Research Review NS 17.2 (2001)45-61 PUBLIC VERSUS PRIVATE RESOURCES FOR CHILDBIRTH AND CARE IN GHANA Delali Badasu Family planning has been at the centre of Ghana's population programs that have been pursued since the adoption of a national population policy in 1969. A national family planning programme launched in 1970 and relaunched in 1991 and 2001 has aimed at achieving a high prevalence of modern contraceptive use in Ghanaian society. Unfortunately the desired impact has not been achieved. In fact, modern contraception has not even contributed significantly to Ghana fertility transition that began over a decade ago. The Ghana Statistical Service has attributed fertility decline in the country partly to abortion. Based on some theoretical considerations, and secondary sources of data and information to provide the empirical evidence, this essay suggests that abortion and other alternative resources available for managing childbirth and care probably undermine efforts at promoting modern contraception in Ghanaian society. It recommends policies to influence these alternative practices, which may not only improve the prospects of increased modern contraceptive use, but also resolve some underlying health and developmental issues associated with these private means of managing childbirth and care. Background The past three to four decades have witnessed increasing recognition by many governments in developing countries that rapid population growth exacerbates socioeconomic development problems. Subsequently, many of these countries have adopted and implemented population policies and family planning programs with a main aim of curbing rapid population growth (Piotrow et al. 1997: 8-9). One of the main objectives of many family planning programs has been to encourage couples to adopt the practice of modern contraception and thereby limit their family size. Increased contraceptive prevalence rates have been recorded in the populations of a number of developing countries since the 1960s and have even been recognized as the primary cause of fertility decline in some of them (Feyisetan and Casterline 2000: 100; Westoff and Bankole 2000; Hardee et al. 1999; Bongaarts and Watkins 1996). This is particularly so in Latin American and Asian countries. In the Sub-Saharan African region, however, contraceptive prevalence rates are still low, except in a few Eastern and Southern African countries (Kenya, Botswana, and Zimbabwe), where successful family planning programs have contributed to fertility decline (Agyei-Mensah 1996). The failure or modest achievement of many family planning programs in Sub-Saharan Africa and in other developing countries has been attributed to many factors, including their narrow focus on demographic targets, poor quality of services, and inadequate geographical coverage of facilities. (Sai 1997). Following the adoption of the Programme of Action, drafted at the 1994 International Conference on Population and Development (ICPD) in Cairo, some countries have adapted their family planning programs to fit the reproductive health paradigm. The main objective has been to make modern family planning/reproductive health services more accessible, affordable and within a wider scope of cultural, social and economic contexts (Hardee et al 1999; Sai 1997). Indeed the ICPD official document, Plan of Action, recognized that there is a large number of women worldwide who have 45 need for contraception but were not using any and may do so if more accurate information and affordable services were easily available to them. The concept, unmet need for contraception, has been used to describe such categories of women whose fertility preferences show that they need contraceptives to stop or postpone childbearing but are not using any. Some studies show that an estimated 100 million married women have an unmet need for contraception in the developing world alone and the level of unmet needs is highest in Sub- Saharan Africa (Govindasamy and Boadi 2000:1; Feyisetan and Casterline 2000). The unmet need, measured in terms of discrepancies between fertility preferences and contraceptive behaviour, has therefore been the main index of the degree of failure of family planning programmes to reach their intended populations and apparent reason for continuing some of the programmes. In Ghana, for example, high rates of unmet need for contraception exist among women at all ages. The level of total unmet need was 35% among married women according to the 1998 GDHS. The same survey report showed that as low as 22% of currently married women reported that they were using any contraceptive method. Only 13% were using any modern method, a proportion that is far below a level of 25% suggested by Caldwell et al. (1992) as one of the preconditions for fertility decline in African populations. Commenting on this paradox, the Ghana Statistical Service (GSS) (1998: iii) has noted that, considering the low prevalence rate of modern contraception, "...couples are using means (including induced abortion) other than family planning methods to reduce fertility." Anarfi (2002) has also suggested that the role of herbs in the fertility decline needs to be investigated, given the high prevalence of their use among Ghanaian women, even though the potency of some as contraceptives is not scientifically established. These observations may not be speculations. Data on induced abortion indicates that though abortion is illegal in many countries, its incidence has been increasing for a number of reasons including the desire fbr smaller families and increase in premarital sexual activity (Coeytaux 1993 cited by Mundigo and Indriso 1999: 24). Furthermore, there is no country where abortions do not occur although the frequency and distribution vary. Estimates for the developing world put an overall figure of around 30 million induced abortions annually, and accounting for up to one-third of births averted by either contraceptives or abortion (Mundigo and Indriso 1999: 23). In Ghana, abortion though illegal and allowed only under some circumstances is observed to be widely practised clandestinely and accounts for a high proportion of maternal deaths. Against this background, the present paper tries to explore a number of secondary sources of data to establish that the practice of abortion and other means for managing childbirth and care probably undermine efforts directed at increasing the contraceptive prevalence rate in Ghanaian society. It argues that, quite apart from the problems associated with implementing Ghana's Family Planning Program, these alternatives might have also contributed to low contraceptive prevalence in the population and must therefore be investigated. It refers to all these other means or methods as "private" resources for managing childbirth and care while referring to modern contraception (Government's preferred method for limiting birth) as "public". Some Theoretical Considerations The design and implementation of family planning programs have been generally informed by theories of the determinants of fertility decline or transition. Though there has been considerable disagreement among demographers about the determinants of fertility decline, generally it has been viewed as a consequence of reduced demand for live births (Feyisetan and Casterline 2000: 100). Family planning programs therefore seek to influence the reproductive behaviour of couples or individuals by providing knowledge about and access to contraceptives or the innovation of family 46 Research Review NS 17.2 (2001) 45-61 limitation practices or technology and establishing the social legitimacy of contraceptive use (Feyisetan and Casterline 2000:100). Changes in reproductive behaviour result from change in individuals' or couples' fertility preferences. What brings about the change in fertility preferences, the desire to postpone or avoid pregnancy altogether, is what has been debated by demographers and modelled in many frameworks or theories resulting in what Mason (1997: 443) describes as "too many formal theories, none of which seems wholly satisfactory." Even more relevant to the present study is Mason's (1997: 445-449) assertion that despite the numerous theories on fertility transition that demographers can boast of there are some errors in their thinking about fertility transitions due to some of the assumptions they make about the phenomenon. One of them is the assumption that the regulation of fertility is fundamentally different in pre- transitional and post-transitional populations, that in the former, decision about fertility is wholly governed by social constraints or cultural norms rather than by the individual. She refers to the use of the concept 'natural' fertility by demographers to describe fertility in pre-transitional populations. Associated with this is demographers' interpretation of non-numerical and seemingly fatalistic responses such as 'it is up to God' as an absence of individual rational choice of desired number of children. Refuting this notion, Mason (1997:447) pointed to the, "new consensus among anthropologists regarding culture as a set of rules individuals are ceaselessly reinterpreting and renegotiating," which may result in cultural change, though the gatekeepers of powerful institutions such as the church may have a disproportionate say in determining social norms. Violations of cultural norms sometimes show how individuals 'interpret' the rules of behaviour to suit their needs. She further emphasized that, given the importance of survival and well-being of children in traditional societies (where infant and child morbidity and mortality are high) it would be surprising that parents in pr&-transitional populations did not think, plan and strategize about number of children. Stressing again the weakness of the natural fertility concept with respect to early stages of fertility decline, Mason also pointed out that it is a failure to recognize the possibility of multiphasic response to "excess numbers of surviving children," as noted by Davis (1963). Elaborating on Davis' idea, Mason (1997:446) explained that: When families begin to experience the economic stress brought on by survival of too many children, their first response need not be to limit the number of births. Rather, in many historical situations, Davis suggests, the easiest way to cope with increased child survival may be to send children to work as servants in other households or to have them migrate to urban areas or overseas destinations (both of these having been common in the history of Western Europe) ... alternatives to family limitation vary by historical and cultural context... In settings with few alternatives to birth limitation, fertility control may begin while mortality levels are relatively high... In settings with many alternatives, fertility control may be postponed until the alternatives are exhausted and mortality has fallen to low levels. Wrigley (1978:148) shares similar views when he states that: When the demographic transition occurred it did not take the form of a move from a situation in which fertility was uncontrolled to one in which it was reduced by the exercise of prudent restraint. Fertility is under constraint in almost all societies. The key change was from a system of control through social institution and custom to one in which the private choice of individual couples played a major part in governing the fertility rate. 47 In a study on behavioural and biological determinants of fertility transition in the Sub-Saharan African region, Frank and Bongaarts (1991) also stated that eventual adoption of modern contraceptives in the region will be preceded by abandoning of traditional birth control practices. The implication: some methods of birth control were being practiced before the introduction of modern methods. Mason (1997: 446) has cited a number of ethnographic, historical, and demographic studies which provide additional evidence that people in different historical and cultural contexts did think about and also controlled offspring configuration such as sex composition, though they often did so postnatally rather than before birth and the practices did not result in fertility decline. She noted a variety of birth control methods (both pre- and post-natal) in the western world that have been documented: infanticide, abandonment of children, loaning of children to other families or giving them for adoption or fostering, marrying them off at a young age and sending them overseas as migrants. Similar practices in Africa have been documented in ethnographies on African societies (Oppong 1974, Fortes 1949, Raum 1940). Anthropological, sociological and other studies indicate that many traditional societies in Ghana also held strongly to taboos about terminal and postpartum abstinence and associated long periods of breastfeeding that aimed at enhancing the health of the mother and child and also control the timing of birth (Amankwaa et al. 2001; Oppong 2001; Oppong and Bleek 1982). Abortion has also been practiced in many traditional societies to limit or space births (Anarfi 1996; Ampofo 1970). Traditional practices of fostering as an institution and now becoming a response to crisis in the family or irresponsibility on the part of fathers especially to raise or care for their children (Oppong and Bleek 1982) provide a means by which births are managed. It is these and other alternative practices apart from modern contraception in Ghana and their possible effects on modern cotraception that the presnt paper is concerned about. Before we examine these alternative practices we look at the status of family planning in the country. Family Planning In Ghana In 1969, Ghana adopted a comprehensive national population policy with the main aim of curbing the rapid growth rate of her population, which was 2.6% at that time. Though the policy document encompassed all areas of human and social behaviour which directly or indirectly affect the demographic process of the country (National Population Council (NPC) 1999:2), considerable attention has been given to the growth rate of the population. A national family planning program was launched in 1970 with the principal aim of providing couples with information on and access to modern contraception, which has been government's preferred means of birth control in the country. It was relaunched in 1991, redesigned under the revision of the National Population Policy in 1994, shifted to the reproductive paradigm following the ICPD recommendations and relaunched again in 2001, as it could not achieve its objectives (NPC 1999:1, PIP 1995:2). The population growth rate, which was around 2.6% per annum at the time of adoption of the policy and was targeted to reduce to 1.7% by the year 2000, rather increased to around 3% since the 1980s. The 2000 Ghana National Population and Housing Census has put the growth rate at 2.6%, showing some decline in the inter- censal growth rate. About a decade ago when fertility levels began to decline significantly in the population level of contraceptive prevalence rate was low and has not still increased appreciably since then. Contraceptive Knowledge In Ghana creation of awareness of contraception is perhaps the greatest achievement of the Family Planning Program. Knowledge of modem family planning methods increased at each of the 48 Research Review NS 17.2(2001)45-61 three successive GDHS conducted in 1988, 1993 and 1998. The highest increase occurred between 1988 and 1993 when the proportion of currently married women who knew any modern method increased from 79% at the 1988 GDHS to 91% at the 1993 survey. It then rose rather slowly to 94% at the 1998 survey. The rapid increase between 1993 and 1998 has been attributed to gains made by relaunching of the program in 1991 (Ministry of Health 1992). At the 1998 GDHS, 92.5% of all women and 93.1% of currently married women indicated that they know1 at least one modern contraceptive method. The proportion of all men who said so was 94.6% and the corresponding figure for currently married men was 96.0%. Knowledge about any traditional method is also quite high, 73.2% among currently married women and 79.4% among their male counterparts. Lower percentages of all women (69.4%) and all men (69.5%) know about at least one traditional method. Knowledge about a source or a place where family planning services are available was also high. Eighty-one percent of currently married women and 86% of currently married men knew where to obtain a modern family planning method as recorded at the 1998 GDHS. This implies that phisical access to family planning services is probably not a barrier to contraceptive use for the most married men and women. Contraceptive Use The surveys, however, recorded wide gaps between knowledge and use of contraceptives. While over 90% of currently married women stated that they know of at least one modern method, only 50.8% said they have ever used any contraceptive method, and 37.8% indicated that they have ever used a modern method. There were 30.6% who have ever used a traditional method. The current figures on contraceptive use show high rates of discontinuation. Among married women, the proportion who use any family planning method rose sharply from 13% at the 1988 GDHS to 20% at the 1993 survey (also as a result of the relaunch in 1991) and slowly to 22% at the 1998 survey. The proportions using any modern method at the three consecutive surveys were 5% (1988), 10% (1993) and 13% (1998). The corresponding figures for the traditional methods also were 8%, 10% and 9%. A comparable proportion of married women use traditional methods of family planning methods. Contraceptive use has not been uniform among all subgroups of the population. Overall, the contraceptive prevalence rate is higher among currently married women than all women together. This must be expected, since results of the surveys and other researches have shown that there are lower rates of contraceptive use among young girls and boys, though they may be sexually active. Unmarried young people have limited access to family planning services. Society generally disapproves of contraceptive use by young people because premarital sex is frowned upon in Ghanaian society. For such categories of individuals, alternative sources of birth control such as abortion have been important for doing away with unwanted pregnancy (Population Impact Project (PIP) 1995). See also PIP (1997) for a collection of several works on adolescent sexual and reproductive behaviour in Ghana. But there has been increasing need for contraception among young people due to a rising (though slowly) age at first marriage but a relatively stable age at first intercourse. The medium age at first marriage has risen slowly over the last two decades from 18.7 years for women aged 40-49 years to 19.6 years for women aged 25-29 years, according to the 1998 GDHS results. The median age at first sexual intercourse has not changed much during the same period. It rose insignificantly from 17.5 years for women aged 40 years and over to 18 years for women aged 25-29 years. Men assume sexual activity later and also marry later than women. The survey results show that only around one in four men were married by age 22, compared with three in four women. Men's median age at first 49 sexual intercourse for ages 25-29 is 19.4 while for those aged 30-34 years, the median age is 24.8 years. Sexual activity precedes marriage for them also, and low contraceptive use rates imply unwanted or mistimed pregnancies which have to be terminated through abortion or managed by some postnatal means. At the 1998 GDHS, more urban women (17%) than rural women (11%) were using a modern family planning method. There were differentials in contraceptive use by education, with the highest rates of use of modern methods among women who have education up to secondary level and beyond (20%) compared with 9% among those with no education. The difference between all men (20%) and all women (13%) is also significant. It has been speculated that most men use condoms, not with their spouses but with other partners. If men were using condoms with their wives it could reduce the rate of unmet need for contraception among married women and unwanted pregnancy too. Unmet Needs for Contraception Analysis of fertility preferences and contraceptive practices from the various GDHS indicate that there are many women who either say that they do not want any more children or want to delay the next birth but are not using contraceptives or need more effective contraceptives—women with unmet need for contraception. The level of total unmet need for contraception was 39% (25% for spacing and 13% for limiting) at the 1993 GDHS and 23% (11% for spacing and 12% for limiting) at the 1998 survey. The level of unmet need is still high despite the decline. The differences between Wanted (3.7) and Total Fertility Rate (4.6) at the 1998 survey also indicate that not all demand for contraception has been met (Ghana Statistical Service 1999: 74, 1994:74).Total demand for family planning in 1998, made up of the proportion of currently married women with unmet need for family planning (23%) and the proportion currently using family planning (22%) was 45%. If all women with unmet need for contraception were also using some method, the contraceptive prevalence rate would have been 45%. The reasons given for non-use are varied as indicated at the surveys, but health concerns and fear of side-effects are the most often stated. At the 1998 survey, fear of side effects was the most frequently cited reason, accounting for 16% compared with far lower rates of response for other method related reasons: lack of access/availability (0.5%), cost (2.2%) and inconvenience of use (4.0%). Husband's opposition (5.2%) and religious opposition (3.1%) were not significant reasons for non-use at the 1998 survey. Earlier, these were prominent reasons for limited contraceptive use among women. The introduction of strategies in the 1980s to involve men in family planning was a response to the high rate of spousal opposition reported at earlier surveys (Piotrow et al 1997, Ministry of Health 1992). The involvement of religious bodies in the National Family Planning Programme is also aimed at removing religious barriers to contraception, especially modern contraception. Alternative Means of Childbirth and Care Management The recent fertility decline in the population, from a Total Fertility Rate of 6.4 in 1988 to 4.6 in 1998, is a sign that couples are deciding to have fewer children than before. The contraceptive prevalence rate in the population has however been lower than expected. We therefore examine the practice of abortion which has been suggested to be an alternative means of birth control in Ghana. Induced Abortion In Ghana Abortion is a sensitive issue in Ghanaian society, and it is generally deplored. The law on abortion was highly restrictive until 1985 when it was relaxed. Earlier, "the Abortion Law in Ghana was derived from the United Kingdom Statute of Offences Act of 1861" (Adanu et al. 2001: 4). 50 Research Review NS 17.2 (2001) 45-61 Since 1985, following the amendment of the criminal code, abortion is not an offence when performed in the following circumstances: when it is performed by a medical practitioner specializing in gynaecology or a registered practitioner in a government hospital or registered private hospital or clinic or when the pregnancy is a result of rape, incest, or defilement of a female idiot. Abortion is also allowed when the health of the pregnant woman—physical or mental—is risked by the continuation of the pregnancy or if the baby would suffer from or later develop serious physical abnormality or disease (Ahiadeke 2001: 96). The relaxation of the law on abortion has not brought about much change in the availability of the service. Women seeking abortion often obtain it from a number of unqualified health practitioners: traditional healers, quack doctors, qualified nurses who use unhygienic methods, including insertion of foreign bodies into the uterus, or herbs, brand medicine and many other "crude" methods (Anarfi 1996, Oppong and Bleek 1982, Ampofo 1970). Complications, sometimes resulting in death, are associated with these clandestine abortions. They may or may not be reported for treatment at hospitals. Clandestine abortion may or may not be included in hospital abortion statistics depending on the procedures involved and ease of access to a hospital or obtaining medical care on the part of the person who has undergone the abortion. It is, therefore, difficult to guess or estimate the magnitude of abortion in the whole country. Low rates of contraception among people of reproductive age and high rates of unwanted fertility suggest that abortion rates might be higher than the estimates available. At the 1998 GDHS, data on pregnancy outcome was collected at a national level for the first time. The report on the survey indicates that 9% of all births during the five-year period before the survey were not wanted, while 28% were mistimed (that is, wanted later). The same survey results also show that 10 years before the survey, 12% of all pregnancies were lost as early pregnancy loss2 (9.7%) or stillbirths3 (2.0%) or terminated through spontaneous or induced abortions (Ghana Statistical Service/Macro International Inc. 1999:30). The age and residence characteristics of the pregnancy outcomes show that generally younger women aged below 25 years and older women aged 44 years and above and urban women are more likely to have pregnancies resulting in non-live births. For the age group 15 to 19 years, for example, nearly one in four pregnancies was lost every year during the ten-year period before the survey. Apart from the GDHS, which is a national survey and quite recent, a number of studies have been done on the subject of abortion which give some record on the practice over the past four or more decades (see Anarfi 1996 for a discussion of several of such works). A few of them, which examine a number of aspects of the practice are cited here. Ampofo (1970) carried out a study on abortion at the Korle Bu Teaching Hospital (Ghana's biggest referral hospital) and examined the characteristics of the women who had induced abortion and were admitted to the Korle-Bu Teaching Hospital. The results of his investigation showed that the incidence of induced abortion was higher among younger women (15-24 year group) who accounted for 75% of the 116 induced abortion cases admitted to the hospital. Out of the total number, 60% were nulliparous. The majority of the cases were young girls attending colleges (40%) or higher institutions (20%). The remainder had had primary education or were in primary school. Analysis of the marital status of the cases investigated showed that almost half (48%) of the total number of the cases were single girls. Housewives formed 34% while the rest (18%) were living with their male partners but did not regard them as their husbands. Knowledge about contraceptives was low among the women. Only 15 out of the 116 gave the impression that they had adequate knowledge about contraception but were however not adequately motivated to use contraceptives. 51 Other characteristics investigated included methods, the cost of procuring abortion, partner's knowledge about seeking abortion, and where abortion was performed. The methods included oral medication, herbal pessary and intrauterine instrumentation. The study noted that there was an increasing trend towards using intrauterine instrumentation, which would lead to increased incidence of complications due to sepsis, tetanus and perforation of the uterus in future. The cost of abortion ranged between 6-10 new cedis, and in some cases, 20-40 new cedis. Forty percent of partners of the women did not know about the women's decision to seek abortion. As many as 58% help the women to secure the abortion, through self-induction at home or at a clinic, forced mainly through information from a friend. Abortions were performed in the provider's home/clinic or in the woman's own house. A very recent study at the Korle Bu Teaching Hospital by Adanu et al. (2001) investigated the reasons for undergoing induced abortion using a case study approach.4 The study also compared spontaneous and induced abortion cases. The relationship with sexual partner or spouse and financial support dominated the reasons for seeking abortion. The results of the study showed very few of those who had induced abortion were married. Twenty-six percent described their relationship as unsteady. While 86.4% of spontaneous abortion cases indicated that their partners were a significant source of financial support for them, only 21.2% of induced abortion cases said so. The majority of the induced abortion cases were nulliparous. While only 27.5% of those who had spontaneous abortions had parity of zero, the corresponding figure for those with induced abortion was 61%. The study indicated also that most of the induced cases had formal education beyond the first cycle. Among both groups, contraceptive knowledge was high, 81,0%, but only 31.3% said they had ever used a modern contraceptive method. These figures are however higher than those observed in Ampofo's sample. Like Ampofo's study, Adanu et al.'s also showed that costs of procuring abortion varied between c60,GOO and 0400,000. Herbal preparations cost less than other methods. The main sources of information on where to seek abortion were friends, partners, relatives, co-workers, local drug peddlars, and own knowledge. In another recent study on abortion among adolescents in Ga Mashie, Accra Yoder and Henry() also found out that the main reasons given for abortion were financial. Collectively, the 29 girls aged 15-24 years who participated in the study had experienced 64 pregnancies, half of which were aborted and half resulted in live births. A study by Oppong and Bleek (1982) among the Kwahus in the 1970s on fertility decision- making in the context of profound socioeconomic change and differentiation reveal the importance of familial role systems (particularly uncertain and changing context of lineage support and marital instability) on the one hand and occupational opportunities, townward migration and upward mobility on the other fertility behaviour. Early childbearing is a hindance in such context. They observed that "some individuals will terminate a pregnancy at all cost, even risking life and health to do so" to take advantage of economic opportunities, though others have a fatalistic or casual approach to their fertility behaviour and outcome (Oppong and Bleek 1982:15). Another work on abortion in Ghana by Ahiadeke (2001) is a multistage random sampling survey on 18,301 ¥/omen aged 15-49 years screened for pregnancy in eight communities in four of the ten regions of Ghana.5 The rate of abortion in the study areas was 17 induced abortions per 1000 women of childbearing age, 19 abortions per 100 pregnancies or 27 abortions for every 100 live births. This study also reveals that the majority (60%) of the women who had an abortion were younger than 30, and 36% were nulliparous. Most of the women in this study were married and indicated that they did not want to have any more children at the time they became pregnant. Only 11 of the 18,301 women studied said they used contraceptives prior to their current pregnancy. This reflects the extremely low contraceptive use among married women in the sample and also explains the indication of previous 52 Research Review NS 17.2(2001)45-61 abortion among the respondents. The main reasons for seeking abortion include financial constraints (also the main reason for previous abortion), unplanned pregnancy and uncertainty about keeping the pregnancy. Only 12% of the women said they had obtained their abortion from a physician. Anarfi (1996) has also examined another aspect of abortion in the country recently: the role of local herbs in the recent fertility decline in Ghana. He listed seventy-nine methods and substances, including herbs, that have been used in the Ghanaian society as identified in a number of works done on the subject. Herbs, according to Bleek (1986, cited by Anarfi 1996), constitute the largest group of the abortifacients. In a later study, Anarfi (2002) concluded that obviously the enterprise has existed for a long time and may be more familiar for some people than modern family planning methods. But modern contraceptives may be safer6 than abortion, especially the clandestine type which is often practiced in Ghana. In an earlier study, Anarfi (1996) has also observed that the current method mix of contraceptives casts doubt on their adequacy in accounting for the fertility decline in the country. He discusses, on the other hand, the possible role of local herbs in birth control in Ghana and noted: In conclusion, it has been established that a large number of known medicinal herbs are being used for birth control in Ghana. There is overwhelming evidence that most of the herbs act as abortifacients although this conclusion still remains speculative. In collaboration with the GPTHA (Ghana Psychic and Traditional Healers* Association), known herbal medicines should be scientifically analyzed, not only to determine their efficacy, but also their toxic content so as to advise users accordingly. If this is carefully done, local herbs stand the chance of filling in cheaply the unmet needs for contraceptives in Ghana and hopefully, much of Africa, in the medium term. Whether all Ghanaians will prefer abortion to modem contraception is debatable as the practice is supposed to be unacceptable, at least openly. The evidence from the few studies examined here however suggest that abortion is not uncommon in the country. Many women who have abortions have had a previous history of abortion. Financial factors dominate reasons for seeking abortion, and for the unmarried and young adults issues of partner relationships and the need to continue one's education account for induced abortion as a means of getting rid of unwanted or mistimed pregnancy. The practice is quite common, though shrouded in secrecy. Both the unmarried and married and both young and old practice it, though the young and unmarried are more likely to have an abortion than the older and married women. In the majority of the cases studied, abortion is done with the knowledge and help of partners. The clandestine nature of most abortions in the country makes it risky. For that reason, it is not be a preferred choice for some women or couples, who may have to resort to postnatal means of care for their unwanted births. Results of the 1998 GDHS showed that 9% of births were not wanted while 28% were mistimed or wanted later (Ghana Statistical Service 199:79). This means that some couples who may be uncertain about keeping their pregnancy have had to make decisions postnatally to care for these unwanted or mistimed births. In the next section, we examine some postnatal alternative means for managing childbirth and care. Postnatal Fertility Regulation As discussed earlier, contrary to the notion that traditional society was helpless in the control of offspring, anthropological studies provide evidence that couples did think about their offspring 53 configurations and also did something about them. Some of these practices have survived to date and are even reinforced by some modernization processes. The process of modernization has been associated with rising costs of raising children. Child- rearing costs, such as costs of education and declining benefits from children due to their reduced contribution to family labour have emerged as contexts within which couples have had to examine their decisions on childbirth and care (Makinwa-Adebusoye 1994, Oppong 1987, Caldwell 1977). Parental aspirations for children's schooling and perceived education costs under economic hardships influence parents' fertility decisions. This is particularly true of urban areas. Among the poor and low-income communities, where economic transformations are not significant and levels of living are low, fertility levels are still high. Fertility decline in Ghana started among the urban elite and is now spreading to those in the lower socioeconomic strata within the population (Agyei- Mensah 1997, Caldwell 1977). In rural areas also, where low levels of living abound in the agrarian communities and the cost of raising children is relatively low due to kin support and lack of enforcement of laws on children's education, large family sizes are still maintained (Addai-Sundiata 1997). Here children are still an important source of family labour. In urban areas too children offer many services to their parents and also partake in many economic activities, especially in the informal sector. Child Labour Child labour attracting some remuneration is prevalent today in both rural and urban areas. Most child labour is concentrated in the rural areas. In urban areas children engage mainly in trading activities and paid domestic work and/or baby-sitting and many types of employment in the urban informal sector. Those in rural areas work predominantly in the agricultural sector and undertake some trading activities. In the agricultural sector commercial production has not been accompanied by mechanization. Casual labour needed during peak periods, such as weeding and harvesting, is provided by children whom employers may prefer to adults as they pay them less. Though some popular views seem to suggest that the issue of child labour is a recent one and may be attributed to increasing poverty and child neglect, there is ample documentation of its occurrence as far back as the colonial period. Hear (1982), for example, has traced back the emergence of child agricultural labour to the introduction of cocoa and commercial agriculture in Ghana, in a work titled "Child Labour and Development of Capitalist Agriculture in Ghana". He discusses the "incorporation of children into the cocoa economy during the colonial period" and "employment of children in capitalist agriculture that developed in the north of Ghana", specifically rice production in the Dagomba District around Tamale and in the Fumbisi Valley further north, from the 1960s (Hear 1982: 498). Labour legislation, according to Hear, was ineffective in stemming the flow of child labour that was predominantly from the northern parts of the country to the south. Children cut out school attendance in order to work on the rice fields even though the Nkrumah Government had then inaugurated a campaign of mass education in the country. The various governments that came into power after Nkrumah's regime also encouraged mass education as part of general socioeconomic development programme. The Free Compulsory Universal Basic Education (FCUBE) programme being implemented in the country since the early 1990s has not yet been able to keep all children of school-going age at school and away from economic activities. A third of children in the rural areas who are of primary school-going age are not enrolled (Ghana Statistical Service 2000: 9). The proportion in urban areas that are not in school is around 20%. In both rural and urban areas a significant proportion of children of school-going age who are not in school are engaged in economic activities. Even those in school also engage in some 54 Research Review NS 17.2 (2001) 45-61 economic activity outside school hours. Children work to maintain themselves or supplement family income. The political will to enforce child labour laws under the United Nations Convention on the Rights of the Child to which Ghana is a signatory and the Children's Act 560 of the 1992 Constitution of Ghana has been weak. This is due to the socioeconomic context within which child labour is thriving. Increasing poverty resulting from unemployment and the inadequate income of many families compel parents to send their children to work to supplement the family income. In some cases, the children themselves decide to engage in economic activity due to unfavourable conditions at home (UNICEF 2000, Yeboah 1998, Apt and Grieco 1995). A UNICEF report, a Situational Analysis of Children & Women in Ghana (2000), also indicates that the incidence of child labour is higher in rural areas, fishing settlements, and areas where animal rearing is a main economic activity. It cited a number of studies on child labour, for example, one by Fentiman et al. (1999 cited by UNICEF 2000) which shows that there is a very high rate of child labour in the fishing villages along the Volta Lake in the Afram Plains. It referred also to a survey by the Afram Plains Development Organization (APDO) which also revealed that 50% of children are out of school and engaged in fishing, farming, shepherding of cattle and other types of activities. A report by the National Commission on Children (NCC) which is also contained in the UNICEF report indicates that an estimated half of all school-going age children in Elmina do not attend school regularly because they engage in a lucrative fishing activity that fetches them about 020,000 a day compared with the national minimum daily wage of 05,000 (UNICEF 2000: 123, 125, 127). With respect to regional differences in child labour participation rates the northern regions have higher rates than the southern ones because of their lower school enrollment rates and higher drop- out rates, greater need for child labour for farming activities and higher incidence of poverty (UNICEF 2000). In these predominantly agricultural areas, where 8 or 9 out of every 10 persons are poor, (Ghana Statistical Service 2001), children's contribution to family income has been very important. Increasing opportunities for women to work in the formal sector without corresponding adequate facilities for childcare and familial tasks have also created the need for domestic servants/house- helps. Most of these are young girls who have little or no education. A study by Yeboah (1998) on house-helps in Sekondi-Takoradi Metropolis found out that there were house-helps as young as nine years employed by working mothers to take care of their children, do domestic chores, and in some cases, trade for them. In their study of 112 randomly selected street girls in Accra, Apt and Grieco (1995:41) established that most of them are Northerners (75%), especially Dagomba, and have come to Accra to work for money. Many of them come from large sized families. The streets of Accra and other big towns in particular are often full of child vendors throughout the day. One does not need a survey to conclude that children form a conspicuous proportion of the labour force engaged in trading. A number of cases reported in the media also portray a picture of children engaged in certain activities to earn income, not only to cater for themselves, but even for their families.7 It may not be the best choice for some parents to send their children to the streets to beg for alms but the availability of such means of making a living and other forms of work that attract children and has renumeration attached may not augur well for the adoption by some parents of means to limit their family size. Some may even use child labour as a survival strategy to maintain large family size under conditions of poverty. Many cases of child trafficking have stories about parental consent. 55 Early Marriage In traditional Ghanaian societies, girls marry soon after they attain puberty and undergo the puberty rites. With the introduction of western education, girls who attain high levels of education marry at older ages. According to the 1998 GDHS, the proportion of women who are married by age 15 has declined from 11% among women aged 40-44 years to 4% among those aged 15-19 years at the time of the survey. The decline in the percentage of first births occurring before age 18 from 30% in the cohort age 45-49 to 20% for the cohort age 20-24 is another evidence of decline in early marriage and consequent childbearing8 (Ghana Statistical Service 1999: 36). Despite the rise in age at first marriage, a significant proportion of girls give birth to children outside marriage and consequently enter motherhood early. The 1998 GDHS indicates that the percentage of teenage girls who have begun childbearing increases with age from 2% for those aged 15 years to 32% for those aged 19 (Ghana Statistical Service 1999: 36). Analyses of the data with respect to educational status and residence reveal the following. Adolescents with little or no education are about seven times more likely to have begun childbearing earlier than those with education up to secondary level or beyond. Adolescent fertility is more prevalent in rural than urban areas. This is due mainly to low rates of schooling and early marriage among rural girls. It can be concluded from the foregoing analyses that educational cost and other expenses on children may force urban dwellers and educated couples to limit the number of children they desire to have. In rural areas on the other hand, the lower cost of raising children and early marriage among girls may prevent family planning programs from motivating couples to reduce their family size. When children marry early they reduce the span of years over which their parents must care for them. Child Fostering Evidence from anthropological studies, sociological studies and some recent demographic surveys indicate that fostering is still practised widely in West Africa. Children are given out or exchanged among families—to be raised and cared for by foster parents and not their biological parents. The maintaining of kinship ties was once the main reason for fostering in such societies, though some children were fostered by strangers or friends (Appiah 2001:8-9, Page 19891 Isiugo- Abanihe 1985; Goody; 1975 Oppong; 1974; Oppong and Abu 1987). The practice of child fostering now probably acts as a safety net for families that may have more children than they can cater for. In a study on childbearing and childrearing in Sub-Saharan Africa, Page (1989) noted that there seems to be higher incidence of fostering in West Africa than in other parts of Africa, that more foster children leave home between age four and ten and may stay with foster parents for a few months or until marriage. His study also indicates that fostering among children under 15 years was as high as over 20% in some African countries, including Ghana. The 1998 GDHS results show that the proportion of children under 15 years living with both parents was 49.1%, showing a slight decline from the 1993 level of 49.9%. Those not living with both parents or either of them formed about 15% (Ghana Statistical Service 1999:10, 1994:13). A wide range of specific reasons or functions exists within various contexts or cultures for the practice institutionalized fostering. Goody (1975) suggests two broad classifications, the institutionalized, purposive type and those arising out of needs, for example divorce and death of parent(s) (as crisis fostering). The functions that they both play may be similar, easing the burden of families in need. Among the southern Ewe of Ghana, for example, fostering was done to help children from poor families and to give relief to less fortunate families (Appiah 2001, citing Fiawoo 1978). Generally however, the practice was primarily to maintain kinship ties or train young people in some trade. 56 Research Review NS 17.2 (2001) 45-61 In recent times, increasing poverty has been found to have rendered the practice "a prevalent means of easing the burden of parents who cannot cope with childcare responsibly" as a result of large family size or economic constraints, as observed by Ardayfio-Schandorf and Amissah (1996). In a study, characteristics of fostered children, foster parents and some aspects of the practice in three areas of Accra where the study was carried out, Ardayfio-Schandorf and Amissahs (1996) found out that high fertility and economic constraints on families to provide for their offspring account largely for fostering among the low income families in the study areas. The high income families on the other hand have their children fostered mainly for social reasons, such travel by parents and the desire to provide the child worth access to better educational facilities. Thus the study also observed that a larger percentage of children from low income families (10%) had four or more biological siblings, in contrast with those from high income families where none of the fostered children had more than three siblings. It is therefore not surprising that almost all low income families studied were found to foster only one additional child while almost half of the high income families fostered two or three children. Commenting on these findings Ardayfio-Schandorf and Amissah (1996:189) observed that it seemed to be an interesting trend which provoked the question of the part small families may be playing in encouraging high fertility among others. In answering this question one may say that the small and high income families, according to this study, are essentially offering opportunities to children from low income families. Their acts of kin obligation or friendship (where foster children belong to friends) may, however, indirectly reduce motivation for contraception among the parents of fostered children to limit their family size. We must also consider that fostering in another sense does not imply residence with the person providing assistance. Flow of resources from persons who are better off to their relatives and their children is a common practice in Ghanaian societies. Thus dependants of a person may be with their biological parents while they are provided for by a kin or another person as indicated in several studies. Oppong's(1981) study on middle class migrant matrilineal Akans in Accra indicates brothers assist their sister's children as an obligation. In another study among 398 teachers of different ethnic groups, Oppong (1987) found out that half of them had responsibility mainly for their sister's children. Associated with both types of fostering is what Oppong and Bleek (1982) referred to as the behaviour of "free-riders". These are biological parent(s) usually fathers who fail to take up responsibility for the care of their offspring. Instead, a relative or a kin (usually the grandparents) foster the child. This is due mainly to changing cultural values and inability of society to enforce sanctions against the shirking of such responsibility. Fostering and other means exist for parents to care for their children who have more children than they can care for. With the availability of such means the motivation to use family planning methods may not be strong enough for some couples. In a period of agricultural adversity, fertility among the Kassena-Nankana of Northern Ghana was found to have dropped relative to other years as a result of increased desire to space or limit births (Appiah-Yeboah et al., 2001). This is evidence that couples may begin to adopt family planning practices under certain circumstances. An earlier study by Oppong (1987: 174) among some 398 teachers of different ethnic backgrounds (a national sample) indicated that the respondents who said they were contracepting were those who have "greater propensity to save and yet perceive that what they had saved was inadequate. They also demonstrated anticipations of the strains of having children due to increase in flows of resources from parents to children (costs of parenting) and dwindling kin support and perceived idea that parental cost and tasks should be individualized. Migration 57 One of the main objectives for the implementation of Ghana's population and family planning policy has been to maintain a balance between population and resources available to meet the needs of the nation. Hence the considerable attention paid to birth and the population growth rates of the country vis-a-vis the pace of socioeconomic development. However, in those parts of the country (the Upper East Region, for example) where the population - resource relationship argument might find support, migration has to some extent challenged its validity for at least the individual or the family. The northern regions have had population pressure on land as well as very low levels of socio-economic development. It might be expected that modern family planning would be embraced there. On the contrary, migration has been a mechanism by which individuals and families have responded to the problem. In a study on population and land degradation in the Upper East Region, Nabila (1997: 81) noted: The level of out-migration is quite high in almost all the districts ... Population pressure on the environment resulting in lack of land and other rural opportunities, in the form of development projects, continue to be a major reason for the out-migration of a majority of the people resulting to some extent in depopulation of certain households and localities. Out-migration in the northern regions dates back to the colonial period when it was a deliberate colonial policy to supply cheap labour to the south to work in the agricultural and mining sectors (Dickson 1969 cited by Nabila 1997). The trend has continued to present times and is now dominated by young persons who migrate into the urban centres to work in the informal sector (Seidu 2001). The present movement of young girls who work as head load carriers (kayayoo) in the urban centres in the south has become a national issue. Despite the negative effects of migration on the place of origin of migrants it has been an important feature of Ghana's population, providing opportunity for persons in less endowed areas to improve their livelihood (Nabila 2001, Ghana Statistical Service 1997). For example, the growing informal sector in urban areas which attracts low skilled labour, may continue to provide the opportunity for parents to send off their children to urban centres, even when they do not have any skills or career. Migration from deprived rural communities and regions into Ghana's towns and other destinations that offer better socioeconomic opportunities perhaps promotes the maintenance of large families since children and even adults can access resources in better endowed areas. The trends in and opportunities for migration outside the country and the flow of resources back home may be considered another challenge to the population-resource argument (see Anarfi 2001; Manuh 2001), Analysis of the regional differentials in fertility transition in Ghana indicates that the northern regions, with higher rates of fertility and a greater degree of economic deprivation also have higher out-migration rates and have been source regions of interregional migrants in the country ( Nabila 2001, Hear 1982). Lack of development in these regions delays fertility decline but opportunities for migration also can undermine efforts at encouraging couples to limit their family size. Though the potential for doing so (measured in terms of unwanted or mistimed pregnancies) exists, alternatives to contraception are available for them. Conclusion Evidence from a number of sources indicates that resources for managing childbirth and care other than family planning are widely used in the Ghanaian society. These, collectively referred to in this paper as private resources for fertility management, range from abortion to post-natal control measures, such as, child fostering, early marriage of girls, early entry into the labour market and migration. These practices potentially take the burden of caring for children off their parents, 58 Research Review NS 17.2 (2001)45-61 especially in the case of "free-riders". People by virtue of cultural practices that provide support for children from kin, may not consider the outcome of their sexual practices as issues to be concerned about. Parenting is shifted to others by biological parents who avoid caring for their offsprings.Any attempts to improve contraceptive prevalence in the Ghanaian society must, therefore, not focus on improving the family planning/reproductive health delivery alone. A political economy approach to implementing family planning programs which emphasizes child welfare policies and programs must also be seriously considered. Improving the socioeconomic conditions that can motivate couples to invest in other economic ventures and not only in children must be given attention. Enforcement of laws on child education and addressing the issue of child labour will have to become essential components of the family planning/reproductive health programme. Until these are considered seriously, private means of managing childbirth and care may continue to undermine efforts at increasing the contraceptive prevalence rate in the population. The practice of abortion also needs to be tackled since it has been a major health issue and need not be a preferred method of averting births when other means are available. Post abortion counselling covering family planning or reproductive health issues will be a practical means of helping people to choose contraception rather than to choose abortion. Responsible parenthood will have to be at the centre of all efforts and activities aimed at reducing births. Free riders, child exploiters (both parents and others) will have to be sensitized to recognise the right of children to care and maintenance. When this is achieved the number of children people want will not be a game but a matter of choice and contraception will be an option for many more. References Adanu, R., E. Tweneboah and M. Ntumi, 2001. Abortions in Korle Bu Teaching Hospital. Paper prepared under Women's Health in the City of Accra Project for presentation of findings, Institute of African Studies, University of Ghana and University of Michigan. Addai-Sundiata, J.H., 1996. Family Dynamics and Residential Arrangements in Ghana. In Ardayfio- Schandorf, ed, The Changing Family in Ghana.. Agyei-Mensah, S., 1997. Fertility Change in a Time and Space Perspective: Lessons from Three Ghanaian Settlements. Ph.D. Geografisk Institute, Norges Teknisk Naturvitenskapelige Universitetet, Trondheim. — 1996. New Perspective on the Fertility Situation in Sub-Saharan Africa. Norsk Geografisk Tidsskrift 50(1996): 101-112. — and A. Aase, 1998. Patterns of Fertility Change in Ghana: A Time and Space Perspective. Geografiska Ahiadeke, C, 2001. Incidence of induced Abortion in Southern Ghana. International Family Planning Perspectives, 27(2): 96-101&108. —2001. Some Determinants of International Migration in Ghana during the period 1989-99: Application to the United States Immigration Data. Paper Presented at the national Conference on Migration and Livelihood in Contemporary Ghana. November 12-14, 2001, ISSER, University of Ghana Ampofo, D. A., 1970. The Dynamics of Induced Abortion and the Social Implications for Ghana. Ghana MedicalJournal pp 295-302. Anarfi, J. 2001. The Determinants of International Migration in Ghana in Contemporary Times Paper presented at the National Conference on Migration and Livelihood in Contemporary Ghana. November 12- 14, University of Ghana. —1996 The Role of local herbs in the recent fertility decline in ghana: contraceptives or abortifacients? Paper presented at the Seminar on Sociocultural and Political Aspects of Abortion from an Anthropological Perspective, Trivandrum, India, March 25-28. 59 Appiah, E.N., 2001. Health and Nutritional Status of Children Left Behind by Female Migrants: A Case Study of Amansie - East District of Ghana. M.Phil (Population Studies), Regional Institute for Population Studies (RIPS), University of Ghana. Appiah-Yeboah, S. et al, 2001. The impacts of Agricultural Adversity on Fertility Among the Kassena- Nankana of Northern Ghana. Paper presented at the Population Association of America Conference, Washington, D.C. March 29-31. Apt, N.A. and M. Grieco, 1995. Listening to Girls on the Street Tell Their Own Story: What will help them most? Accra: Centre for Social Policy Studies, University of Ghana. Ardayfio-Schandorf, E. and M. Amissah, 1996. Incidence of Child Fostering Among School Children in Ghana. In E. Ardayfio-Schandorf ed. The Changing Family in Ghana. Accra: Ghana Universities Press. Bongaarts, J. and S.C. Watkins. 1996. "Social Interactions and Contemporary Fertility Transitions." Population and Development Review 22(4):639-682. Bortei-Doku Aryeetey, E., 2001. Fishers Across borders. Paper presented at the National Conference on Migration and Livelihood in Contemporary Ghana. November 12-14,. ISSER. University of Ghana. Caldwell, J.C., 1977. Population Growth and Family Change in Africa: The New Urban Elite in Ghana. London: C. Hurst & Company. —, and P. Caldwell and I. Orubuloye, 1992. Fertility Decline in Africa: A New Type of Transition? Population and Development Review. 18.2:211 -242. Cohen, B. 1993. Fertility Levels, Differentials and Trends. In K.A. Foote et al. (eds.) Demographic Change in Sub-Saharan Africa. Washington D.C: National Academy Press. Pp 8-67. Davis, K., 1963. The Theory of Change and Response in Modern Demographic History. Population Index 29 (October): 345-66. Feyisetan, B. and J.B. Casterline, 2000. Fertility Preferences and Contraceptive Change in Developing Countries. International Family Planning Perspectives 26(3): 100-109. Fortes, M., 1949. The Web of Kinship Among the Tallensi: The Second Part of an Analysis of the Social Structure of a Trans-Volta Tribe. London: Oxford University Press. Frank, O. and J. Bongaarts, 1991. Behavioural and Biological Determinants of Fertility Transition in Sub- Saharan Africa. Statistics in Medicine 10:161-195. Ghana Statistical Service, 2000. Policy Trends in Ghana in the 1990s. Accra: Ghana Statistical Service. Ghana Statistical Service, 1998. Family Planning Service Delivery in Ghana: Comparison Report of 1993 & 1996 Situation Analysis Study (SAS) Results. Accra: GSS. —/Macro International Inc., 1999. Ghana Demographic and Health Survey, 1998. Accra: GSS/Calverton (Maryland): Macro International Inc. Goody, E., 1975. Delegation of Parental Roles in West Africa and the West Indies. In J. Goody, ed. Changing Social Structure in Ghana: Essays in Comparative Sociology of a New State and an Old Tradition. London: International African Institute. Government of Ghana, 1994. National Population Policy (Revised Edition, 1994). Accra: National Population Council. —-1969. Population Planning for National Progress and Prosperity: Ghana Population Policy. Accra: Government of Ghana. Govindasamy, P. and E. Boadi, 2000. "A Decade of Unmet Need for Contracetion in Ghana: Programmatic and Policy Implications." Calverton, Maryland: Macro International Inc/Accra: National Population Council Secretarait. Gyepi-Garbrah, B. 1988. Some Implications of Early Childbearing in Ghana. Accra: PIP. Hardee, K., K. Agarwal, N. Luke, E. Wilson, M. Pendzich et al., 1999. Reproductive Health Policies and Programmes in Eight Countries: Progress since Cairo. International Family Planning Perspectives 25 (Supplement): 52-59. Hear, V.N.H. 1982. Child Labour and the Development of Capitalist Agriculture in Ghana. Development and Change. Isiugo-Abanihe, U.C, 1985. Child fosterage in West Africa. Population and Development Review 1 l.l: 53-97. Makinwa-Adebusoye, P., 1994. Changes in the Costs and Benefits of Children to their Parents. In T. Locah and V. Hertrich, The Onset of Fertility Transition in Sub-Saharan Africa. 1USSP. Liege, Belgium: Derouaux Ordina Editions. 60 Research Review NS 17.2(2001)45-61 Manuh, T., 2001. Ghanaian Migrants in Toronto, Canada—Care of Kin and Gender Relations. Paper presented at the Seminar on Globalization and Changes in the Cultures of Survival and Care. University of Ghana, October 10-12. Mason, K.O. 1997. Explaining fertility Transitions. Demography,, 34(4): 443-454. Ministry of Health, 1996. Health Sector S Year Programme of Work, Accra: Ministry of Health. Ministry of Health (Health Education Division) / The Johns Hopkins University (Population Communication Services), 1992. Partners in Family Health Communication. Accra: Ministry of Health. Mundigo, A. I. and C. Indriso, eds. 1999. Abortion in the Developing World. London: Zed Books. Nabila, J.S., 2001, The Migrant as an Asset and/or a Liability: Challenges and Opportunities for Development. Inaugural Lecture, University of Ghana, Legon, May 17. — 1997. Population and Land Degradation in the Upper East Region of Ghana. Bulletin of the Ghana Geographical Association 20:74-86. National Population Council, 2000. Adolescent Reproductive Health Policy. Accra: NPC. Nuhhu-Billa, B. 2001a. Children for Hire? Junior Graphic No. 47, August 1-7. —2001b. Child Bread Winner: He begs to Feed Mum and Self Junior Graphic No. 58 October 17-23. Oppong, C. 1987, Responsible Fatherhood and Birth Planning. In C. Oppong, ed.} Sex Roles, Population, and Development in West Africa: Policy Related Studies on Work, and Demographic issues. London: James Curry Ltd. — and W. Bleek, 1982. Economic Models and Having Children: Some Evidence from Kwahu. Africa 52(4): 16-32. — 1981, Middle Class African Marriage. London: George Allen & Unwin. —, 1974. Growing Up in Dagbon Accra: Ghana Publishing Corporation. Page, H., 1989. Childrearing Versus Childbearing: Coresidence of Mother and Child in Sub-Saharan Africa. In RJ. Lesthaeghe ed. Reproduction and Social Organization in Sub-Saharan Africa. Berkeley: University of California Press. Piotrow, P.T., D.L, Kincaid, J.G. Rimon II, and W. Rinehart, 1997, Health Communication: Lessons from Family Planning and Reproductive Health. Westport, Connecticut; Praeger, Population Impact Project (PIP) 1995. Adolescent Fertility and Reproductive Health in Ghana. Accra: PIP. Raum, O.F., 1940. Chaga Childhood: A Description of Indigenous Education in an East African Tribe. Sai, F.T. 1997 Where is the International Population Movement Going? Health Transition Review. London: Oxford University Press. Supplement to Volume 7: 1 -5 Seidu, A., 2001, The Dynamics of Migration in Northern Ghana. Paper presented at the National Conference on Migration and Livelihood in Contemporary Ghana. November 12-14. ISSER. University of Ghana. UNICEF/Republic of Ghana, 2000. Situation Analysis of Children and women in Ghana. Accra: UNICEF/ Republic of Ghana. Westoff, C.F. and A. Bankole, 2000, Trends in the Demand for Family Limitation in Developing Countries. International Family Planning Perspectives 26(2): 56-52,97. Yeboah, S., 1998. Househelps in Ghana: A Case Study of the Sekondi-Takoradi Metropolis. B.A. Dissertation, Department of Geography, University of Ghana. Yoder, S. and R, Henry, 2002. Coping with Pregnancy: Experiences of Adolescents in Ga Mashie, Accra, Ghana. Calverton, MD: ORC MACRO, MEASURE DHS+ Notes 1 Knowledge of a family planning method, both modern and traditional, at the 1998 GDHS is defined simply as having heard of a method (Ghana Statistical Service 1999:39). 2 The survey defines early pregnancy losses as pregnancies of less than 28 weeks that are terminated through spontaneous or induced abortions (Ghana Statistical Sendee 1999:30). 3 Stillbirths are defined as children born dead after a gestation of 28 weeks or more (Ghana Statistical Service 1999:30). 4 This was one of the studies done under the Women's Health in the City of Accra Project by the Institute of African Studies, University of Ghana and the University of Michigan in 2001, 61 5 The study was conducted in the Eastern, Volta, Greater Accra and Central regions between January and March 1997 under the Maternal Health Survey by the Institute of Statistical Social and Economic Research (ISSER), University of Ghana. 6 This is not to overlook the fear of side-effects and other health concerns which have been major reasons for nonuse of modem contraceptives in Ghana. These reasons accounted for over 20% of all reasons among currently married women with unmet need for contraception at each of the GDHS. But clandestine abortion (which is used more than abortion performed by qualified practitioners) also has health risks, even death. 7 In a story captioned "Children for Hire" in the Junior Graphic, Nuhhu-Billa (2001a) tells of the case of children in the city of Accra, aged between six and fifteen years whom their parents hire to blind people to help to collect alms and in return receive between 01,000 and 03,000 per day. In another story, captioned "Child Bread Winner ... He Begs to Feed Mum and Self, she writes about an eight year old school dropout who begs on the street to take care of himself and his unemployed single mother (Nuhhu-Billa 2001b). 8 In traditional Ghanaian societies, childbearing soon follows marriage. Thus any reduction in the proportion of the former could be due to a drop in the latter. This situation is not exactly the same now as more childbearing is occurring outside marriage. 62