The Alan Guttmacher Institute   search  
home home about contact e-lists support agi buy help
publications article archive state center media center tablemaker
Issues in Brief

An Overview of Clandestine Abortion in Latin America

Each year, more than four million women in Latin America undergo an induced abortion. Because most abortions are illegal, these procedures are performed under clandestine and often dangerous conditions. As a result, the region faces a serious public health problem that threatens women's lives, endangers their reproductive health and imposes a severe strain on already overextended health and hospital systems.

The practice of induced abortion in Latin America is shrouded in secrecy, a direct result of the stringent legal limitations on abortion throughout most of the region. Induced abortion is punishable by law in almost every country except Cuba and a few other Caribbean nations. In most of the region, doctors may legally terminate a pregnancy that threatens the life of the woman, that results from rape or incest, or that is characterized by fetal deformity,1 but these options are rarely used.

Concern over the high level of clandestine abortion in Latin America is not new. Policymakers and health professionals have been aware for the past 20-30 years that unsafe procedures were being performed in most countries of the region, and at a leve l with serious consequences for women's health and for the cost of national health care services.

Community surveys conducted in Chile in the early 1960s were the first attempt to measure the extent of the problem. These surveys found that women were likely to have two or three abortions over the course of their childbearing years.2 And studies in the 1970s in countries as diverse as Brazil, Colombia, the Dominican Republic, Mexico, Peru and Venezuela indicated that women averaged 0.5-1.5 induced abortions over their reproductive lifetime in these coun tries, and 2.0 or more induced abortions in Chile and Cuba.3

Although most Latin American health experts were aware of the general scale of clandestine abortion and related problems, until recently they had little reliable information with which to answer many questions. What methods are used to induce abortion? Who are the major practitioners? How many women are hospitalized for the treatment of complications, and what proportion is this of the actual numbers of women experiencing induced abortion? Which women are most likely to have induced abortions, and for what reasons?

This report presents an overview of the practice of induced abortion in Latin America. It draws upon a number of sources: a collaborative study on clandestine abortion in six countries; a large-scale study in urban Colombia; an in-depth study of trends i n abortion and contraception in three countries; and a number of smaller studies. Findings from many of these studies were presented at the first regional meeting on induced abortion in Latin America, held in Colombia in 1994.4

Current Level of Abortion

Estimated rates of abortion are highest in Peru and Chile (each year, almost one woman in every 20 aged 15-49 has an induced abortion), intermediate in Brazil, Colombia and the Dominican Republic (about one woman in 30), and lowest in Mexico (approxi mately one in 40). If these rates continue to prevail throughout the 35 years of a woman's reproductive lifetime, the average woman in Mexico is likely to have had at least one abortion by the time she is 50, compared with about 1.6 abortions among women in Chile and about 1.8 among women in Peru.

If the annual number of abortions estimated to occur in these six countries (2.8 million in the early 1990s) is extrapolated to the entire region, then about four m illion induced abortions are being carried out each year in Latin America (Table 1).

Table 1. Abortion Incidence
Country/year Annual no. of abortions Rate per 1,000 women Average per women
Total 2,768,150 33.9 1.2
Brazil, 1991 1,433,350 36.5 1.3
Chile, 1990 159,650 45.4 1.6
Colombia, 1989 288,400 33.7 1.2
Dom. Republic 82,500 43.7 1.5
Mexico, 1990 533,100 23.2 0.8
Peru, 1989 271,150 51.8 1.8
Latin America* 4,000,000 33.9 1.2
*Estimated on the assumption that these six countries account for 70% of the population of Latin America and that all countries in the region have similar hospitalization levels; rounded to the nearest 100,000.

Most Abortion Seekers Are Married and Have Children

Profiles of women obtaining abortions based on national data are useful. Knowledge about the women most likely to terminate an unwanted pregnancy is important because it helps identify subgroups in the population most in need of contraceptive protection. In addition, the information indicates that clandestine abortion is common at many levels of society and that unsafe procedures have consequences for the health of a broad cross-section of women.

One way to obtain information about women having abortions is to examine data on women hospitalized as a result of complications of clandestine procedures. However, few national systems provide data on the background characteristics of hospitalized women, and hospitalized women are not necessarily representative of all women having induced abortions.

Nonetheless, these and other data show that most Latin American women having induced abortions are in their 20s or older, married and already mothers. In most developed countries, by contrast, women having induced abortions are often very young, the vast majority are single and about half are childless.5

In Colombia, for example, national statistics from the 1980s show that about 87% of all hospitalized abortion patients were older than 20.6 In a 1992 Colombian survey representative of urban households, 23% of all women between the ages of 15 and 55 reported that they had had an induced abortion at some point in their lives; the proportion was somewhat higher than this among women who had not completed primary school (29%, see Table 2), among women with some un iversity education (28%) and among those in the labor market (27%).7

Table 2. In Colombia
Educational level of urban women % who had an abortion
All levels 23
Incomplete primary 29
Complete primary 22
Incomplete secondary 21
Complete secondary 18
Incomplete university 28
Complete university 24
Source: See reference 7.

A 1990 study carried out by the Latin American Federation of Obstetricians and Gynecologists in four countries (Bolivia, Colombia, Peru and Venezuela, see Table 3) found that the vast majority of hospitalized abortion patients (79%) were married women, ha lf (51%) had two or more children, and half had also had seven or more years of schooling. Some 86% were aged 20 or older.8

Table 3. In Four Countries
Characteristics of abortion patients Percent distribution
Marital status
Single 21
Married or in a consensual union 79
No. of children
0 26
1 23
2-4 40
>=5 11
Years of schooling
0-4 21
5-6 27
7-9 28
>=10 24
Note: Hospital data from Bolivia, Colombia, Peru and Venezuela.

Another 1990 study of Chilean women hospitalized for abortion complications in nine hospitals in Santiago also found that few were young women (only 11% were younger than 20), that the vast majority (78%) were married or in a consensual union, that 76% al ready had children, and that more than 80% had had seven or more years of schooling.9

A 1992-1993 study of abortion patients in two large maternity hospitals in the poor northeast state of Fortaleza, Brazil, found that 77% were older than 20 and that 68% had had five or more years of schooling. But in this study, 53% of women had never been married or in a stable union, and 9% were separated, divorced or widowed.10

In the Dominican Republic, in-depth interviews in 1992 with women hospitalized for abortion complications in two large hospitals in Santo Domingo (Table 4) revealed that 84% were in their 20s or older, 88% were in a union, and 73% had had five or more years of schooling.11

Table 4. In Santo Domingo
Characteristics of abortion patients Percent distribution
<20 16
20-24 33
25-29 25>
30-34 15
>=35 11
Marital Status
Not in a union 12
In visiting union 9
In consensual union 66
Married 13
Source: See reference 11.

Abortion Techniques Used

A 1993 study in Brazil, Colombia, Chile, the Dominican Republic, Mexico and Peru showed that women everywhere are familiar with teas and infusions made from herbs and other vegetable products that are believed to induce abortion (Table 5 ).12

Table 5. Abortion Methods
Voluntary trauma
•Falls, punches, excessive physical activity, etc.
Natural products taken orally or vaginally
•Seaweed, vegetables, seeds
•Teas and infusions prepared from herbs and vegetables
Manufactured products taken orally or vaginally
•Beer, wine, vinegar
•Soapy substances
•Caustic substances, including bleach, dye, potassium salts
Physical objects inserted in uterus
•Catheter (rubber tubing) alone
•Catheter used to flush toxic fluids
•Sharp objects such as wire, knitting needles and sticks
Pharmaceutical products administered orally or vaginally
•Misoprostol, prostaglandins, estrogen, laxatives, quinine, oxytocins, etc.
Medical techniques
•Dilatation and curettage
•Vacuum aspiration
Source: 1992 AGI Survey of Opinions on Abortion Practice in Brazil, Colombia, Chile, Dominican Republic, Mexico and Peru (see reference 12).

If these products do not have the desired effect (and nobody knows whether they are genuine abortifacients), women are then likely to resort to riskier methods: the insertion of a rubber tube, caustic liquids or other foreign objects into the uterus; or the oral or vaginal application of powerful pharmaceutical and hormonal products. In Brazil, a pharmaceutical product usually prescribed for the treatment of gastric and duodenal ulcers, misoprostol (Cytotec), is widely used to terminate pregnancy.13

In the urban areas of some countries, women with financial resources can attend private clinics that perform safe medical abortions using dilatation and curettage (D&C) or vacuum aspiration techniques. And throughout Latin America, middle- or upper-class women are generally acknowledged to obtain safe medical abortions in doctors' offices.

Three broad types of providers perform clandestine abortion in Latin America: doctors; nurses; and untrained persons, such as traditional practitioners or pregnant women themselves. As might be expected, higher income women and urban women are most likely to go to doctors, low-income urban women might avail themselves of any of the three types of providers, and poor rural women commonly depend on untrained practitioners or use traditional methods that they themselves can use to induce abortion.14

The procedures on which poor rural women depend often occur under dangerously unsanitary conditions. These women frequently experience complications as serious as hemorrhage and infection that require hospital treatment.

Researchers estimate that in the six countries in question, more than half a million women are hospitalized each year to obtain treatment for abortion complications. If this number is extrapolated to all of Latin America (assuming similar conditions in ot her countries and given that the six study countries account for 70% of the total population of Latin America), almost 800,000 women each year are likely to obtain hospital treatment for the complications of induced abortion (Table 6). These data highlight the need to improve the treatment of abortion complications and to reduce the unhealthy consequences.=paragraph

Table 6. Unsafe abortions
Country/year Number of hospitalizations
Total 555,630
Brazil, 1991 288,670
Colombia, 1989 57,680
Chile, 1990 31,930
Dom. Republic, 1992 16,500
Mexico, 1990 106,620
Peru, 1989 54,230
Latin America* 800,000
*Estimated on the assumption that these six countries account for 70% of the population of Latin America and that all countries in the region have similar hospitalization levels; rounded to the nearest 100,000.

Complication Rate Declines

Obviously, the safer the abortion procedures used in any setting, the fewer the women who will require hospitalization. However, the numbers hospitalized also depend on the availability of hospital services. Women living in rural areas are probably less likely than those living in urban areas to have access to hospital treatment, even though they are precisely the women most likely to be exposed to unsafe techniques, untrained practitioners and unhygienic settings.

Most health professionals believe that the risks of complications from induced abortion are lower in Latin America today than they were in the past. Reasons for the reduced risk are that more women have access to safe medical procedures, more women are aw are of the need to seek medical treatment quickly when a complication arises and more practitioners—even those without medical training—routinely prescribe antibiotics for their patients.

The number of hospitalized women with a severe infection is believed to have declined dramatically as a result of these trends. On the other hand, the growing use of hormonal methods and of techniques like misoprostol, which induce heavy bleeding, has pro bably increased the numbers of women seeking treatment for hemorrhage. =paragraph

Community surveys carried out in Chile in the 1960s, when the conditions for pregnancy termination were much less safe than they are now, found that one in three women undergoing induced abortion were hospitalized. Today, researchers and health profession als estimate that one in five abortion patients require hospitalization for the treatment of complications.15

What Leads Women To Seek an Abortion?

Most studies of women hospitalized for abortion-related complications conclude that the major reasons women choose not to give birth are that they are unable for economic, personal or family reasons to have or support the child. Many women are not married or are in unstable relationships. If they are single or very young, they are likely to decide that they cannot take care of a child alone and without financial support.

But the majority of the patients appear to be married women who already have all the children they feel able to care for. The difficult social and economic conditions facing many millions of families in Latin America's poor rural areas and vast city slums spur the desire of couples to have fewer children. So do increases in women's level of education and women's increased participation in the labor market.

Yet, despite the desire for smaller families and for better timing of births, women are not always able to attain control over their childbearing. Surveys of women in Latin America find substantial proportions who do not want to become pregnant but are no t using a contraceptive method or are relying on traditional methods, such as periodic abstinence or withdrawal, which have high failure rates. These women, often characterized as having an "unmet need" for contraception, represent anywhere from 17% of wo men aged 15-44 in Colombia to 43% in Bolivia.16

In addition, some women are having unplanned pregnancies because of ignorance about when they are most likely to conceive,17 and because even women who use effective contraceptive methods may do so irregularl y or incorrectly.18

Surveys in 10 countries in Latin America show that as a result of all these factors, 24-65% of women who had had a child in the previous five years said they had not wanted their most recent birth at the time it occurred or that they had not wante d another child ever.19

Combining estimates of the number of induced abortions with survey information from six countries on the proportion of births that are unplanned indicates that 60% of pregnancies in Mexico end as wanted births—the highest proportion in any of the cou ntries—and 17% end as induced abortions. In contrast, approximately 30% of all pregnancies in the other five countries end in induced abortions (Figure 1).

Figure 1
Outcome of Pregnancies

These levels of abortion are not unique to Latin America: In the United States, for example, almost 30% of all pregnancies are resolved by abortions, and in China and Japan, the proportions are 30% and 25%, respectively.20 Most developing regions of the world have lower levels, however.

Women Use Both Abortion And Contraceptive Methods

Although very little information exists on historical trends in abortion in Latin America, the level of abortion was probably quite low prior to the 1960s, when there was little attempt to control family size and the average woman had six or more children. It is only in the context of rapidly changing socioeconomic conditions and a rise in families' motivation to achieve a better standard of living that women increasingly have turned to both abortion and contraception to help them achieve a small family.

In the late 1960s and in the 1970s, researchers theorized that if the practice of abortion is widespread before the use of contraceptive methods becomes extensive, then the incidence of abortion will remain high as the motivation to limit family size intensifies, even while contraceptive practice rises.21 According to this theory, recourse to abortion will not decrease until enough years have passed for the widespread, habitual and effective use of contraceptives to become well established at all levels of society.

Research now exists that compares national and regional levels of abortion and of contraceptive use in Brazil, Colombia and Mexico at three points in time: the late 1970s, the mid-1980s and the early 1990s.22 The comparison shows that as levels of contraceptive use increased in the late 1970s, abortion had reached a moderate level (20-30 abortions per 1,000 women per year) in all three countries. Abortion peaked in Colombia and Mexico in the mid-1980s, and there were indications of a slight drop in some urban areas by the early 1990s. During the same period, abortion levels in Brazil were still rising.

The more organized national family planning programs in Colombia and Mexico may have begun to create a permanent "culture of contraception." In Brazil, on the other hand, actual and desired family size are both small, but no national family planning program exists, many rural women lack access to high-quality contraceptive services and many women who obtain methods from commercial sources use them incorrectly or only sporadically. Reductions in levels of abortion are likely to be slower in coming to Brazil than to the other two countries.

Abortion Will Not Disappear

Induced abortion remains a widespread practice in Latin America, primarily because women's use of contraceptives and their ability to use methods consistently and correctly have not kept pace with the growing intensity of the desire for smaller families, a desire probably fueled by rising aspirations and, often, unstable family situations.

While the true level of abortion is unknown, it is evident that at least 800,000 of the estimated four million women who have induced abortions each year require hospitalization for the treatment of abortion complications. Although abortion rates are lower in some countries than in others, the practice of abortion is not disappearing and levels may still be going up in some countries.

Despite this pessimistic scenario, abortion is probably a safer procedure for women now than it was in the past. Because of the wider use of medical techniques and the more common use of antibiotics, fewer women in the region are likely to die from abortion complications or to suffer grave health consequences, although both outcomes are still far too frequent.

Improved contraceptive services and improved contraceptive use can go a long way toward reducing levels of unplanned pregnancy in Latin America. There is evidence, for example, that abortion rates may be leveling off or declining in parts of Colombia and Mexico, two countries in which contraceptive use is widespread. Nevertheless, studies indicate that even where family planning services are available and accessible, many women have difficulty using methods consistently and effectively, and contraceptive discontinuation and failure rates can be high.

Moreover, many circumstances conspire to make effective contraceptive decision-making and practice difficult, especially for women lacking social and family support. In this rapidly developing region of the world, the roles and responsibilities of women continue to change; life in large cities and declining agricultural areas is often precarious and economic stability elusive; migration is seen as a necessity for survival and family dissolution is commonplace.

Many of these conditions and constraints on women can be improved only by fundamental social and economic change. However, policymakers and program managers could usefully focus on services that might have a more immediate impact on women's ability to plan their families better and to have fewer unplanned pregnancies and abortions. Improvements in the availability, delivery and quality of contraceptive services and the establishment of postabortion contraceptive counseling services in large public hospitals are direct steps that could be taken immediately.

Over the longer term, governments should consider reviewing laws that contain punitive measures against women who have undergone illegal abortions and improving services for the management of complications arising from abortion. Both these initiatives were among recommendations approved at the International Conference on Population and Development in Cairo in 1994 and the Fourth World Conference on Women in Beijing in 1995 and agreed to by almost all nations of the world.


1. S.K. Henshaw, "Induced Abortion: A Worldwide Review," Family Planning Perspectives, 22:76-89, 1990.

2. T. Frejka and L. C. Atkin, "The Role of Induced Abortion in the Fertility Transition of Latin America," in J.M. Guzman et al., eds., The Fertility Transition in Latin America, Oxford University Press, New York, 1996, Table 9.1, p. 182.

3. Ibid.; and T. Frejka, "The Role of Induced Abortion in Contemporary Fertility Regulation," in Proceedings of the Conference of the International Union for the Scientific Study of Population, Montreal, 1993.

4. Research Conference on Induced Abortion in Latin America and the Caribbean, Universidad Externado de Colombia, Bogotá, Nov. 15-18, 1994; and L. C. Remez, "Confronting the Reality of Abortion in Latin America," International Family Planning Perspectives, 21:32-36, 1995.

5. S.K. Henshaw and E. Morrow, Induced Abortion: A World Review, 1990 Supplement, The Alan Guttmacher Institute (AGI), New York, 1990.

6. E. Prada Salas, "Aborto Complicado: La Punta del Iceberg de la Práctica Clandestina," paper presented at the Research Conference on Induced Abortion... (see reference 4).

7. L. Zamudio, N. Rubiano and L. Wartenberg, "La Incidencia del Aborto Inducido en Colombia," paper presented at the Research Conference on Induced Abortion... (see reference 4).

8. F. Pardo and G. Uriza, "Estudio de Morbilidad y Mortalidad por Aborto en 36 Instituciones de Bolivia, Colombia, Perú y Venezuela," Revista Colombiana de Obstetricia y Ginecología, 42:287-297, 1993.

9. P.A. Lavin, "Informe Preliminar Sobre la Caracterización de los Casos y Costo del Tratamiento del Aborto Hospitalizado en Santiago de Chile," paper presented at the Research Conference on Induced Abortion... (see reference 4).

10. C. Misago et al., "Determinants of Induced Abortion Among Poor Women Admitted to Hospitals in Fortaleza, North Eastern Brazil," paper presented at the Research Conference on Induced Abortion... (see reference 4).

11. D. Paiewonsky, "Algunos Hallazgos en Torno a los Condicionantes Sociales del Aborto Inducido en la República Dominicana," paper presented at the Research Conference on Induced Abortion... (see reference 4).

12. AGI, Clandestine Abortion: A Latin American Reality, New York, 1994; and S. Singh and D. Wulf, "Estimated Levels of Induced Abortion in Six Latin American Countries," International Family Planning Perspectives, 20:4-13, 1994.

13. S.H. Costa and M. Vessey, "Misoprostol and Illegal Abortion in Rio de Janeiro, Brazil," Lancet, 341:1258-1261, 1993; and H. Lutécia Luna Coêhlo et al., "Misoprostol and Illegal Abortion in Fortaleza, Brazil," Lancet, 341:1261-1263, 1993.

14. AGI, 1994, op. cit. (see reference 12), p. 14.

15. S. Singh and D. Wulf, 1994, op. cit. (see reference 12).

16. AGI, Hopes and Realities: Closing the Gap Between Women's Aspirations and Their Reproductive Experiences, New York, 1995, appendix table 7, columns 6 and 7, p. 52.

17. Ibid., appendix table 6, column 3, p. 50.

18. Ibid., appendix table 6, column 16, p. 51.

19. Ibid., appendix table 5, columns 18 and 19, p. 49.

20. Ibid., Chart 13, p. 25.

21. M. Requena, "Condicciones Determinantes del Aborto Inducido," Revista Medica de Chile, 94:714-722, 1966; and C. Tietze and J. Bongaarts, "Fertility Rates and Abortion Rates: Simulations of Family Limitation," Studies in Family Planning, 6:114-120, 1976.

22. S. Singh and G. Sedgh, "Trends in Abortion, Contraception and Fertility in Brazil, Colombia and Mexico," International Family Planning Perspectives, vol. 23, 1997 (forthcoming).

Susheela Singh conducted the research for this report and Deirdre Wulf wrote the text.

back to top