Abortion Estimates: Difficulties and Limitations

The Dimension of Abortion: Levels and Trends

What Are the Indicators Used for Measurement?

The main indicators conventionally used for measuring abortion trends and levels are as follows:

Abortion Rate: the number of abortions per 1,000 women of childbearing age.

Abortion Ratio: refers to the number of abortions in relation to the number of live births or the number of pregnancies. The number of abortions for every 100 pregnancies, excluding spontaneous ones, can be interpreted as the likelihood that a woman would opt for an abortion if she were pregnant, which does not necessarily follow the same pattern as the abortion rate. For example, the abortion ratio may increase while the abortion rate remains stable or declines if the total level of pregnancies falls.

Index of Incidence: theproportion of women that have had at least one abortion

Total Abortion Rate: thetotal number of abortions a woman has in the course of her lifetime.

Estimates of Abortion in Latin America and the Caribbean within the World Context

According to the WHO, it is estimated that the incidence of abortion worldwide in 2000 totaled 46 million interrupted pregnancies (World Health Organization, 2004). These figures are based on various sources of information, mainly hospital registers that are adjusted and corrected, as well as the revision of data from household surveys and service providers, sometimes taking into account the opinions of experts from each country. Of the total number of abortions considered by this organization, nearly 27 million were performed legally, while 19 million took place outside the legal system and were therefore much more likely to have been high risk (unsafe abortions).

The majority of these unsafe abortions occur in developing countries in which there are greater legal restrictions on access to abortion. In this respect, access to abortion is easier in industrialized countries, where 86% of the population lives under more liberal legislation. Conversely, in developing countries, excluding China, only 39% live in these conditions, and 44% live in areas where abortion is totally illegal or only permitted under certain conditions.  The conditions are usually extremely restrictive, such as saving the woman’s life (Deidre, 1999).

As one can see from Table 1, the gap between developed and developing countries is extremely wide: the rate of unsafe abortions around the year 2000 was 2 for every 1,000 women aged 15 to 44 for the former and was eight times higher for the second group. At the same time, the abortion ratio was 1 abortion for every 25 live births and 1 for nearly 7 live births respectively. In Latin America and the Caribbean, the ratio was one abortion for every 3 live births. For Asia and Africa, this was 1 abortion for nearly 7 live births. Estimates undertaken for the early 1990s, seen in the same Table, show that the total number of abortions worldwide has remained virtually unchanged. Although the difference between the two periods is nearly one million, this should not be regarded as a precise trend, given the relative imprecision and uncertainty of the data provided by the sources and derived from the estimates. Singh and Ratnam (1998), as well as WHO (WHO, 1993), also point out that according to worldwide estimates, there was one abortion for every 10 pregnancies and one abortion for every seven live births, both at the beginning and at the end of the 1990s.

In Latin America and the Caribbean, it is estimated that this same decade saw approximately 4 million unsafe, illegal abortions a year (4.6 million at the beginning and 3.7 million at the end of that decade). Moreover, it is estimated that between one out of every 20 and one out of every 40 women had an abortion every year during this period (AGI, 1994). This region has the highest rate and ratio of unsafe abortions: 29 abortions per thousand women aged 15 to 44 and nearly one abortion for every three live births in the year 2000. Likewise, regardless of the periods considered, at both the beginning and the end of the 1990s Latin America was the region with the highest rates of unsafe (and also illegal) abortions: between 41 and 29 abortions per thousand women.  This figure is two to almost three times higher than the world average, which is approximately 15 abortions per thousand women, and between 5 and 8 times higher than that of developed countries.

As this table shows, in Latin America and the Caribbean, the region that has seen the highest decrease in fertility rates in recent years, the abortion ratio in 2000 varied from 15 abortions per 100 live births in the Caribbean to over twice in South America (39 abortions per 100 live births).  This latter sub-region is also that with the lowest fertility levels, with widespread modern contraceptive use and yet the highest ratio of abortions in the world. For its part, Central America occupies an intermediate position, with 20 abortions per 100 live births, and is the sub-region with the third largest proportion of abortions.

In Latin America and the Caribbean, the synthetic abortion index is 0.8 abortions per women, in comparison with Africa and Asia, where it is 0.7 and 0.6 respectively (Shah and Ahman, 2004). The first of these three indices is particularly high, considering that the overall fertility rate in the region is approximately 2 children per woman.

Diversity of Abortion Estimates within the Region

In a restrictive legal context, with limited access to abortion and an intense clandestine practice, as in the case of the majority of countries in the Latin American and Caribbean region, it is difficult if not impossible to have an accurate view of the incidence of this practice for each country. There are several and diverse estimates on abortion, which vary according to the various sources and methods of information gathering and analysis. Very few countries in the region, however, have data on the topic obtained using different methods that can be compared to each other, and therefore very little data is reliable and sufficient for analyzing the evolution of abortion over time.

Comparison between the various information sources is often difficult, since the population of reference is often not the same. Data obtained from hospital records are used to estimate the number of “risky abortions” and sometimes, through the application of a correction factor, to estimate the abortion level nationwide. At the same time, published survey data provide estimates on abortion rates and ratios, but these are frequently inaccurate, in addition to the fact that they fail to clearly explain aspects of the information presented (the  population of reference, the date of the study, the level of reliability of the data, the sample procedure etc.). Their results should be considered as a base line estimate of the prevalence of abortion, and therefore, only as an approximation of it (Faúndes, 2005).

In the bibliographical review presented here, very few references were found on the various countries in the region that included estimates on abortion, either by using direct, indirect or secondary methods and sources. They are often studies on specific social contexts that do not show the trends in this practice either over time or at the national level. Many of these studies cite the same sources, proving and confirming the lack of studies on the subject. As can be seen from Table 2, for some countries, no information is available, whereas in others, varying estimates are provided. This proves the inaccuracy of the estimates on the intensity of this phenomenon in the region. Their variability, as well as the diversity of figures estimated, even for similar periods, is due to problems that are inherent to the data sources consulted and the methodology used in the estimates.

Therefore, it is important to note and reiterate the fact that the overview obtained from the various sources merely provides an approximation of the prevalence of abortion in the various countries and within some of them.

Although information on abortion in Latin America is not sufficiently precise, the evidence provided shows that it is a phenomenon of enormous magnitude in the region. This should serve to alert and make various sectors of the population aware of the consequences of abortion, particularly those responsible for designing and implementing policies and programs aimed at improving women’s health and living conditions. It should also serve to promote the undertaking of research with scientific rigor, preferably using similar methodologies that would provide more reliable estimates of the occurrence of this phenomenon in the various countries in the region.

An analysis of the information available in these countries (Table 2) shows an initial contrast between Cuba, where abortion is available at the woman’s request, legal and performed by trained physicians in hospitals, and other countries with a different degree of legal restrictions and a usually clandestine practice. In 1990, Cuba reported 54.5 abortions per 1,000 women of reproductive age or approximately 44 abortions for every 100 pregnancies (Bernstein and Rosenfield, 1998). According to Álvarez Vásquez, (1994), official figures from 1974 to 1985 show a variation of between 40 and 50 abortions per thousand women of reproductive age. From this date onwards, a slight yet steady decline is observed, as a result of which, by 1990, the abortion ratio was 8 cases for every 10 live births. The abortion rate for 1993 was 26.6 for every thousand women. Nevertheless, in these cases, abortions performed through menstrual regulation, estimated to account for at least 50% of these procedures, are not counted. Another study shows that over 140,000 abortions are performed annually in Cuba, a figure that totaled 186,658 in 1990, with an abortion ratio of 8 for every 10 live births (as in the study mentioned earlier) and an abortion rate of 59.4 per 1,000 women of reproductive age. This estimate includes abortions performed through menstrual regulation procedures (Álvarez et al., 1999). Other estimates for 1996 give much higher figures, with a 78‰ rate and a ratio totaling 58.6 abortions for every 100 pregnancies, with each woman experiencing an average of 2.3 abortions during her reproductive life. The estimated abortion rate includes both menstrual regulation, which some authors believe accounts for 60% of both early abortions performed without pregnancy tests and the termination of proven pregnancies (Henshaw et al., 1999). According to Álvarez, in 2004, the abortion rate was 20.9 for every 1,000 women ages 12 to 49, and 34.4 abortions for every 100 pregnancies; the differences between these estimates and the ones mentioned above raise problems since they do not consistently include menstrual regulation (Álvarez, 2005).

By contrast, Puerto Rico, which also has non-restrictive legislation and reliable information systems, has a considerably lower abortion rate: 22.7 for the period from 1991-1992. Cuba and Puerto Rico are both states with low fertility rates, yet they present significant differences as regards the use of modern contraceptives and above all, the quality of family planning services. Whereas Cuba has a limited range of contraceptive methods (the IUDs are of poor quality and the availability of this type of products is usually irregular), Puerto Rico has a high prevalence of modern contraceptive use and the majority of the population has access to quality health services (Henshaw et al., 1999).

In the French Antilles, where abortion is legal, the rate per thousand women aged 15 to 49 in 1997 was 49‰ in Guadeloupe, 26‰ in Martinique and 27‰ in French Guyana. Moreover, Guadeloupe has a high proportion of repetitive abortions (43% in that country, 25% in Martinique and 37% in French Guyana) (Le Corre and Thomson, 2000). Between 1995 and 1999, the increase in the number of abortions was 4% in Guadeloupe, 17% in Martinique and 13% in French Guyana (Boudan, 2000). In 2003, the rate for women aged 15 to 49 was estimated at 41.5‰ in Guadeloupe, 22.9‰ in Martinique and 37.1‰ in French Guyana (Vilain, 2005).

For other Latin American countries, with legislations ranging from totally restrictive to allowing abortion under certain conditions, it is estimated that during the 1990s, according to the hospital registers and estimations of the Alan Guttmacher Institute (AGI), (1994), approximately 2.8 million abortions were performed a year in six countries in the region – Brazil, Colombia, Chile, Mexico, Peru and the Dominican Republic (see Table 2). Extrapolating this figure for the entire region, and given that the population of these countries accounts for 70% of the total in Latin America, a total of 4 million abortions annually is estimated in this region. According to Henshaw et al., (1999), during this period, the proportion of abortions for every 100 pregnancies oscillated between a minimum of 17% for Mexico and a maximum of 30% for Brazil and Peru and 35% for Chile. According to another study, in 2000, Peru had a higher abortion ratio, with 54 cases for every 100 live births (Ferrando, 2002).

The estimates of Henshaw et al. (1999) suggest that nearly three abortions are reported for every 10 pregnancies in four countries: Brazil, Colombia, Peru and the Dominican Republic. The estimates for Chile are slightly higher, 3.5 abortions for every 10 pregnancies, and far lower for Mexico, where there are fewer than 2 abortions for every ten live births. Peruvian women have the highest risk of having an abortion, 56 of the 1,000 women aged 15 to 49, followed by women from Chile (50‰) and the Dominican Republic (47‰). The estimates for Brazilian, Colombian and Mexican women are fairly similar to each other, yet lower than the previous ones (30‰, 26‰ and 25‰ respectively). However, they also suggest a fairly high level. An average of 1.2 abortions per woman throughout her reproductive life is estimated for all these countries.

On the other hand, as we shall see later on, surveys, records and other empirical evidence from other studies show varied and conflicting estimates of the incidence of abortion in certain countries.

On the basis of hospital records, and considering that only one out of every 7 abortions is performed in a hospital (AGI uses this same figure), in a study conducted in Peru, it was estimated that in 1994 there were 271,150 abortions, as opposed to 351,813 in 2001. It was also estimated that the ratios of abortions for every 100 live births for each of these years, were 43 and 54 respectively (Ferrando, 2002). In Mexico, the National Population Council estimated that there were 220,000 abortions a year during the period from 1990 to 1992 (Lerner and Salas, 2003), whereas in the period from 1995 to 1997, 196,000 were registered. Likewise, according to this organization, the overall abortion rate fell from 1.2 in 1976 to 0.1 in 1997. This decrease coincided with a rapid increase in contraceptive use. However, adds CONAPO, the proportion of woman who have ever been pregnant and who have experienced at least one abortion remained virtually constant between 1992 and 1997 in the national sphere (between 19.6% and 19% respectively) (CONAPO, 2000). The low estimated rate of abortion, compared with that of other countries in the region with similar levels of development, can be largely explained by the favorable impact of family planning policy in Mexico. Nevertheless, there is also a possibility that abortion levels in the country may have been underestimated (Ojeda et al., 2003). Official data in Mexico contrast with those of the AGI, according to which 533,100 abortions were performed in that country in the 1990s (Alan Guttmacher Institute, 1994).

Although the Dominican Republic and Colombia are in an intermediate position in relation to other Latin American countries, abortion is a fairly widespread practice in both. In the former, the number of annual abortions in late 1980 was estimated at 65,000, equivalent to the interruption of one pregnancy for every three births. By early 1990, it was estimated that the number of annual abortions had risen to 82,500 (Paiewonsky, 1999; Alan Guttmacher Institute, 1994). It is estimated that in 1992, 16,500 women were hospitalized for abortion complications in this island country (Ferdinand, 2000). Other authors report that in Colombia, hospital abortions appear to have remained stable in recent years, whereas extra-hospital abortions have increased (Prada, 1994, quoted in Zamudio and Rubiano, 1994,).

Chile, where abortion is considered illegal under all circumstances, is one of the countries in the region with the highest abortion levels, with rates oscillating between 45 and 50 for every thousand women and just over three abortions for every ten pregnancies. This condition has prevailed over the past three decades (Henshaw et al., 1999; Bay et al., 2004).

In a summary of the practice of unsafe abortion in Latin America, Paxman et al., (1993), highlight the long history of the practice of abortion and its high rates, despite its illegality. Between 1940 and the late 1980s, abortion rates in Chile varied from 32 to 77 cases for every thousand women of reproductive age. After achieving a peak in the 1960s (at 77‰), this rate fell, due to social changes and greater access to family planning. As another study points out (p. 194), “the Chilean status regarding fertility is paradoxical: abortion is prohibited, contraceptive prevalence is high and modern methods are both available and affordable. Nevertheless, the number of abortions is high and similar to the incidence of unplanned or unwanted pregnancies” (Den Draak, 1998, quoted by Bay et al., 2004). For other authors (Women’s Health Journal, 1999; and 2003), the high abortion rate in Chile indicates that current legislation on abortion contradicts public opinion in that country, where a high percentage of men and women support the female population’s right to decide on its reproduction. This fact also suggests that the country’s legislation in this matter fails to reflect women’s health needs and demands. As repeated in the literature on the topic, the legal prohibition on obtaining an abortion, as well as the moral sanctions imposed by the Catholic Church and other faiths, are not restrictions that prevent this practice.

In Ecuador, where there is very little information available on abortion, Fassin and Delafosse (1992) conducted a study on the basis of hospital statistics. According to their research, in 1990, the abortion ratio was 11 to 12 for every 100 pregnancies that took place in the various regions, but these data failed to distinguish between miscarriages and abortions. Another article makes it clear that the number of abortions is unknown. It is estimated, however, that if one abortion is performed for every two live births, then the number of abortions a year would be approximately 135,000 (Voluntad, 1993). According to the Demographic and Health Survey quoted by Remez (1996), 8% of women had an abortion or a miscarriage, with women of a higher socio-economic level reporting a higher number of abortions than miscarriages (17% as opposed to 8%).

Dewart estimates that in the late 1980s in Guatemala there was a ratio of between 4 and 18 abortions for every 100 pregnancies (Dewart, 1992). According to Singh, (2005), in 2003, 55,000 abortions were performed in the country, meaning a rate of 20 abortions per 1,000 women and a ratio of 12 abortions for every 100 pregnancies. In Haiti, where existing information is scarce and unreliable, several authors believe that abortions are widely performed; nevertheless, data from the 1994-1995 Demographic and Health Surveys (DHS) “do not corroborate this assumption, since they find that only 6% of women in Port-au-Prince and 3.5% from other urban areas had had any experience of abortions since the start of their sexual activity” (p. 194) (Bay et al., 2004).

In Nicaragua, with a female population of over two million, nearly a quarter of which are of reproductive age, over 14,700 women were hospitalized every year for the treatment of incomplete abortions between 1992 and 1996 (Blandón et al., 1998). According to the United Nations Population Fund (UNPFA), in 1998, 6,000 Nicaraguan women had risky abortions, and half of these women were teenagers (McNaughton, 2002). Rayas and Cattoti (2004) indicate that during the period from 1995-2000, there were 191,466 abortions in this country.

It is estimated that by 2000, approximately 33,000 clandestine abortions were performed in Uruguay, meaning a clandestine abortion ratio of 38.5 for every 100 pregnancies. In other words, out of every 10 conceptions, 4 ended in illegal abortions (Sanseviero, 2003). However, Sanseviero does not regard this figure as an underestimate, despite previous estimates of 50,000 to 300,000 abortions: “these last figures seem excessive for a population of just over 3 million inhabitants in the entire country”.

In Argentina, several authors indicate that abortion is an extremely widespread phenomenon. They judge there to be no accurate data on this, although, according to some experts, the number of abortions performed there every year is estimated at between 335,000 and 500,000 (Checa, 1996; Gogna et al., 2002).

In Guyana, according to the medical records of 7 family planning clinics in Georgetown for 1992, nearly half the women declared having had at least one abortion, while a quarter stated they had had more than one. The abortion ratio was 41 abortions for every 100 live births, although doctors assume that this figure could be almost double, in other words, 1:1 (Núnes and Delph, 1995).

The Variability and Intensity of the Estimated Occurrence of Abortion within Countries

Research based on surveys and medical records, whether national or applied in specific local contexts or population groups, provides relevant data on the intensity and variability in the patterns of the practice of abortion in certain countries. This undoubtedly reflects the heterogeneity of the conditions prevailing in different geographical and social areas.

In 1992 in Colombia, Zamudio et al. (1999) carried out a study of the urban areas with over 100,000 inhabitants where 73% of the population was concentrated. Their results show that nearly a quarter of all the women interviewed between the ages of 15 and 55 had had at least one experience of abortion, and that among women that had been pregnant at least once, the proportion of abortions was 12.4 for every 100 pregnancies. In 1992, the abortion rate for all women in these regions was 25 for every thousand. The differences in abortion incidence between the various regions considered in this study varied from 18% to 30% for all the women that have had an abortion. This variation ranged from 19% to almost twice that (37%) for those that had had at last one pregnancy, with a ratio of 7 to 15 abortions for every hundred pregnancies. According to the authors, these differences are due to the different ages at first intercourse, the greater or lesser influence of the Catholic Church in each of the urban areas, and obviously to the women’s degree of willingness to report this type of event. The breakdown of the specific abortion rates by age or generation show that the average annual proportion of women with an experience of abortion rose sevenfold from the period 1952-1956 to 1988-1991. This increase may be due to the fact that recent abortions are better reported than those that took place in previous years.

In a study conducted in Mexico (Núñez Fernández, 2001), the author presents the results of two surveys designed using alternative methodologies to estimate the incidence of abortion and record the population’s views on it. In the first survey, conducted in 1989 in four neighborhoods of low socio-economic status in Mexico City, women of fertile age declared that of all the first pregnancies lost, 17% were abortions, a figure that rises to 46% according to men’s declarations. In the second survey, conducted in 1991 in three of the four neighborhoods where the previous survey was undertaken, and using the conventional procedure of directly asking the woman about the occurrence of abortions, it was found that one out of three women reported having had an abortion. Conversely, when a different method of collecting information was used, in which the woman herself provided the answer and placed it in a sealed box (in other words, gave a confidential response), the number of abortions declared rose by 50%.

Ojeda et al. (2003) provide information on Mexican women and those of Mexican origin living in the trans-border region of Tijuana, Mexico and San Diego, United States, using data from two sources. The “Survey on the Social Status of Women and Reproductive Health in Tijuana, BC” includes the answers of 2,706 women affiliated with the Mexican Social Security Institute, hospitalized for various reasons related to pregnancy, during the spring and summer of 1993. Hospital records from a San Diego clinic were also used. The first survey found that 7.4% of the women received medical treatment for abortion-related causes and 23% declared that they had had at least one pregnancy. The majority declared that they had had only one abortion (78%) and the rest more than one, without specifying whether these were miscarriages or induced abortions. Conversely, according to the San Diego Hospital records and during the same period of study as the previous survey, 69% of the Mexican women that received treatment related to their pregnancies had been admitted for abortion. In San Diego, interrupting a pregnancy is regarded as a constitutional right. For this reason, it is performed in safe medical conditions. This contrasts with what happens in Mexico where, although the legislation permits abortion under certain conditions, access to this procedure is restricted in practice due to social, cultural, moral and even political reasons. This encourages the clandestine nature of abortions, meaning that they are performed in extremely risky conditions for health.

According to Álvarez et al. (1999), the Cuban province of Havana has the highest abortion ratio in the country – 53.3 for every 100 live births – while in other provinces this ratio does not exceed 49.5. The findings by these same authors in a survey undertaken in 1990 in a municipality of Havana indicate that, of all the women interviewed aged 13 to 34, 9.7% had had an abortion in the year before the interview and 54% had interrupted a pregnancy at some time in their lives. Over a third of the abortions had been carried out as part of the menstrual regulation procedure.

On the basis of data from a survey conducted in three marginalized communities in the north of Santiago, Molina et al. (1999), found that, of all the women interviewed in the Chilean capital, 30% had experienced an abortion in their lives: of these, 72% reported them as miscarriages, a figure which is over twice that estimated in the literature for women of childbearing age. The abortion rate for every 1,000 women in the three communities was 53.8 prior to this study, conducted between 1987 and 1988, and 28.1 after this period, in 1989 and 1990. The abortion ratio was 154.8 for every hundred pregnancies, with a very similar ratio after the intervention, which is also reported in an article by Paxman et al. (1993).

In a survey undertaken in four provinces in Bolivia (Cochabamba, Chuquisaca, La Paz and Santa Cruz, in which approximately 7,500 women were interviewed, the prevalence of abortion was low, 5.7%, and higher in urban (7.2%) than in rural zones (1.3%). This prevalence varied from 1% to 2% in Cochabamba and Chuquisaca to 5% in La Paz and 10% in Santa Cruz (Tinajeros, 2005).

On the basis of various sources, it is estimated that of the total of 33,000 abortions performed annually in the year 2000 in Uruguay, just over two thirds were performed on women from the interior of the country and the rest on residents of the capital, Montevideo. A total of 16,000 abortions are performed annually in the city’s clandestine clinics and 11,000 in those in the interior. Through information from hospital discharges in the public sector, it is estimated that there were another 6,000 abortions performed in hospitals (Sanseviero, 2003).

A study undertaken in Brazil, Colombia and Mexico presents estimates of abortion trends in the different regions of each of these countries, during three different periods ranging from 1976 to 1991 (Table 3). The data are taken from indirect estimates based on hospital records and in which the levels of under-registration considered in 1990 for each of these nations were – in the order above – 15%, 17% and 20% respectively, although it was much higher for previous years (Singh and Sedgh, 1997). Nationwide, the estimated abortion rate for the first period, 1976-1980, was 22 for every thousand women of childbearing age in Brazil and Mexico and 31 for every thousand in Colombia. The authors consider these rates to be moderate compared with those in other countries in the world. It should be borne in mind, however, that these cases involve illegal abortions, possibly with higher levels of under-declaration during the first time period studied. By the mid-1980s, rates had only risen slightly in these three countries. Since the start of this decade and until the beginning of 1990, the abortion rate continued to rise in only Brazil (to 39 per thousand). In Mexico and Colombia, the evidence leads the authors to assume that rates stabilized during the three periods considered.

Patterns related to abortion also vary within each country. In Brazil, for example, the abortion rate rose above the national average in the north and northwest, which are characterized by higher poverty levels. Although the rate also rose in Rio de Janeiro, it did not change significantly in Sao Paulo. In the south, the most developed region and one with a higher percentage of population of European origin, a slight yet stable reduction was observed after 1980. By 1991, the rate here had fallen to below the national average. In the Atlantic region of Colombia, abortion rates stabilized and there were only slight variations. But in the central and eastern regions, there was a marked increase from the mid-1980s onwards. The Pacific and Bogotá regions of this country showed a progressive reduction from the mid-1970s onwards. In the case of Mexico, the increase in the abortion rate between 1977 and 1987 was almost exclusively the result of a change in trends in the southwest and the Federal District (which includes the metropolitan zone), which were the only ones that experienced a sharp increase during this period. In other regions of the country, with the exception of the northeast, abortion rates remained reasonably stable, with a low-to-moderate rate during the 1980s. From the middle of this decade until the early 1990s, the southeast and the Federal District were the only regions in which the abortion rate dropped sharply; conversely, in the other regions, there were slight increases. According to Singh and Sedgh (1997), this pattern suggests that certain characteristics of metropolitan zones may influence the phenomenon of abortion, such as women’s higher educational attainment and the better quality of family planning services, which leads to better contraceptive use. In the case of Brazil, however, this tendency is not as clear. As Ahman and Shah (2002) suggest, this fact could be linked to the frequent practice of sterilization in the country. The study suggests that this could lead a larger number of women to seek abortion to space their pregnancies rather than limit their pregnancies though sterilization.

Conversely, as shown in Table 3, during the period studied, abortion ratios increased consistently in the three countries indicated, especially in Brazil during the last period. This upward trend may be associated with a drop in the number of live births, combined with an increase in or the stability of abortion rates in many regions in these countries. Whereas from the mid- to the late 1970s, the abortion ratio in each of the countries mentioned was 10 to 18 cases for every hundred pregnancies, in the early 1990s, it rose to 18 to 31 abortions for every 100 pregnancies.

An examination of the results within each country individually shows that abortion ratios increased in 5 out of the 6 regions considered in Brazil, in 3 out of the 5 regions examined in Colombia and in 4 out of 5 in Mexico. In certain regions, where abortion rates began to decline or stabilized (the south of Brazil, Bogotá and the Pacific zone in Colombia, and the southeast and the Federal District in Mexico), abortion ratios also became more stable or declined during the 1980s. In the early 1990s, however, a significant proportion of pregnancies in the three countries – almost a third in Brazil, a quarter in Colombia and nearly a fifth in Mexico – ended in abortions.

^ Top of page

Home | Summary | Acknowledgements |