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.
1 Introduction THE CONTRACEPTIVE REVOLUTION IN THE DEVELOPING WORLD Over the past 25 years the world has experienced a contraceptive revo- lution (Donaldson and Tsui, 1990~. Contraceptive prevalence the percent- age of women of reproductive age, married or living in union, that use some type of contraceptive method-has risen from less than 10 percent around the world in the early 1960s to an estimated 55 percent in the late 1980s and early 1990s (Bongaarts et al., 1990; Population Reference Bureau, 1992~. This increase is by no means limited to the developed countries. Although prevalence levels are higher in the industrial than in the developing world (72 versus 51 percent), it is noteworthy that more than half the women of reproductive age in developing countries currently use some form of contra- ception (Population Reference Bureau, 1992~. The importance of this phenomenon lies in the close relationship be- tween contraceptive prevalence and fertility (Mauldin and Segal, 1988~. Contraceptive use is one of four key factors that determine fertility, the other three being involvement in sexual union, postpartum nonsusceptibility, and induced abortion (Bongaarts, 1978~. Of the four, contraception has the strongest effect on fertility in most developing countries (Donaldson and Tsui, 1990~. Africai is different, in that in the past the duration of postpar 1Throughout this volume, the terms sub-Saharan Africa and Africa are used interchangeably. The countries north of the Sahara (Egypt, Libya, Algeria, Tunisia, and Morocco) are labeled as s
6 FA CTORS AFFECTING CONTRACEPTIVE USE turn nonsusceptibility has been the key determinant of fertility levels. How- ever, with an increase in the use of modern contraception, the relative im- portance of the two factors is changing (Chapter 7; Jolly and Gnbble, 1993~. In short, for those interested in current and future fertility rates, it is essen- tial to examine the issue of contraceptive prevalence. Even so,. the contraceptive use rate in Africa, 13 percent, is low relative to other regions of the developing world (Population Reference Bureau, 1992; United Nations, 1993~. The prevalence rates for Asia and Latin America are surprisingly similar and much higher, namely, 56 and 57 per- cent, respectively. However, these regional figures mask considerable variation among countries of a given continent: for example, Thailand (66 percent) versus Pakistan (12 percent); Costa Rica (70 percent) versus Bolivia (30 percent); and Zimbabwe (43 percent) versus Mali (3 percent) (Population Reference Bureau, 1992; Demographic and Health Survey data tapes). No doubt there are also marked differences within each country. The moderate to high levels of contraceptive use found in most Asian and Latin American countries generally reflect both existing social condi- tions (as measured by literacy, female education, life expectancy, infant mortality, and related indicators) and access to family planning services through the public or private sector (Mauldin and Ross, 1991~. The coun- try-by-country differences in contraceptive prevalence notwithstanding, family planning has become a widely practiced, culturally acceptable behavior throughout many countries in Asia and Latin America. CONTRACEPTIVE USE IN SUB-SAHARAN AFRICA The situation is markedly different in sub-Saharan Africa, where high birth rates have been the norm, and access to modern contraception was extremely limited (except in pilot program areas) prior to 1980. Among the factors that have contributed to sustained high fertility in Africa are a large percentage of the population living in rural areas, low levels of socioeconomic development, high rates of infant and child mortal- ity, and patterns of social organization and deeply ingrained cultural values that maintain the demand for large families. Moreover, until recently, the majority of African government officials expressed little support for "popu North Africa. The Sudan, listed by the United Nations as North Africa, is nonetheless included in our analysis. Except where noted, discussion is limited to the 39 mainland sub-Saharan countries plus the large island-nation of Madagascar. Mauritius, which has a relatively high modern contraceptive prevalence rate of 46 percent, is excluded, given its unique sociohistori- cal background and the ethnic composition of its population (Mauritius Ministry of Health et al., 1992).
INTRODUCTION 7 ration control," as enunciated at the World Population Conference in Bucharest in 1974; their position then was that "development is the best contracep- tive" (Donaldson and Tsui, 1990~. African governments questioned the motivation of Western nations concerned with limiting the population growth of African nations, and they viewed foreign assistance for population pro- grams as a poor use of resources, given the other development needs of their countries. The successful implementation of family planning programs seemed so unlikely in most African countries that prior to the 1980s, most international agencies working in family planning chose to invest their fi- nancial and human resources in the more promising areas of Asia and Latin America. This allocation served as an additional constraint to contraceptive use: the lack of access to family planning services. Historical Factors Leading to High Fertility In virtually all societies, even those where contraception is not prac- ticed, human fertility falls considerably short of its biological maximum as a consequence of cultural practices or physical impairments that curb repro- duction. Such constraints were certainly in place in sub-Saharan Africa during the colonial and precolonial periods; prolonged breastfeeding coupled with long periods of postpartum abstinence supported a marked pattern of birth spacing. In some areas, there prevailed high levels of sterility, which further reduced fertility. Although there was considerable regional variation, the mean length of the nonsusceptible period in western Africa before the l950s following each live birth was commonly on the order of 2.5 years, due primarily to postpar- tum taboos on sexual relations (Page and Lesthaeghe, 19814. Where there was such a taboo, other practices worked in support of this prolonged absti- nence (e.g., a woman's returning to her native village at the time of the baby's birth and remaining there until the child was l to 2 years old). Similarly, the practice of polygyny (having more than one wife) provided males with alternate sexual partners during the postpartum abstinence pe- riod. In several regions where Islam was prevalent, birth spacing was more a function of lactational amenorrhea (i.e., nonsusceptibility to conception due to breastfeeding) because the accompanying period of postpartum absti- nence followed the direction of the 40-day rule, common to Islamic custom. In eastern Africa, periods of postpartum abstinence were also shorter than in western Africa, but the overall nonsusceptible period was still on the order of 1.5 to 2.0 years as a consequence of prolonged lactation (Page and Lesthaeghe, 19811. Very few African societies had no postpartum taboo at all. Historically, the basic pattern of fertility limitation hinged on these birth-spacing practices.
8 FACTORS AFFECTING CONTRACEPTIVE USE Birth spacing continues to be widely accepted and practiced in many African societies. Even today, one can go to remote villages throughout the continent and find women with little or no education who recognize the importance of birth spacing for the health of their children. In fact, women who fail to observe this practice in some societies may find themselves the target of scorn or ridicule by other members of the community (Caldwell and Caldwell, 1981~. Although birth spacing results in a delay in preg- nancy, the motivation has not been one of achieving a smaller family size. To the contrary, spacing may have arisen to enhance the probability that each child would survive through childhood and beyond. Thus, although Westerners tend to view family planning as a means of achieving a small family norm, birth spacing in Africa has been used to attain what many Africans consider the ideal: a large number of healthy children. Reports from the early 1980s indicated that the practice of postpartum abstinence was on the wane throughout Africa (Page and Lesthae`ghe, 1981~. Increased female education and urbanization have affected patterns of union in a number of countries, making prolonged periods of sexual abstinence more difficult to observe. For example, with urbanization, returning to one's village for childbirth may be logistically or economically difficult. Some women reduce the abstinence period in an effort to keep their hus- bands closer to home. However, available data suggest no further decline in the components of the nonsusceptible period during the decade of the 1980s (Lesthaeghe et al., 1992~. If postpartum practices remain fairly stable in the coming years, any decrease in fertility will necessarily result from increased contraceptive use. In many instances, women have moved from postpartum abstinence to modern contraception as a means of achieving the desired spacing without the inconvenience of abstinence. For example, Hill and Bledsoe (1992) cite the case of The Gambia, where women appear to be using Western contra- ceptive technologies to achieve a characteristically African goal: long birth intervals between children. By contrast, traditional methods of birth spac- ing and modern contraceptive use are viewed by many Africans as being two very distinct practices: The former is perceived to be highly beneficial to the health and welfare of the family; the latter has been less readily embraced. In addition to postpartum practices, pathological sterility (the inability to bear a first birth caused mainly by sexually transmitted diseases (STDs), particularly gonorrhea) resulted in diminished levels of fertility in a variety of populations from the coast of the Indian Ocean to the western Sahel. The eastern coastal zones of Kenya and Tanzania were particularly affected, together with several lake regions. Also, nomadic populations in the west- ern and central Sahel exhibited higher than normal levels of sterility. How- ever, the zone with the most severe sterility was unquestionably Central
INTRODUCTION 9 Africa, including southern Sudan, northern Zaire, the Central African Re- public, Gabon, and parts of Cameroon (Page and Coale, 1972; Frank 1983a,b). Because of these factors, fertility levels in sub-Saharan Africa in the l950s were low compared to many other developing countries. This picture started to change in the following decades as a result of declining levels of pathological sterility (in part due to the introduction of antibiotics) and especially the reduction in the durations of both breastfeeding and postpar- tum abstinence. Increased education, urbanization, and access to baby-milk formulas were typically associated with reduction of the postpartum nonsusceptible period. In many areas the resulting rise in fertility, in tandem with declin- ing mortality, yielded record rates of population growth (e.g., Kenya had an annual population growth rate of 4.0 percent in the late 1970s). Family Planning Initiatives In the 1980s, major economic and political crises occurred that dramati- cally changed the climate for population policies and family planning pro- grams. Government leaders and other policymakers in some African coun- tries became increasingly concerned that rapid population growth would have a detrimental effect on socioeconomic development. Moreover, gov- ernment officials recognized the potential health benefits of birth spacing and were willing to support family planning service delivery as a maternal and child health (MCH) intervention. In an unprecedented move, African leaders collectively endorsed family planning and the necessity of integrat- ing it into MCH programs in the 1984 Kilimanjaro Action Program (Eco- nomic Commission for Africa, 1984~. The same year they joined with leaders from developing countries around the world at the World Population Conference in Mexico to advocate increased support for family planning service delivery by donor nations. This shift in position was reflected in the results of a recent United Nations (UN) survey on population issues: 27 of the 45 governments in the region estimated their country's level of fertility to be too high, 15 judged the level to be satisfactory, and only 3 judged the level to be too low (United Nations, 1989a). A number of countries Ghana, Kenya, Nigeria, Rwanda, and Senegal currently have explicit population policies with pre- cise demographic objectives. Even governments not supportive of popula- tion policies have tended recently to turn a blind eye to private family planning initiatives. Laws in some francophone countries forbidding all publicity and distribution of contraceptive products are rarely enforced. This change in political climate was coupled with significant increases in the quantity of resources allocated to developing and strengthening fam- ily planning service delivery in Africa. The population budget dedicated to Africa by the U.S. Agency for International Development, the primary funder
10 FA CTORS AFFECTING CONTRA CEPTI VE USE of family planning programs worldwide, increased from $21.6 million in 1983 to $128 million in 1991 (U.S. Agency for International Development, Africa Bureau and Office of Population, personal communication, 19921. Over the same period, the United Nations Population Fund's support for population activities in Africa increased from $16.9 to $55 million (R. Cornelius, U.S. Agency for International Development, Office of Population, personal communication, 19934. Other bilateral, multilateral, and nongovernmental programs also contributed to this effort. This dramatic increase in inputs to family planning activities is re- flected in the Family Planning Program Effort Index, developed originally by Lapham and Mauldin (1984) and revised by Mauldin and Ross (1991~. Between 1982 and 1989, the program effort score for Africa improved from 15 to 36 (of a possible score of 120), representing the largest increase for any of the regions examined. However, even with this increase, Africa lagged well behind Asia and Latin America. By the late 1980s, the results of population-based surveys showed that family planning was gaining acceptance in at least three countries: Zimba- bwe, Botswana, and Kenya (with prevalence levels for all methods com- bined of 43, 31, and 27 percent, respectively). Although low by the stan- dards of Asia and Latin America, the prevalence data from these few countries have led population experts to reconsider their views regarding the potential acceptance of contraceptive use in sub-Saharan Africa. Potential Effect of AIDS on Contraceptive Use The potential effect of acquired immune deficiency syndrome (AIDS) must be considered in any examination of trends in contraceptive use and fertility in African populations. AIDS may affect contraceptive use in two ways. In regions where mortality from AIDS is high, couples may decide to have more children than they otherwise would in order to increase the likelihood of a certain number of surviving children. This resulting in- crease in the demand for births would lower the likelihood of couples adopting a contraceptive method to control their fertility. On the other hand, efforts to prevent the spread of AIDS may result in increased condom use. Data necessary to elucidate such effects are not yet available. Prelimi- nary findings from several African settings suggest increasing use of condoms for AIDS prevention (Plummer et al., 1988; Mony-Lobe et al., 1989; Musagara et al., 1991), a trend that may also reduce birth rates. One study found that serotesting and counseling of couples, where one partner was HIV-positive, resulted in increased condom use. Condoms were used more consistently when the man was the HIV-negative partner (Allen et al., 1992~. However, the rise in condom use has been small or limited to very specific popula
INTRODUCTION 1l lions (such as commercial sex workers and their clients), so that- effects on fertility have likely been minimal to date. In the Rakai district of southwestern Uganda, a high seroprevalence area, there has been some resistance to condom use, because it reduces fertility at a time when entire populations are worried about their long-term survival as a result of AIDS (Musagara et al., 19911. Resistance to condom distribution has been found even among trained community health workers, who see childbearing as an essential countermeasure to the perceived effect of the epidemic on the survival of the clan and the tribe. Qualitative infor- mation from Rakai indicates that older men are seeking to have sexual relations with younger women as a means of avoiding HIV infection (Serwadda et al., 1989~. Consistent with this trend is the finding that 25 percent of girls aged 13 to 14 in Rakai trading centers were found to be HIV positive in a 1989 serological survey, whereas the boys in this age group were all seronegative (Serwadda et al., 1990~. The trend of much lower HIV seroprevalence among males persists up to age 20 and for all geographic strata of the district, from the trading centers to the most rural villages (Wawer et al., 1991a). Bledsoe (1989) reports that some parents may be- come reluctant to send their daughters to school for fear of their beginning relationships with older, wealthier, married men who may be HIV positive. Instead, they may encourage their daughters to marry early, increasing their exposure to pregnancy (which may lead to higher fertility) and possibly reducing their interest in contraception. Implications of the Rise in Contraceptive Use Is Africa on the brink of a contraceptive revolution that would mark the onset of widespread fertility decline? Or is the experience of a few coun- tries unlikely to be repeated elsewhere, making them an exception to the reluctance to use modern contraceptives that has characterized much of Africa? Will the current economic crisis in many parts of Africa bring on a "crisis-driven" reconsideration of the desired number of children at the in- dividual level or simply reinforce the demand for many children that ac- companies high levels of infant and child mortality? This volume addresses these key questions by exploring the determi- nants of contraceptive use and how they operate in sub-Saharan Africa. The multiple levels of socioeconomic organization that influence fertility decision making are spelled out in the next section, and these form the basis of the chapters that follow.
12 FA CTORS AFFECTING CONTRA CEPTIVE USE LEVELS OF SOCIOECONOMIC ORGANIZATION AFFECTING CONTRACEPTIVE USE Although fertility decision making is sometimes viewed as an indi- vidual matter, it is strongly influenced by factors at several levels. As illustrated in the top box of Figure 1-1, factors that affect contraceptive use can be organized into four main categories: . national; regional; community, kinship, and household; and individual. The different factors at each level (shown in boxes in Figure 1-1) are described in the sections that follow, with an indication of the chapters in which each topic is more fully developed. In our framework, these factors influence contraceptive use (bottom box of Figure 1-1) via their effects on the demand for and supply of births (middle box of Figure 1-1~. A more thorough discussion of the links among demand, supply, and contraceptive use, based on the synthesis model of Easterlin and Crimmins (1985), can be found in Appendix A. National Level The social policy environment of a country has direct implications for reproductive practices. For example, policies regarding education and, more specifically, female education are likely to influence contraceptive use and fertility. Similarly, policies regarding age of marriage may affect the po- tential reproductive span for childbearing. A second major factor at the national level is the economic situation. The demographic literature is replete with examples of the links between low levels of socioeconomic development and sustained high fertility. It is not coincidental that industrialized countries have low birth rates, whereas much of the developing world and sub-Saharan Africa in particular has a much higher level of fertility. It has been argued that without improve- ments in standards of living, little change in fertility can be expected. However, others have argued that deteriorating economic conditions could in fact diminish the demand for children (Lesthaeghe, 1989a; Caldwell et al., 1992; Working Group on Demographic Effects of Economic and Social Reversals, 1993~. In Chapter 3 we examine this question with respect to Africa. A third significant factor at the national level is government and donor support for family planning. Over the past decade there has been a growing recognition of the importance of political will in determining the success of family planning efforts (Keller et al., 1989; Donaldson and Tsui, 1990~. Private family planning associations have played a pioneering role in pro
INTRODUCTION 13 NATIONAL REGIONAL COMMUNITY, INDIVIDUAL KINSHIP, AND HOUSEHOLD Social policy Local political and envi ronment soci al organ i zat i on HousehoLd and kinship structures and decision nuking Economi c I of rest ructure situation (schools, health) ;~nH I Scar m~ric~t~ Con j uga l bond and spousal r~cr~llr~ Government and Infant arid chi Ed donor support morta l i ty for fami ly p lane, rig Family planning program implementation Individual know l edge, attitudes, and access to fami ly planning Wider kin resources Group norms regard) ng fert i l i ty Fami ly planning servi ces Costs and | benef i ts of I i nvestments | in chi ldren I Costs of access to fami ly planning Demand for births (spacing, limiting) Potential supply of births (postpartum nonsusceptible period, union status) Contracept i ve use FIGURE 1-1 Factors affecting contraceptive use. mating family planning worldwide, but strong government support has been an important element in most developing countries with high levels of con- traceptive use. This political will is often reflected both explicitly, in the form of a national population policy with demographic goals, and implic- itly, through policies affecting female education and occupational opportu- nities, legal restrictions on the importation of contraceptives, and policies governing access to specific contraceptive methods.
4 FA CTORS AFFECTING CONTRACEPTIVE USE A fourth national-level factor is family planning program implementa- tion. For a successful program, the political will cited above cannot be limited to verbal support. Rather, it must consist of a concerted effort to diffuse this support throughout the government bureaucracy and, moreover, to harness the available infrastructure for the delivery of services. Those involved in international family planning programs argue that access to services is a major determinant of contraceptive use and that the resources needed to ensure access are controlled at the national level. These two interrelated issues of political will and program implementation are the focus of Chapter 5. Regional Level2 Countries are used as the unit of analysis in many demographic studies for obvious reasons. Yet there may be substantial variation across regions within a country. The socioeconomic conditions and social policies that are influential at the national level by no means wield uniform influence at the regional level. Differences in economic activity, standards of living, trans- portation and communications infrastructure, and ethnic composition con- tribute to regional variation. Although the situation at the national level will determine in part what occurs at the regional level, there are other factors that influence regional outcomes. Local political and social organization comes into play in a variety of distinctive ways in sub-Saharan Africa. Most African countries are charac- terized by relatively centralized systems of government. Nonetheless, much of the power at the local level often lies outside the government in more traditional patterns of social organization. Especially in matters related to fertility decision making, the village chief whose power is bestowed by locally sanctioned means may have far greater influence than government officials who may not be from the region. There are certain aspects of social organization at the regional level that may work to the benefit of family planning. Community networks may serve as channels for the dissemination of information about contraception, thereby providing a vehicle for the diffusion of fertility control (Lesthaeghe, 1989a; Watkins, 1991~. These networks, which often precede public sector service provision, operate at the grass-roots level enabling them to reach people in remote areas. This quality is especially important given research 2Many of the variables we identify as operating at the regional level can also operate at the national or community level. Our purpose in assigning variables to different levels is to point out very general levels of disaggregation of the factors affecting contraceptive use. l
INTRODUCTION 15 questioning the ability of the public sector to provide adequate services at the local level, particularly in rural areas. Differences in infrastructure at the regional level may also influence contraceptive prevalence rates. These differences are manifested in the degree of urbanization, the density of roads, and the quantity and quality of schools and health services. Urbanization and the density of roads reflect the level of economic development, which in turn affects desired family size. Access to education affects the literacy level of the population and the costs of children. Access to health services is associated with lower infant and child mortality, thereby increasing the proportion of children surviving and decreasing the number of children couples need to bear to meet their fertility goals. The nature of labor markets may also vary greatly by re- gion, depending on the type of economic activity, proximity to major lines of transportation, female participation in the labor force, and so on. Both the standard of living and the infrastructure for health and social services at the regional level are reflected in the level of child and infant mortality. Where parents fear that one or more of their children may not survive to adulthood, the response may be to have a large number of chil- dren to ensure that some will reach adulthood. The political, social, and economic organization of a region has a direct bearing on the group norms regarding fertility. The demand for a large number of children is likely to be sustained in regions where children are needed as a source of labor, where infant mortality is high, and where women's sole means of gaining status and economic security is by having a large number of children. The strong influence of religions (Islam in par- ticular) in parts of western Africa supports the widespread practice of po- lygyny and early entry into marriage, which may be both a cause and an effect of low levels of female education. Until the recent past, these diverse economic and social factors have tended to reinforce pronatalist norms. An important factor that counterbalances prevailing high fertility norms is access to family planning services at the regional level. It is generally the case that services are far more accessible in urban than in rural areas of Africa. Also, they tend to be more accessible in those regions with a higher standard of living, because of improved infrastructure for health and social services, higher levels of female education, a greater degree of westerniza- tion, and thus greater receptivity to birth limitation. Community, Kinship, and Household The factors affecting contraceptive use outlined in Figure 1-1 are by no means unique to sub-Saharan Africa. Indeed, much of the above narrative relating to national- and regional-level factors could be applied equally well to Asia or Latin America. Similarly, factors operating at the community,
16 FACTORS AFFECTING CONTRACEPTIVE USE kinship, and household level play a role in fertility decision making in societies around the world. Yet it is the nature of the familial relations and their overriding influence on fertility decision making in sub-Saharan Af- rica that is often cited as the defining characteristic of Africa's demographic experience. Chapter 4 is devoted to exploring ways in which the household, kinship, and community context of fertility in Africa is different and how its influences on fertility may be changing. To understand reproductive decisions in sub-Saharan Africa, it is im- portant to understand the relation of individual couples to broader house- hold and kinship structures. A number of studies have emphasized the interests of kin in the reproduction of an individual couple. A wife and husband may be under pressure from other relevant decision makers in the wider family who have a stake in maintaining the couple's continued fertil- ity. Furthermore, the couple may distribute some of the costs of childrearing among its kin through the mechanism of child fostering, particularly where educational costs are concerned. Some authors (notably Caldwell and Caldwell, 1987) suggest that a primary source of pronatalist pressure is from kinship structures, which have their roots in traditional African religions, as dis- cussed in Chapter 4. Similarly, the weak conjugal bond that is evident throughout most of sub-Saharan Africa has direct implications for fertility decision making. In the conventional economic-demographic view of the household, the indi- vidual couple is implicitly regarded as a well-defined decision-making unit, within which all economic resources are pooled and from which emanate well-defined reproductive demands. Three aspects of social organization in sub-Saharan Africa undermine the applicability of this model: high rates of marital dissolution, polygyny, and an unusually sharp separation of male and female budgets and childrearing responsibilities. It has been argued (see among others, Caldwell and Caldwell, 1987; Frank and McNicoll, 1987) that wives and husbands in sub-Saharan Africa hold sharply differing views regarding the costs and benefits of childbearing, with the net consequences being pronatalist. Differences in views give rise to the possibility of intrahousehold conflicts of interest and implicit bargaining strategies. Another determining factor of contraceptive use at the level of commu- nities and households involves the costs and benefits of investments in chil- dren, particularly with respect to schooling. In societies that have under- gone rapid transitions in fertility and contraceptive use, changes in the perceived benefits and costs of schooling have played an important role (see Knodel et al., 1987, 1990, for Thailand). As economists have noted, parents may think in terms of a "quantity-quality trade-off," where quantity refers to the number of children and quality to the level of human capital investment per child. The scope for a quantity-quality transition in Africa is not yet known,
INTRODUCTION 17 but some researchers (Kelley and Nobbe, 1990) argue that the trade-off will prove to be a key to future fertility decline in Kenya and elsewhere. The cost of access to family planning services is also a factor at this level. On one hand, there are financial costs for clinic visits, contraceptive supplies, transport to and from the site, child care, and so forth. On the other, there may be opportunity costs such as losing time from one's other occupations. In addition, where contraceptive use remains controversial, there may be high psychological costs: fear of rejection or reproach by one's spouse, in-laws, and friends, and self-doubt over the wisdom of the . . . aeclslon. Individual Level When all is said and done, it is the individual who must make the decision to use or not to use a means of pregnancy prevention. Three categories of factors come into play here. First, does the individual want more births and, if so, when? The personal preferences of the individual, as well as group norms regarding family size and appropriate spacing, will have a bearing on the demand for births. Second, is the individual or the individual's partner at risk of preg- nancy? Contraceptive prevalence rates estimated from the Demographic and Health Surveys are generally based on all women of reproductive age, married or living in union. Such measures are very useful for cross-na- tional comparisons, but they do not necessarily focus precisely on the popu- lations at risk of conceiving, as discussed in Appendix A. For example, women in the postpartum period who are amenorrheic or abstaining from sexual relations are at little or no risk of pregnancy. Likewise, women who are infecund as a consequence of STDs find their supply of births compro- mised and would be less likely to consider contraception. Third, does the individual know of contraception and want to use it? This issue is examined in Chapter 2. Knowledge of modern contraception is by no means universal in sub-Saharan Africa, though it is increasing quickly with the expansion of family planning programs. Attitudes toward family planning are closely linked with the demand for births. That is, contracep- tion is valued and sought after in those situations in which the individual wants to prevent a pregnancy; it is rebuked by those who believe that preg- nancy prevention is undesirable if not immoral. Collectively, these factors at the national, regional, community, house- hold, and individual levels determine rates of contraceptive use in a given country. Generally, conditions and policies at the national and regional levels have repercussions at the community, household, and individual lev
18 FACTORS AFFECTING CONTRACEPTIVE USE els, though the flow may occur in the opposite direction (e.g., government leaders may be reluctant to tale a strong stance on family planning, fearing the backlash of a strongly pronatalist population). ORGANIZATION OF REPORT In this report we begin by examining the current levels and trends in contraceptive use in Africa (Chapter 2~. From there we proceed to explore the factors that explain these levels and trends in terms of the socioeco- nomic context (Chapter 3) and of community, kinship, and household struc- ture (Chapter 4~. Together, Chapters 3 and 4 describe numerous factors that affect (either sustaining or reducing) the demand for large families. Chap- ter 5 on population policies and family planning programs indicates how improvements in the family planning supply environment affect fertility decision making. In Chapter 6 we present the results of multivariate analy- ses of the relative importance of sociodemographic and economic factors in contraceptive use at the regional level. In Chapter 7 we analyze the impor- tance of contraceptive use relative to the traditional methods of birth spac- ing in determining fertility levels in selected African countries. Finally, in Chapter 8 we present the conclusions that emerge from this examination of contraceptive use in Africa.
