Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
7 Conclusions Since World War II, there have been major improvements in the health of women and children in most developing countries. These improvements, how- ever, have been unevenly distributed: they have been dramatic in some countries, moderate in others, and small in many countries, particularly the poorest countries of Africa and South Asia. Overall, the incidence of poor health and of infant, child, and maternal mortality remains unacceptably high throughout the develop- ing world. Many developing countries have also experienced significant declines in fertility over the last 40 years. Other countries with the highest rates of infant, child, and maternal mortality also have high fertility rates. This report has examined the relationship between fertility and health during the course of this transition in fertility and mortality and has assessed the impact of changes in reproductive patterns on the health of women and children. As discussed throughout the report, assessing the health effects of reproduc- tive patterns and changes in them is not a straightforward task. First, as outlined in Chapter 2, the relationships between fertility and health are very complex. For this evaluation, we have focused on what we term direct effects, a subset of the 'possible associations between reproductive patterns and women's and children's health. For example, we hypothesize that child spacing directly affects child mortality through mechanisms such as maternal depletion. By contrast, an indi- rect effect could occur if mothers who space their children more closely were less likely to be able to work for pay and thus improve the economic status of their family. Other indirect relationships between fertility and health may be as important as the direct effects. 90
CONCLUSIONS 9 1 Second, the data on which studies cited in this report are based are often seriously deficient. For example, data on maternal health and mortality are scarce, and research examining the relationship between birth spacing and mater- nal mortality in developing countries has yet to be carried out. Most studies of contraceptive risks and benefits are based on data from industrialized countries, and conclusions about the safety of contraceptive use in developing countries must in part be made by extrapolation. Information on gestational age, birth- weight, and maternal and infant nutritional status, which is necessary to sort out the association between birth spacing and child health, has been difficult to collect in developing countries. Furthermore, most data on which analyses of maternal and child health are based come from observational rather than experimental studies, a situation that complicates analytic designs, making it more difficult to draw inferences about causality. Third, the limitations of the analytic strategies of many studies make fox conclusions difficult to draw. Many studies have not adequately considered alternative factors that may account for the observed relationships. Problems such as relevant, unmeasured influences and the joint operation of causal factors affect many studies of human behavior, but may be particularly troublesome in the associations discussed in this report because fertility and health are complex, interrelated processes. Few of the studies on which this report is based have attempted to deal with these issues in a comprehensive manner. While the shortcomings of the evidence are clear and should be kept in mind, we believe that the available evidence is sufficient to draw important conclusions about how reproductive patterns affect women's and children's health. REPRODUCTIVE PATTERNS AND WOMEN'S HEALTH: RISKS FOR INDIVIDUAL WOMEN Matemal mortality has declined significantly in this century in the developed world. Some developing countries have also witnessed declines in maternal mortality because of improvements in prenatal care, better midwifery, widespread use of aseptic delivery procedures, the introduction of antibiotics, improvements in the provision of health services, and overall advances in women's social position and standard of living. These experiences provide unequivocal evidence that mortality and morbidity due to reproductive causes can be reduced. What role can changes in reproductive patterns play in this process? It is clear that reductions in the number of pregnancies women have in their lifetimes and in the incidence of high-risk pregnancies will substantially reduce the risk of maternal mortality and morbidity for individual women. Furthermore, the positive effects of lower fertility and reductions in the frequency of high-risk pregnancies on women's health are likely to be greatest in populations in which fertility rates are high, health facilities are poor or unavailable, and the incidence of reproductive morbidity is high.
92 CONTRACEPTION kD REPRODUCTION Each time they become pregnant, women face a risk of morbidity and mortality, and these risks are higher in societies in which health conditions are poor. A reduction in the number of times a woman becomes pregnant during her life will reduce her lifetime risk of dying from reproductive causes. If, in addition to reducing the total number of pregnancies she has, a woman uses contraception to avoid high-risk pregnancies, the beneficial effect will be reinforced. Pregnan- cies that are particularly high risks for women include those that occur when a woman has a previous gynecological or obstetrical illness or problem, such as postpartum hemorrhage, or has a preexisting health problem, such as diabetes, while she is pregnant. In addition, pregnancies to very young and older women, and first- and higher-order pregnancies (fifth and higher-order) appear to be riskier than others. While first births cannot, of course, be avoided if a woman chooses to have children, the risks appear to be attenuated if the first birth is delayed beyond the high-risk early teenage years. Risks Associated With Induced Abortion Unsafe induced abortion is an important cause of reproductive morbidity and mortality. As noted in Chapter 3, unsafe induced abortion is a primary cause of maternal mortality. Family planning services have the potential to reduce abor- tion-related health problems by reducing unwanted pregnancies. Countries in which safe abortion is not available have the greatest obligation to provide all needed contraceptive and medical services to reduce unintended pregnancy and to treat the complications of unsafe abortion. Contraceptive Risks and Benefits According to a United Nations (1989) estimate, over 400 million women in developing countries are using some form of contraception. This diffusion of modern contraceptives has facilitated widespread regulation of fertility. The most important conclusion to be drawn from the extensive literature examining the noncontraceptive health risks and benefits of contraception is that risks associated with contraceptive use are significantly less than the risks associated with preg- nancy and childbirth. This conclusion is especially true in many countries in the developing world where childbirth and pregnancy risks are high. Although contraception in general is safer than pregnancy, it is nonetheless important to consider contraceptive risks and benefits in relation to the character- istics of different women. For example, to recall Chapter 4, decisions to use oral contraceptives should consider factors such as age and whether a woman smokes cigarettes, and decisions to use an intrauterine device should consider a woman's pattern of sexual activity. Both contraceptive needs and the risk-benefit profiles of women change over the course of their reproductive lives. Information about
CONCLUSIONS 93 risks and benefits, including information on contraceptive efficacy, at different life-cycle stages is necessary for informed decision making and for safer, more effective contraceptive practice. An ongoing program to evaluate the health risks and benefits of contracep- tion use is needed in both developing and developed countries. Such research will help scientists to understand the effects of specific methods under different conditions of use. Such studies would be especially useful in the case of oral contraceptives and other steroid or hormonal methods because the~evolution of methods (e.g., changes in dosages in oral contraceptives) and long latency period of potential health problems (e.g., cancer) means that potential risks and benefits can be understood only through long-term studies. Ongoing evaluation of oral contraceptives and other methods is also needed in both developing and devel- oped countries to provide the depth of knowledge necessary for the refinement and improvement of contraceptive methods and guidelines for their use. REPRODUCTIVE RISKS AND CHILDREN'S HEALTH: RISKS FOR INDIVIDUAL CHILDREN Parents can maximize the chances of survival and good health for each of their children by lengthening the intervals between births, avoiding pregnancies at very young and older ages, and avoiding higher-parity pregnancies. Firstborn children also appear to have higher risks of morbidity and morality. However, parents obviously cannot increase their children's chances of survival by avoiding having a first birth, although they may be able to improve their firstborn's survival chances by delaying the birth until their early twenties and having their second child more than two years after their first. The association between birth spacing and child survival has been observed in developing countries, in high-mortality historical populations, and in contem- porary industrialized countries. In this wide range of historical and contemporary populations, children born after short birth intervals have higher mortality than children born after longer intervals. Furthermore, the relationship remains impor- tant in studies that hold some of the potentially confounding factors constant. However, when potentially confounding factors are held constant, it appears that the detrimental effects of older maternal age and higher birth order on children's health are less important than previously thought. Our understanding of the mechanisms affecting the observed associations between health and birth spacing as well as for maternal age, birth order, and family size remains incomplete. While a relationship between close birth spacing and higher infant and child mortality is widely observed, we do not yet know whether this relationship is due to birthweight or gestational age or some other factor. In contemporary developing countries, families who use contraception to space their births may also be more likely to use health services when their children are ill. Until funkier research has been completed, caution is necessary
94 CONTRACEPTION ED REPRODUCTION when drawing conclusions about the amount of improvement in children's sur- vival that may result from changes in reproductive patterns. Nonetheless, given the breadth of the available evidenceas well as the likelihood that there are important indirect benefits of lower fertility on family health and well-being we believe that parents who wish to improve their children's chances of survival should avoid short birth intervals, births at very young and at older ages, and higher-order births. The effects of this strategy are likely to be greater where living standards are poor, the incidence of disease is high, and parents do not have access to adequate health care. Given the strength of the observed relationship between short birth intervals and infant and child health, policy makers, health and family planning program managers, and concerned citizens in developing countries can improve children's health by encouraging both breastfeeding and contraceptive use in order to lengthen birth intervals. With this approach the beneficial effects of breastfeeding for children's health are reinforced by the contraceptive effect of breastfeeding in delaying the next birth. In many developing countries, breastfeeding has declined during the course of development, and many mothers apparently decide to discon- tinue breastfeeding when they adopt contraception. It also appears that some family planning programs may discourage breastfeeding for women adopting contraception out of concern about the effects of hormonal contraception on breast milk. While more research is needed on the relationship between contra- ceptive use and breastfeeding and, in particular, on the extent to which women substitute one for the other, programs designed to encourage both breastfeeding and contraceptive use for birth spacing are likely to have important benefits for the health of children. Because of the clear relationship between a reduction in the number of pregnancies a woman has and reduction in her lifetime risk of dying from reproductive causes, another potential health benefit of lower fertility for children is a reduced risk of losing their mother to illness or death. Documentation of the effects of maternal mortality on the health and well-being of children in develop- ing countries is mostly anecdotal. However, since women remain primary care- ~kers for children in most countries, maternal death or illness almost certainly has severe consequences for the health and survival of young children. AGGREGATE-LEVEL EFFECTS Reducing fertility and avoiding high-risk pregnancies are important strate- gies to reduce the risk of mortality and morbidity for individual women and children. Determining the implication of widespread individual changes in repro- ductive patterns on aggregate measures of health, such as the infant mortality rate, however, is complex and has been the subject of considerable debate (see Trussell and Pebley, 1984; Winikoff, 1983; Bongaarts, 1988; Trussell, 1988; Palloni, 1988~.
COI¢CLUSIONS 95 There are at least two reasons to believe that a reduction in fertility will bring about an unambiguous decline in mortality rates and an improvement in the health of a population. First, maternal mortality rates, which measure the frequency of death due to reproductive causes, are likely to decline as a consequence of a fertility decline, because women will be exposed to risks associated with preg- nancy less often. In other words, if fertility rates decline, there will be fewer pregnancies, and thus fewer women exposed to the risk of dying from reproduc- tive causes each year. Second, at a given level of funding for health care, lower fertility rates are likely to mean better health care for each pregnant woman and child because more resources per capita will be available. As a consequence of better health care, maternal and child morbidity and mortality would be expected to decline. Measuring the impact of change in the level and pattern of childbearing for a society is complicated by the fact that mortality sates reflect in part the distribu- tion of pregnancies and births among high-risk and low-risk groups in the popula- tion. A change in reproductive patterns is likely to change this distribution of pregnancies and births in several ways at the same time. The end result, in some populations, may be that there is relatively little change in mortality rates as a direct consequence of change in reproductive patterns, even though the rates for specific groups may decline. Insofar as these changing patterns of reproduction are associated with lower levels of fertility, they will also be associated with lower numbers of infant deaths. As shown in Chapter 6, one of the major effects of a fertility decline is to reduce the proportion of births to high-parity women in a population and, at the same time, to increase the proportion of first births. Currently available evidence indicates that infant mortality rates are higher for first births as well as for higher- order births. Since both first births and higher-order births have elevated risks of mortality, the effect of a fertility decline, if everything else remains the same, may be small or neutral. Because the available data indicate maternal mortality ratios are higher for first pregnancies as well as for higher-order pregnancies, this argument also applies to maternal mortality ratios, which measure the average risk of death for women associated with each pregnancy in a population. Another type of distributional change may mean that, holding everything else constant, mortality rates might actually increase somewhat during the early part of a fertility decline. The reason is that it is often women who are at relatively low risk for infant, child, or maternal mortality who first adopt contraception as a means of reducing fertility. Thus, a larger proportion of pregnancies and births would occur to women who were at higher risk in the initial period of the fertility decline than had been the case before the decline began. The potential scope for improving health at the societal level through changes in reproductive patterns also depends on the distribution of pregnancies and births among high- and low-risk groups in a society. In South Asia and some African countries, for example, many births occur to very young women. There is,
96 CONTRACEPTION ED REPRODUCTION therefore, considerable potential for improving the health of women and children by delaying the onset of childbearing. The potential for improving child health through changes in patterns of birth spacing patterns is likely to be larger in Latin American countries than in other areas of the developing world because of the high proportion of short birth intervals. The current scope for improvements in child health through birth spacing is considerably more limited in South Asian and sub-Saharan African countries because short birth intervals are relatively rare. A central concern for policy makers in South Asian and sub-Saharan African countries is that birth intervals may become shorter during the course of moderni- zation because of the decline in breastfeeding and postpartum abstinence. If such changes occur, the effect may be to slow the pace of infant and child mortality decline relative to the pace that would be achieved if longer birth spacing had been maintained. In reality, fertility declines and changes in reproductive patterns do not occur in isolation. They are accompanied by (and brought about by) a variety of other important social and economic changes as well as by governmental policy initia- tives. These changes themselves are likely to have major effects on health and mortality independent of their relation to fertility, and declines in fertility may in part be responsive to these improvements in health and mortality conditions. In fact, most countries have experienced fairly continuous mortality declines while undergoing fertility transitions, and interruptions in the mortality decline have generally been due to natural disasters, economic calamity, or major epidemics. INDIRECT EFFECTS OF REPRODUCTIVE CHANGE ON THE HEALTH OF WOMEN AND CHILDREN Lower fertility and changing reproductive patterns may also have important indirect effects on the health of women and children. These effects include shifting attitudes away from fatalism, making it feasible for women to develop roles independent of motherhood, and increasing the resources available for each member of the family because of smaller family sizes. These indirect effects are difficult to document, but in the long run they may be equally or more significant than the direct effects of changing reproductive pattems. Considerable work remains before we have a clear understanding of the indirect effects of changing reproductive patterns on maternal and child health. There is particular need to understand how family structure and the process of family decision making in developing countries adjusts to changing economic, social, and demographic situations. Seemingly minor changes in one element of reproductive patterns may have long-term consequences that may improve the health and well-being of all family members. For example, Ryder (1976) argues that delaying the age at which women in developing countries have their first birth is of particular importance to these societies because it allows time for women to participate in other, nonfamilial social roles, such as that of a student or worker.
CONCLUSIONS 97 Exposure to modern medical services may also have significant long-term consequences on the attitudes of women and children beyond the immediate effects of treatment. For example, women may be more likely to continue treatment if their first exposure to modern medicine produces positive results. Increasing our understanding of how attitudes and family health care are influ- enced by exposure to modern health care, including family planning services, is crucial to policy makers. FAMILY PLANNING AND THE HEALTH OF WOMEN AND CHILDREN One aim of this report is to evaluate the potential for family planning to bring about additional improvements in the health of women and children. At several points, we have emphasized the complexity of the relationships involved, but the implications of the available evidence are clear. Maternal, infant, and child mortality and morbidity remain important problems throughout the developing world and are clearly related to reproductive patterns. Although there is a great deal of variation in the effect that reproduction has on the health of individuals, families, and countries, the reduction of high-risk pregnancies typically would have a positive impact on the health of mothers and children throughout the developing world. Contraceptive use and controlled fertility are safer than unregulated child- bearing. Unsafe abortions are a significant cause of maternal mortality in many developing countries, a finding that must be considered by countries debating the merits of making safe abortions available. Greater control of reproduction would improve maternal and child health by reducing births, especially high-parity births, and by reducing closely spaced pregnancies. Easy access to contraceptive services should be encouraged, particularly in conjunction with efforts to increase prenatal care, to improve breastfeeding practices, and to advance other health services. Efforts to increase education, especially female literacy, and to improve nutritional status may act synergistically with family planning and health services to improve maternal and child health. It should also be clear from this report that additional research is needed in many areas before we understand adequately the causal linkages between repro- duction and women's and children's health. While this research is being carried out, however, government officials, policy makers, public health practitioners, and individuals everywhere must make decisions about the best ways to improve the health of individuals and families. Family planning activities have an impor- tant potential role as a component of health programs directed toward improving the health of women and children.