THEME PAPERS
Making abortions safe: a matter of good public health policy and practice
Eliminer les risques liés à lavortement devoir dune bonne politique de santé publique
Abortos sin riesgo: un componente indispensable de las políticas y prácticas adecuadas de salud pública
M. Berer
Editor, Reproductive Health Matters; and Chairwoman, Gender Advisory Panel, Department of Reproductive Health and Research, World Health Organization
ABSTRACT
Globally, abortion mortality accounts for at least 13% of all maternal mortality. Unsafe abortion procedures, untrained abortion providers, restrictive abortion laws and high mortality and morbidity from abortion tend to occur together. Preventing mortality and morbidity from abortion in countries where these remain high is a matter of good public health policy and medical practice, and constitutes an important part of safe motherhood initiatives. This article examines the changes in policy and health service provision required to make abortions safe. It is based on a wide-ranging review of published and unpublished sources. In order to be effective, public health measures must take into account the reasons why women have abortions, the kind of abortion services required and at what stages of pregnancy, the types of abortion service providers needed, and training, cost and counselling issues. The transition from unsafe to safe abortions demands the following: changes at national policy level; abortion training for service providers and the provision of services at the appropriate primary level health service delivery points; and ensuring that women access these services instead of those of untrained providers. Public awareness that abortion services are available is a crucial element of this transition, particularly among adolescent and single women, who tend to have less access to reproductive health services generally.
Keywords: abortion; maternal mortality and morbidity; public health, law and policy, abortion service provision; abortion training; midwifery
RÉSUMÉ
Selon lOMS, 25% environ de toutes les grossesses dans le monde sont interrompues volontairement, soit environ 50 millions chaque année. Vingt millions de ces avortements se déroulent dans des conditions dangereuses, soit que les personnes qui les pratiquent nont pas la formation voulue, soit quelles utilisent des méthodes à risque, voire les deux à la fois. La mortalité liée aux avortements représente au moins 13% de la mortalité maternelle dans le monde. La prévention de la morbidité et de la mortalité dues aux avortements dans les pays où elles restent élevées relève de la santé publique et elle fait partie des initiatives pour une maternité sans risque. Le présent article examine les changements à apporter à la politique générale et à la prestation des services de santé pour éliminer les risques associés aux avortements. Il sinspire darticles publiés dans les pays en développement où ces questions ont été eéudiées et où certains de ces changements ont été effectués.
Bien que la fécondité ait baissé de moitié ou presque dans les pays en développement depuis les années 60, la volonté de contrôler et despacer les naissances sest développée plus rapidement que la pratique de la contraception. La baisse de la fécondité est parfois imputée à la seule efficacité de la contraception mais linterruption volontaire de grossesse est un important facteur complémentaire. Au-delà de leur existence en tant quinstruments juridiques, les bonnes lois et les bonnes politiques témoignent de la reconnaissance publique de la nécessité pour les femmes de lavortement, au même titre que la contraception. Elles traduisent le respect pour la responsabilité des femmes en tant que mères et la conscience du coût des avortements dangereux, non seulement pour les femmes mais aussi pour leurs enfants, leur partenaire et leur famille.
Lélimination des risques liés à lavortement passe par la légalisation de lavortement. Si des mesures visant à réduire les risques liés aux avortements ont été prises dans de nombreux pays avant ou sans que la loi ait été modifiée, la sécurité nest garantie quune fois la loi révisée. La « sécurité » ne se limite pas aux procédés médicaux utilisés par chaque praticien. Elle consiste aussi à écarter tout risque de sanction contre les femmes et les praticiens, même là où lavortement illégal est sciemment toléré par les pouvoirs publics.
Toute une gamme de mesures positives ont été prises en vue de réduire la mortalité et la morbidité associées aux avortements par de nombreux pays. Pour être efficaces, ces mesures doivent tenir compte des raisons pour lesquelles les femmes avortent, du type de services requis et à quel stade de la grossesse, du type voulu de dispensateurs de soins, et des aspects relatifs à la formation, au coût et au conseil. Trois conditions essentielles doivent être remplies : des changements doivent être apportés au niveau de la politique nationale ; une formation doit être dispensée aux personnels appelés à pratiquer les avortements et les services doivent être assurés au premier niveau approprié des services de santé ; il faut veiller à ce que les femmes sadressent à ces services et non à ceux de praticiens non qualifiés. Il est indispensable que le public soit informé de lexistence des services pratiquant les avortements, en particulier les adolescentes et les femmes célibataires, qui ont moins accès aux services de santé génésique en général.
La santé dune femme est dautant moins exposée et la tâche du praticien dautant moins compliquée que lavortement est pratiqué à un stade précoce de la grossesse. Les réglementations qui tendent à retarder lintervention sont à éviter ; lavortement, par exemple, doit être autorisé à la demande dune femme, en labsence de tout autre intermédiaire. Avec la formation voulue, les infirmières/sages-femmes seraient les personnes les plus indiquées pour pratiquer les avortements. Il faudra former des instructeurs, fournir le matériel nécessaire et assurer une formation aux techniques daspiration et à la façon de pratiquer les avortements prescrits pour raison médicale. Dans les pays industrialisés, les femmes qui sollicitent un vortement ont rarement besoin dun véritable « conseil » pour ce qui est de la décision davorter mais elles ont besoin dinformations. Des informations doivent ainsi leur être fournies sur la méthode choisie, sur la reprise des rapports sexuels, sur la prévention de la transmission du virus de limmunodéficience humaine (VIH) et des autres maladies sexuellement transmissibles et sur divers moyens de contraception et lusage du préservatif.
La pratique des avortements dangereux se caractérise par linégalité du coût, de la sécurité et de la qualité des soins. La sécurité signifie laccessibilité financière pour les femmes les plus pauvres comme pour celles qui ont la possibilité de payer, avec des soins de qualité pour toutes. La prise en charge du coût des avortements pratiqués en toute sécurité dans les services de santé publics ne constitue pas en soi une dépense nouvelle ; elle consiste à remplacer les dépenses liées aux complications soignées dans les hôpitaux du niveau tertiaire par des actes simples pratiqués en toute sécurité dans des dispensaires accessibles localement.
Beaucoup de progrès peuvent être accomplis avec des médecins et des hôpitaux, des promoteurs de la santé des femmes et des tribunaux déterminés, sans même que les lois changent. Mais la responsabilité déliminer les risques liés aux avortements est un aspect de la santé publique qui relève avant tout des gouvernements. La réduction de la mortalité maternelle due aux avortements dangereux fait partie de lengagement international pris au Caire en 1994, et réaffirmé en 1999 (Le Caire + 5). Le présent article décrit les mesures pratiques à la disposition des pays pour réaliser ce changement.
RESUMEN
La OMS ha calculado que, a nivel mundial, un 25% de los embarazos terminan en un aborto provocado, lo que representa aproximadamente 50 millones de abortos cada año. De ellos, 20 millones son practicados en condiciones peligrosas, debido ya sea a la intervención de personas no adiestradas, al uso de técnicas peligrosas o a ambas cosas. La mortalidad por aborto representa al menos el 13% de la mortalidad materna a nivel mundial. La prevención de la mortalidad y la morbilidad por aborto en los países donde esas tasas siguen siendo altas es una responsabilidad de salud pública y un componente de las iniciativas de Maternidad sin Riesgo. Este artículo analiza los cambios que es necesario introducir en las políticas y la prestación de servicios de salud para acabar con los abortos peligrosos. Se basa para ello en un examen de artículos de países en desarrollo en los que se ha estudiado este problema y en los que se han introducido algunos de esos cambios.
Aunque la fecundidad se ha reducido casi a la mitad en los países en desarrollo desde los años sesenta, el interés por controlar y espaciar los nacimientos ha aumentado más rápidamente que las tasas de utilización de métodos anticonceptivos. La disminución de la fecundidad se atribuye a veces únicamente a unas prácticas anticonceptivas eficaces, pero el aborto provocado es un complemento importante. Unas leyes y políticas adecuadas, además de proporcionar un instrumento jurídico, demuestran la aceptación pública del hecho de que las mujeres necesitan poder recurrir al aborto, así como a métodos anticonceptivos; demuestran también respeto hacia la responsabilidad de las mujeres como madres, y conciencia de los costos que representan los abortos peligrosos, no sólo para las mujeres, sino también para sus hijos ya nacidos, su pareja y su familia.
Legalizar el aborto es una condición esencial para hacerlo más seguro. Aunque en muchos países la tendencia a la disminución de la peligrosidad del aborto comenzó antes de que se cambiaran las leyes, o sin necesidad de tales cambios, la modificación de la legislación es indispensable para mantener la seguridad lograda. Por seguridad no se entiende sólo la utilización de procedimientos médicos sin riesgos. Se trata también de evitar que se adopten medidas punitivas tanto contra las mujeres como contra el personal sanitario, incluso cuando el Estado tolera el aborto ilegal.
Se han tomado diversas medidas para reducir la mortalidad y la morbilidad por aborto en un número creciente de países. Para ser eficaces, tales medidas deben tener en cuenta las razones que llevan a las mujeres a recurrir al aborto, el tipo de servicios que deben prestarse y en qué fases del embarazo, el tipo de dispensadores de atención necesarios, y diversos aspectos de la capacitación, los costos y el apoyo psicológico. Se requieren tres acciones cruciales; a saber : introducir cambios en la política nacional, adiestrar en las prácticas de aborto a los dispensadores de asistencia oportunos y ofrecer servicios en los puntos apropiados del sistema de atención primaria, y asegurar que las mujeres utilicen esos servicios y no recurran a prestadores de asistencia no capacitados. La sensibilización del público respecto a la existencia de servicios de aborto es fundamental, sobre todo entre las adolescentes y las mujeres solteras, cuyo acceso a los servicios de salud reproductiva es en general menor.
Cuanto más al comienzo del embarazo se provoque el aborto, menos riesgo supondrá para la salud de la mujer y menos complicado resultará para quien lo practique. Deben evitarse las medidas de regulación que tiendan a retrasar el procedimiento; por ejemplo, la intervención se debe poder realizar simplemente a petición de la interesada, sin la tutela de ninguna otra persona. Con el adiestramiento oportuno, las enfermeras-parteras serían el personal más adecuado para esos fines. La capacitación de los instructores, el suministro de equipo de aspiración por vacío y la enseñanza de su manejo y de la manera de realizar abortos médicos... todo ello es necesario. En los países desarrollados, pocas de las mujeres que quieren abortar necesitan en realidad «apoyo psicológico» para decidirse a hacerlo, pero sí necesitan información. Ello incluye la información relativa a los métodos abortivos y la elección de uno de ellos, los consejos sobre la reanudación del coito, la prevención de la propagación del virus de la inmunodeficiencia humana (VIH) y de otras enfermedades de transmisinón sexual, y la elección de anticonceptivos y preservativos.
El aborto peligroso se inscribe en una situación de falta de equidad en el costo, falta de seguridad y mala calidad de la atención, mientras que el aborto sin riesgos es accesible tanto para las mujeres más pobres como para las que pueden costearlo, y se asocia a una alta calidad de la atención para todas. Para costear los abortos sin riesgo en los servicios de salud pública no es necesario crear una nueva partida de gastos, basta con destinar parte de los fondos dedicados a tratar los casos complicados en los hospitales de nivel terciario a introducir técnicas sencillas y seguras en los dispensarios locales.
Es mucho lo que puede lograrse gracias a la intervención de los médicos y los hospitales comprometidos en esas iniciativas, así como a los activistas y tribunales que defienden la salud de la mujer, incluso en ausencia de cambios legislativos. Sin embargo, convertir los abortos peligrosos en seguros es ante todo una cuestión de salud pública que incumbe a los gobiernos. La reducción de la mortalidad materna por abortos peligrosos forma parte del compromiso internacional contraído en El Cairo en 1994, y reafirmado en El Cairo+5 en 1999. El artículo esboza las medidas prácticas que los países pueden adoptar para propiciar esos cambios.
Introduction
WHO estimates that about 25% of all pregnancies worldwide end in an induced abortion, approximately 50 million each year. Of these abortions, 20 million are being performed under dangerous conditions, either by untrained providers or using unsafe procedures, or both. Deaths as a result of unsafe abortions in developing countries are estimated at 80 000 annually, i.e. 400 deaths per 100 000 abortions. This figure hides substantial regional variation, however, with unsafe abortions in Africa being at least 700 times more likely to lead to death than safe abortions in developed countries (Table 1). Although over the past 10 years there have been improvements in the safety of the abortion procedures used and access to treatment for complications for some women in developing countries, the number of women requiring treatment for serious complications of unsafe abortion remains very high and many women never receive care at all (18).
Unsafe abortion procedures, untrained abortion providers, restrictive abortion laws and high mortality and morbidity from abortion tend to occur in one and the same countries. This article outlines ways in which good public health policy and good quality medical practice can make abortions safe. It is based on a review of published and unpublished sources identified in MEDLINE and POPLINE searched, as well as articles in a range of journals, newsletters, books and other publications in the field.
Fertility decline and the need for abortion
Unplanned and unwanted pregnancies constitute a serious public health responsibility. While fertility has declined by almost half in developing countries since the 1960s (9), the motivation to control and space births has risen faster than the rates of contraceptive use. Once people decide they want fewer children, they use a combination of approaches to achieve this, including modern and/or traditional methods of contraception and abortion. Fertility declines are sometimes attributed only to effective contraceptive practice, but abortion is also an important element (1013). The extent and effectiveness of contraceptive use has an influence on abortion rates, but all contraceptive methods can fail despite correct and consistent use (14, 15). In Italy, Turkey and the United Kingdom, failure to use the withdrawal method correctly, failure of condoms, and inconsistent pill use were the commonest reasons for abortion (1618). Greece is one of the few countries where, although women have access to contraception, many prefer to use abortion as a primary method of birth control (19). This has also been the case in the Russian Federation, the former Yugoslavia (20) and many of the countries of the former Soviet Union, where lack of access to contraceptive methods as well as preference for abortion as a method have contributed to high abortion rates.
The increasing gap between age at menarche and age at marriage means that there is a longer period during which single women may have an unwanted pregnancy. Lack of access to family planning services for young and single women greatly contributes to the rate of abortions. In Argentina, Chile, Ghana, Kenya and Viet Nam, single women with unwanted pregnancies describe poor communication with partners on sexuality, poor knowledge of fertility and contraception, low contraceptive use rates and/or irregular and ineffective use of contraceptive methods (2126).
Increased use of condoms, rather than of a more efficacious contraceptive method, particularly among those who need dual protection from both sexually transmitted diseases and unwanted pregnancy, can also result in higher abortion rates (27). Higher abortion rates are often considered a negative outcome of condom use but, if rates of sexually transmitted diseases, especially human immunodeficiency virus (HIV) infection, are reduced by the use of condoms, this should be seen as a positive outcome and good public health policy, provided abortion is safe.
Most women use contraception and abortion when necessary because they want to be good mothers to the children they already have (28). Some women are simply not ready or able to have children, and a small but growing number, whose position deserves respect, do not wish to have children at all (29). Concerns about womens health, family welfare and poverty are common reasons for abortion, especially among women with several children (14, 30, 31).
In Asian countries, the rate of abortions has been directly influenced by national population policies. In China, after the one-child policy came into effect, there was a great increase in the number of abortions (32). Similarly, in Viet Nam, the abortion rate is influenced by the two-child policy, the desire for smaller families and inadequate contraceptive services (33). In China, including the Province of Taiwan, and the Republic of Korea, preference for sons also influences decisions about abortions in the context of small family norms (34).
Even in countries where contraceptive prevalence is very high, there are still unplanned pregnancies and abortions. The Netherlands, with a small family norm, has a comprehensive programme of sex education, good contraceptive and emergency contraceptive services and safe, legal abortion services: as few as 6% of first births may be unplanned and the abortion rate is one of the lowest in the world (at 6 per 1000 in 1994, compared with 26 per 1000 in the USA) yet abortion remains a necessary part of fertility control (35).
The transition from unsafe to safe abortions
A range of positive steps has been taken to reduce deaths and morbidity from abortion in a growing number of countries over the past 15 years. Since 1980, abortion laws have been liberalized in some form in Albania, Algeria, Barbados, Belgium, Botswana, Bulgaria, Burkina Faso, Cambodia, Canada, China (Province of Taiwan), Czechoslovakia, Spain, Ghana, Greece, Guyana, Hungary, Indonesia, Malaysia, Mongolia, Pakistan, Romania, South Africa, Spain, and Turkey (3643). In other countries there have been attempts to change highly restrictive abortion laws and/or major national debates on abortion (37, 4448). In Brazil, for example, Congress considered 46 bills on abortion between 1946 and 1995; 13 of 16 bills over the period 199195 were favourable towards making abortion legal under some circumstances (49).
Even in the absence of legal change, it has become easier for women to obtain treatment for abortion complications in large hospitals, at least in urban areas. The high cost and poor quality of care available in many developing country public hospitals is also being addressed. Manual vacuum aspiration techniques are finally beginning to replace dilatation and curettage as the standard of care for incomplete abortion; which in itself reduces complications (18, 50, 51). Furthermore, there are more abortion providers who have received some training and have greater awareness of safer procedures and practices so that, for example, in Bangladesh there are fewer serious complications and deaths than there were 10 years ago (7).
In some countries, women themselves are starting to use safer methods for self-induced abortions, in particular, intravaginal application of the prostaglandin misoprostol, resulting in fewer complications and shorter hospital stays. This has been well documented in Brazil (5256) and is thought to be common elsewhere in the region (57). Such changes have succeeded in reducing at least some of the more appalling examples of morbidity and mortality arising from the insertion of sticks, roots and sharp instruments into the uterus.
Measures carried out on a project basis, however, may not have an impact on public health. For example, in Latin America successful pilot projects to improve the quality of post-abortion care have not always been scaled up or sustained (58). Real progress is dependent on legal and other changes in national policy and practice.
The need to legalize abortion
Making abortion legal is an essential prerequisite to making it safe. In this respect, changing the law does matter and assertions to the contrary are ill conceived and unsupported in practice. Although, in many countries, trends towards safer abortion have often occurred prior to or in the absence of changes in the law, legal changes need to take place if safety is to be sustained for all women.
Safety is not only a question of safe medical procedures being used by individual providers. It is also about removing the risk of exposure and the fear of imprisonment and other punitive measures for both women and providers, even where illegal abortion is tolerated. Health professionals providing safe but clandestine abortion in urban Latin America have described a lack of medical support, the need for secrecy, as well as threats of violence, extortion and prosecution (5). In Bolivia and Chile, the interrogation of women seeking treatment for abortion complications in public hospitals is or has been routine (30, 58). In Nigeria, illegal abortion carries a sentence of up to 14 years imprisonment except where the life of the woman is at risk (59). Moreover, although illegal abortion has been tolerated in a number of countries over the past 10 years, arrests have been made without warning.
Safety is also about making sure that abortions will not be carried out by clandestine and unskilled providers who operate in situations that endanger womens lives, even if they have the best of intentions. A woman has little or no recourse when abortion is illegal, even if she is seriously injured, badly treated, refused pain relief, sent home in a poor condition, charged a large amount of money, or suffers any other form of negligence or malpractice. Continuing pregnancies following attempts at self-induced abortion are not uncommon and women may need follow-up care for other reasons, but they may be impossible to contact because they have given a false address (6062)1.
Good laws and policies on abortion, in addition to being legal instruments, are a sign of public acceptance of fertility control and of womens need for abortion. They signify an acceptance of the limitations of contraception and contraceptive use, and of womens right to decide the number and spacing of their children. Further, they mark respect for and acknowledgement of womens responsibility as mothers. Not least, they indicate a public health awareness of the costs of dangerous abortions, not only to women but also to their existing children, partners and families, and to health services and society as well.
Both the content of the law and the policy that defines how the law will be implemented matter. It is often in the details that service delivery is facilitated or blocked. Zambia (63) and India (64) are often erroneously cited as examples of why changing the law does not matter, as both are classified as countries where abortion is legal but where abortion mortality remains high. However, the term legal does not necessarily mean that the law is appropriate for the circumstances in which it must be implemented. Abortion mortality remains high in Zambia and India because of obstacles to putting the law into practice, including provider unwillingness, lack of training for providers, failure to authorize providers and facilities, and a lack of resources for and commitment to delivering good services at the primary care level. In Zambia, the law requires several doctors signatures for an abortion when in most places there are few or no doctors and lack of resources is an important issue. One study found that legal abortion services were inaccessible or unacceptable to schoolgirls, among whom more than half of abortion deaths were occurring, because providers did not respect confidentiality. The young women were apparently required to reveal who made them pregnant, which they were unwilling to do, and feared being expelled from school (63). Abortion was legalized in India for broad social and medical reasons in 1972, when experience in providing safe abortions was more limited. Today, many of the annual 6.7 million abortions performed in the country are still carried out by untrained providers in unapproved sites. Approved abortion clinics are concentrated in the cities (64), and are unevenly distributed. A total of 1632% of approved primary health centres in four states have never offered abortions since they lack trained providers and functioning equipment. In one state, acceptance of sterilization following abortion has been required (65), although this is not stipulated in law. Furthermore, women are often expected to attend without an appointment, and if the clinic is too busy they are told to come back another day, again without an appointment. Women may also be charged for an abortion according to the numbers of weeks of pregnancy, although the procedure is supposed to be free (T.K. Sundari Ravindran, personal communication, 1997). Thus, women may be discouraged or prevented from seeking bona fide services in ways never intended by the law.
Changing laws and policies
To make abortion safe, restrictive laws need to be annulled, amended or replaced; traditional and, in some cases, religious laws may also require attention when legal change is being contemplated (66). Countries have taken three main routes to this end: liberalizing the existing law within the penal or criminal code; partially or fully legalizing abortion through a positive law or a court ruling; and decriminalizing abortion by taking it out of the law altogether. These changes have already occurred in almost all industrialized countries and are happening in a growing number of developing countries as well.
Different laws permit different combinations of individual choice and state commitment versus state control over abortion. Sweden has been described as having an enabling law, in that the state grants freedom of choice and provides the means to implement it. Irelands law is restrictive, in that the state both denies private discretion and discourages or forbids implementation; thus, Irish women often travel to the United Kingdom for abortions, while in most countries with such laws, abortions are performed clandestinely. The USA is described as having a hindering law, in that the state grants individual choice on abortion but provides no services. Israel, on the other hand, is said to have an intrusive law, wherein the state limits individual discretion by requiring approval for all abortions, but provides all abortions that have been approved (67).
Abortion mortality and morbidity tend to be highest in countries where abortion laws are most restrictive. Many such laws originate from colonial times and are no longer operative in the countries that drew them up. Restrictive laws allow abortion only when a woman can be seen as a victim of circumstances, i.e. in a medical emergency or cases of fetal abnormality or following rape or incest (68).Yet the great majority of women need abortions for family planning reasons and on economic and social grounds.
The least fundamental form of abortion law reform is to add limited grounds for abortion to an already restrictive criminal law. In Ghana, for example, a 1960 law allowed abortion only to save a womans life, while a 1985 amendment allowed abortion to protect a womans physical or mental health as well as on juridical and fetal impairment grounds (37). However, in 1995 unsafe abortions and high abortion mortality were still common in Ghana (69); little had changed in practice.
Broader grounds for abortion, to achieve partial legalization, may be added to an existing law. For example, Malaysia now permits abortion within 120 days of conception when the continued pregnancy poses a threat to the womans life or to her physical or mental health greater than would the termination of the pregnancy (70). Since in the first trimester an induced abortion is always safer than pregnancy, such wording is open to liberal interpretation, as has occurred in the United Kingdom, although it can also be applied narrowly.
Canada is the only country to date that has decriminalized abortion entirely (37). In 1988, Canadas highest court struck down the federal law on abortion and the parliament did not replace it. Although there are abortion regulations at the state level, any recriminalization of abortion would be illegal. This represents the most complete form of normalization and de-politicization possible, bringing abortion in line with all other medical procedures and making good medical practice and quality of care in service provision the only issues involved. Any breaches of medical practice are punishable under other existing laws.
Who decides and when
The earlier in pregnancy that an abortion takes place, the safer it is for the womans health and the less complicated for the provider. Hence, on public health grounds, regulations that tend to delay the procedure should be avoided. Such regulations include putting the abortion decision into the hands of people other than the woman herself, weighting conscientious objection clauses in favour of providers who want to opt out, or requiring a waiting period between obtaining permission for and having an abortion.
In India the law requires a medical practitioners authorization for an abortion. In addition, the public health services sometimes ask women for their husbands signature of consent, even though this is not stipulated in law. In the Punjab, the High Court allowed a man to divorce his wife on grounds of cruelty because she had had two abortions against his wishes, which implies acceptance of husband consent (64, 65). In 14 countries2, the husbands authorization is required (37), and this can only be bypassed by a court or designated medical person, e.g. on health grounds. In contrast, South Africas new law stipulates that the only consent required for abortion is that of the woman herself (40). Young women are required to have parental consent for abortion in 27 countries, most commonly in eastern and western European countries, but also in China, India and some states of the USA (36). Again, consent may be waived with a courts authority, but this is very burdensome. Under the new South African law, on the other hand, a medical practitioner or registered midwife must advise a young woman to consult with her parents, guardian, family members or friends but abortion is not denied if she does not (40).
In countries where safe abortions are the norm, more than 90% of women have abortions in the first trimester of pregnancy3. Western European laws commonly allow abortion upon the womans request in the first trimester of pregnancy, while in the second trimester the permission of one or more doctors or a designated medical committee is required and/or more restricted grounds pertain. Laws of this type were passed some 2030 years ago, and along with good service delivery, have reduced abortion mortality and morbidity to a rarity (37). One of the unintended consequences, however, is the creation of aminority of excluded women who have difficulty obtaining second trimester abortions and may have to travel to another country for this. Moreover, fewer clinics offer second trimester procedures (71)4. Thus, obstacles and delays in obtaining abortions after 12 weeks often contribute to making the procedure later and more complicated than necessary.
Women who need abortions after the first trimester of pregnancy include the following: those who are not aware that they are pregnant or who deny the pregnancy until it begins to show (most often young women); those who think they are too old to get pregnant; those whose personal circumstances change dramatically during the pregnancy (e.g. the husband leaves or dies); those who develop medical reasons for abortion; and those who find out that the fetus is seriously damaged. Where abortion has previously been illegal and clandestine, women needing second trimester abortions also include those who are unaware that the law has changed, those living far from facilities, those who need more time to find out where to obtain a safe abortion, and those who have attempted self-induced abortion unsuccessfully and who have a continuing pregnancy.
Cuba is an early example of a developing country that legalized abortion on broad indications. In the context of sweeping changes in the countrys health services in 1959, a 1936 law which had made abortion legal on grounds of serious risk to a womans health was officially interpreted to encompass the WHO definition of health as a total state of well-being. Abortion services were extended to all obstetricgynaecology hospitals. In 1979, when a new Penal Code was drafted, instead of specifying when abortion was legal, it specified when abortion was illegal: if it were carried out without the womans consent or in other than hospital premises, if the provider failed to comply with established norms, or if it were carried out for profit. Since there were hospitals throughout the country providing abortions free, these conditions did not create obstacles for women. Furthermore, the law specified that menstrual regulation was not equivalent to abortion, since delay in menses may be due to causes other than pregnancy (72).
South Africas 1996 law is a more recent example, and has been accompanied by efforts to develop good service provision nationwide. However, it imposes incremental limits after the first trimester: a pregnancy may be terminated upon request of the woman during the first 12 weeks of pregnancy, but in weeks 1320 termination must be approved by one medical practitioner and, after week 20, two medical practitioners (or one medical practitioner and a registered midwife) (40). These restrictions may prove problematic, especially in rural areas.
In Bangladesh, menstrual regulation is available only up to 10 weeks of pregnancy. A 1990 study of women seeking this treatment found that almost 20% were turned away because they were over this time limit (73). Many others would not try to seek such services but go straight to an unregistered provider, probably an important source of continuing mortality and morbidity.
In Sweden, abortion is available at the womans request up to 18 weeks of pregnancy and with the agreement of a medical board after that (74). This allows almost all abortions to be the womans decision alone, a facilitating policy, which has developed on the basis of experience and an evolving awareness of womens needs on the part of medical professionals and policy-makers.
A medical board and individual medical practitioners can be either supportive or restrictive. However, by putting the decision into the hands of anyone except the woman who is seeking an abortion, countries risk perpetuating the need to seek unregistered providers and unsafe procedures, thus maintaining the public health problem they hoped to reduce. The examples of Sweden and Canada show that criminal law and complicated restrictions on abortion are not necessary. They offer unambiguous models, worth emulating.
Striking a balance
Some laws stipulate that health professionals can refuse to participate in a legal abortion on grounds of conscience. In Great Britain, this is the case unless the womans life is at risk. Moreover, no one applying for a gynaecology post in Great Britain can be asked his or her views on abortion as part of the job interview, even if abortion provision appears on the job description. However, providers may opt out for less than conscientious reasons, leaving women vulnerable and putting the onus on them to find a provider, which can be difficult and time-consuming (81). In South Africa, a draft of the new law made it mandatory for anyone who objected on grounds of conscience to refer the woman to another provider, but this was omitted in the final text (75), which says only that the woman shall be informed of her rights (40). A balanced law would protect both a true conscientious objection and a womans right to obtain a legal abortion without delay. Nevertheless, some advocates argue that health professionals have an obligation to perform all socially sanctioned medical services, including abortion (76). Similarly, it can be argued that abortions should be carried out by dedicated services and only those who are sympathetic to womens need for abortion should be employed in them, as a matter of quality of care and respect for womens feelings.
A certain proportion of women change their minds and decide to continue their pregnancies after having arranged for an abortion (77). This does not justify imposing a waiting period between arranging an abortion and having the procedure, as is the case in France and the Netherlands, where this regulation is also meant to prevent women crossing their national borders for abortions.
Where counselling is to be provided, laws may specify what it should consist of and whether it is mandatory. Counselling can be directive, to try to influence or control a womanÆs decision. Thus, anti-abortion organizations sometimes offer counselling services in some developed countries. In Singapore, the abortion law was liberalized in 1974 as part of national policy to encourage small families; in 1986, mandatory counselling was introduced in order to encourage those who could afford it to have more children, which led to a decrease in the number of abortions (78). In contrast, the aim of non-directive counselling, which is considered the most ethical form of counselling, is to help women to decide what is best for them. The new South African law says that the state shall promote non-mandatory and non-directive counselling before and after the termination of pregnancy (40).
In Viet Nam, very few women who have had abortions receive information on how to avoid future pregnancies, although they would like to have this. Some are forced to seek information elsewhere, others are left knowing as little as they did before their abortions (26). In Guyana, in contrast, the 1995 law stipulates counselling before and after abortion, stresses the importance of use of contraception, suggests including the womans partner in counselling to foster male responsibility, and spells out in detail the content of counselling (information on alternatives to abortion, methods of abortion, possible adverse effects, contraception and sexually transmitted infections, and psychosocial guidance). It even imposes a 48-hour waiting period before an abortion is carried out, to allow time for counselling (39). In the year after the law was changed, however, doctors records showed that counselling was concentrated almost exclusively on offering contraception (57). This is not surprising in that these doctors were not trained as counsellors.
Striking a realistic balance and finding out what women require is advisable. Abortion services that are openly available have the opportunity to offer family planning and sexual health information and services, to give women the means to protect themselves. In developed countries, experience has shown that few women who seek an abortion actually need counselling as regards the abortion decision, but they do need information. This includes information on and choice of abortion method before the abortion and on what happens during the procedure, information on possible complications and seeking help for these afterwards, information about resuming sexual intercourse, prevention of HIV infection and other sexually transmitted infections, and the offer of a choice of contraceptive methods. The involvement of partners should be possible, but only at the womans request, so as to protect her right to privacy.
Paying for safe abortions
Where abortion is clandestine and unsafe, women (or their partners or families) are buying drugs and other means of self-induced abortion and/or paying clandestine providers, while public health services and women are paying for the treatment of abortion complications, often in tertiary level hospitals, where costs are highest. Costs (economic and social) incurred for unsafe abortions include not only acute care, however, but also the longer-term complications of damage to reproductive organs, pelvic inflammatory disease, and secondary infertility. Moreover, the need for blood transfusions to deal with haemorrhage and other complications of unsafe abortion should be considered against a background of increasing HIV seroprevalence in many developing countries. Costs for families, especially for a womans existing children, also include those that result from a maternal death.
Unsafe abortion situations are characterized by a lack of equity in cost, safety and quality of care. In some Bolivian hospitals, women who present with signs of induced abortion are charged higher fees for treatment of complications than women who appear to have miscarried, which contributes to delays in obtaining care (58). In Egypt, as elsewhere, the price for a clandestine abortion increases in proportion to the level of safety provided (79).
Most authors agree that treating abortion complications in sub-Saharan Africa consumes a disproportionate amount of hospital resources (50), while in Bangladesh, up to 50% of hospital gynaecology beds are reported to be taken up with abortion complications (6). Women tend to wait until complications become severe before seeking help, increasing both the cost and complexity of treatment. Furthermore, women attending untrained providers have been found to make more visits for care and spend more overall than women attending trained providers in the first place. (7). A study in United Republic of Tanzania estimated that the cost per day of treating abortion complications, including the costs of drugs, meals, staying costs and surgical procedures, was more than seven times the Ministry of Healths annual per capita budget. Only 3 of 455 women were treated and discharged on the same day; 25% needed one day, almost 50% needed 2 days and the remainder needed 35 days to recover (80). In Guyana, about 25% of the blood available at the main public hospital was used to treat abortion complications before the law was changed (57).
Covering the cost of safe abortions in public health services is therefore not about incurring entirely new costs, but about shifting expenditure away from complicated cases in tertiary level hospitals to safe, simple procedures that can be provided in primary clinics. Women may or may not be charged a fee at the point of service, but safety should mean affordability for the poorest of women as well as for those who can pay, with one high standard of care for all.
Requirements for safe abortions
Most developed countries still require that gynaecologists carry out abortions, yet this is not necessary, particularly not for abortions performed under 14 weeks of pregnancy, given that the skills needed have been greatly simplified and the rate of complications is so low (81). With appropriate training, nursemidwives or those with comparable training would be the most appropriate abortion providers.
Training of trainers, provision of equipment and training in vacuum aspiration techniques, and in how to provide medical abortions are needed. In many countries, one of the consequences of the long-standing illegality of abortion is that many providers are still using dilatation and curettage and other outdated methods, which have not been in use in developed countries for many years since they have a higher rate of morbidity. In the past decade, manual vacuum aspiration, which is considerably safer and less costly, has been used to treat incomplete abortions following unsafe procedures in a growing number of developing countries (82). In Nigeria, manual vacuum aspiration has been used on an outpatient basis for most cases of abortion complications, reducing waiting time for women from 4872 hours to 1015 minutes (83). In addition to reducing mortality and morbidity from unsafe abortions (84), vacuum aspiration can also be used for safe, early abortions up to 14 weeks of pregnancy preventing the consequences of unsafe abortions altogether. In South Africa, the training of midwives in manual vacuum aspiration is a key activity of the new national abortion programme. (85).
The guidelines for training midwives prescribe a 160-hour course combining theory and clinical practicals. The curriculum includes an overview of the [law] and the problem of unsafe abortion, professional practice and ethics, communication skills and counselling techniques, patient assessment and preparation, pharmacology, the MVA technique, infection control, management of abortion complications, post-abortion family planning, emergency contraception, identification and treatment of sexually transmitted infections (STIs) and strategies for dual protection against unwanted pregnancy and STIs.... (85).
First trimester abortions using surgical or medical methods5 can be provided on an outpatient basis in primary care facilities; newer second trimester methods also require less skilled management than in the past (e.g. intravaginal misoprostol in weeks 1222 of pregnancy) (86, 87). Where prevalence of sexually transmitted infections is high among women seeking abortion, prophylactic antibiotic treatment prior to surgical abortion can prevent infection of the upper reproductive tract (88).
A study comparing medical abortion using mifepristonemisoprostol with early surgical abortion in China, Cuba, and India found medical abortion to be safe, efficacious and acceptable under a range of conditions (89). Fully established services for routine surgical abortion are not required prior to introducing medical abortion, although vacuum aspiration is a necessary back-up to both first and second trimester medical procedures for the small number of cases of incomplete abortion (90). It has been persuasively argued that medical abortion can be largely self-administered as long as the woman considers the method acceptable, is early enough in pregnancy (up to 9 weeks LMP), can adhere to the protocol, is able to manage minor adverse reactions and seek help for more serious ones, can notice and cope with the expulsion of the embryo, and can recognize a complete abortion, return for a follow-up visit or use a home pregnancy test (91).
Along with safe methods and trained providers, programmes require locally accessible services in both rural and urban areas. In Zambia, gynaecologists were found to be a major obstacle to the setting up of safe abortion services (92). The ambivalence of doctors was also found to have hampered the implementation of a revised abortion law in Indonesia (41). In Bangladesh and India, untrained providers, who are often more easily accessible in rural areas, have never actively been stopped from practising (7, 64). In South Africa, in contrast, as cases have arisen where unlicensed providers have continued to offer services despite the changed legal status of abortion, criminal charges have been pressed against them. A nationally coordinated programme aims to ensure that throughout the country, primary and secondary care facilities are prepared to perform abortions. All nine provinces in South Africa have developed provincial plans, each in collaboration with a medical school or a tertiary training hospital. A national advisory group has been set up to coordinate and monitor implementation of the new law, including health service managers, representatives from medical schools, academics and specialists, the nursing council, researchers and the nongovernmental sector. This group plans to meet every 46 months and make recommendations to government on relevant issues (85).
Quality of care
Bringing abortion services out into the open is a precondition for ensuring quality of care, accessibility, availability and affordability, especially for the poorest women. This encourages health professionals to provide a defensible service. In Guyana, for example, although most clandestine abortion providers were medical professionals before the law was changed in 1995, septic abortion was the third highest cause (19%) of hospital admissions. After the law changed, this same group of providers organized themselves and voluntarily began to give prophylactic antibiotics. Admissions to the main public hospital for septic and incomplete abortions fell by 41% within 6 months of this decision (57).
Public visibility in service provision means that women will have a more open choice of providers and can take action if their rights are violated or care is substandard; legalization also ensures that providers who attempt to sexually molest clients, anecdotally a not uncommon problem for women seeking clandestine abortions, can be prosecuted (57). Sympathetic treatment on the part of service providers is important. Uncaring treatment and verbal abuse on the part of health care staff towards women seeking treatment for complications of clandestine abortion has been well documented in Latin America (29, 51). Indeed, lack of sympathy is a problem in many countries, even those with safe abortion services such as Great Britain and the USA (81, 93), where abortion has been legal for 30 years. South Africa is trying to confront this problem through workshops for service providers, to clarify values and increase empathy and respect for women with unwanted pregnancies (85).
Other ways to monitor and ensure quality of care in developing (57, 85) and developed countries alike include the following: oversight by an independent national advisory body, decisions as to whether or not the procedure will be covered by national health insurance, the standards that approved institutions must meet, regulation of fees for services and requirements for record-keeping, and the collection of data. In Guyana, data collection includes relevant demographic information about the woman, length of pregnancy, reasons for abortion, type of procedure, any complications and whether and what kind of contraceptive method was provided. Where deaths from dangerous abortions were previously high, and to ensure that the mortality rate declines towards zero, the collection of baseline data (57) and regular audit of all reported abortion-related deaths, as part of broader maternal mortality audits, will reveal continuing risks, allowing discussion and action to reduce these6.
Guidelines for health service professionals are valuable for ensuring equity of access and quality of care. The United Kingdom Royal College of Obstetricians and Gynaecologists is currently preparing evidence-based guidelines that cover organization of services, information for women, pre-abortion assessment, abortion procedures, management of complications and after-care (94).
Raising public awareness
Although abortion has been legal in India since 1972, interviews with 67 women in rural Maharashtra in 1997 found that only 18% knew that this was the case, while 64% thought it was not, and the remainder were unsure. Even those who knew it was legal sometimes had inadequate or incorrect information about whether husbands consent was needed eligibility for abortion, and the time limits within which abortion is permitted (64).
In Puerto Rico, although abortion has been legal for 20 years (a consequence of its common-wealth status with the USA), there is still a widespread perception that abortion remains illegal. Public information on where women can get an abortion is very limited, and clinics still use euphemisms to suggest that abortion services are provided. Medical students also know very little about what is permitted (95).
In Mozambique, although abortion has not been legalized as such, safe terminations have been available at the main hospital in Maputo since 1981 in order to reduce mortality from unsafe abortions. However, a study reported in 1997 found that young women who had recently migrated to the city, who did not have a steady partner, who were from poorer socioeconomic groups, who did not use contraception and who had no previous abortions were less likely to know that they could obtain a safe abortion at that hospital. They were more likely to seek clandestine abortions and experience complications, for which they went to that same hospital. Thus, the women who were most at risk were also those who had the least information (96).
Hence, widespread public awareness is an important component in making abortion safe where it has previously been unsafe; women need to know that safe abortions are not only permitted but available.
Conclusions
Much can be done despite the difficulties of changing national abortion laws. Womens health groups and other advocates, parliamentarians and health professionals, can work together to support the right of women not to die from unsafe abortions and to ensure that they receive treatment for complications. They can urge hospitals not to report women and legitimate service providers to the police, as well as advocate for the decriminalization of abortion. In countries where the letter of the law is not a primary obstacle, they can also campaign for a choice of safe abortion methods, improvements in regulations governing the registration of providers and facilities, and for better training for providers. Additionally, they can monitor accessibility, affordability, and quality of care in these services (97, 98).
Even where legal change has not taken place or is not likely to happen quickly, improvements in the abortion methods used and in the responsiveness of providers to the demand for safer abortions can improve the situation to some extent. Committed doctors can make an important difference by providing treatment for abortion complications (99), interpreting the law in a liberal way and providing safe services where these are legal (92), and training providers in the safest techniques to reduce mortality and morbidity (96, 100). Furthermore, courts can pass judgements that result in shorter prison sentences and fewer prosecutions of women for having abortions, and help to open the way to law reform. This occurred in Nepal in the 1990s (101) as it did in countries such as Spain in the 1970s.
Abortion law reform is a necessary condition for making abortion safe, though it is not sufficient in itself. Women remain vulnerable where safe abortion is not legally sanctioned because quality of care cannot be assured, abuses cannot be challenged and both women and providers remain at risk of prosecution, blackmail, and social and professional stigma. The dedication of individuals to providing treatment for abortion complications or safe abortions in a context of clandestinity, important as it is, cannot make up for the absence of a legal framework and national programmes. In the long run, abortion needs to be decriminalized in order for it to be made safe.
Although law, policy and womens rights are central to this issue, making abortions safe is above all a public health responsibility of governments. Moreover, reducing maternal mortality by making abortions safe is also an important part of the international commitment made in Cairo in 1994 at the UN International Conference on Population and Development (ICPD) and re-affirmed at the Cairo+5 meeting in 1999. The practical steps to bring about the changes outlined in this article could be achieved by most countries in a few short years once they have committed themselves to making abortion safe. 
References
1. Unsafe abortion: global and regional estimates of incidence and mortality due to unsafe abortion. Geneva, World Health Organization, 1998 (unpublished document WHO/RHT/MSM/97.16).
2. Sharing responsibility: women, society and abortion worldwide. New York, NY, Alan Guttmacher Institute, 1999: 35.
3. Sharing responsibility: women, society and abortion worldwide. New York, NY, Alan Guttmacher Institute, 1999: 36.
4. Wulf D, Jones H. Health professionals perceptions about induced abortion in South Central and Southeast Asia. International Family Planning Perspectives, 1997, 23 (2): 5967.
5. Rodriguez K, Strickler J. Clandestine abortion in Latin America: provider perspectives. Women and Health, 1999, 28 (3): 5975.
6. Ahmed S et al. Induced abortion: whats happening in rural Bangladesh. Reproductive Health Matters, 1999, 7 (14): 1929.
7. Makinwa-Adebusoye P, Singh S, Audam S. Nigerian health professionals perceptions about abortion. International Family Planning Perspectives, 1997, 23(4): 148154.
8. Mahomed K et al. Improved treatment of abortion complications and post-abortion family planning in Zimbabwe. Paper presented at: 120th Annual Meeting, American Public Health Association, Washington, DC, 812 November 1992.
9. Bongaarts J. Trends in unwanted childbearing in the developing world. Presented at: Annual Meeting, Population Association of America, Washington, DC, 2729 March 1997.
10. Singh S, Henshaw S. The incidence of abortion: a worldwide overview. Presented at: Seminar on Socio-Cultural and Political Aspects of Abortion from an Anthropological Perspective, Trivandrum, India, 2528 March 1996.
11. Desgrées du Loû A et al. Le recours à lavortement provoqué à Abidjan. Une cause de la baisse de la fecondité? [Recourse to induced abortion in Abidjan. A cause of the reduction in fertility?]. Population, 1999, 54 (3): 427446 (in French).
12. Laughrin AK et al. Abortion and contraception in Kazakhstan, Kyrgyzstan, and Uzbekistan: the provider perspectives (preliminary results), 1994 (unpublished data).
13. Bankole A, Singh S, Haas T. Reasons why women have induced abortions: evidence from 27 countries. International Family Planning Perspectives, 1998, 24 (3): 117127.
14. Frejka T, Atkin LC. The role of induced abortion in the fertility transition of Latin America. In: Guzman JM et al. eds. The fertility transition in Latin America. Oxford, Oxford University Press, 1996: 179191.
15. Huntington D. Advances and challenges in postabortion care operations research: summary report of a global meeting, 1921 January 1998. New York, NY, Population Council, 1998.
16. Bastianelli C et al. Contraccezione e interruzione volontaria della gravidanza. Indagine conoscitiva su di un campione di 500 donne. [Contraception and induced abortion. Study of a sample of 500 women.] Minerva Ginecologica, 1996, 48 (9): 359363 (in Italian).
17. Price SJ et al. Use of contraception in women who present for termination of pregnancy in inner London. Public Health, 1997, 111 (6): 377382.
18. Pile J et al. The quality of abortion services in Turkey. In: Huntington D. Advances and challenges in postabortion care operations research: summary report of a global meeting, 1921 January 1998. New York, NY, Population Council, 1998.
19. Georges E. Abortion policy and practice in Greece. Social Science and Medicine, 1996, 42 (4): 509519.
20. Raevic M. Yugoslavia: abortion as a preferred method of birth control. Reproductive Health Matters, 1994, 2 (3): 68 74.
21. Geldstein RN, Pantelides EA. Double subordination, double risk: class, gender and sexuality in adolescent women in Argentina. Reproductive Health Matters, 1997, 5 (9):121131.
22. Palma I, Quilodrán C. Adolescent pregnancy in Chile today: from marriage to abortion. Reproductive Health Matters, 1995, 3 (5): 1221.
23. Adjase ET. Hu MAni So Ma Me Nti (Two heads are better than one): teenage sexuality, unwanted pregnancy and the consequences of unsafe abortion in Sunyani district, Ghana. Amsterdam, Royal Tropical Institute, 1997 (MA thesis).
24. Lema VM, Rogo KO, Kamau RK. Induced abortion in Kenya: its determinants and associated factors. East African Medical Journal, 1996, 73 (3):164168.
25. Bélanger D, Khuat TH. Young single women using abortion in Hanoi, Vietnam. Asia Pacific Population Journal, 1998, 13: 326.
26. Bélanger D, Khuat TH. Single womens experiences of sexual relationships and abortion in Hanoi, Vietnam. Reproductive Health Matters, 1999, 7 (14): 7182.
27. Calves AE, Meekers D. Gender differentials in premarital sex, condom use, and abortion: a case study of Yaoundé, Cameroon. Washington, DC, Population Services International Research Division, 1997 (Working Paper No. 10).
28. Petchesky RP, Judd K, eds. Negotiating reproductive rights: womens perspectives across countries and culture. London, Zed Books, 1998.
29. Gillespie R. Voluntary childlessness in the UK. Reproductive Health Matters, 1999, 7 (13): 4353.
30. Casas-Becerra L. Women prosecuted and imprisoned for abortion in Chile. Reproductive Health Matters, 1997, 5 (9): 2936.
31. Johansson A et al. Husbands involvement in abortion in Vietnam. Studies in Family Planning, 1998, 29 (4): 400413.
32. Wu T, Parish WL. The one-child policy and induced abortion in Jilin, China. Presented at: Annual Meeting, Population Association of America, Miami, 57 May 1994.
33. Johansson A et al. Abortion in context: womens experience in two villages in Thai Binh province, Vietnam. International Family Planning Perspectives, 1996, 22 (3):103107.
34. Westley SB. Evidence mounts for sex-selective abortion in Asia. Asia-Pacific Population and Policy, 1995, 34: 14.
35. David H, Rademakers J. Lessons from the Dutch abortion experience. Studies in Family Planning, 1996, 27 (6): 341343.
36. Henshaw SK. Induced abortion: a world review, 1990. Family Planning Perspectives, 1990, 22 (2): 7689.
37. Rahman A, Katzive L, Henshaw SK. A global review of laws on induced abortion, 19851997. International Family Planning Perspectives, 1998, 24 (2): 5664.
38. Gursoy A. Abortion in Turkey: a matter of state, family or individual decision. Social Science and Medicine, 1996, 42 (4): 531542.
39. Nunes FE, Delph YM. Making abortion law reform happen in Guyana: a success story. Reproductive Health Matters, 1995, 3 (6): 1223.
40. Choice on Termination of Pregnancy Act, 1996. Republic of South Africa. Reproductive Health Matters, 1996, 5 (9): 116118.
41. Djohan E et al. The attitudes of health providers towards abortion in Indonesia. Reproductive Health Matters, 1993, 1 (2): 3240.
42. Barbados Medical Termination of Pregnancy Act, 19834, Supplement to Official Gazette. Barbados, 17 February, 1983.
43. Tietze C, Henshaw SK. Abortion: a world review. New York, NY, Alan Guttmacher Institute, 1986: 26.
44. Gabriel A. Amathila goes it alone. Sister Namibia. September/October, 1997.
45. Bill proposes to allow some abortions under health care system. Conscience, 1997, 18 (3): 31.
46. Bill to ease abortion restrictions narrowly defeated: Portugal. Conscience, 1997, 18 (1): 34.
47. Abortion debate in Mexico. Press release, Grupo de Información en Reproducción Elegida. Mexico City, October, 1998.
48. Tornaria C. Aborto: debate reabierto. [Abortion: a debate reopened]. Mujer/Fempress, 1994, 148/149: 4 (in Spanish).
49. Baltar da Rocha MI. The abortion issue in Brazil: a study of the debate in Congress. Estudos Feministas, 1996, 4 (2): 505522.
50. Benson J et al. Complications of unsafe abortion in sub-Saharan Africa: a review. Health Policy and Planning, 1996, 11 (2): 117131.
51. Langer A et al. Improving post-abortion care in a hospital in Oaxaca, Mexico. Reproductive Health Matters, 1997, 5 (9): 2028.
52. Arilha M, Barbosa RM. Cytotec in Brazil: At least it doesnt kill. Reproductive Health Matters, 1993, 1 (2): 41 52.
53. Costa SH. 1998. Commercial availability of misoprostol and induced abortion in Brazil. International Journal of Gynaecology and Obstetrics, 1998, 63 (suppl. 1): S131S139.
54. Fonseca W et al. [Sociodemographic, reproductive and clinical characteristics of abortion patients hospitalized in southern Brazil.] Cadernos Saúde Publica, 1998, 14 (2): 279286 (in Portuguese).
55. Faúndes A. Data from the Sistema de Informações Hospitalares do Sistema Unificado de Saúde, 1998, (cited in ref. 2).
56. Faúndes A et al. Post-abortion complications after interruption of pregnancy with misoprostol. Advances in Contraception, 1996, 12 (1): 19.
57. Nunes F, Delph Y. Making abortion law reform work: steps and slips in Guyana. Reproductive Health Matters, 1997, 5 (9): 6676.
58. Diaz J et al. Evaluation of a strategy for improving the quality of services and contraceptive acceptance in the post-abortion period in three public sector hospitals in Bolivia. In: Huntington D. Advances and challenges in postabortion care operations research: summary report of a global meeting, 1921 January 1998. New York, NY, Population Council, 1998.
59. Okagbue I. Pregnancy termination and the law in Nigeria. Studies in Family Planning, 1990, 21 (4): 197208.
60. Kissling F. Abortion: the link between legality and safety. Reproductive Health Matters, 1993, 1 (2): 6567.
61. Costa SH, Vessey MP. Misoprostol and illegal abortion in Rio de Janeiro, Brazil. Lancet, 1963, 341: 12581261.
62. Hajaj Gonzalez C et al. Congenital abnormalities in Brazilian children associated with misoprostol misuse in first trimester of pregnancy. Lancet, 1998, 351: 16241627.
63. Koster-Oyekan W. Why resort to illegal abortion in Zambia? Findings of a community-based study in Western Province. Social Science and Medicine, 1998, 46 (10): 13031312.
64. Gupte M, Bandewar S, Pisal H. Abortion needs of women in India: a case study of rural Maharashtra. Reproductive Health Matters, 1997, 5 (9): 7786.
65. Khan ME et al. Situation analysis of medical termination of pregnancy (MTP) services in Gujarat, Maharashtra, Tamil Nadu and Uttar Pradesh. In: Huntington D. Advances and challenges in postabortion care operations research: summary report of a global meeting, 1921 January 1998. New York, NY, Population Council, 1998.
66. Banwell SS, Paxman JM. The search for meaning: RU486/PG, pregnancy and the law of abortion. American Journal of Public Health, 1992, 82 (10): 13991406.
67. Yishai Y. Public ideas and public policy: abortion politics in four democracies. Comparative Politics, 1993, 25 (2): 207228.
68. Smyth L. Narratives of Irishness and the problem of abortion: the X case 1992. Feminist Review, 1998, 60: 6183.
69. Lassey AT. Complications of induced abortions and their prevention in Ghana. East African Medical Journal, 1995, 72 (12): 774777.
70. Malaysia Act A727 of 1989. Annual Review of Population Law, 1989, 16: 32.
71. Abortion statistics 1997: England and Wales. Pro-Choice Alliance, London. From: Office for National Statistics Monitors AB 98/3 & 4, 26 June 1998.
72. Alvarez-Lajonchere C. Commentary on abortion law and practice in Cuba. International Journal of Gynecology and Obstetrics, 1989, 3 (suppl.): 9395.
73. Kamal MG et al. Study on interventions necessary for preventing rejection of MR clients. Dhaka, Bangladesh Association for Prevention of Septic Abortion, 1990.
74. The Swedish Abortion Act. Stockholm, Ministry of Health and Social Affairs, 1995.
75. Stevens M, Xaba M. Commentary. Reproductive Health Matters, 1997, 5 (9): 119.
76. Meyers C, Woods RD. An obligation to provide abortion services: what happens when physicians refuse? Journal of Medical Ethics, 1996, 22 (2): 115120.
77. Soderberg H et al. Continued pregnancy among abortion applications. A study of women having a change of mind. Acta Obstetrica Gynecologica Scandinavica, 1997, 76 (10): 942947.
78. Singh K et al. Abortion trends in Singapore: a 25-year review. Journal of Pediatric and Adolescent Gynecology, 1996, 9 (1): 2730.
79. Lane SD et al. Social Science and Medicine, 1998, 47 (8): 10891099.
80. Mpangile GS et al. Factors associated with induced abortion in public hospitals in Dar es Salaam, Tanzania. Reproductive Health Matters, 1993, 1 (2): 2131.
81. Roe J, Francome C, Bush M. Recruitment and training of British obstetrician-gynaecologists for abortion provision: conscientious objection versus opting out. Reproductive Health Matters, 1999, 7 (14): 97105.
82. Johnson BR et al. Costs of alternative treatments for incomplete abortion. Washington, DC, Population and Human Resources Department, World Bank, 1993 (Policy Research Working Papers WPS 1072).
83. McLaurin KE. A pro-active approach: meeting womens needs for abortion care in restrictive environments. Paper presented at : 119th Annual Meeting, American Public Health Association, Atlanta, GA, 1114 November 1991.
84. Rogo KO et al. Strategies to reduce morbidity and mortality due to unsafe abortion at the community level. East Africa Medical Journal, 1999, 76 (Suppl.).
85. Dickson-Tetteh K, Rees H. Efforts to reduce abortion-related mortality in South Africa. In: Berer M, Ravindran TKS, eds. Safe motherhood initiatives: critical issues. London, Blackwell Science, 1999.
86. Bugalho A et al. The effectiveness of intravaginal misoprostol (Cytotec) in inducing abortion after 11 weeks of pregnancy. Studies in Family Planning, 1993, 24 (5): 319323.
87. Carbonell JL et al. Vaginal misoprostol for abortion at 1013 weeks gestation. European Journal of Contraception and Reproductive Health Care, 1999, 4 (1): 3540.
88. Blackwell A et al. Health gains from screening for infection of the lower genital tract in women attending for termination of pregnancy. Lancet, 1993, 342: 206210.
89. Winikoff B et al. Safety, efficacy and acceptability of medical abortion in China, Cuba and India: a comparative study of mifepristone-misoprostol versus surgical abortion. American Journal of Obstetrics and Gynecology, 1997, 176: 431437.
90. Population Council. Medical methods of early abortion in developing countries: a consensus statement. Contraception, 1998, 58: 257259.
91. Ellertson C, Elul B, Winikoff B. Can women use medical abortion without supervision? Reproductive Health Matters, 1997, 5(9): 149161.
92. Castle MA, Likwa R, Whittaker M. Observations on abortion in Zambia. Studies in Family Planning, 1990, 21: 231235.
93. Joffe C. Doctors of conscience: the struggle to provide abortions before and after Roe v. Wade. Boston, MA, Beacon Press, 1995.
94. The care of women requesting induced abortion. Evidence-based Guideline No. 7. London, Royal College of Obstetrician-Gynaecologists, 2000.
95. Azize-Vargas Y, Avilés LA. Abortion in Puerto Rico: the limits of colonial legality. Reproductive Health Matters, 1997, 5(9): 5665.
96. Hardy E et al. Comparison of women having clandestine and hospital abortions. Reproductive Health Matters, 1997, 5(9): 108115.
97. Lamas M. The feminist movement and the development of political discourse on voluntary motherhood in Mexico. Reproductive Health Matters, 1997, 5 (10): 5867.
98. Bellucci M. Womens struggle to decide about their own bodies: abortion and sexual rights in Argentina. Reproductive Health Matters, 1997, 5 (10): 99106.
99. Mora M, Villarreal J. Unwanted pregnancy and abortion: Colombia. Reproductive Health Matters, 1993, 1 (2): 11 20.
100. Rogo K, Orero S, Oguttu M. Preventing unsafe abortion in Western Kenya: an innovative approach through private physicians. Reproductive Health Matters, 1998, 6 (12): 7783.
101. Ramaseshan G. Women imprisoned for abortion in Nepal: report of a Forum Asia fact-finding mission. Reproductive Health Matters, 1997, 5 (10): 133138.
Correspondence
M. Berer
Reproductive Health Matters
444 Highgate Studios, 5379 Highgate Road, London NW5 1TL, England
1 In one hospital in Brazil, 18% of 803 women giving birth had made unsuccessful attempts to abort (61). Also in Brazil, a recent study of abortion failure after misoprostol use involving 42 infants with congenital abnormalities found these to be consistent with vascular disruption in the fetus in utero caused by contractions from the prostaglandin, although the association needs to be confirmed (62).
2 Egypt, Guinea-Bissau, Iraq, Japan, Kuwait, Malawi, Morocco, Nicaragua, Republic of Korea, Saudi Arabia, Syria, Taiwan, Turkey, and United Arab Emirates.
3 Weeks of pregnancy are sometimes measured from first day of last menstrual period (LMP) and sometimes from the estimated date when conception probably took place, which is about 2 weeks later. Sources are not always clear about the distinction. Abortions using vacuum aspiration can be undertaken up to 14 weeks LMP.
4 Some 2000 women from outside Great Britain had an abortion in England and Wales in 1997, almost all of them from countries where there is a first trimester limit for on-request abortions (71).
5 Surgical abortion here refers to vacuum aspiration and manual vacuum aspiration. Medical abortion refers to the intravaginal application of a combination of mifepristone and a prostaglandin, either gemeprost (which requires refrigeration) or misoprostol (which does not).
6 For example, Why mothers die: report on confidential enquiries into maternal deaths in the United Kingdom 1994 1996. London, Her Majestys Stationery Office, 1998. These reports are published every 3 years.