Unsafe abortion is a health problem of global concern. In many developing countries, women know little about and have limited access to effective contraceptive methods, and as a result may experience unwanted pregnancies. Because of restrictive abortion laws and poor access to safe methods of pregnancy termination, many women resort to obtaining abortions under unsafe conditions. Worldwide, an estimated 20 million unsafe abortions and 70,000 maternal deaths due to unsafe abortions occur each year, and most of these events take place in developing countries. In Africa alone, an estimated 3.7 million unsafe abortions are performed annually, and approximately 23,000 deaths result from these procedures.1
Under Nigerian law, performing an abortion is a criminal offense unless the pregnancy threatens the woman's life, and penalties for the offense are severe.2 Because of these legal restrictions and because of religious and social norms opposing abortion, the practice of abortion is shrouded in secrecy; abortions are typically performed clandestinely, often by unskilled providers under unsanitary and dangerous conditions.
Furthermore, because of abortion's illegal status and the social stigma that accompanies it in Nigeria, precise information about it is difficult to obtain. Clearly, however, abortion is a major public health problem: Nigeria's maternal mortality rate is estimated to be 800 deaths per 100,000 live births,3 and in West Africa overall, an estimated 14% of maternal deaths are attributed to abortion.4 Partial data on the number of complications and deaths related to abortions indicate that the consequences for women's health are serious, and the cost of treating women for complications places an additional burden on an already troubled health care system.5 Hospital-based studies reveal that induced abortions represent a substantial proportion of all gynecologic admissions.6
The problems related to abortion may be particularly severe for adolescent women.7 As increasing numbers of rural families migrate to urban areas, parental control and supervision are weakened, and young people are exposed to modern influences that encourage sexual activity in relationships that may not lead to marriage. Because contraceptive knowledge is low in Nigeria and access to services is poor, unplanned pregnancies among young unmarried women are increasingly common. Among adolescents, reasons for having an abortion include the desire to remain in school, financial concerns and fear of social reprisal because of an out-of-wedlock pregnancy. 8
Adolescents face unique barriers to obtaining a safe abortion: When they become pregnant, they are slower to recognize and accept the pregnancy; they are less likely than older, more experienced women to know where to seek advice and help; they may use ineffective methods to attempt to induce an abortion; and they may be unable to afford a physician's fee.9 All of these factors may cause delays, and the later an adolescent seeks an abortion, the more likely she is to suffer complications that may lead to hospitalization. Two hospital-based studies found that of the women who said they had had an induced abortion, or whose symptoms indicated that they had had an induced abortion, about 70% were adolescents.10
Adult Nigerian women's fertility preferences are also likely to be affected by increased urbanization and modernization, and increasingly, older women may want smaller families and greater control over the timing of their births. They, too, will be at risk of having an unwanted pregnancy if their family planning needs are not met. In a recent community study, 42% of abortions occurred after the first birth, and most of these were among married women.11 A common reason given by women for these abortions was poor timing of the pregnancy or the need to space births better. The use of abortion by married women to space births not only is a likely response to social change in recent decades, but is a traditional practice in northern Nigeria.12
Existing research provides insights into various aspects of abortion in Nigeria. Many studies describe the characteristics of women who have been hospitalized for complications of abortion;13 a few community-based studies describe all women who sought abortion.14 This body of research provides information about service providers, their techniques and the possible consequences of different procedures. However, these studies have not generated the kinds of nationally representative data on the provision and incidence of abortion that are necessary to formulate effective policies, programs and strategies to deal with the issue. Most available information has been derived from highly localized community surveys and studies based in large teaching hospitals. Moreover, research has been disproportionately concentrated in the highly urbanized southwestern health zone.
We report on the findings of a 1996 survey of health professionals from most of Nigeria's 31 states concerning their perceptions of abortion practices and the conditions under which women obtain abortions. The results are qualitative and give an approximate picture of the conditions of abortion provision. Given the scarcity of information on the subject, these findings add a valuable perspective.
Methodology
Sample Selection and Data
A self-administered questionnaire in English (Nigeria's official language, which is widely spoken) was pretested in Lagos and Ibadan among doctors, nurses, midwives and chemists (unlicensed patent medicine dealers). The survey was carried out with the assistance of university-based collaborators who distributed and collected the questionnaires.
A purposive sample of 76 health professionals who were considered knowledgeable about abortion practices were selected from urban and rural areas throughout the country. Emphasis was placed on finding chiefs of obstetrics and gynecology in major hospitals and clinics, private physicians practicing obstetrics and gynecology, government officials, representatives of women's groups, academics conducting research on sociological and reproductive health issues, and activists in nongovernmental organizations concerned about this issue.
Five of the selected professionals could not be located because of changes in address. Of the 71 questionnaires that were returned, four were excluded because of extensive missing data or apparent misinterpretation of some questions; these respondents were not different in any significant manner from the 67 who constituted the final sample.
Socioeconomic conditions and health care infrastructures vary substantially by region in Nigeria. For example, a much higher proportion of the population in the North than in the South are Muslim, rural, poor and poorly educated; the North also is less well served in terms of health facilities. Consequently, results are presented for either the four health zones (Northeast, Northwest, Southeast and Southwest) or, if information on an item was missing for many respondents, the two large regions (North and South). Area-specific results were obtained by aggregating responses according to the region in which participants live and work, because their answers are likely to reflect the conditions in the areas with which they are most familiar.*
The survey included questions on methods used to induce abortions, the types of practitioners who perform induced abortions, the approximate cost of procuring an abortion, the reasons why women have abortions and the probability that women who have abortions will have complications that may be severe enough to require hospitalization or that women with complications will receive treatment in hospitals. For most questions, respondents were asked to describe the situation as it pertains to four subgroups of women: poor and relatively well-off women residing in urban centers and rural areas.
The detailed information obtained from respondents for all of these subgroups was combined to produce averages for all women aged 15-49, using proxy weights for the proportional size of the four subgroups from the 1990 Nigerian Demographic and Health Survey (NDHS). Because data are not available on the income distribution of the population by region and by place of residence (urban or rural), a woman's educational attainment was used as a proxy indicator of her relative economic position. In the South, women with seven or more years of education were considered to be nonpoor, and in the North, where the level of education is much lower, women with any schooling were classified as nonpoor. Using the NDHS distribution of women by educational attainment, we estimated that nationally, nonpoor urban and nonpoor rural women each represent 12% of all women aged 15-49, poor urban women make up 13% of this age-group and poor rural women account for the remaining 63%.
Weights were obtained from the NDHS in the same manner for the four health zones and for the two major regions. To produce subnational estimates for all women of reproductive age, we weighted aggregated responses for the four subgroups of women by their relative size within each health zone or region. To obtain national values for each subgroup, we weighted values for the health zones or regions (as appropriate for the measure) by the proportion of women of reproductive age who live in each area. Overall aggregate estimates for Nigeria were then obtained by further weighting the values for the four subgroups, using the national distribution of women according to the proxy measure of income within urban and rural areas.
Characteristics of Respondents
In all, 51% of the respondents were medical professionals (43% physicians and 8% nurses, midwives and other health care workers); the remainder had backgrounds in teaching, research, or health administration or policy. Two out of three had had experience with abortion practice through work in the public sector, and the rest in the private or nonprofit sector.
Respondents were asked about the ways they had been exposed to the practice of abortion and could report more than one way. Some 58% had had experience with abortion in their practice in public hospitals, 32% in private hospitals and 22% in their private office practice. About one-third had had experience that came from outside their workplace, often from involvement in youth programs or in providing educational or counseling services.
About half of the respondents reported that they had previously worked for six or more months in rural areas. This is a reasonable level of exposure to conditions in rural areas, in light of the relative paucity of health care facilities there. In addition, given that Nigeria is predominantly rural (75% of the population lived in rural areas in the early 1990s15) and that most respondents were born in rural areas and maintain ties through regular visits to these areas, most urban-based health professionals have some knowledge of the issues and conditions concerning the provision of abortion in rural areas.
The respondents were highly educated. Some 42% had medical degrees, 33% had master's or doctoral degrees, 10% had nursing diplomas and 8% had bachelor's degrees; 7% (including one with less than a bachelor's degree) had other qualifications.
Despite the small sample size, the respondents represented 22 of Nigeria's 30 states and the Federal Capital Territory. Eighteen were from the Northeast, 23 from the Northwest, 11 from the Southeast and 15 from the Southwest. This geographic spread provides some basis for generalizing from this study to the country as a whole.
Results
Methods Used to Induce Abortion
Respondents were given a comprehensive list of methods of inducing abortion and were requested to check all that they believe are used. Dilation and curettage was checked by almost all respondents. Manual vacuum aspiration, hormonal methods and traditional methods were cited by more than half of respondents. Misoprostol (a prostaglandin marketed under the name Cytotec) was checked by slightly more than one-third of respondents, and all other methods (e.g., massage and insertion of a catheter or other object into the vagina) by fewer than one-third.
Another question asked which methods are most commonly used by physicians and by nonphysicians. As may be expected, respondents think that these two major categories of providers differ markedly in the techniques they use to induce abortion (Table 1). Dilation and curettage was mentioned by more than three-quarters of respondents as being most widely used by physicians. Other methods believed to be commonly used by physicians are vacuum aspiration (mentioned by more than one-quarter of respondents) and prostaglandins and injectables (about one in seven).
Nearly half of respondents think that nonphysicians favor herbal solutions and indigenous medications. Close to one in five think that nonphysicians also use hormones and other drugs, and about one in seven say they use injectables. This finding probably reflects the relative ease with which prescription drugs are regularly sold over the counter. However, although a substantial proportion of respondents listed Cytotec as a known method, it was not reported as commonly used by either nonphysicians or physicians. The introduction of objects into a woman's vagina was mentioned as a method used by nonphysicians only.
Type of Provider
Abortion providers were classified into five categories: physician; nurse or midwife; traditional provider (birth attendant, herbalist, healer or other lay practitioner); chemist; and the woman herself (when the abortion is self-induced). The last two categories probably overlap because some women induce their abortions by using drugs purchased from chemists.
Since women's place of residence and relative affluence are important factors in the choice of an abortion provider, respondents were asked to estimate the proportion of induced abortions performed by each type of provider, for each of the four subgroups of women. Overall, slightly more than half of women who seek abortions are thought to go to medically trained providers (doctors, nurses or midwives), about one in six to traditional providers and nearly one-fourth to chemists; fewer than one in 10 are believed to attempt to induce the abortion themselves (Table 2). The proportion expected to seek abortions from medically trained providers is substantial, but a recent study found that an even higher proportion of women who reported having had an abortion said they had obtained it from a medically trained provider (about 80%).16
As might be expected, respondents believe that among relatively well-off women who are seeking an abortion, a high proportion in both urban and rural areas go to a physician (two-thirds and two-fifths, respectively). Respondents also think that next to trained physicians, nonpoor women are most likely to go to a nurse or midwife. However, some nonpoor women (one in five in urban areas and two in five in rural areas) are thought to seek abortions from traditional providers or chemists, or to induce their abortions themselves.
About half of poor women in urban centers are also believed to go to a physician, nurse or midwife to obtain an abortion. Thus, it is not always true, as some anecdotal evidence suggests, that poor women, presumably unable to afford physicians' fees, invariably resort to traditional providers, and invariably face the increased risks of an abortion performed under dangerous and unsanitary conditions.
Poor women in rural areas are believed to be the least likely of all four subgroups to go to trained medical professionals: About one-third are expected to do so. They are thought to be more likely to seek an abortion from a traditional provider or a chemist (more than half do so) or to induce the abortion themselves (about one in nine).
Poor rural women in the Southwest are thought to be somewhat more likely than those in other parts of the country to utilize medically trained providers for induced abortion: About half are believed to do so, compared with one-third or fewer elsewhere. This is partly because of the greater access to doctors in this region.
Cost of an Abortion
Overall, the cost of an abortion is believed to be lowest if it is done by a traditional provider and highest if it is performed by a doctor. Respondents' estimates of the average cost of an abortion for poor women in urban areas range from about 160 naira (roughly US $2) for herbs purchased from a traditional provider to about 1,900 naira (approximately US $24) for an abortion performed by a doctor at a private clinic. Even the lowest price is a significant amount to a poor woman: For example, 160 naira is the price of basic meals for two days for one person.
Nonpoor urban women are believed to pay about twice as much as their poor counterparts for an abortion, regardless of the type of provider—for example, about 260 naira for herbs and about 4,000 naira for the service of a doctor at a private hospital. The pattern of costs and the differences among poor and nonpoor women is similar in rural areas. This finding is plausible and may be partly a result of providers' charging more to those whom they consider better able to pay, as well as better-off women's paying more for a higher quality and more confidential health service.
Abortion-Related Complications
In assessing the hazards that clandestine abortion poses to a woman's health and life, we considered both extremely serious complications (excessive loss of blood, lacerations or tears in the vagina or cervix, perforation of the uterus, and sepsis or septic shock) and incomplete abortions, which are usually identified by heavy bleeding and present a somewhat less severe health risk but may require hospitalization. Respondents were presented with a list of these conditions and were asked: "Out of 10 poor [nonpoor] women in urban [rural] areas who have an abortion performed by each type of provider ..., about how many would experience a medical complication that should receive medical treatment?"
Respondents perceive that the procedure is likely to have the highest risk of a medical complication if it is performed by traditional providers or chemists, or is self-induced: More than half of women using these means are expected to have a complication. The proportion of women expected to have complications is also quite high (up to two-fifths, depending on the subgroup) when the procedure is performed by a nurse or midwife. Even among women obtaining abortions from physicians, as many as one-fifth are thought to experience complications.
The study participants estimate that slightly more than two-fifths of all Nigerian women having abortions experience a medical complication (Table 3). This proportion is lowest (about one-fifth) among nonpoor urban women, substantially higher (about one-third) among nonpoor women in rural areas and even higher among poor women (two-fifths to one-half).
Respondents see relatively little regional disparity. They feel that women in the North are only slightly more likely than those in the South to have medical complications. The fact that poorer women and women in the North tend to obtain abortions from traditional providers explains their slightly higher risk of medical complications. By comparison, nonpoor and southern women are more likely to use trained physicians, minimizing their risk of complications.
Hospitalization Following Abortion
Nationally, respondents estimate that almost half of women who experience an abortion complication are treated in a hospital (Table 3). The chance that women with medical complications will be treated in a hospital depends on their relative affluence and whether they live in an urban or rural area. About two-thirds of relatively well-off women in both urban and rural areas are considered likely to be hospitalized if an abortion results in a complication, compared with only about one-half of urban poor women and about one-third of rural poor women. The very low proportion for poor rural women results partly from the fact that the majority of secondary and tertiary health facilities and physicians are located in cities and towns. This pattern is found in both the North and the South, and differences in this measure are relatively small between the two regions.
The overall probability of hospitalization among women obtaining abortions reflects the effect of both the level of unsafe abortion and access to hospitals. Thus, we estimate this probability by combining two sets of information: the proportion of women expected to have a complication and the proportion expected to be hospitalized.
Respondents expect that almost one-fifth of all Nigerian women obtaining abortions are hospitalized because of a medical complication (Table 3). The probability of being hospitalized ranges from about one in seven among nonpoor urban women who have an abortion to almost one in four among nonpoor rural women. About one-fifth of poor women in both rural and urban areas who obtain an induced abortion are expected to be hospitalized.
The relatively low level of hospitalization for nonpoor urban women reflects that they are more likely than any other subgroup to use physicians, and are therefore likely to have fewer medical complications. By comparison, the higher levels of hospitalization among nonpoor rural women are probably due to different factors in each region: in the North, higher proportions of women obtaining an unsafe abortion, given the more common use of traditional providers; and in the South, a combination of relatively easy access to hospitals and greater use of safe sources of abortion.
Postabortion Counseling
Evidence from the literature indicates that many Nigerian women have had more than one abortion, heightening their risks of morbidity and mortality.17 Postabortion contraceptive counseling is an important factor in preventing repeated abortions. The survey asked respondents whether they think that most, some or few women having abortions receive any counseling from the provider, and whether they think that most, some or few of those hospitalized for complications receive counseling from hospital staff.
An overwhelming proportion of respondents think that most women having an abortion are not advised about contraceptive sources and use by the abortion provider. About one in 10 believe that most women receive counseling from their abortion provider. Two-thirds think that some or most women in the South receive postabortion counseling, compared with one-third in the North; the difference may be attributable to the higher proportion of women in the South who go to medically trained professionals for abortions.
On the other hand, respondents believe that in all regions, women who are hospitalized for complications are more likely to receive counseling. Close to half think that most of these women obtain counseling from hospital staff, and almost another third think that some do.
Women's Reasons for Having Abortions
The general consensus among respondents is that the most frequent cause of abortion is an out-of-wedlock pregnancy. This reflects the widespread belief in Nigeria that childbearing should take place only within marriage. Other reasons that are considered frequent causes of abortion are that the woman is too young or the pregnancy was unplanned. Consistent with the findings of earlier, hospital-based studies,18 respondents perceive that the typical woman seeking an abortion in Nigeria is young and unmarried and has an unwanted pregnancy. These reasons pertain to women in all four health zones.
Another important reason why a woman may seek abortion is her inability to support a child. More than half of the respondents think that a woman may have an abortion for this reason. It is noteworthy that in the North, where cultural norms and religious beliefs reduce the chances that women will be working for pay, about two-fifths of respondents think that women have abortions for economic reasons. In contrast, in the South, where a high proportion of women have major financial responsibility for themselves and for their children, and are engaged in remunerative jobs, almost three-quarters of respondents believe that economic reasons lead to induced abortions.
No respondents believe that medical problems with the fetus lead women to seek abortions. This may be an indication that facilities for detecting medical abnormalities in a fetus are not readily available in Nigeria.
Support for Policy Changes
An overwhelming proportion of respondents—all of those in the South and nine in 10 of those in the North—characterize abortion as a serious or severe public health problem in Nigeria. Survey participants nearly unanimously recommend that the risks of unsafe abortion be publicized, with the hope of reducing its level. Almost all respondents also support postdelivery and postabortion contraceptive counseling, and improved sexuality education in schools.
A large proportion of respondents recognize the importance of improving family planning services. However, while all participants in the Southeast and Southwest support this recommendation, only two-thirds in the Northwest and about eight in 10 in the Northeast agree.
Existing laws specify severe penalties both for abortion providers and for women obtaining abortions, but legal action is rare. No respondents in the South think that the existing laws should be enforced, and in the North, only about one in seven recommend strict enforcement. Yet, only about half of respondents recommend that the law on abortion be liberalized or that the provision of abortion be made legal. Support for liberalizing the abortion law is lowest in the Southeast (about one in three), where the majority of residents are Roman Catholic. However, support is also relatively low in the Northwest (about half), probably because of the influence of the Hausa culture and its interpretation of the Muslim religion.19
Discussion
Unsafe abortion is a serious health problem in Nigeria, and one of long standing. Studies in the 1970s that focused on the costs of hospitalization to women's health and to the public health system demonstrate that unsafe abortion was already a problem of some magnitude at that time.20 Research on women's beliefs and attitudes concerning abortion and the types of methods that have been used to induce abortion confirm that the procedure was historically known and practiced in Nigeria.21 A 1994-1995 community study found that in each of two local government areas, Jos (in the North) and Ife (in the South), about 11% of women reported ever having had an abortion;22 since many women probably do not acknowledge their abortions in face-to-face interviews, the true proportion is probably substantially higher.
Recent increases in urbanization and improvements in education are changing the context of reproductive behavior and perceptions about planning pregnancies and births, and may lead to an increase in pregnancies that are considered unplanned. Small-scale studies point to a high likelihood of increased use of abortion by unmarried adolescents who accidentally become pregnant but wish to finish school.
Although economic circumstances have been less of a focus, an increasingly important reason to seek abortion may be the desire to improve one's standard of living, combined with the inability to afford to bring up an additional child; this is most likely in the South, where women play a greater role in supporting the household. Research in a northern Nigerian Muslim community suggests that in communities where births have traditionally been spaced at least two years apart and women have used abortion as one means to achieve such spacing, the use of abortion may decrease, at least in the short run, as women gain some education and reject such a large birth interval as an old-fashioned norm, although it is the most beneficial for the health of women and their infants. However, the use of abortion by schoolgirls is likely to be increasing in this region.23
Even though the Nigerian pattern of differing access to safe abortion according to women's residence and socioeconomic circumstances is intuitively plausible and expected, its consistency with findings in other regions of the world where abortion is legally restricted24 adds validity to the results. The proportion of poor urban women seeking abortions who are thought to go to a doctor is slightly higher in Nigeria (one-third) than in Latin America (two in 10) or in some Southeast Asian countries (one in 10 in the Philippines, Myanmar, Laos and Cambodia). The profile of providers for relatively well-off urban women in Nigeria is similar to that for the regions of South Central and Southeast Asia as a whole, but again, the proportion of women who are believed to go to doctors or trained nurses is higher in Nigeria (more than three-quarters) than in some Asian countries where abortion is legally restricted (about half in the Philippines, Myanmar, Laos and Cambodia). However, the proportion of nonpoor urban women who go to physicians is somewhat higher in Latin America (eight in 10) than in Nigeria (two-thirds).
Medically trained practitioners, including private doctors, have probably played an increasing role in abortion provision in Nigeria over the past 20-30 years: Despite the legal restrictions, trained medical practitioners perform the procedure because the demand for it is increasing, awareness of the risks of unsafe abortion is growing, and the practice is clandestine and is rarely prosecuted.25 The fact that medical personnel may be underemployed or underpaid increases the likelihood of their providing abortions. The 1994-1995 study in Jos and Ife found that about 80% of women who reported ever having had an abortion said they had gone to a private doctor; most of the remainder said they had attempted to induce the abortion themselves.26
By comparison, the health professionals in our survey believe that on average in Nigeria, only a little more than half of women seeking abortions go to medically trained providers, and the rest go to traditional providers or chemists, or induce the abortion themselves. While health professionals may exaggerate the extent of unsafe provision of abortion, women who report their abortions in a face-to-face interview may be better educated, and therefore more likely to have gone to a medical doctor, than women who do not report their abortions. The actual profile of abortion provision in Nigeria may lie between the pictures presented by these two studies.
Information from the 1990 NDHS suggests that low levels of knowledge of modern contraceptives and method sources, as well as poor access to sources and supplies, contribute to women's need to seek abortion. According to the survey, only about 44% of all women aged 15-49 knew of at least one modern contraceptive method, and only 33% knew of a source (Table 4); the proportions were lower still (35% and 29%, respectively) among women aged 15-19.27
Only 9% of all women aged 15-49 and 4% of teenagers had ever used a modern method; about 4% of all women and 2% of adolescents were using one at the time of the survey. It is encouraging, however, that 31% of sexually active unmarried adolescent women were using a method, even though only 26% of users were relying on an effective modern method.28 Levels of knowledge and use are substantially higher in the South than in the North, but even in the South, only 12% of married women were using a modern method at the time of the survey.
The total fertility rate in Nigeria remains high—six births per woman nationally, although it is substantially lower in the South than in the North. Relatively few births are characterized as unplanned. Overall, 12% of NDHS respondents reported that their most recent birth had been unwanted or mistimed. By comparison, in the 1994-1995 study of Ife and Jos, 20% of women of reproductive age reported having had at least one unwanted pregnancy.29 However, very high proportions of married NDHS respondents aged 15-49 either wanted no more children or wanted to delay the next birth by two or more years (15% and 49%, respectively). Combined with low levels of contraceptive use, these high levels of desire to postpone or prevent childbearing mean that 28% of all women 15-49 have an unmet need for effective contraception.
With new attention being given to reproductive health, social concern about unsafe abortion has been expressed in many ways. Professional medical organizations have focused on this issue, and the Nigerian Medical Association established the Campaign Against Unwanted Pregnancy about 20 years ago. Nigerian physicians have spearheaded initiatives to modify the existing legislation concerning the conditions under which abortion is allowed, though without success so far.30 Other groups that focus on improving health, researchers in large medical teaching hospitals and universities, public health activists and women's organizations are trying in varied ways to bring public attention to unsafe and clandestine abortion. Areas in which these groups are active include improving medical services for the treatment of women with abortion complications, providing contraceptive counseling and services for women who have had an abortion, expanding contraceptive service provision and, possibly, broadening the conditions under which abortion is permitted.
This survey has made available more detailed information on induced abortion practices at the national level than has been available so far. The findings may be useful in guiding policy formulation and the development of programs that will decrease the adverse consequences of clandestine, unsafe abortions for women and society in Nigeria and that will improve women's ability to prevent unplanned pregnancies. There is great need for more research on abortion, especially at the national level, but also in-depth studies covering smaller areas. As social and economic conditions, as well as the provision of abortion services, in Nigeria change, the need for research that is representative of a broad cross-section of women in the country becomes even greater.