Previous - Next
This is the old United Nations University website. Visit the new site at http://unu.edu
Ted Greiner
Abstract
Breastfeeding can enhance care of infants. In most cultures there is an age after which breastfeeding is no longer considered socially acceptable. In Western countries, shorter periods have long been the norm. Researchers may underestimate how common sustained breastfeeding is in both industrialized and low-income settings. Little is known about the contributions breastfeeding may make to the emotional and physical health of mothers and infants when sustained for longer periods. It clearly offers economic and certain nutritional advantages, such as preventing severe vitamin A deficiency, and it reduces fertility. Sustained breastfeeding is often associated with slower child growth. This is probably not often a unidimensional causal relationship. Part of the explanation may be that sustained breastfeeding increases infant survival under extreme conditions of poverty and food insecurity.
The duration of breastfeeding, especially exclusive breastfeeding, may have decreased in some areas in recent decades due to inappropriate messages from health workers, in part due to a lack of careful definitions. A schematic figure depicts four separate processes, each referred to at times as "weaning." To reduce the incidence of early cessation of breastfeeding, it is important to separate the "complementation" and "replacement" components.
Modernization processes such as urbanization can occur so rapidly that new ideas for achieving infant care goals may be needed. However, in relatively stable resource-poor settings, care strategies such as sustained breastfeeding are likely to be well adapted, and outsiders would be wise to focus on protecting them. Indeed, breastfeeding programmes should place priority on protection (marketing codes) and support (breastfeeding-friendly practices at delivery and support measures for women in the market labour force) before promotion (mass media).
Advantages of early breastfeeding for care
The younger the child, the more his or her physical as well as emotional welfare is dependent on care. Yet, the time and knowledge required for proper feeding increase to a maximum when the child is 6 to 18 months old. Up until 6 months of age, breastfeeding can meet the infant's nutritional needs; any additional time and resources spent on feeding of supplements are usually unnecessary and may be harmful. Most of the basic knowledge required in these first months is "automatically" transferred as a part of growing up and becoming a mother in traditional cultures. As the child grows older, language capabilities and motor skills enable him or her to better express and independently respond to his or her own hunger signals.
Breastfeeding contributes to care by fostering mother-infant bonding, stimulation, and skin and eye contact, as well as providing high-quality nutrients hygienically and countering infection. Human milk appears to contain factors that promote brain growth and development, particularly visible in infants born pre-term [1]. Breastmilk is rapidly digested. When breastmilk forms all or nearly all of the infant's food, the infant will want the breast open, and this will naturally lead to frequent contact between mother and infant.
Bottle-feeding levels reached their height in the West by about 1970. By that time Western culture was so bottle-oriented that it was assumed that there were no differences, even psychologically, between bottle-feeding and breastfeeding, as long as the bottle-feeding mother looked at and fondled the infant. This no doubt influenced the type of research done at the time. Jelliffe and Jelliffe [2] illustrate this by citing a book on mother-infant attachment that does not even refer to breastfeeding [3]. Newton and Newton [4] and Klaus and Kennell [5] in some of their bonding research began to question this assumption. Research suggests that, via the effects of oxytocin and gastrointestinal hormones, breastfeeding appears to change the psychological profile of the mother to make her more open, flexible, and "service-oriented" [6, 7].
Advantages of sustained breastfeeding for care
Breastfeeding for three years or longer is not as uncommon as most researchers assume, either in developing or in industrialized countries, though clearly prevalences are higher in the former. Among La Leche League members in the United States, even during the 1970s, when breastfeeding rates were at their lowest, extended breastfeeding was practiced but kept "in the closet" [8]. Even in developing countries, little attention is given to breastfeeding that takes place for several years. Some researchers seem unconsciously to adhere to norms that lead them to expect that little if any breastfeeding is taking place after a certain age (often two to three years). In both Ghana and Lesotho, I have observed children in school uniforms breastfeeding. These children, usually standing or kneeling beside mothers who were sitting, took the breast themselves from compliant mothers who went on with their business. Neither the mothers nor bystanders paid any attention to these children's breastfeeding behaviour.
I have not come across much discussion of very long breastfeeding. Perhaps these children have younger siblings who are breastfed and, if asked in a survey, their mothers might not consider the older children still to be breastfeeding. Even if they did, the investigator might not. Jelliffe [9] cites Oomen as writing, "In the case of the small boys at Jobakogl [Papuan village] who strolled to the women's house at dawn to have their morning drink, it requires some twisting of the term to consider them still 'breastfed.'" Thus, even in developing countries the true extent of breastfeeding sustained for many years may be underestimated.
A major reason for practicing sustained breastfeeding in industrialized countries in the face of social disapproval has been the belief that it provides a closer bond between mother and child. These children are often said to be more secure and more independent. They continue to remember this close bond, and their mothers believe that it continues, in some sense, even into adolescence, easing the difficulties in the mother-child relationship [8,10].
There are unquestionable nutritional and economic advantages of sustained breastfeeding [11]. Even beyond infancy, young children return to the breast for comfort when they are sick and anorexic and thereby passively receive more food [12,13]. In poor countries, breastmilk can play a key role in vitamin A nutrition, irrespective of the child's age [14]. Its effects in promoting child survival seem to be more distinct than its effects in promoting child growth [15]. For older children in very poor situations where household access to food is highly insecure, breastfeeding may have a trade-off effect, providing an increased chance for survival but at the cost of a reduced growth rate. If so, this would be an exception to the usual situation, in which increased growth is usually assumed to be a proxy for health and survival.
The duration of breastfeeding (as well as its exclusivity) does contribute to longer birth spacing. An extreme example of the importance of this for child survival is seen in Yemen, where combined breastfeeding and bottle-feeding was the norm according to the 1979 National Nutrition Survey. The 1979 World Fertility Survey estimated that only 2% of couples practiced a modern family planning method. Abstinence in this traditional Muslim setting was mandated for only the first 40 days. Thus breastfeeding practices were the major determinant of birth spacing. Mortality rates were much higher for younger infants when birth spaces were shorter, as well as for children one to four years old: when the birth space for the subsequent child was less than 24 months, the one- to four-year-old death rate was 141/1,000 alive at that age; with a birth space of two to three years it was 18/1,000; for three to five years it was 2/1,000; and for longer birth spaces it was 3/1,000 [16].
As the recent adoption of "triple nipple" (combined breast and bottle) feeding has led to shorter birth spaces in Yemen, many women now are attempting to care for three or even four children under the age of five and cannot cope. Many who did breastfeed longer than average said they did so to achieve longer birth spacing [17]. This birth spacing effect of breastfeeding has long been recognized by women in many countries, although it may be less clear in areas where early supplementation is now the norm. In Ethiopia many women reported stopping breastfeeding in order to have more children [18].
A manual for slave owners in the Caribbean, "Practical Rules for the Management and Medical Treatment of Negro Slaves in the Sugar Colonies," published in 1811 and cited by Henriques [19] stated:
Negroes are universally fond of suckling their children for a long time. If you permit them, they will extend it to the third year ... your business is to counteract their designs, and to oblige them to wean their children as soon as they have attained their fourteenth or sixteenth month ... If you neglect to do this, you not only lose some of your mothers' labour, but you prevent their breeding as soon as they otherwise would do, in all probability.
Constraints related to care
The many factors that lead to less than optimal infant feeding patterns can be divided into "ideational" (knowledge, attitudes, and beliefs, often culturally informed) and "external" constraints. It is commonly assumed that "external" constraints are mainly responsible for the fact that exclusive breastfeeding is rare, particularly its high opportunity cost, at least in modern settings.
Any other kind of infant feeding requires someone to devote time specifically to food preparation, feeding this food to the infant, and maintaining hygiene during preparation of the food and cleaning of utensils (especially time-consuming where clean running water and modern cooking and refrigeration facilities are lacking). In artificial feeding, some of these time-consuming steps are often cut down beyond what good hygiene demands. Even when women are educated and make an effort to clean the bottle properly, resource constraints can prove impossible to overcome [20].
Other efforts to save time include offering older infants gruels either in a bottle (with the nipple cut open to allow a thicker fluid to pass through) or in a feeding cup with a lid and a perforated spout. The reason often given for adding solid foods early is that it reduces the frequency of infant crying, allowing the mother to get on with her work. Pacifiers (also called "dummies" or "soothers") are used for similar reasons. Much of this infant crying may be due to hunger or inherent sucking needs, but part is probably related to needs for care and comfort. Thus some of the "premature" supplementation seen in the early months of life throughout the world probably reflects an attempt to cope with time constraints that prevent mothers from providing as much care as their infants need.
In using an economic model, it was pointed out that these ways of saving time incur other costs [21]. Bottle-propping deprives the infant of body and eye contact and stimulation and may lead to increased ear infections. Older infants who carry the bottle around with them make little effort to keep it free from dirt and flies. Increased illness results in high costs for extra care. However, individuals are usually not aware of the trade-offs involved (in part due to lack of understanding of the causes of disease and malnutrition in infants) or feel they have no choice. Piece workers, for example, even if they work at home, may consciously reduce breastfeeding to increase the time available for earning money [22].
Furthermore, although other forms of feeding require more time than breastfeeding, they do not necessarily require the mother's time. The availability of very low-cost forms of child care probably leads to decreased breastfeeding in situations where opportunity costs for child care by the mother increase (e.g., when new demands are placed on the mother's time or when new opportunities arise for income earning) [23]. Then grandmothers, sisters, or others take over more of the care and feeding responsibilities for the young child. However, potentially negative trade-offs are involved here, too, particularly when young girls stop school to take over child care responsibilities.
The poor caring capabilities of uneducated younger siblings and housemaids are also sometimes cited by mothers and researchers as a cause of malnutrition [24]. In a study in Sierra Leone, children who were sent away from their mothers suffered from higher mortality rates only if they were young at the time [25], suggesting that the biological mother's role in care is superior only at earlier ages, perhaps due in part to breastfeeding.
Breastfeeding may explain the evolution of patterns of child care based on the mother as the major caregiver at least during the early months of life. In traditional settings it is rarely perceived as something separate from or additional to her other child-care responsibilities. Breastfeeding commonly is done at the same time as the hands are busy with something else. Young babies are swung around from the back to the front to breastfeed. Older children take the breast on their own when it is easily available. In either case the mother may pay no attention and continue undisturbed with her work or sleep. When women do choose to take time off for breastfeeding, they sometimes describe this as a necessary rest and an advantage of breastfeeding. Breastfeeding even provides women with special status and benefits in some cultures.
In a study of several subsistence cultures, it was found that women tend to perform tasks compatible with child care [26]. These tasks characteristically take place in an environment not likely to pose dangers to a young child, are repetitive and can be easily interrupted, and are carried out not too far from home. However, women lose power over the nature and location of their work as needs for earning cash increase.
In addressing this problem, attention commonly focuses on the need to overcome constraints for many employed women workers. This is an important strategy, especially for women working in the health and education sectors, since they are influential in society and could help lead the way towards change for others if enabled to care for and breastfeed their own infants better during the first year or so of life [27]. Women doing paid agricultural work and employed in the informal sector also need to be enabled to breastfeed as much as possible, although little attention has been given to how to meet their needs.
"Ideational" factors are also important in explaining the lack of exclusive breastfeeding [22]. In many cases, women need not only to be "enabled" through correct information (rarely available where health workers are inadequately educated regarding breastfeeding or where the infant food industry is the major source of information) and assisted with health and lactation management problems that may interfere with breastfeeding. They also need to become "empowered" through emotional and practical support from their peers, spouses, employers, and others.
In traditional rural settings, approaches dealing with ideational aspects alone may be able to increase rates of exclusive breastfeeding substantially. In these settings, neither the financial nor the opportunity costs of breastfeeding are nearly as great as those of supplemental feeding [21], particularly where women's economic activities tend not to conflict much with breastfeeding. Information on the value of exclusive breastfeeding and the dangers of feeding unnecessary supplemental fluids is rarely available in appropriate or credible forms.
Like other aspects of infant and child care, breastfeeding is often considered unimportant or at least something simple that women can take care of alongside other tasks society expects them to handle. Women have been left to cope as best as they can, often expected to achieve some kind of "supermother" ideal of combining productive and reproductive work, with little support for either. If the importance of exclusive breastfeeding were appreciated, and if the trade-offs for not doing it were explicitly visible to all, society would make an effort to ensure that ideational and external constraints did not interfere with it.
In some settings the major resource available that could increase support for the breastfeeding mother would be the free time that fathers tend to have more of than mothers. However, models are needed to encourage men to provide a wide range of support in child care and household chores. Lacking this, the main model being offered in many places now is the advertisers' image of the father bottle-feeding his baby.
The complementation process
In much of the English-language technical literature, it is not always clear that authors are aware that complementation and replacement of breastmilk are two separate components of the "weaning process." For example, they are indistinguishable in the diagram in figure 1, commonly used to illustrate "weaning."
Thus mothers are rarely advised how to achieve complementation, that is, to avoid unintentionally replacing breastmilk by providing so much additional food and fluid that breastmilk production is reduced. Advice commonly a part of nutrition education, like, "Feed solids to your baby x number of times starting at age y months," does not even indicate the desirability of complementing rather than replacing breastmilk. Attention almost never focuses on how much breastmilk the child receives after the period of exclusive breastfeeding. It is assumed that breastmilk quantity gradually declines from high levels a few months after delivery to low levels a few months later, and that both of these levels are somehow biologically predetermined rather than the result of largely behavioural factors (e.g., frequency and intensity of suckling).
The components of the overall "weaning process" can best be illustrated by comparing a purely schematic plot of the infant's approximate total daily nutritional requirements with the amount of these nutrients that might be provided if the mother breastfed exclusively for the first six months and continued to breastfeed fully but with adequate complementary foods for many months thereafter (fig. 2). Four of these components are sometimes individually referred to as "weaning," but often the meaning is uncertain or vague. Numbers 1 and 2 refer to the initiation of breastfeeding and the period of exclusive breastfeeding.
The zone containing the number 3 illustrates complementation, and the dotted plateau at 4 illustrates the desirability of continuing to breastfeed at the same level even once complementation begins. The 5 is located in the area that illustrates replacement. Finally, the word "wean" has commonly been used to refer to the cessation of breastfeeding (number 6), something else to which the child eventually must accustom himself.
When solid foods are added to the diet of exclusively breastfed infants in the United States, a partial replacement of breastmilk occurs [28], even when the mothers are "advised to maintain the same nursing pattern, not to decrease nursing frequency, and to feed solid foods after nursing" [29]. However, it cannot be assumed that such advice is optimally effective. Whether efforts to provide more effective communication and support can help women to achieve complementation without replacement needs to be tested.
Achieving conceptual clarity on this issue has been complicated by the fact that in industrialized countries (and among the urban elite in much of the developing world), many women begin reducing how much they breastfeed already by six months of age or earlier. They often do not sleep with their baby, carry the baby on their body, or provide the breast very often for comfort or other purposes besides feeding. For them (and the health care establishment that advises them to care for their infants this way), how to achieve complementation without replacement has never been an issue. Replacement feeding with various liquids is intentionally started in the early weeks of life.
In a controlled experiment in Honduras, infants offered complementary foods at four months of age consumed slightly less breastmilk than those who were offered nothing extra [30]. By six months of age, there was no difference in growth rates between those who received complementary feeding and those who continued exclusive breastfeeding. Since the foods in the experiment were sterile and of high nutritional quality, complementation before six months of age in a poor setting would probably have a negative effect on growth.
However, until there is widespread cultural support for exclusive breastfeeding for six months (including among women working for pay), it is likely that very few women will be able to achieve it. The question of how many women may be physically or nutritionally unable to achieve it can be studied only incompletely until these cultural barriers are removed. The net effect of delaying complementation even later than six months also needs to be studied under different conditions if possible.
Based on what we know today, the following recommendation might be more appropriate than weaning advice given currently:
Introduce your baby gradually to solid foods starting at about six months of age. Once he or she accepts them, continue breastfeeding as often as before and add solids as the baby's appetite seems to increase. Once or twice a day is enough in the beginning, but gradually increase them. The child should continue breastfeeding just as often during the second year, but offer solid foods a few times a day. Once you do start to breastfeed less often, remember that you must make even greater efforts to ensure that your child eats several meals of nutritious food each day.
FIG. 2. Components of the "weaning process"
Care aspects of the cessation of breastfeeding (weaning)
The potential implications for care when breastfeeding does not take place or ends early were referred to in the first section of this paper. This section deals with weaning in relation to sustained breastfeeding. Cessation of the breastfeeding relationship is seen by psychologists as a positive developmental step for the child [31]. Where sustained breastfeeding is considered deviant behaviour, weaning may actually be a relief for the child who receives negative messages, for example, that this is something "only babies do" [32]. Investigators who studied data on US breastfeeding pairs in the 1970s [33] wrote that a toddler who is still breastfeeding "leads the weaning process, with some help from a mother who no longer offers the breast without vigorous request." However, where no such societal biases exist, weaning is quite different. When the child is older, the mother can discuss weaning with the child or let external circumstances determine when and if occasional breastfeeding continues.
In many developing countries, weaning is usually occasioned by the next pregnancy. Breastmilk from a pregnant woman may be considered harmful or breastfeeding may be thought to harm the foetus.
Where intercourse is forbidden with a lactating woman, the need for its resumption will influence the duration of breastfeeding is anecdotal evidence that some women have been breastfeeding less in recent years in order to reduce the period of abstinence. They hope that this will reduce the risk that their partners will contract AIDS.
In many cases weaning may be rapid, achieved by placing something bitter on the nipples or sending the child to live with a relative for some time. The psychological impact of abrupt weaning on the infant has long been regarded as one cause of kwashiorkor [34] and was once considered so important that it was studied prospectively on 16 Zulu "volunteers" whose children were given dried milk powder "in an attempt to offset nutritional ill effects" [35]. These children become extremely sad and frustrated. To soothe them, their caregivers may put them to the "dry" breast. Not infrequently, this rapidly leads to the appearance of milk, so-called relactation [36], even among grandmothers [37]. The psychological effect of weaning on the mother can apparently be substantial [38], but I have not seen research on it in developing countries.
Recommendations for action and research
People living in a relatively stable, resource-poor setting have developed approaches for care that are in many respects superior to anything that an outside agency or even a change agent from the nearest town could improve on much. Under conditions of rapid change, however, such as urban migration, much assistance and support may be needed.
Research on care could advantageously be done jointly or in "mirror studies" where similar protocols are used in both developing and industrialized countries. Care is an area in which anthropologists from developing countries could probably offer a lot to those dealing with the serious care deficiencies existing in most industrialized countries. The care aspects of exclusive breastfeeding as well as sustained breastfeeding deserve attention in both contexts, where policy makers, health workers, and others may be uninformed.
Intervention to increase the duration of breastfeeding is a huge topic integral to the entire question of how best to protect, support, and promote breastfeeding [39]. Research should focus on the cultural factors that protect traditional practices of sustained breastfeeding. The advice and counsel of older women should be sought in cultures where the duration of breastfeeding appears to have declined in recent decades. Although protective actions deserve first priority and are least likely to do harm, this does not mean that everything traditional is rational or needs to be protected.
Breastfeeding appears to work best in a carefree environment where it is given little specific attention by most women except when they perceive problems. Except in these self-perceived problematic situations, there is a danger that intervention will change perceptions about breastfeeding, with results that are difficult to predict. We know from historical experience that breastfeeding is biologically extremely robust but psychologically vulnerable. Thus breastfeeding projects should pilot promotional and supportive approaches before implementing them on a large scale, especially in countries where the median length of breastfeeding is still greater than one year or so. Even in countries where the duration of breastfeeding is already short, breastfeeding programmes have a responsibility to document what actions are taken and their impact.
Emphasizing the importance of breastfeeding may be a useful component of a breastfeeding programme, but must avoid putting pressure directly on women to breastfeed. If such advocacy does succeed in convincing leaders and decision makers of the importance of breastfeeding, it must be followed up with demands that society meet the needs of breastfeeding women and children.
The breastfeeding component of good infant care can be achieved only when adequate attention is given to the care of the mother. Women wanting to practice exclusive breastfeeding and full breastfeeding with complementation thereafter should be provided with correct information and relieved of nutritional and work burdens that stand in their way. Both the men in their families and society at large have roles to play in providing this support. The perception that these kinds of demands are radical or unrealistic may be characteristic of societies in which awareness about the importance of breastfeeding is lacking or where the needs of women and children are considered to be of secondary importance.
References
1. Morley R. Lucas A. Influence of early diet on outcome in preterm infants. Acta Paediatr Suppl 1994;405:123-6.
2. Jelliffe DB, Jelliffe EFP. Human milk in the modern world. Oxford: Oxford University Press, 1978.
3. Bowlby J. Attachment and loss. London: Hogarth Press, 1969.
4. Newton M, Newton M. Psychologic aspects of lactation. N Engl J Med 1967;277:1179-88.
5. Klaus MH, Kennell JH. Maternal-infant bonding: the impact of early separation or loss on family development. St Louis, Mo, USA: Mosby, 1976.
6. Uvnäs-Moberg K, Widström A-M, Marchini G. Winberg J. Release of Gl hormones in mother and infant by sensory stimulation. Acta Paediatr Scand 1987;76: 851-60.
7. Widström A-M, Wahlberg V, Matthiesen A-S, Enroth P. Uvnäs-Moberg K, Werner S. Winberg J. Short-term effects of early suckling and touch of the nipple on maternal behaviour. Early Hum Dev 1990; 21:153-63.
8. Avery JL. Closet nursing: a symptom of intolerance and a forerunner of social change? Keeping Abreast J 1977;2:21227.
9. Jelliffe DB. Culture, social change and infant feeding. Am J Clin Nutr 1962;10:1945.
10. Wrigley EA, Hutchinson SA. Long-term breastfeeding, the secret bond. J Nurse-Midwifery 1990;35:3541.
11. Rhode JE. Human milk in the second year. Nutritional and economic considerations for Indonesia. Paediatrica Indonesia 1974;14:198-207.
12. Hoyle B. Yunus C, Chen LC. Breastfeeding and food intake among children with acute diarrhea! disease. Am J Clin Nutr 1980;33:2365-71.
13. Brown KH, Stallings RY, Creed de Kanashiro M, Lopez de Romaña G. Black RE. Effects of common illnesses on infants' energy intakes from breast milk and other foods during longitudinal community-based studies in Huascar (Lima), Peru. Am J Clin Nutr 190;52: 1005-13.
14. Mahalanabis D. Breast feeding and vitamin A deficiency among children attending a diarrhoea treatment centre in Bangladesh: a case-control study. Br Med J 1991 ;303:493-6.
15. Dualeh KA, Henry FJ. Breast milk - the life saver: observations from recent studies. Food Nutr Bull 1989; 11 (3):43-6.
16. Suchindran CM, Adlakha AL. Levels, trends differentials of infant and child mortality in Yemen. Unpublished USAID paper.
17. Greiner T. The planning, implementation and evaluation of a project to protect, support and promote breastfeeding in the Yemen Arab Republic. Doctoral thesis, Cornell University, Ithaca, NY, USA, 1983.
18. Knutsson KE, Mellbin T. Breastfeeding habits and cultural context. J Trop Pediatr 1969;15:48-55.
19. Henriques F. Jamaica. London: MacGibbon and Kee, 1957.
20. Greiner T. Regulation and education: strategies for solving the bottle feeding problem. Cornell International Nutrition Monograph Series no. 4. Ithaca, NY, USA: Cornell University, 1977.
21. Greiner T. Almroth S. Latham MC. The economic value of breastfeeding (with results from research conducted in Ghana and the Ivory Coast). Cornell International Nutrition Monograph Series no. 6. Ithaca, NY, USA: Cornell University, 1979.
22. Marchione TJ, Helsing E. Project report, results and policy implications of the cross-national investigation: rethinking infant nutrition policies under changing socioeconomic conditions. Acta Paediatr Scand 1984;Suppl 314:161.
23. Greiner T. Breast-feeding in decline: perspectives on the causes. In: Jelliffe DB, Jelliffe EDP, Sai FT, Senanayake P. eds. Lactation, fertility and the working woman. London: International Planned Parenthood Federation, 1979:61-70.
24. Kumekpor T. Mothers and wage labour employment. Home Sci 1973;2(4):16-30.
25. Bledsoe CH, Ewbank DC, Isiugo-Abanihe UC. The effect of child fostering on feeding practices and access to health services in rural Sierra Leone. Soc Sci Med 1988; 27:627-36.
26. Brown J. A note on the division of labor by sex. Am Anthropol 1970;72:1073-8.
27. Van Esterik P. Greiner T. Breastfeeding and women's work: constraints and opportunities. Stud Fam Plan 1981 ;12:18497.
28. Stuff JE, Garza C, Boutte C, Fraley JK, Smith EO, Klein ER, Nichols BL. Sources of variance in milk and caloric intakes in breast-fed infants: implications for lactation study design and interpretation. Am J Clin Nutr 1986;43:361-6.
29. Stuff JE, Nichols BL. Nutrient intake and growth performance of older infants fed human milk. J Pediatr 1989;115:959-68.
30. Cohen RJ, Brown KH, Canahuati J. Riveral LL, Dewey KG. Effects of age of introduction of complementary foods on infant breast milk intake, total energy intake, and growth: a randomised intervention study in Honduras. Lancet 1994;344:288-93.
31. Emde RN. Moving on: weaning development for both mother and baby. Keeping Abreast J 1977;2:176-7.
32. Kadushin A. Breastfeeding and weaning a preschool child. Keeping Abreast J 1977;2:208-11.
33. Reame SB, Sugarman M. Breast feeding beyond six months: mothers' perceptions of the positive and negative consequences. J Trop Pediatr 1987;33:93-7.
34. Geber M, Dean RFA. The psychological changes accompanying kwashiorkor. Courrier 1957;6:3 (cited in ref. 9).
35. Albino RC, Thompson VJ. The effect of sudden weaning on Zulu children. Br J M Psych 1956;29:177 (cited in ref. 9).
36. Phillips V. Relactation in mothers of children over 12 months. J Trop Pediatr 1993;39:45-8.
37. Slome C. Non-puerperal lactation in grandmothers. J Pediatr 1956;49:550 (cited in ref 2).
38. Amsel PL. The need to wean-as much for mothers as for baby? Keeping Abreast J 1977;2:188-92.
39. Greiner T. Infant and young child nutrition: historic review from a communication perspective. In: Koniz-Booher P, ed. Communication strategies to support infant and young child nutrition. Cornell International Nutrition Monograph nos. 24/25. Ithaca, NY, USA: Cornell University, 1993:7-15.
