SEARCH

SEARCH BY CITATION

Keywords:

  • contraception;
  • induced abortion;
  • community-based study;
  • Ghana

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Study setting and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. References

This article describes the results of a community-based survey on birth control in a rural district in western Ghana. Structured interviews with 2179 women and men aged 15–49 years were used to study the prevalence of contraceptive methods and induced abortion. In addition, the influence of induced abortion on reported fertility in relation to residence and education was analysed. The results show that 59.8% had used a contraceptive method at some time in life, while use of induced abortion was reported by 22.6% of respondents. Prevalence of ever-use of any method to avoid childbirth, contraception and/or induced abortion was 67.1%. Urban residence and higher education were associated with more induced abortions and higher use of contraceptive methods. Differences in use of induced abortion were partly responsible for the education and residence-related changes in fertility. Induced abortion needs to be considered when discussing methods in use to avoid childbirth in developing countries.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Study setting and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. References

Birth control in developing countries has a long and controversial history (Shane 1997). It is usually regarded as synonymous with contraception. Although both contraception and induced abortion are used to achieve the desired number of children, induced abortion has only recently been recognized as a public health issue (Kulczycki et al. 2000).

Africa is lagging behind other world regions in the use of contraceptives (Arkutu 1995). Only 20% of African married women currently use contraception vs. 58% of married women worldwide (Alan Guttmacher Institute 2000). But in Africa, too, people wish to limit their family size, and it is estimated that 20–40% of married women have an unmet need for contraception (Shane 1997). Collection of data on induced abortion is often hindered by legal and moral restrictions (Coeytaux 1988; Barreto et al. 1992). The World Health Organization estimates that around 20 million unsafe abortions annually occur worldwide, more than 95% of which in less-developed countries. As a result, some 80 000 women die each year of abortion related complications (WHO 1997). Abortion complications contribute to a large proportion of maternal morbidity (Rogo 1993; Benson et al. 1996; Rasch et al. 2000). In the setting of this study, Berekum District in rural Ghana, abortion complications form the largest single cause of maternal mortality (Geelhoed et al. 2002). Therefore, this community-based survey was carried out to obtain an estimate of knowledge and ever-use of contraceptive methods and of induced abortion. In addition, the influence of induced abortion on reported fertility in relation to educational level and residence was analysed.

Study setting and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Study setting and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. References

The study took place in Berekum District, Brong Ahafo Region, in rural Ghana. This district is relatively small with 424 km2 and 96 757 inhabitants, who are mainly occupied with agriculture (cocoa, subsistence farming, timber). Two-thirds of the population live in about 30 villages scattered around the district, the remaining people live in the central town of Berekum.

Traditionally, the Ghanaian culture was pro-natalist. Contraception was rare, although outside marriage it was possibly used to avoid exposure of a secret relationship (Bleek 1976). Induced abortion is thought to have been rare in the past, but seems rather common in recent times (Bleek & Asante-Darko 1986). Since the 1960s family planning programmes have been introduced in Ghana, which have mostly been directed towards married women (Twum-Baah et al. 1997). In Berekum District oral and injectable contraceptives, intrauterine devices and condoms are supplied by several government and private health facilities. Natural family planning, the cervical mucus method, is offered by the (mission-run) district hospital. Tubal ligation is possible at this district hospital in the case of severe obstetric complications, or on request in a government hospital in a neighbouring district. Contraception is promoted in mother and child clinics, focused on spacing and limiting births.

Induced abortion is legal in Ghana (Mensa-Bonsu 1996), if performed by a registered medical practitioner and when the physical or mental health of the pregnant woman is threatened, when the child is likely to be born with a serious physical abnormality, or when the pregnancy resulted from rape or incest. In all other situations it is illegal to procure or assist in an induced abortion. Induced abortion is legally available in one private clinic in Berekum District since 1994.

This community-based survey was performed between February and July 1999. A structured questionnaire in Twi, the local language, was designed. Questions concerning personal characteristics, pregnancy history, contraceptive use and infertility were included (appendix, English version). The place of interview was assumed to be the same as the place of residence. Thirty-five male and female interviewers were recruited from staff of the district hospital, and trained. As no lists of inhabitants or detailed community maps were available in the district, systematic random sampling of houses was used. Men were often absent from home at the time of the visit. Therefore, some houses were replaced by local gathering points in their vicinity (such as a shady area under a near tree, sport field, or bar). The interviewees were selected by choosing the first person encountered of the appropriate age and sex. Interviews were held anonymously in a private and confidential setting, and the option for refusing to co-operate or to refrain from answering certain questions was explicitly given before each interview.

Table 1. Appendix  English version of the questionnaire (questions relevant for this article only)Thumbnail image of

Based on an estimated prevalence of 30% of induced abortion among women of reproductive age, a sample of nearly 900 women would be necessary to obtain an accurate estimate of prevalence (with a 95% confidence limit of ±3%) from a total of 20 000 women of reproductive age living in the district. To allow for refusals and to increase statistical validity the aim was set at interviewing a minimum of 1000 women. It was decided to interview an equal number of men in the same age group. The sample size was thus set at 2000 interviews. The number of interviews scheduled for each community corresponded proportionally with the population number as reported by the District Health Authorities.

Political and traditional leaders in the district co-operated with the survey. Communities were informed of planned visits to their area for interviews through a local radio station, and through a letter with date and time of the intended visit.

Data were analysed with EPI-Info (version 6.04). Statistical analysis was performed using the chi-square test for calculating differences in proportions. Risk ratios with confidence limits were calculated. Stratified analysis was used to assess confounding. Differences in means were compared using Student's t-test for normally distributed continuous variables, or the Kruskal–Wallis test when a continuous variable did not satisfy the assumptions underlying the t-test.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Study setting and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. References

In the study 2179 interviews were conducted. While nobody refused to co-operate, or refrained from answering certain questions, 42 forms (1.9%) were incorrectly completed or represented persons from a wrong age group. Excluding these left 2137 interviews for further analysis. Thus, the response rate was 98.1%. Socio-demographic characteristics of the study population are shown in Table 1.

Table 1.  Socio-demographic characteristics of the study population (n=2137) Thumbnail image of

Knowledge about methods to prevent unwanted pregnancies was reported by 1593 respondents (74.5%). Of these, 1277 persons (80.2%) reported having ever used such a method. Thus, of all 2137 respondents, 59.8% had used a contraceptive method at some time. Induced abortion was reported by 482/2137 respondents (22.6%). The number of induced abortions reported by one person ranged from 0 to 10. The induced abortion rate was 150 per 1000 births (798/5306 births). The prevalence of ever-use of any method of birth control, contraception and/or induced abortion, in the study population was 67.1%.

Knowledge about contraception, use of contraceptives and of induced abortion were significantly correlated with five factors: gender, age, education, residence and parity (Table 2). Although several of these five factors were related to each other (e.g. age and education, age and parity, gender and education, education and residence), the association of any determinant with contraception and induced abortion could not be solely explained by confounding of the other factors.

Table 2.  Proportion of all respondents reporting knowledge of prevention of unwanted pregnancies, ever-use of contraceptives and ever-use of induced abortion (n=2137) Thumbnail image of

Gender

Women were better informed: 833/1073 women (77.6%) mentioned knowledge of contraception vs. 760/1064 men (71.4%). Men, however, reported more use of contraception (670/1064, 63.0%) than women (607/1073, 56.6%), while induced abortions were more often reported by women (303/1073, 28.2%) than by men (179/1064, 16.8%).

Age

Persons below the age of 20 years mentioned knowledge of prevention of unwanted pregnancies significantly less often: 105/188 teenagers (55.8%) vs. 1488/1949 adults (76.3%). Use of contraceptives increased with age, from 75/188 teenagers (39.9%) to 451/697 people of 35 years and older (64.7%). Induced abortion was more common among young adults: 327/1252 respondents aged 20–35 (26.1%) vs. 128/697 persons aged 35 or older (18.4%).

Education

People who had received secondary education more often had knowledge of contraception (1181/1542, 76.6%) than persons without formal education (214/339, 63.1%). They also reported higher usage of contraceptives (975/1542, 63.2%) than respondents without formal education (150/339, 44.2%). Induced abortion was more frequently reported by people with secondary education (380/1542, 24.5%) than by people without formal education (47/339, 13.9%).

Residence

Respondents from Berekum town reported knowledge about contraception as regularly as in other communities, 582/760 persons (76.6%) in town vs. 1011/1377 persons (73.4%) in the other communities. Use of contraception was similar as well, 478/760 (62.9%) in the town and 799/1377 (58.0%) elsewhere in the district. People reporting induced abortion, however, were more likely to live in Berekum town: 236/760 of the Berekum town respondents (31.1%) reported induced abortion, 217/1137 of the respondents in near villages (19.1%) and 29/240 of the respondents in remote villages (12.1%).

Parity

Contraceptive knowledge increased with increasing parity, from 1240/1695 (73.2%) respondents with fewer than five children to 353/442 (79.9%) with five children and more. Use of contraception increased from 984/1695 (58.0%) to 293/442 (66.3%) in these groups. Induced abortion, however, was more regularly used by persons with fewer than five children, in 414/1695 (24.4%) vs. 68/442 (15.4%) by persons with five children and more.

When asked to name a method to prevent pregnancy 1680 replies were given by 1593 respondents. They described modern methods of contraception, often summarized under the term `family planning', in 57.2% (962 times). Traditional methods were mentioned in 0.6% of cases (10 times), abstinence in 0.8% (14 times) and withdrawal in 0.6% (10 times). The natural family planning method was mentioned in 13.4% of cases (225 times). Contraceptive methods most frequently used were modern reversible methods for women, such as oral and injectable contraceptives and intrauterine devices (Table 3). These methods were used by 54.0% (690 times). The persons who reported use of these methods were usually older (56.7% were 30 years of age or older vs. 49.5% of the general study population), and more highly educated (47.5% had secondary education or more vs. 42.1% in the general study population). They also were more likely to have five or more children (22.6%) than the general study population (16.6%). Both men and women, in town and in villages, equally often reported having used these methods.

Table 3.  Reported ever-used methods of contraception (n=1277) Thumbnail image of

Another frequently used method was natural family planning (Table 3), which was used by 35.6% (454 times). Men and women, in all age groups used this method. Respondents with higher education (50.0% had secondary education or more) mentioned its use more often, as well as persons in Berekum town (41.4% lived in Berekum town vs. 35.6% of the general study population). Its use was also more frequently reported by people with five or more children (23.1%).

Condom use (Table 3) was nearly exclusively mentioned by men (91.2%). Its use was reported by 20.8% (265 times). Condoms were used more often by younger people (64.2% were below the age of 30 years vs. 50.5% of the general study population), by persons who had higher education (62.7% had at least secondary education), and by persons residing in Berekum town (46.8%). Of these respondents, 57.3% had no children.

A small number of respondents had made use of tubal ligation (1.5%, 19 times), traditional methods (0.9%, 12 times) and a variety of other methods, such as withdrawal or abstinence (2.1%, 27 times; Table 3).

The most commonly stated reason for non-use was fear of negative health consequences or side-effects of modern contraceptive methods (89 times; 28.2%; Table 4). These concerns were mostly expressed by women (84.3% of these respondents was female), and by persons with lower education, as 45.0% had no formal education or only primary school vs. 27.9% in the general study population. It was mentioned by respondents from all age groups and all communities.

Table 4.  Reasons for non-use of contraceptives (n=316) Thumbnail image of

Desire to conceive was mentioned by 21.5% (68 times) as reason for non-use, and the lack of need for contraception, for example, when a person was not sexually active, accounted for 29.4% (93 times; Table 4). Church influence against contraceptive use was named by 13.6% (43 times), while partner and family influences were reported by 5.1% (16 times) and 3.8% (12 times), respectively (Table 4). Contraceptives not being affordable or available was mentioned by 2.2% (seven times) as reason for non-use (Table 4).

Respondents who did not use contraception for health concerns practised induced abortion slightly more frequently than the general population (23/89 vs. 482/2137; ns). Respondents who did not use contraception for religious reasons, used induced abortion slightly less than the general population (6/43 vs. 482/2137; ns).

As place of residence, level of education and age were interrelated in the study population, the influence of education on women's fertility was assessed in residence- and age-strata (Table 5) and of residence in age- and education-strata (Table 6). The mean number of children among women in Berekum town and in the villages, aged below 30 years as well as aged 30 years and older, was smaller when they had received more than primary education (Table 5a). However, the mean number of pregnancies did not always show such education-related difference. Among older women in town the mean number of pregnancies did not differ with educational level, and the smaller number of children in this group depended rather on an increased use of induced abortion among educated women (Table 5b). Among village women the mean number of pregnancies did clearly show an education-related reduction similar to that observed in the mean number of children (Table 5c, d).

Table 5.  Mean number of pregnancies, children, pregnancy losses and induced abortions reported by women of different educational level, specified per age and residence strata (n=1073) Thumbnail image of
Table 6.  Mean number of pregnancies, children, pregnancy losses and induced abortions reported by women with different residence, specified per age and education strata (n=1073) Thumbnail image of

Residence in Berekum town was associated with a smaller mean number of children, and a similarly smaller mean number of pregnancies, at least among less-educated, older women (Table 6b). Among educated older women the mean number of children was smaller when they lived in town; however, their mean number of pregnancies did not differ with residence. The urban residence-related reduction in number of children was because of an increased use of induced abortion, which was also observed among educated younger women (Table 6c, d).

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Study setting and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. References

The use of anonymous, privately conducted interviews yielded a very high response rate. Thorough preparation of the District population is likely to have assisted in the respondent's good faith. Although contraceptive use and induced abortion are sensitive topics in the local culture, it seems that the interviews were acceptably conducted, and that respondents considered it appropriate that attention was given to these topics. However, some degree of recall error may have occurred. Unfortunately re-interviews to assess the consistency of the answers were not possible. Occasionally pregnancies may have been reported twice, by the man who caused the pregnancy as well as the woman who carried it. In the study area partners do not necessarily live in the same house or even in the same community, and polygamy is common. Checking for double reporting was not considered feasible.

It is not clear whether the sampling method used in this survey indeed provided each individual with an equal probability of selection. A house in Berekum District generally lodges several households consisting of various members of an extended family, and often unrelated households too. There is only a very limited number of houses in which one nuclear family lives, and these are generally families of a higher socio-economic status with possibly a smaller number of children. Individuals of such families would have had a higher probability of selection. The replacement of an unknown number of houses in the sample frame by local gathering points, when men were not found at home, may have influenced their selection as well. Yet the preservation of anonymity was considered of utmost importance, and therefore tracing of men not found at home was not feasible.

Nevertheless, when comparing our results with data known for Berekum District or for Ghana, as recently reported in District Health Reports and the Ghana Demographic and Health Survey (GDHS; Ghana Statistical Service 1999), we conclude that the results adequately represent the prevalence of ever-use of contraception and induced abortion in Berekum District.

Respondents in our study reported contraceptive knowledge in 74.5% and use of contraception in 59.8%. GDHS reported a higher knowledge of contraceptives at 95%, but only 44.7% of the women in that survey reported ever to have used a contraceptive method. These differences may be related to the phrasing of questions. GDHS used the term `family planning', thus explaining what was meant with `a way or method to delay or avoid a pregnancy'. The question in our study asked for `a method to avoid a pregnancy when you do not wish it'. Both questions measure knowledge only superficially. It seems, however, that one cannot consider contraceptive knowledge the equivalent of knowledge of methods to prevent unwanted pregnancies. Contraceptive education in Ghana emphasizes spacing and limiting births (Twum-Baah et al. 1997). Contraception may thus be seen as a method to achieve a small family, rather than as a method to avoid unwanted pregnancies.

The higher ever-use of contraception observed in our study area suggests that contraceptive knowledge was more often put into practice in the study district than nationwide. Female respondents in our study reported ever-use of any contraceptive method more frequently than in GDHS in all age groups, especially women aged 15–19. GDHS reported 18.6% ever-use in this age group, while our study found 40.8% (RR 2.2, 95% CI 1.6–3.0).

The pattern of contraceptive methods used here was similar to the pattern in GDHS, with reversible female methods being most popular among modern contraceptives, and a relatively high popularity of the natural family planning method. The increased use of modern methods with increasing education observed in our study was also reported by GDHS, as well as the increased use of modern methods when respondents had more children.

In our study and in GDHS the natural family planning method was the most commonly used method of contraception. While in many African countries modern contraceptives as oral contraceptives or intrauterine devices are most often used, in West Africa use of the natural family planning method is more common (Population Reference Bureau 1998). If properly used, this method can be quite effective in preventing pregnancies (Klaus 1995). Other advantages are its low cost and lack of provider dependence. Whether this method is always properly applied is, however, disputable, as in many countries in West Africa knowledge of the correct timing of ovulation tends to be patchy, and male co-operation can be unreliable (Kirk & Pillet 1998).

The results of this study suggest that the use of contraceptives for male use such as condoms may be under-reported when only women are interviewed. Women in this study hardly ever mentioned use of this method, while men described its use regularly. Condom use is an important method, as it currently is one of the few possibilities for protection against both pregnancy and sexually transmitted diseases.

Health concerns as reason for non-use was at 27.1% the most common reason found in this survey, excluding the perceived lack of need for contraception. This was similar to findings in GDHS. Apart from lack of knowledge, health concerns are an important reason for non-use of contraceptives (Bongaarts & Bruce 1995). This survey found that the same women who said they were afraid to use contraceptives did use induced abortion. This suggests that health concerns related to contraception are based upon misunderstandings. However, the difference between prevention and `cure' for an unwanted pregnancy may play a role in this issue. Preventive measures need, in the mind of users, to be safer than `curative' emergency measures. Objectively, modern contraceptives are known to be safer than pregnancy (Population Reference Bureau 1988; Shane 1997) and safer than induced abortion (Benson et al. 1996; Alan Guttmacher Institute 2000). Counselling about contraceptive safety needs to be encouraged to reduce health concerns. Relatively safe emergency measures, such as postcoital contraceptive pills and postcoital IUD-insertion might be introduced into the range of locally available contraceptives.

The prevalence of induced abortion in this study (22.6%; 150/1000 births) is higher than previously reported rates, although only few induced abortion rates for Africa, with considerable variance, have been reported. Data comparison is difficult as no single unit of prevalence has been used. The estimated annual unsafe abortion rate in West Africa is among the highest in the world, at 37 per 1000 women aged 15–44 years (Alan Guttmacher Institute 2000). Elsewhere, a lower estimated annual unsafe abortion rate of 15 per 1000 women of reproductive age is given for sub-Saharan Africa (Rogo 1993). In Benin a low prevalence of 1.5 induced abortions per 1000 births was observed (Fourn et al. 1997), while a Nigerian study reported a prevalence of 11% for women aged 15–45 years (Okonofua et al. 1999). In Zaire a prevalence of 15% for ever-pregnant women of reproductive age was reported (Shapiro & Oleko Tambashe 1994). An early study in Ghana found a high prevalence of more than 50% of women, and 15% of pregnancies (Bleek & Asante-Darko 1986). Recently, a annual induced abortion rate of 17/1000 women of reproductive age was found in another population-based survey in southern Ghana (Ahiadeke 2001).

Use of any method of birth control was quite common for 67.1% in the study population. The survey results show that the same factors which are known to increase contraceptive use, such as higher education and urban residence, also increase induced abortion. The comparison of mean numbers of reported pregnancies, children, pregnancy losses and induced abortions (Tables 5 and 6 indicates that part of the reduction in fertility associated with higher education and urban residence has to be attributed to induced abortion. In Zaire the situation is similar, with differences in fertility were partly caused by differences in incidence of induced abortion (Shapiro & Oleko Tambashe 1994). Both higher education and urban residence may increase the use of birth control (including induced abortion) in several ways. People with higher education and urban residence often have easier access to both contraceptives and induced abortion. Facilities for higher education are usually located in urban areas. Young people attending these facilities often do not wish to interrupt their studies for childbirth, and thus need to delay the start of childbearing. The need to delay is not adequately met by the current family planning programmes in Ghana. Although, compared with GDHS, teenagers in our study reported a relatively high ever-use of contraceptives, their knowledge and ever-use was considerably lower than those of older respondents. Additionally, respondents with none or a small number of children had a lower level of contraceptive knowledge and ever-use than respondents with a higher number of children.

Female education in Africa is vital for many health issues (Harrison 1997). The observed association with increased use of induced abortion should not obstruct the pursuit of good education for women. It rather shows how important it is for young people to have the knowledge and means to control their fertility during the years of their education.

Conclusions

  1. Top of page
  2. Abstract
  3. Introduction
  4. Study setting and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. References

Both contraceptives and induced abortions are used to achieve the desired number of children in Ghana and other sub-Saharan countries. To increase use of contraceptives and prevent use of induced abortion, family planning programmes need to address the need to delay, space and limit births, among both women and men, regardless of their age or marital status. Public education concerning contraceptive safety should be improved.

Care ought to be taken in assigning any observed reduction in fertility associated with education or residence entirely to the increased or more reliable use of contraceptives, as induced abortion may act as a confounder. Despite ethical and legal restraints, induced abortion as a method of birth control has to be assessed scientifically in order to reduce its health hazards.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Study setting and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. References
  • 1
    Ahiadeke C (2001) Incidence of induced abortion in Southern Ghana. International Family Planning Perspectives 27, 96101, 108108.
  • 2
    Alan Guttmacher Institute (2000) Sharing Responsibility: Women, Society and Abortion Worldwide. Alan Guttmacher Institute, New York.
  • 3
    Arkutu A (1995) Family planning in sub-Saharan Africa: present status and future strategies. International Journal of Gynecology and Obstetrics 50 (Suppl. 2), S27S34.DOI: 10.1016/0020-7292(95)02482-r
  • 4
    Barreto T, Campbell OM, Davies JL et al. (1992) Investigating induced abortion in developing countries: methods and problems. Studies in Family Planning 23, 159170.
  • 5
    Benson J, Nicholson LA, Gaffikin L & Kinoti SN (1996) Complications of unsafe abortion in Sub-Saharan Africa: a review. Health Policy Planning 11, 117131.
  • 6
    Bleek W (1976) Sexual Relationships and Birth Control in Ghana. A Case Study of a Rural Town. Department of Cultural Anthropology, University of Amsterdam, Amsterdam, The Netherlands.
  • 7
    Bleek W & Asante-Darko NK (1986) Illegal abortion in Southern Ghana: methods, motives and consequences. Human Organization 45, 333344.
  • 8
    Bongaarts J & Bruce J (1995) The causes of unmet need for contraception and the social content of services. Studies in Family Planning 26, 5775.
  • 9
    Coeytaux FM (1988) Induced abortion in sub-Saharan Africa: what we do and do not know. Studies in Family Planning 19, 186190.
  • 10
    Fourn L, Fayomi EB & Zohoun T (1997) Prevalence des interruptions de grossesse et de la regulation des naissance au Benin. Journal de Gynecologie, Obstetrique et Biologie de la Reproduction (Paris) 26, 804808.
  • 11
    Geelhoed DW, Visser LE, Asare K, Schagen van Leeuwen JH & Van Roosmalen J (2002) Trends in maternal mortality: a 13-year hospital-based study in rural Ghana. European Journal of Obstetrics and Gynecology and Reproductive Biology, in press.
  • 12
    Ghana Statistical Service (1999) Demographic and Health Survey 1998. Ghana Statistical Service, Accra.
  • 13
    Harrison KA (1997) The importance of the educated healthy woman in Africa. Lancet 349, 644647.
  • 14
    Kirk D & Pillet B (1998) Fertility levels, trends, and differentials in Sub-Saharan Africa in the 1980s and 1990s. Studies in Family Planning 29, 122.
  • 15
    Klaus H (1995) Natural Family Planning: A Review. Natural Family Planning Center of Washington, Bethesda.
  • 16
    Kulczycki A, Potts M & Rosenfield A (2000) Abortion and fertility regulation. Lancet 347, 16631668.
  • 17
    Mensa-Bonsu HJAN (1996) The Annotated Criminal Code of Ghana, 2nd edn. Black Mask Ltd, Accra.
  • 18
    Okonofua FE, Odimegwu C, Ajabor H, Daru PH & Johnson A (1999) Assessing the prevalence and determinants of unwanted pregnancy and induced abortion in Nigeria. Studies in Family Planning 30, 6777.
  • 19
    Population Reference Bureau (1988) Contraceptive Safety: an Overview. Population Reference Bureau, Washington DC.
  • 20
    Population Reference Bureau (1998) 1998 World Population Data Sheet. Population Reference Bureau, Washington DC.
  • 21
    Rash V, Muhammed H, Urassa E & Bergstrom S (2000) The problem of illegally induced abortion: results from a hospital-based study conducted at district-level in Dar es Salaam. Tropical Medicine and International Health 5, 495502.
  • 22
    Rogo KO (1993) Induced abortion in sub-Saharan Africa. East African Medical Journal 70, 386395.
  • 23
    Shane B (1997) Family Planning Saves Lives. Population Reference Bureau, Washington DC.
  • 24
    Shapiro D & Oleko Tambashe B (1994) The impact of women's employment and education on contraceptive use and abortion in Kinshasa, Zaire. Studies in Family Planning 25, 96110.
  • 25
    Twum-Baah KA, Ameka EK, Okrah EA & Ohene-Okai A (1997) Second Round Situation Analysis Study of Family Planning Service Delivery Points in Ghana. Ghana Statistical Service, Accra.
  • 26
    WHO (1997) Unsafe Abortion: Global and Regional Estimates of Incidence of Mortality Due to Unsafe Abortion with a Listing of Available Country Data, 3rd edn. WHO/RHT/MSM/97.16, World Health Organization, Division of Family Health, Geneva, Switzerland.