually provides the sole source of nutrients during the early months of life. At some point during the first year of life, complementary foods should be introduced, but breastfeeding often continues through the second year of life (109). The advantages of breast-feeding are well recognized and will not be further elaborated here.
The age at which breast milk alone becomes an inadequate sole source of nutrients appears to be influenced by maternal nutritional status and is subject to great individual variation. Under favourable circumstances, breast-feeding alone is capable of meeting an infant's nutritional requirements and of supporting normal growth for the first four to six months of life (112, 119, 120).
Where undernutrition and deprivation are prevalent, available data indicate that growth faltering in exclusively breast-fed infants may occur before three months although it is generally not evident until after three months (112, 121). Thus, under some conditions breast milk may be an insufficient sole source of nutrients by three months. However, the situation is complicated by increasing exposure of the child to infection at a time when its passive immunity is dwindling (120).
There has been considerable controversy regarding the age at which complementary foods should be introduced. (Complementary foods are any foods an infant or child receives in addition to breast milk.) Some advocate categorically that in well-nourished and healthy populations complementary food should not be introduced until an infant is older than six months, unless growth is faltering. Another view considers the risks and long-term consequences of early growth failure and decreased immunological competence caused by undernutrition to be potentially grave. Consequently, this second view favours introduction of complementary foods even before growth failure is evident. The problems attendant on introduction of complementary foods are potentially solvable, and therefore this view contends that the prudent approach is to concentrate on making the introduction of complementary foods safer for the infant.
The age at which complementary foods become necessary varies greatly, as mentioned earlier, with variation among individuals probably being larger than variation among communities. An individualized approach in populations or in communities at risk is therefore necessary. As has been emphasized by Scrimshaw and Underwood (122), the best way to determine when complementary foods are necessary is through monitoring of body weight. Although there are logistic problems associated with frequent weighing, potential benefits can be substantial. Thus, regular monitoring of body weight should be encouraged and facilitated.
Traditional complementary foods are often of low caloric density and (123) low protein content, contain little or no fat, and are often limited in micronutrient content. Such foods are not well suited to supplement the breast-fed infant's diet. Development and use of appropriate foods need to be promoted. Representative complementary foods are described in Cameron and Hofvander (124).
Weanling Diarrhoea
The introduction of complementary foods is one factor, along with environmental contamination and poor hygenic practices, commonly associated with diarrhoea. Typically, a child will already be in a sub-optimal nutritional state (as evidenced by sub-optimal weight) at the time he/she is first given complementary foods. These foods are often contaminated (116,125) and are fed under poor sanitary conditions, e.g., from unwashed hands or after having been dropped on the floor. This increases even further the force of infection in a child who often has repeated infectious episodes. Moreover, the child's defences against infection may be impaired as a result of his/her poor nutritional state, and the child often develops diarrhoea with this additional burden. Diarrhoea entails impaired intestinal absorption and often anorexia, which aggravate the existing nutritional state. Soon another episode of diarrhoea follows, resulting in more nutritional deficit and further impairment of host defences. Thus, a vicious cycle is set in motion that is difficult to break once it has been established.
Recurrent diarrhoea is responsible for much of the growth deficit seen among children in developing countries (116, 125). Clearly, measures effective in preventing or ameliorating the malnutrition-diarrhoea cycle are potentially a major influence on childhood nutrition. Efforts to prevent establishment of the cycle should thus receive high priority.