Unsafe Abortion: Worldwide Estimates for 2000

Abstract

Abstract

Unsafe abortion is preventable and yet remains a significant cause of maternal morbidity and mortality in much of the developing world. Over the last decade, the World Health Organization has developed a systematic approach to estimate the regional and global incidence of unsafe abortion. Estimates based on figures around the year 2000 indicate that 19 million unsafe abortions take place each year, that is, approximately one in ten pregnancies ended in an unsafe abortion, giving a ratio of one unsafe abortion to about seven live births. Almost all unsafe abortions take place in the developing world. In Latin America and the Caribbean, 3.7 million unsafe abortions are estimated to take place each year, with an abortion rate of 26 per 1000 women of reproductive age, almost one unsafe abortion to every three live births. Asia has the lowest unsafe abortion rate at 11 per 1000 women of reproductive age, but 10.5 million unsafe abortions take place there each year, almost one unsafe abortion to every seven live births. However, excluding East Asia, where most abortions are safe and accessible, the ratio for the rest of Asia is one unsafe abortion to five live births. In Africa, 4.2 million abortions are estimated to take place per year, with an unsafe abortion rate of 22 per 1000 women, or one unsafe abortion per seven live births. In contrast, there is one unsafe abortion per 25 live births in developed countries.

Résumé

L'avortement non médicalisé peut être évité, mais demeure pourtant une cause importante de morbidité et de mortalité maternelles dans la plupart des pays en développement. Ces dix dernières années, l'Organisation mondiale de la santé a élaboré une méthodologie pour estimer l'incidence régionale et mondiale de l'avortement non médicalisé. Les estimations fondées sur les chiffres pour 2000 indiquent que 19 millions d'avortement non médicalisés se produisent chaque année, c'est-à-dire qu'une grossesse sur dix s'est terminée par un avortement non médicalisé, soit une proportion d'un avortement non médicalisé pour environ de sept naissances. La quasi-totalité des avortements non médicalisés ont lieu dans le monde en développement. En Amérique latine et aux Caraı̈bes, 3,7 millions d'avortements non médicalisés sont réalisés chaque année, avec un taux d'avortement de 26 pour 1000 femmes en âge de procréer, soit près d'un avortement non médicalisé pour trois naissances vivantes. L'Asie a le taux le plus faible d'avortement non médicalisé, 11 pour 1000 femmes en âge de procréer, mais 10,5 millions d'avortements non médicalisés s'y produisent chaque année, presque un avortement non médicalisé pour 7 naissances vivantes. Pourtant, àl'exception de l'Asie Orientale oú la plupart d'avortements sont sûrs et accessibles, l'Asie a un taux d'un avortement non médicalisé pour 5 naissances vivantes. En Afrique, 4,2 millions d'avortements sont effectués chaque année, avec un taux d'avortement non médicalisé de 22 pour 1000 femmes, soit un avortement non médicalisé pour 7 naissances vivantes. En contraste, les pays développés montrent un avortement pour 25 naissances vivantes.

Resumen

A pesar de ser evitable, el aborto inseguro sigue siendo una de las causas mayores de la morbi-mortalidad materna en muchas partes del mundo en desarrollo. Durante la última década, la Organización Mundial de la Salud produjo una metodologı́a para calcular la incidencia aproximada de aborto inseguro a nivel regional y global. Usando como base las estimaciones de la populación para el año 2000, se calcula que ocurren 19 millones de abortos en condiciones de riesgo cada año; es decir, aproximadamente 1 de cada 10 embarazos terminó en un aborto inseguro, dando una razón de 1 aborto inseguro por cada 7 nacimientos vivos. Casi todos los abortos inseguros tienen lugar en paı́ses en vı́as de desarrollo. En América Latina y el Caribe, se calculan 3,7 millones de abortos inseguros anualmente, con una tasa de aborto de 26 por 1000 mujeres en edad reproductiva, o casi un aborto inseguro por cada 3 nacimientos vivos. Asia tiene la tasa de aborto inseguro más baja, de 11 por 1000 mujeres en edad reproductiva; no obstante, en esta región ocurren 10,5 millones de abortos inseguros anualmente, casi un aborto seguro por cada 7 nacimientos vivos. Sin embargo, excepto en Asia Oriental donde el aborto seguro es común y accesible, la tasa de abortos para el resto de la región es de un aborto por cada 5 nacimientos vivos. En Africa, se calcula que ocurren 4,2 millones de abortos cada año, con una tasa de aborto inseguro de 22 por 1000 mujeres, o un aborto inseguro por cada 7 nacimientos vivos. En contraste, en los paı́ses desarrollados ocurre un aborto inseguro por cada 25 nacimientos vivos.

Key Words:

• unsafe abortion
• contraception and unplanned pregnancy
• research methodology

Induced abortion is one of the most difficult indicators to measure. Surveys show that under-reporting [1][2][3][4][5] occurs even where abortion is legal. In countries where abortion is legal but difficult to obtain or legally restricted and inaccessible, little information is available on abortion practice. Because of the difficulty of quantifying and classifying abortion in such circumstances, when it is taking place mainly in clandestine conditions, its occurrence tends to be under-reported, not reported at all or reported as spontaneous abortion (miscarriage). The language used for induced abortion sometimes reflects this ambivalence, for example, the term induced miscarriage (fausse couche provoqu é) is used, while menstrual regulation may be carried out when pregnancy has not been confirmed. In addition, the large national population and health surveys, for example, the Demographic and Health Surveys, collect information on pregnancy loss that does not distinguish between miscarriages and induced abortions.

The World Health Organization (WHO) is concerned with the public health aspects of abortion. WHO defines unsafe abortion as a procedure for terminating an unwanted pregnancy either by persons lacking the necessary skills or in an environment lacking the minimal medical standards or both [6]. This is the third time that global estimates on unsafe abortion have been calculated. In 1993 and 1997, regional and global estimates were issued for the first [7] and second time [8]. Since then further data have become available, and the incidence of and mortality due to unsafe abortion globally and by region have been re-estimated, updating earlier estimates. It should nevertheless be borne in mind that such estimates will always have a degree of uncertainty and should be considered only as the best estimates based on information available at the time. The current calculations use the most recent population estimates from the United Nations (UN) Population Division [9].

Estimating the incidence of unsafe abortion around the year 2000

A systematic strategy to estimate the incidence of unsafe abortion has been developed by WHO. Published and unpublished reports are collected and scrutinised for relevant information, together with the assessment of the study design, coverage and analysis of data. Different approaches were used to arrive at country estimates. Reliable national estimates were used along with community studies of abortion incidence where available. The annual unsafe abortion ratio was estimated after adjustment for under-reporting. Adjustments took into account the existing abortion law (de jure) and its application (de facto), information on the providers of unsafe abortions and cultural and rural/urban differences. Careful consideration was also given to total fertility rates (TFRs), and reported contraceptive prevalence in countries.

In other instances, unsafe abortion was estimated from hospital data by adjusting for spontaneous abortion. The abortion to birth ratio was adjusted following a procedure suggested by Singh and Wulf to account for spontaneous abortion [10]. Inter-country differences in the propensity to use hospitals for pregnancy-related conditions were taken into account to the extent that they applied equally to the propensity to seek treatment of abortion complications and use hospitals for deliveries. The abortion to birth ratio was further adjusted on the assumption that half or more of induced abortions are accomplished without complications requiring hospitalization. If available data were not national in scope, the abortion:birth ratio was corrected for a lower abortion ratio in rural areas, as indicated above.

The estimated numbers of unsafe abortions were then aggregated to come up with the global figure by applying the adjusted ratios to the birth estimates of the UN Population Division [9], giving the estimate for the annual incidence of unsafe abortion around the year 2000.

The incidence of unsafe abortions globally

Unsafe abortions are taking place all over the world (Table 1

Table 1 Global and regional annual estimates of incidence of unsafe abortion by United Nations regions, around the year 2000^a

	Estimated number of unsafe abortions	Incidence ratio (unsafe abortions per 100 live births)	Incidence rate (unsafe abortions per 1000 women aged 15–49)
World	19000000	14	12
More developed regions^b	500000	4	2
Less developed regions	18500000	15	15
Africa	4200000	14	22
Eastern Africa	1700000	16	29
Middle Africa	400000	9	20
Northern Africa	700000	15	15
Southern Africa	200000	16	16
Western Africa	1200000	13	24
Asia^b	10500000	14	11
Eastern Asia^b	^c	^c	^c
South-central Asia	7200000	18	20
South-eastern Asia	2700000	23	19
Western Asia	500000	10	11
Europe	500000	7	3
Eastern Europe	400000	14	5
Northern Europe	10000	1	1
Southern Europe	100000	7	3
Western Europe	^c	^c	^c
Latin America and Caribbean	3700000	32	26
Caribbean	110000	15	11
Central America	680000	20	19
South America	2900000	39	30
North America	^c	^c	^c
Oceania^b	30000	12	15

a Figures may not add up to equal totals due to rounding.

b Japan, Australia and New Zealand have been excluded from the regional estimates, but are included in the total for developed countries.

c For regions where the incidence is negligible no estimates are shown.

), with the exception of countries where abortion is legal, safe and relatively accessible. Worldwide the estimates indicate that 19 million unsafe abortions take place each year, i.e. approximately one in 10 pregnancies ends in an unsafe abortion, giving a ratio of one unsafe abortion to about seven live births. For the period around 1997, 20 million unsafe abortions per year were estimated [8]. Because of possible differential reporting and coverage, however, it is difficult to attribute the difference between the two figures to a trend.

Almost all unsafe abortions take place in the developing world. Despite liberalization of abortion laws in some developing countries, the infrastructure to provide legal and safe abortion lags behind, and many women still resort to abortion outside approved facilities or rely on unskilled providers when faced with an unwanted pregnancy.

Variations within sub-regions may be important. For example, abortion is legal and unsafe abortions are rare or non-existent in some countries in Asia, e.g. China, Vietnam and Turkey, as well as in Tunisia in north Africa. On the other hand, legal abortion and menstrual regulation occur in parallel with large numbers of unsafe abortions in India and Bangladesh, a situation that is occurring to a greater or lesser extent in other parts of the world as well.

Figure 1 shows the ratio of unsafe abortions to births, together with the TFR per 10 women, and the percent using all or modern contraceptive methods for all developing country regions except eastern Asia, where the availability of legal and safe abortions means that unsafe abortions have been all but eliminated.

Fig. 1 TFR, contraceptive use (all methods and modern methods) and unsafe abortion ratio (unsafe abortions per 100 live births) by developing region Note: TFR=total fertility rate; CCP=% using any contraceptive; CCP modern=% using modern contraceptives; and unsafe abortion ratio = number of unsafe abortions to 100 live births

In Latin America, where fertility has declined to low levels, close to 4 million unsafe abortions are estimated to take place each year, almost one unsafe abortion for every three live births. The incidence ratio ranges from 15 unsafe abortions to 100 births in the Caribbean to more than double that in South America. The lowest TFR in the developing world is found in South America, which also has the highest incidence ratio of unsafe abortions worldwide, almost 40 to 100 births, in parallel with a high reliance on modern contraception. However, sterilization is the most commonly used method of family planning in South America, which may make many women turn to unsafe abortion for spacing births before limiting family size through sterilization.

Among developing regions, Asia has the lowest unsafe abortion rate at 11 per 1000 women of reproductive age. This is because unsafe abortion is rare in several of the most populated countries of the region, notably China. Because of the very large number of women of reproductive age in Asia, however, half of the world's unsafe abortions take place there, more than one third in south-central Asia alone. Unsafe abortion rates in south-central and southeast Asia are about 20 per 1000 women of reproductive age. Southeast Asia has a low TFR, similar to South America, but because of a greater reliance on reversible contraceptive methods, has about two-thirds of South America's abortion ratio and rate.

In Africa, over 4 million unsafe abortions are estimated to occur each year, and this is probably more of an under-estimation than that in other regions. East, Central and West Africa stand out as having the highest TFR of the developing world (around 6 children per woman) and the lowest contraceptive use. Because of high birth rates these parts of Africa seemingly have modest unsafe abortion ratios of 9–16 to 100 live births. However, with unsafe abortion rates of 20–29 per 1000 women, this incidence falls between that of South America and southeast and south-central Asia, showing that women have a similar propensity to turn to unsafe abortion in all these regions. North Africa and southern Africa show similarly lower unsafe abortion ratios and rates, and these are expected to decrease further as the infrastructure and skills become available to meet the demand for safe abortions in South Africa, which now has Africa's most liberal abortion law.

Although the evidence remains incomplete, it is becoming increasingly apparent that the incidence of unsafe abortion may be rising among unmarried adolescent women in urban areas, particularly where abortion is illegal and fertility regulation services are inaccessible, inadequate or inappropriate to meet their needs. The need for readily available information and services for abortion is not confined to young women, however. In several countries the legalization of abortion has not been followed by the systematic elimination of unsafe abortions through safe abortion services, for a variety of reasons.

Great progress has been made in improving some areas of reproductive health, most notably in contraceptive use. However, although unsafe abortions are entirely preventable, they are still occurring in all developing country regions. The major public health implications include but are not limited to maternal morbidity and mortality, as well as financial costs to women and to health services for treating complications. Preventing unintended pregnancies and unsafe abortion must therefore continue to be a high priority for improving women's reproductive health. It also remains important to study and monitor the extent of unsafe abortion globally, regionally and by country, so that the public health impact can continue to be assessed.