Throughout this report, many unanswered questions were encountered concerning the nutritional status of lactating women and links between maternal nutrition and both infant and maternal health. The subcommittee concluded that research is needed to develop indicators of nutritional status during lactation, to identify groups at risk of nutritional problems, and to determine effects of maternal nutritional status on various measures of lactation performance (such as the volume and composition of milk, the duration of lactation, and infant growth and health).
In the process of reviewing data concerning the nutrition of breastfeeding women, the subcommittee could not ignore the very different likelihood of breastfeeding given race or ethnic group, income, and region of residence in the United States. Indeed, the most disadvantaged groups, which would be the groups most likely to benefit from breastfeeding, were found to have the lowest breastfeeding rates. Many factors other than maternal nutrition influence both a woman's desire to breastfeed and her success in doing so. To implement the subcommittee's recommendations for breastfeeding, it is desirable to know much more about the determinants of breastfeeding. To investigate the effects of maternal nutrition on the milk, the infant, and the mother herself, much more information is needed on the interactions between milk production, infant demand, and maternal nutrient intake and stores, as well as on the transfer of nutrients from mother to milk and the factors regulating this process.
The following recommendations are not listed in order of priority.
RECOMMENDATIONS THAT FOCUS PRIMARILY ON NUTRITION
Following are the subcommittee's research recommendations that concern mainly the nutrition of the lactating woman or of her breastfed infant.
Indicators of Maternal Nutritional Status
Research is needed to develop indicators of nutritional status for lactating women.
First, the identification of normative values for nutritional status should be based on observations of representative, healthy, lactating women in the United States. In addition, indicators are needed both of (1) risks of adverse outcomes related to the mother's dietary intake and (2) the potential of the mother or her nursing infant to benefit from interventions designed to improve their nutritional status or health.
Identification of Groups of Mothers Who Need Nutritional Intervention
Research is needed to identify groups of lactating women in the United States who are at nutritional risk or who could benefit from nutrition intervention programs.
In general, it has been difficult to identify groups of mothers and infants in the United States with nutritional deficits that are severe enough to have measurable functional consequences. Priority should be given to the study of lactating women in subpopulations believed to be at risk of inadequate intake of certain nutrients, such as calcium (black women) and vitamin A (low-income women). The potential influence of culture-specific food beliefs on nutrient intake of lactating women should be included in any such investigations.
Maternal Nutrition and Lactation Performance
Intervention studies of improved design and technical sophistication are needed to investigate the effects of maternal diet and nutritional status on milk volume; milk composition; infant nutritional status, growth, and health; and maternal health.
The nursing dyad (the mother and her infant) has seldom been the focus of studies. Thus, a key aspect of this recommendation is concurrent examination of the mother, the volume and composition of the milk, and the infant. The design of such research should be adequate for causal inference; thus, if possible, it should include random assignment of lactating subjects to treatment groups. Appropriate sampling and handling of milk for the valid assessment of energy density, nutrient concentration, and total milk volume are essential, as is accurate measurement of nutrient concentrations.
With regard to energy balance of lactating women, the threshold below which energy intake is insufficient to support adequate milk production has not yet been identified. Resolution of this question will probably require supplementation studies of women in developing countries whose diets are chronically energy deficient. Although such deficient diets are not common in the United States, identification of the level of energy intake that is too low to support lactation will be useful in establishing guidelines for women who want to breastfeed but also want to restrict their energy intake to lose weight. Although chronically low energy intakes by women in disadvantaged populations may not be completely analogous to acute energy restriction among otherwise well-nourished women, ethical considerations limit the kinds of investigations that could directly address the influence of energy restriction. In supplementation studies, measurements should be made of lactation performance and of any impact on the mother's nutritional status and health, including the period of lactation amenorrhea.
With regard to specific nutrients, the impact of relatively low intakes of folate, vitamin B6, calcium, zinc, and magnesium during lactation on the mother's nutritional status and health needs to be assessed in more detail. As a part of this assessment, studies of the absorption of calcium, zinc, and magnesium during lactation will be useful. There is also a need to identify a reliable indicator of vitamin B6 status of infants and to document the relationships between this indicator, maternal vitamin B6 intake, and vitamin B6 content in milk. Finally, resolution of the conflicting findings concerning the impact of maternal protein intake on milk volume would be desirable.
Physical Activity, Energy Intake, and Lactation
The impact of high levels of physical activity on milk volume, milk composition, and duration of lactation requires further study, especially in populations in which energy intake is low relative to total need. Such research should be designed to identify the relative energy deficit imposed by high levels of physical activity.
With greater numbers of women involved in physically demanding work and with increased interest in physical fitness, an increased number of women in the United States may need or want to resume heavy physical activity post partum. The potential impact of such activity on lactation is unknown.
Maternal Nutrition and the Infant's Immune Function
Studies should be conducted to determine relationships, if any, between the nutritional status of the mother, the concentrations and functions of the components of the immunologic system in human milk, and the susceptibility of the recipient infant to common infectious agents.
Desirable design features of these studies include categorization of sufficient numbers of breastfeeding women according to their nutritional status, measurement of the levels and functions of key immunologic agents in human milk, monitoring of the immune system of the infants, and determination of the incidence and severity of common infections in the nursing dyad during the study periods.
Currently, there are conflicting reports concerning the effects of maternal nutritional status upon the immunologic system in human milk. If the nutritional status of the mother alters the levels or functions of those defense agents in human milk, it would be important to investigate how low dietary intake by the mother influences the recipient infant's risk for infectious diseases compared with that of breastfed infants of well-nourished mothers, as well as that of formula-fed infants. If relationships between maternal diet and infant risk are found, studies are warranted to investigate whether nutritional supplementation of the mother corrects the problem.
Indices of Nutritional Status of Infants
Improved indices are needed for the evaluation of specific nutrient status.
Most estimates of nutrient requirements of infants are based on measurements of specific nutrient intakes of exclusively breastfed infants. This approach often is the result of inadequate alternatives for the assessment of the status of specific nutrients (e.g., vitamin B6, folate, and zinc). Improved indices of nutrient status depend upon a better understanding of the role of specific nutrients in the normal development of functional capacities. The need for population-based studies and more detailed metabolic approaches is evident for both practical purposes (clinical recommendations and management) and an improved understanding of the physiology of normal development.
Absorption and Utilization of Nutrients from Human Milk
The bioavailability of specific nutrients (especially folate and iron) in human milk merits study, as does the biological basis for the high efficiency of nutrient utilization by breastfed infants.
Reviews of the composition of human milk and the general nutritional status of breastfed infants indicate that nutrients in human milk are highly bioavailable and that they are utilized with a high degree of efficiency. Mechanisms that account for these observations are not understood. An improved understanding of these observations should lead to knowledge of normal nutrient needs, better human milk substitutes, and more effective nutritional management of high-risk infants, such as very-low-birth-weight infants and infants with impaired gastrointestinal and renal function.
RECOMMENDATIONS THAT FOCUS ON NONNUTRITIONAL FACTORS INFLUENCING BREASTFEEDING
Determinants of Breastfeeding
Given the decrease since 1982 in the percentage of mothers who breastfeed, further research is needed to identify the determinants of the decision to both initiate and continue breastfeeding among U.S. women in general and, in particular, among adolescents, those with limited education, and black, Hispanic, and other minority women.
Such research should use methods designed to elucidate maternal beliefs and values, as well as situational factors related to infant-feeding practices. Special attention should be given to the beliefs and attitudes women hold regarding the interaction of maternal diet and nutritional status with breastfeeding.
Current data are limited to a few very small studies. Knowledge of beliefs, attitudes, and situational factors is necessary for both breastfeeding promotion and effective nutritional counseling of lactating women.
Health Care in Support of Lactation
Research is needed to study how various approaches to the health care of lactating women (and those who plan to breastfeed) affect their lactation performance.
Such care would include nutritional screening or evaluation of the mother, nutritional guidance based on this screening, and guidance in establishing breastfeeding. It would be useful to compare approaches used prenatally, at an early (1 to 2 weeks post partum) office or home visit, and during the usual visit at 4 to 6 weeks post partum.
Since mothers are routinely discharged from the hospital within 24 to 48 hours after delivery, little time is available to encourage lactation or to cover other aspects of health education. Breastfeeding problems such as engorgement, sore nipples, delayed milk supply, and weak suck are most common in the first week post partum; they usually occur after the mother has left the hospital and is no longer in direct contact with health care providers. The first week post partum is also a useful time for checking the neonate's weight and for monitoring the neonate for health problems such as jaundice. After delivery, the mother is likely to be highly receptive to educational messages regarding her food intake during lactation and the overall health and nutrition of her infant. Home visits provide opportunities for assessment of resources and environmental conditions that could affect the mother's nutrition or her overall health and ability to breastfeed.
Maternal Anxiety, Stress, and Illness
Studies are needed to investigate the potential influence of maternal anxiety, stress, and illness on milk volume and composition.
Such research would need to include various measures of anxiety and stress—acute and chronic, as well as physical and emotional.
Although certain types of acute physical stress have been associated with impairment of the milk-ejection reflex, there is no information on effects of the more common emotional stresses of caring for a newborn and coping with other demands of family, with work, or with either lack or loss of income.
RECOMMENDATIONS THAT FOCUS PRIMARILY ON GROWTH AND HEALTH
Growth charts are needed that characterize the growth of breastfed infants.
Acceptable approaches for developing such charts include either (1) the collection of longitudinal data, preferably monthly, on length, weight, extent of breastfeeding, and demographic variables from a large, representative sample of infants or (2) the application of appropriate statistical methods to data from a smaller sample of healthy infants who are breastfed throughout the first year of life. Development of the charts should be accompanied by studies designed to distinguish between normal and faltering growth of breastfed infants, using functional outcomes such as physical activity, mental and motor development, and morbidity.
Current infant growth charts were developed using data derived from infants who were primarily formula fed during a period when infant-feeding practices differed considerably from those in use today. Evidence from several studies suggests that growth rates of infants breastfed on demand differ from those illustrated on available growth charts.
Data on normative growth have been recently published for infants followed in Iowa between 1965 and 1987 (Nelson et al., 1989). They are not fully appropriate for constructing growth charts for breastfed infants since those infants were followed only just past the third month of age, the breastfed infants were allowed up to 240 ml of formula per day, and, before 1979, all infants were permitted to receive solid foods beginning at age 1 month.
To interpret the growth charts developed, criteria are needed for identifying a cutoff below which growth rates are unsatisfactory.
Psychological Health of Mother and Infant
Studies should be conducted to investigate the psychological benefits of breastfeeding to the mother and infant.
It is widely believed that breastfeeding has powerful psychological benefits for the mother and infant. Nonetheless, there is relatively little scientific evidence to support that belief. If this belief is correct, the public health implications would be profound. Some of the specific points that should be addressed are the effects of breastfeeding upon the self-esteem of the mother, the mother's concerns with parenting, the ability of the mother to deal with social problems in her family, social interactions between the mother and the infant, the social development of the infant, the ability of the child to adapt to new environmental circumstances, and the possibility that observed effects are related to the transfer of substances to the infant through the milk.
It is essential to determine whether and to what extent breastfeeding protects against infant and early child mortality in populations with generally low rates of infant mortality, and especially in subpopulations with higher than usual infant and child mortality rates.
The rates of infant and early childhood mortality are at or near their historically lowest levels in the industrialized countries; nevertheless, there is every reason to believe that breastfeeding would still confer some benefit on survival. It is unknown at the moment whether infant and early childhood mortality rates are lower among breastfed infants in situations in which death rates are generally low. This information is particularly important among subpopulations, such as ethnic minorities, many of which have higher mortality rates than the population at large.
Use of Substances
The potential influence of maternal smoking and moderate alcohol and coffee consumption on milk production, composition, and infant health requires further investigation.
Data are needed on the direct effects of smoking on milk volume, not just on plasma prolactin levels. Studies should include consideration of dose-related effects.
At present, there is no clear basis for determining the level of cigarette smoking or alcohol consumption that could harm the infant. Although the recommendation is not to smoke at all and to drink alcoholic beverages in moderation, if at all, many women may not follow this advice. Furthermore, preliminary evidence from Costa Rica (Muñoz et al., 1988) suggests that coffee
intake may affect milk iron concentration and infant iron status, but studies of women in other populations and with lower coffee intakes (<3 cups/day) are needed.
Human Immunodeficiency Virus
Investigations should be performed to ascertain whether human immunodeficiency virus (HIV) infection is transmitted to the recipient infant via breastfeeding, whether there are other health effects of breastfeeding upon infants who are otherwise infected with HIV, and whether breastfeeding has harmful or beneficial effects on the health of the mother infected with HIV.
Current public health policies concerning breastfeeding and HIV are based upon a very limited, somewhat anecdotal data base. Moreover, the two principal public health pronouncements are at odds with each other: the first precludes breastfeeding (CDC, 1985), whereas the second encourages breastfeeding in the case of HIV-infected mothers (WHO, 1987). Systematic studies are therefore required to determine the following:
To what extent is HIV excreted in the milk of HIV-positive women?
Is the excretion of HIV in human milk limited to mothers who have received HIV-contaminated blood during the perinatal period?
If HIV is present in human milk, does it present a risk to the recipient infant who was not exposed prenatally, does it increase the risk for an infant who was infected prenatally, or does it immunize against the infection?
Does human milk inhibit the growth of or kill HIV?
Does breastfeeding have a salutary effect upon the HIV-infected infant by defending the recipient against complicating opportunistic agents?
Does the increased nutritional burden of lactation decrease the resistance of HIV-infected women to the infection?
Do other changes during lactation, such as the secretion of lactogenic hormones, affect the ability of the immunologic system of the woman to deal with this retrovirus?
Long-Term Health of the Recipient of Human Milk
Long-term studies should be conducted to determine whether breastfeeding protects against chronic conditions that first become apparent in later life.
These conditions include atopic disorders; type I diabetes mellitus; obesity; atherosclerosis, hypertension, and other chronic cardiovascular diseases; inflammatory bowel disease; and lymphoma and other malignancies. Research on these issues should include measures of the duration of exclusive and partial
breastfeeding, timing of introduction of other foods, and family history of the condition, among other variables.
There are a number of retrospective studies that suggest that breastfeeding may reduce the risks of developing certain chronic conditions later in life. If that is the case, then the public health advantages of breastfeeding transcend the immediate period of breastfeeding. Possible benefits are great since these conditions are not readily amenable to definitive therapy, and their related health care costs are very high. Thus, more definitive prospective studies and more detailed retrospective cohort and case-control studies of groups at high risk for the condition are warranted. If long-term protective effects are confirmed, studies should be conducted to ascertain whether the effects are modified by maternal nutrition. If the positive findings are borne out or extended, they would provide a further indication for promoting breastfeeding among the entire U.S. population.
Health Effects of Lactation on Women with Chronic Diseases
Investigations are needed of the effects of lactation on the health of women who have specific chronic diseases (such as type I diabetes mellitus, cystic fibrosis, and hypertriglyceridemias).
The subcommittee's review indicates that information on women with underlying chronic diseases is particularly limited. Consequently, there is incomplete understanding of adjustments that may be necessary during lactation in the clinical management of those conditions. Surveys and detailed metabolic studies of specific populations are indicated to determine the extent to which current infant feeding recommendations apply to this group of women.
Effects of Chronic Diseases on Lactation
Investigations of the effects of specific chronic diseases (such as diabetes mellitus and cystic fibrosis) on the process of lactation are needed.
Improved understanding of lactation in healthy women provides a basis for investigations of the impact of underlying chronic diseases on that physiologic process. The metabolic adjustments made during lactation involve nearly every maternal organ system. Knowledge of changes in lactation resulting from relevant pathophysiologies will enhance the understanding of normal lactation and provide a basis for the improved management of the lactating woman and her breastfed infant.
Short-Term and Long-Term Health of the Mother
Studies are required that investigate short-term effects (changes in body weight, sexuality, and behavior, as well as incidence of infectious disease) and long-term effects (such as obesity, osteoporosis, and specific cancers) of lactation on the health of women.
Breastfeeding recommendations usually are based on the expected effects on the infant's nutritional status, the infant's decreased susceptibility to infectious illnesses, and improved maternal-infant bonding. Benefits and risks to the mother are considered less often. As reviewed in Chapter 8, there are reasons to suspect that lactation may decrease the mother's risk of developing breast cancer and osteoporosis. Lactation's effects on maternal weight status is of particular short- and long-term interest. Given current recommendations regarding breastfeeding and the physiologic changes associated with lactation, a more thorough examination of maternal outcomes is central to the implementation of current policy and to helping women make fully informed choices.
CDC (Centers for Disease Control). 1985. Recommendations for assisting in the prevention of perinatal transmission of human T-lymphotropic virus type III/lymphadenopathy-associated virus and acquired immunodeficiency syndrome. Morbid. Mortal. Wkly. Rep. 34:721-732.
Muñoz, L.M., B. Lönnerdal, C.L. Keen, and K.G. Dewey. 1988. Coffee consumption as a factor in iron deficiency anemia among pregnant women and their infants in Costa Rica. Am. J. Clin. Nutr. 48:645-651.
Nelson, S.E., R.R. Rogers, E.E. Ziegler, and S.J. Fomon. 1989. Gain in weight and length during early infancy. Early Hum. Dev. 19:223-239.
WHO (World Health Organization). 1987. Breastfeeding/breast milk and human immunodeficiency virus (HIV). Weekly Epidemiol. Rec. 62:245-246.