Unintended Pregnancy And Induced Abortion in Colombia
CAUSES AND CONSEQUENCES

Elena Prada Susheela Singh Lisa Remez Cristina Villarreal

Acknowledgments

This report was written by Elena Prada and Lisa Remez, both independent consultants; Susheela Singh, Guttmacher Institute; and Cristina Villarreal, Fundación Oriéntame. Haley Ball edited the report, and Kathleen Randall supervised layout and production; both are at the Guttmacher Institute.

The authors thank the following current and former Guttmacher Institute staff members for providing assistance at various stages of the report’s preparation: Fernanda Abella, Suzette Audam, Akinrinola Bankole, Ann E. Biddlecom, Patricia Donovan, Gustavo Suarez and Jonathan Wittenberg. They also appreciate the valuable input of the following external reviewers: Deborah Billings, University of South Carolina; Teresa DePiñeres, University of California, San Francisco; Sandra García, Katherine S. Wilson and Rebekah Horowitz, all of the Population Council, Mexico; Pío Iván Gómez, Universidad Nacional de Colombia; Ana Cristina González, independent consultant, Colombia; Guillermo López-Escobar, Academia Nacional de Medicina, Colombia; Axel Mundigo, Senior Fellow, Guttmacher Institute; Melanie Peña, Gynuity; María Isabel Plata, Profamilia, Colombia; Mónica Roa, Women’s Link Worldwide, Colombia; and Luis Távara, Federación Latinoamericana de Sociedades de Obstetricia y Ginecología.

In addition, the project benefited from invaluable advice and direction from a Colombiabased Technical Advisory Committee. The committee’s members were Lenis Urquijo, Victor Hugo Alvarez and Lorenza Ospino, all at of the Ministerio de Protección Social (formerly the Ministerio de Salud); Luis Guillermo Cantor and Teresa Moreno, at the time both at Bogotá’s Secretaría de Salud; Wilma Castillo, Federación Colombiana de Sociedades de Ginecología y Obstetricia; Teresa DePiñeres, University of California, San Francisco; Guillermo López-Escobar, Academia Nacional de Medicina, Colombia; José Molina, Universidad Nacional, Colombia; Alberto Rizo-Gil, Academia Colombiana de Salud Pública; Germán Uriza, Asociación Colombiana de Facultades de Medicina; and Lucy Wartenberg, United Nations Population Fund, Colombia.

1

The report draws on 2008 data from two original surveys conducted in 2009. Elena Prada, Ann E. Biddlecom (formerly with the Guttmacher Institute) and Susheela Singh were co-investigators responsible for overall study design and implementation of the surveys, and for data analysis with valuable support from Suzette Audam. Margoth Mora, Fundación Oriéntame, provided research support in the areas of interviewer training, field supervision and quality control. Sonia Arias, Ida Baena and Adriana Reyes were field work supervisors and conducted the interviews with the professional experts. Mary Baena, Camelia Castro, Gloria Gómez, Meira Horta, Isabel Orjuela, Aurora Peña, Gloria Rivera and Patricia Vásquez served as interviewers for the health facilities survey. Lastly, we would like to thank the 102 medical and nonmedical professionals who participated in the experts survey for sharing with us their invaluable knowledge and insights about induced abortion in the country.

This report was supported by a grant from the Dutch Ministry of Foreign Affairs. The Guttmacher Institute also gratefully acknowledges the general support it receives from individuals and foundations—including major grants from The William and Flora Hewlett Foundation, The David and Lucile Packard Foundation and the Ford Foundation—which undergirds all of its work.

2

Suggested citation: Prada E et al., Unintended Pregnancy and Induced Abortion in Colombia: Causes and Consequences, New York: Guttmacher Institute, 2011.

Guttmacher Institute 125 Maiden Lane New York, NY 10038 USA Telephone: 212‐248‐1111; Fax: 212‐248‐1951 E‐mail: info@guttmacher.org 1301 Connecticut Avenue NW, Suite 700 Washington, DC 20036 USA www.guttmacher.org

©2011 Guttmacher Institute, a not‐for‐profit corporation advancing sexual and reproductive health worldwide through research, policy analysis and public education. All rights, including translation into other languages, reserved under the Universal Copyright Convention, the Berne Convention for the Protection of Literary and Artistic Works and the Inter‐ and Pan American Copyright Conventions (Mexico City and Buenos Aires). Rights to translate information contained in this report may be waived.

English translation of Embarazo no deseado y aborto inducido en Colombia: causas y consecuencias, ISBN: 978-1-934387-09-2.

3

Table of Contents

Executive Summary

5

Chapter 1: The Troubling Reality of Clandestine Abortion

9

Chapter 2: Current Incidence and Practice of Induced Abortion

19

Chapter 3: Abortion Complications and Postabortion Care

34

Chapter 4: The Root Cause of Induced Abortion: Unintended Pregnancy 43

Chapter 5: Conclusions and Implications

52

References

60

Appendix Tables

68

4

Executive Summary Examining the hidden and stigmatized practice of induced abortion is very hard to do. Despite a 2006 Constitutional Court ruling that partially legalized abortion, only a tiny proportion of all induced abortions that take place in Colombia are legal. Those that do not meet the limited legal criteria may pose a grave risk to women’s health and wellbeing. This report presents estimates, derived using an indirect technique, of the levels of induced abortion in the country. It discusses Colombian women’s ability to plan their pregnancies and what happens when they are unable to do so. Trends in abortion over the past two decades are examined, along with trends in what leads directly to women’s recourse to it—namely, unintended pregnancies. The report focuses on the current practice and conditions of abortions, whether they meet the legal criteria or not.

Progress has been made on many fronts • As of 2010, a high proportion—nearly 80%—of Colombian women in a union practice contraception, with an encouraging 73% using a modern method and only 6% using a less effective traditional method. • Increased use of contraceptives is an essential factor underlying the country’s fertility trends: Average family size, which has been falling steadily over the past two decades, is now at replacement level (2.1 children per woman). • A legal breakthrough occurred in 2006 when the Constitutional Court lifted the total ban on induced abortions to legalize the procedure in three circumstances (when a doctor certifies that the life or health of the pregnant woman is threatened, when a doctor certifies that the fetus has an abnormality incompatible with life, and when a pregnancy results from an incident of rape or incest that has been duly reported to the authorities). The Court’s decision was framed in terms of women’s inviolable rights to health and life.

5

Yet unintended pregnancy and unplanned births are widespread • Each year in Colombia, there are 89 unintended pregnancies (i.e., those that are wanted at a later time or are not wanted at all) per 1,000 women of reproductive age. Rates vary widely among regions, from 67 per 1,000 in the regions of Central and Oriental, to nearly twice that in the region of Bogotá (113 per 1,000). • Despite notable gains in contraceptive use over the past two decades, growing motivation to have smaller families means that the proportion of all pregnancies that are unintended rose from one-half to two-thirds during that period. • Unintended pregnancy often leads to unplanned births. The proportion of recent births that were unplanned has risen dramatically, from just 36% in 1990 to 51% in 2010, with notably little difference across regions in 2010.

Many unintended and unwanted pregnancies end in abortion • An estimated two-fifths (44%) of all unintended pregnancies in Colombia end in an induced abortion. • This translates to an estimated 400,400 induced abortions each year. As of 2008, only about 322 (0.08%) of these abortions were reported as legal procedures. • The absolute number of abortions rose nearly 40% from 1989 to 2008, largely because there are many more women of reproductive age today than there were two decades ago. • The country’s annual abortion rate rose slightly over that period, reaching 39 abortions per 1,000 women of reproductive age in 2008, compared with 36 per 1,000 in 1989. Rates of abortion range widely, from 66 per 1,000 women in Bogotá to just 18 per 1,000 in Oriental, likely reflecting regional differences in the strength of women’s motivation to avoid giving birth.

6

•

How the number of abortions relates to the number of births is an indicator of women’s motivation to avoid giving birth when faced with an unwanted pregnancy. There are currently 52 abortions for every 100 live births, a substantial increase from 35 per 100 in 1989.

Unsafe abortions endanger women’s health and burden the health system • An induced abortion performed outside the law can be unsafe. As a result, an estimated one-third of all women having a clandestine abortion develop complications that need treatment in a health facility. The rate of complications is highest for the abortions of poor rural women, compared with the abortions of women in the three other subgroups by poverty and area of residence (53% vs. 24–44%). Unfortunately, one-fifth of all women experiencing abortion-related complications do not receive any treatment at all, and these women are especially likely to suffer debilitating consequences. • Each year, the Colombian health system treats 93,000 women for postabortion complications, and these avoidable complications drain scarce medical resources. Currently, nine women per 1,000 of reproductive age receive facility-based postabortion care. This treatment rate—and the attendant burden on the health system—is highest in the region of Pacífica, where 16 out of every 1,000 women receive treatment each year. • An estimated half of all abortions in Colombia are induced using the drug misoprostol. Providers’ inadequate knowledge of evidence-based protocols, and women’s misunderstanding of when and how to use the method, likely lead to an unnecessarily high complication rate—32%, usually heavy bleeding and incomplete abortion, for which many women seek facility-based care. • Women who are poor and live in rural areas are especially likely to not use misoprostol and turn to traditional midwives or to self-induce. Overall, the highest estimated complication rate for all abortions is 54–66% for those induced by

7

methods other than misoprostol and performed by unskilled traditional providers or by the woman herself.

Action is needed to improve women’s health and lives The recent rise in unintended pregnancies and unplanned births—not to mention persistently high rates of clandestine abortion—point to the need for concerted, unified efforts across the spectrum of Colombian society. Below are some steps to help reduce unsafe abortion’s burden on women and the medical system; improve the provision of legal procedures; and reduce unintended and unwanted pregnancy, the root cause of the vast majority of abortions.

Strengthen contraceptive services. Women and service providers need better information about correct and consistent method use to utilize their current methods as effectively as possible. Access to emergency contraception should be expanded to improve women’s ability to avoid unwanted pregnancy and its consequences. Tailored interventions are needed to meet the contraceptive needs of groups at high risk for unwanted pregnancy. Improve postabortion care services. The coverage of postabortion services needs to be extended and their quality improved. Providers need more accurate information about caring for women who have used misoprostol; they also need training in treating complications with manual vacuum aspiration, a technique far less invasive and less resource-dependent than the widespread dilation and curettage. Improve implementation of the Constitutional Court decision and provision of legal abortions. Public education campaigns are needed to continue to spread awareness of the ruling, as are mechanisms to assure that legal abortion guidelines are strictly followed. It is also vital to research the types of barriers to legal abortion that women and providers currently face.

8

Chapter 1: The Troubling Reality of Clandestine Abortion

Women worldwide cope with the heavy burden of unintended pregnancy. Each country’s particular social, political, cultural and economic context influences a woman’s ability to avoid unintended pregnancy and mediates her response if she experiences one. National law, health policies and services also play an important role. Colombia is no exception.

The country has made great strides in extending throughout the country the means to avoid unintended pregnancy (contraceptive methods), yet the problem of unintended pregnancy has not gone away. As a result, induced abortions are common. And despite a 2006 court ruling that allowed legal abortions under three limited circumstances, nearly all current abortions—at least 99.9%—occur outside the law. Such clandestine abortions, when performed in unsafe conditions by untrained providers, can lead to complications that adversely affect women’s health. Thus, induced abortion continues to threaten the well-being of thousands of Colombian women each year.

In Colombia, as in most of Latin America, induced abortion is contentiously debated, strongly stigmatized and legally restricted. In this environment, religious and moral condemnation of abortion affects attitudes toward it at all levels of society. A decision handed down by the country’s Constitutional Court in 2006 (see law box) lifted the absolute ban on all abortions to allow legal procedures in three circumstances: when a physician certifies that the life or health of the pregnant woman is threatened, when a physician certifies that the fetus has an abnormality incompatible with life, and when a pregnancy results from an incident of rape or incest that has been duly reported to the authorities.1

Among the small proportion of women who seek a legal abortion, many likely experience difficulties actually obtaining one. Recent evidence documents many instances of women encountering daunting institutional and bureaucratic obstacles to obtaining a legal procedure from health facilities that are obligated to provide them.2 Thus, illegal—and potentially unsafe—abortions continue to exact a heavy toll on Colombian women’s

9

well-being. They also represent an avoidable drain on the country’s health system, which is responsible for treating most postabortion cases.3 This report presents estimates of the number of clandestine*A abortions in 2008 and the burden that their consequences posed to women and health care facilities that year. It also examines trends since 1989, the last time such estimates were made. Knowing the current level of induced abortion is essential for informing public policy debate and assessing how well women are able to avoid unintended pregnancy. These estimates are key to determining how to improve contraceptive use, increase access to legal abortion and, when unsafe procedures result in complications, assure better access to postabortion care. The report also assesses current efforts to implement the ruling that partially decriminalized abortion and sheds light on unintended pregnancy—the reason women seek abortion in the first place—and its causes.

Much has happened since the 1989 abortion study. Notably, women are now having nearly one child fewer than they did just two decades ago—a decline from a lifetime average of 2.9 children in 19904 to 2.1 children as of 2010.5 Indeed, Colombia has now reached replacement fertility, which means that the population will no longer grow every year. However, even though women want far smaller families now than they did in the past, many continue to have more children than they want: Colombian women would have an average of just 1.6 children if they could avoid having births they say they do not want to have.5 Women’s inability to achieve their preferred family size inevitably results in unplanned births (i.e., those that are wanted at a later date and those that are not wanted at all). In Colombia, the increase in unplanned childbearing has been dramatic: Some 51% of recent births were unplanned in 2010,6 compared with 36% in 1990.7 This increase in the measure of unplanned births, however, does not show us how the desire for increasingly smaller families may have affected trends in pregnancies that end in abortions. A

*Clandestine abortions are those that do not meet the strict legal criteria and are carried out in secrecy, under medical conditions that may be either safe or unsafe.

10

What societal changes may have affected abortion trends? Many recent developments have likely affected the context in which Colombian women and couples are confronting unintended pregnancy. One of the most important is the landmark Constitutional Court decision of May 2006 (Sentencia C-355/06). In addition to legalizing certain abortions, the decision mandated that the procedure be made available to all women who meet the criteria, regardless of their ability to pay. The case was argued on the basis that Colombia’s absolute ban on all abortions—including those needed to save the life of the pregnant woman—violated the country’s obligations to protect women’s health, as specified in regional and international treaties that the country had ratified and that supersede national law. The ruling was framed in terms of women’s human rights, particularly their rights to health and life. It also lifted the blanket prohibition against abortions with consent among minors younger than age 14 (usually victims of rape or incest).1

Another important development that has likely deeply affected the practice of abortion in Colombia and throughout Latin America8 was the introduction of the drug misoprostol in the 1990s. Misoprostol, also known by the brand name Cytotec, is a prostaglandin that causes uterine contractions and was originally marketed as an ulcer drug. This abortion medication is known for its low cost, its ease of acquisition and the anonymity it affords. Although misoprostol has a high effectiveness rate (84–96%) when administered correctly in controlled clinical studies,9,10 many women who use it lack accurate and complete information about the drug and end up seeking care at health facilities as a result. Many health care providers, researchers and women’s health advocates believe11— and reports in the popular press appear to confirm12,13—that the use of misoprostol as an abortifacient has increased dramatically over the past two decades.

Demographic shifts have also likely contributed to trends in the number of abortions. Due to high population growth rates in the past, there are simply more women of reproductive age exposed to the risk of unintended pregnancy now than there were in 1989. Indeed, the

11

population of reproductive-age women rose from 7.9 million in 1989 to 10.2 million in 2008.14 Moreover, research has shown that women living in cities and those who are well educated tend to have higher abortion rates than others.15 Colombia continues to urbanize at a rapid pace: The proportion of the population living in urban areas, where the desire for small families is especially strong, rose from 68% in 1990 to 75% in 2010.16 And in recent years, the proportion of reproductive-age women who have attended university almost tripled, from 9% in 19904 to 24% in 2010,5 while the proportion with no more than a primary education dropped by almost half, from 41%4 to 22%.6

Some subgroups of women face increased risk of unintended pregnancy The risk of unintended pregnancy—which is closely linked to the likelihood of having an abortion—is highest among women who do not practice contraception, or who do so sporadically or ineffectively. That risk has likely increased among one group of women who have a particular need to avoid pregnancy: unmarried sexually active young adults. Even though their contraceptive use has increased dramatically, single young women’s especially strong motivation to avoid unplanned births does not always match their effective and consistent use of modern methods.

Another group especially vulnerable to unintended pregnancy is women who have been displaced by the social and political violence that has plagued Colombia for decades. An estimated 3.3–4.9 million Colombians were displaced as of the end of 2009; the only country with a larger displaced population at the time was Sudan.17 Given the upheaval in displaced women’s lives and their potential exposure to sexual violence, they are likely to want to postpone pregnancy. Yet displaced women’s abject poverty and social isolation18 may limit their access to sexual and reproductive health care, especially contraceptive services.

Guide to the report This report aims to disseminate findings on the current situation of induced abortion— both legal and illegal—to a wide audience to encourage more informed public discourse

12

and policy formation. Chapter 2 provides estimates for 2008 of the number of abortions in Colombia overall, as well as in each of five main regions—Atlántica, Bogotá, Central, Oriental and Pacífica (see map). These findings result from the application of an established indirect estimation methodology that relies on a range of sources (see data sources box). Chapter 2 also examines trends in abortion incidence and describes the current general context of legal and illegal abortion provision. Chapter 3 details the health consequences of unsafe abortion and presents new information on the incidence of postabortion complications that are treated in health facilities. Chapter 4 discusses a range of factors—social, economic, behavioral and service-related—that contribute to unintended pregnancy, the root cause of induced abortion. Finally, Chapter 5 offers recommendations derived from the current data and earlier analyses.

13

LAW 599 (which issues the Penal Code, with changes from Sentencia C-355/06, 2006) Colombian Congress CHAPTER FOUR Del aborto Artículo 122. Aborto. La mujer que causare su aborto o permitiere que otro se lo cause, incurrirá en prisión de uno (1) a tres (3) años. A la misma sanción estará sujeto quien, con el consentimiento de la mujer, realice la conducta prevista en el inciso anterior. Artículo declarado EXEQUIBLE por la Corte Constitucional mediante Sentencia C355 de 2006, en el entendido que no se incurre en delito de aborto, cuando con la voluntad de la mujer, la interrupción del embarazo se produzca en los siguientes casos: (i) Cuando la continuación del embarazo constituya peligro para la vida o la salud de la mujer, certificada por un médico; (ii) Cuando exista grave malformación del feto que haga inviable su vida, certificada por un médico; y, (iii) Cuando el embarazo sea el resultado de una conducta, debidamente denunciada, constitutiva de acceso carnal o acto sexual sin consentimiento, abusivo o de inseminación artificial o transferencia de óvulo fecundado no consentidas , o de incesto. Artículo 123. Aborto sin consentimiento. El que causare el aborto sin consentimiento de la mujer o en mujer menor de catorce años, incurrirá en prisión de cuatro (4) a diez (10) años. Texto subrayado declarado INEXEQUIBLE por la Corte Constitucional mediante Sentencia C-355 de 2006.
Source <http://www.alcaldiabogota.gov.co/sisjur/normas/Norma1.jsp?i=6388>, accessed

May 30, 2010.

14

FIGURE 1.1

Colombia and its regions

Atlántica Pacífica Central Oriental Bogotá National Territories (not included in the study)

Note

Map does not show the islands of San Andrés and Providencia, which are in the Atlántica region.

15

 Data Sources key informants who were typically heads of gynecology and obstetrics departments or other This report draws on a number of data sources to estimate how many abortions occur in Colombia, the conditions under which they are provided and the consequences of the procedures. Because women are understandably reluctant to openly admit to an illegal and highly stigmatized behavior, the data needed to be collected through an indirect estimation technique, the Abortion Incidence Complications Method (AICM). The data used in the technique come primarily from two surveys, a survey of health facilities and a survey of experts in the field. The surveys were conducted in 2009 but asked about events in the previous year; thus, all data refer to 2008. Other key sources include Colombian National Demographic and Health Surveys (Encuestas Nacionales de Demografía y Salud, or ENDS), conducted from 1990 through 2010, and the country’s National Department of Statistics (Departamento Administrativo Nacional de Estadística, or DANE). • Health Professionals Survey (HPS). A purposive sample of 102 professionals was surveyed over the same months as were the key informants from facilities. (For comparability with previous applications of the methodology, we retain the original adjective • Health Facilities Survey (HFS). After the instrument pretest was completed in February of 2009, study personnel (eight interviewers and three supervisors) collected data during March and April from a nationally representative sample of health facilities considered likely to provide legal abortion services or treat women with abortion-related complications. We conducted interviews with “Health” in the acronym, HPS, despite the trend toward needing to include non-health professionals in the sample.) The knowledgeable professionals were asked about the conditions of abortion provision—if women used misoprostol or, for other types of abortions, which providers women went to and women’s likelihood of suffering complications and of being treated for them in a health senior professionals knowledgeable about services provided at the facility. Each respondent was interviewed in person, using a structured questionnaire. A total of 339 public and private health facilities were selected. The HFS sample frame included all hospitals and between 10% and 100% of primary- and secondary-care facilities, depending on the type of facility and level of specialization. Of this original sample, 39 facilities could not be surveyed, leaving a final sample of 300. Interviews were successfully completed for 289 facilities, resulting in a response rate of 96%. (See article for a full discussion of the sampling technique and eligibility considerations.1)

16

 facility. The primary criteria for selecting respondents were their expertise and extensive knowledge of the conditions of abortion provision and postabortion care. Forty-seven were medical providers who worked in public and private practice, and 55 were professionals from other fields, including researchers, policy analysts and advocates. The study team made a concerted effort to include a sufficient number of HPS respondents who were familiar with the context of abortion provision in rural areas. Respondents came from four of the five regions included in the study: Bogotá, Atlántica, Pacífica and Central. The region of Oriental was not covered because of the difficulty of finding 25 professionals who were knowledgeable about abortion services there and because of high costs in terms of field staff. Thus, for data described in the text that were collected solely through the HPS—for example, types of abortions and their consequences—we present values for the four surveyed regions only. However, the application of the full methodology, which combines data from both surveys, uses Central as a proxy for Oriental because their socioeconomic indicators are similar. References 1. Prada E, Biddlecom A and Singh S, Induced abortion in Colombia: new estimates and change between 1989 and 2008, International • ENDS. The demographic and health surveys for 1990,2 2000,3 20054 and 20105 provide national information on contraceptive use, unplanned births and unmet need for contraception. These studies included 8,644 Perspectives on Sexual and Reproductive Health, 2011, 37(3)114–124. 2. Asociación Probienestar de la Familia Colombiana (Profamilia), Encuesta de Prevalencia, Demografía y Salud, Colombia,
• Other primary data sources. Population data

women aged 15–49 in 1990, 11,585 in 2000, 38,355 in 2005 and 49,818 in 2010. Since not all four surveys covered the region of the National Territories (an area that represents 2.1% of Colombia’s population), all data presented in this report exclude this region. Also, the 2010 ENDS used a different name for the Atlántica region, Caribe; we use the term Atlántica for consistency with all other ENDS reports.

on the number of women aged 15–44 for 1989 are from the United Nations’ regional demographic database, Centro Latinoamericano y Caribeño de Demografía. The number of women in 2008 were interpolated from 2005–2010 data from DANE. The number of live births in 2008 was estimated by applying age-specific fertility rates from the 2010 ENDS to the number of women aged 15–49 (obtained from DANE).

17

 1990, Bogotá, Colombia: Profamilia; and Columbia, MD, USA: Institute for Research Development/Macro International, 1991. 3. Profamilia, Salud Sexual y Reproductiva: Resultados Encuesta Nacional de Demografía y Salud, 2000, Bogotá, Colombia: Profamilia, 2000. 4. Profamilia, Salud Sexual y Reproductiva en Colombia: Resultados Encuesta Nacional de Demografía y Salud, 2005, Bogotá, Colombia: Profamilia; and Calverton, MD, USA: Macro International, 2005. 5. Profamilia, Encuesta Nacional de Demografía y Salud, Colombia, 2010, Bogotá, Colombia: Profamilia, 2011. 6. Centro Latinoamericano y Caribeño de Demografía, América Latina: Proyecciones de Población, 1950–2000, Boletín Demográfico 48, Tabla 12a, 1991. 7. Special tabulations of data from DANE, Proyecciones Nacionales y Departamentales de Población, 2005–2020, Bogotá, Colombia: DANE, 2011.

18

Chapter 2: Current Incidence and Practice of Induced Abortion Although fertility declined substantially over the past two decades, the specific role that abortion played in that decline is unknown. A great deal of speculation surrounds the current level of abortion in Colombia. Articles in the popular press12,19 and even a respected medical journal20 put the annual number of abortions between 320,000 and 450,000, but these estimates are not supported by references to rigorous research.

New abortion estimates for 2008 are now available from an application of the indirect estimation technique, the Abortion Incidence Complications Method (AICM; see methodology box). The technique combines data from two major surveys. The first is the nationally representative Health Facilities Survey (HFS) that covered admissions for treatment of postabortion complications and the provision of legal abortions. The second is the Health Professionals Survey (HPS) conducted among a purposive sample of medical and nonmedical professionals who were highly knowledgeable about the conditions of clandestine abortion. (For comparability with earlier applications of the AICM, we retain the acronym HPS, despite the need to include many non-health professionals in the Colombia sample.) The methodology yields estimates of the total number of abortions in Colombia—legal procedures, as well as those that do not meet the legal criteria and are carried out in secrecy, under both safe and unsafe conditions. This same methodology was used in 1989, allowing us to assess changes over time. However, unlike the 1989 Colombia study that was carried out when all abortions were prohibited, the current one provides the number of legal abortions performed in 2008 and the extent to which eligible facilities offered legal services that year.

What is the incidence of abortion in Colombia? An estimated 400,400 induced abortions occurred in Colombia in 2008, which translates to an annual rate of 39 abortions per 1,000 women aged 15–44 (Figure 2.1).21 Viewed another way, one Colombian woman in 26 has an abortion each year. The country’s abortion rate is somewhat above the average for South America as a whole, which the World Health Organization (WHO) put at 33 abortions per 1,000 women in 2003.22,23 According to scarce national-level data for

19

other Latin American countries with restrictive legislation, Colombia’s rate is slightly higher than Mexico’s (33 per 1,000 women in 2006),24 much higher than Guatemala’s (24 per 1,000 in 2003)25 and much lower than Peru’s (54 per 1,000 in 2000).26

Within Colombia, differences in abortion levels by region likely reflect variations in women’s motivation to time their births and have small families, and in their access to contraceptive services that enable them to do so. Women residing in the region of Bogotá have the highest abortion rate, and those in Oriental have the lowest (66 vs. 18 abortions per 1,000 women).21 This large discrepancy likely reflects the exceptionally strong desire to avoid pregnancy among women living in Bogotá. Compared with the national average, rates are also high in Pacífica and Atlántica (54 and 42 per 1,000 women, respectively). (Estimates for the Pacífica region should be interpreted with caution because the four departments that make up the region vary widely in terms of poverty and women’s vulnerability to unintended pregnancy.)

The abortion rate in Colombia increased by roughly 8% over the past two decades. There were 36 abortions for every 1,000 reproductive-age women in 1989*B and 39 per 1,000 in 2008.21 The absolute number of annual induced abortions, however, rose by roughly two-fifths, from 288,400 abortions in 1989 to 400,400 in 2008. The far higher increase in the number of abortions than in the abortion rate reflects the substantial growth in the population of reproductive-age women over the period.

Another way to assess the incidence of abortion is to relate the number of abortions to the number of births. Nationwide, in 2008, there were 52 abortions for every 100 live births—an increase of nearly half from the 1989 level of 35 induced abortions per 100 live births (Appendix Table 1).21 The ratio went up with time because women of reproductive age continued to have abortions at a fairly steady rate, while their annual number of births declined dramatically. B

*The rate originally published for 1989 was per 1,000 women aged 15–49 (reference 55). To make those data directly comparable to the current rate, we recalculated the 1989 rate for the same population base as the 2008 estimates (per 1,000 women aged 15–44).

20

How is clandestine abortion currently practiced in Colombia? According to the perceptions of well-informed professionals, one-half of all women obtaining an abortion in Colombia use misoprostol, which they obtain from a variety of sources, including drug stores and pharmacies, retail outlets, the black market, health professionals and traditional providers.27 The other half are thought to have their abortion induced by a physician, nurse, pharmacist or traditional midwife using methods other than misoprostol, or to induce the abortion themselves with a physical action or a substance other than misoprostol. It should be remembered that these estimates are based on the HPS, a survey that asked professionals in the field for their expert opinion; no nationally representative empirical evidence from women themselves is available on this topic.

Misoprostol abortions. The predominance of misoprostol as an abortion method is unsurprising, given that the medication is comparatively inexpensive, affords women privacy and is thought to be widely available.8,11-13,28,29 Despite the requirement that misoprostol be sold by prescription, it is easily bought on the black market and over the counter in retail pharmacies. Recognizing the high level of clinical effectiveness and safety of this abortion method, the country’s drug regulatory body, Instituto Nacional de Vigilancia de Medicamentos y Alimentos (INVIMA) approved the use of misoprostol to safely induce legal abortion in June of 2007.30 Two years later, the agency widened misoprostol’s approved uses to include the treatment of incomplete abortions.31

In urban areas, where three-quarters of Colombia’s population resides, all types of providers, except midwives, are believed to use misoprostol more commonly than any other method.27 Some 85% of pharmacist-provided abortions are believed to involve misoprostol, as are 60% of self-induced abortions and 50% of physician-provided abortions. Awareness of the drug has likely spread rapidly through word-of-mouth and informal women’s networks.

A woman’s area of residence and socioeconomic status affect the kind of abortion she obtains. Misoprostol-induced abortions are the most common type of abortion among most subgroups of women: Among urban women (both poor and nonpoor) and nonpoor rural women having

21

abortions, 40–59% are thought to use misoprostol obtained from any source or type of provider (Figure 2.2).27 A smaller proportion (25%) of poor rural women’s abortion involve the use of misoprostol because, compared with other women, they may be less informed about the medication and be less able to afford or obtain it. Examining variations in misoprostol use by region*C presents a slightly different picture. Misoprostol abortions from any source or provider are believed to account for 56–61% of abortions in all regions except Atlántica, where they account for 31% (not shown).27 What might explain the exception in Atlántica? Apart from the region of Bogotá (where 81% of women qualify as nonpoor urban, the subgroup most likely to know about and obtain misoprostol), the proportions in the same subgroup (nonpoor urban) are roughly the same in the four other regions studied (54–56%). The much lower use of misoprostol in just one of these four, Atlántica, suggests that women there have less access to the drug or are less informed about it, but more research is needed to verify that hypothesis.

All other abortions. HPS respondents believe that abortions that do not involve misoprostol are provided in roughly equal proportions by three levels of providers: The riskiest are the 16% of all abortions that are performed in the least safe conditions (9% by untrained traditional midwives and 7% induced by the woman herself with no outside help; not shown).27 The somewhat safe abortions provided by midlevel health professionals (pharmacists and nurses) account for 16% of all abortions. The safest nonmisoprostol abortions are likely the 18% provided by physicians (predominantly surgical abortions). (See notes to Figure 2.2 for examples of the methods thought to be used by each type of provider.) Unsurprisingly, the type of abortion considered to be safest—nonmisoprostol abortions performed in a doctor’s office—is most common among women who are nonpoor and urban (25% of their abortions), and ranges from 4% to 17% in the remaining three subgroups.

C

*Whereas the HFS was conducted in five main regions of the country, the HPS could be carried out in only four. Oriental was omitted because of insufficient sample size; thus, the methodology uses HPS data from a similar region, Central, to represent the relevant measures for Oriental. When the report describes the types of abortions women have and their consequences, it presents HPS data from the four surveyed regions only.

22

The likelihood that women will experience complications from a nonmisoprostol abortion is inversely related to their provider’s level of training. Poor rural women are believed to be the most likely to obtain a nonmisoprostol abortion either by going to an untrained traditional midwife or by inducing the abortion themselves—50% are thought to do so, compared with just 5–25% of other women.27 Thus, this already disadvantaged group is disproportionately exposed to the risk of serious complications that can result from unsafe attempts at interrupting an unwanted pregnancy.

Women in Atlántica are believed to be the most likely to see a licensed physician for a nonmisoprostol abortion (28% vs. 12–22% of women in the other regions with data; not shown).27 Only in Pacifica are a sizable proportion of women thought to have a nonmisoprostol abortion performed by a traditional midwife (12%). Overall, self-induced abortions not involving misoprostol are relatively uncommon. They are especially infrequent in Bogotá, compared with the other three regions with data (4% vs. 5–10%).

Many women cannot afford the expense of a safe clandestine abortion The expectation that the widespread use of misoprostol is a function of its relatively low cost11 was confirmed by the key professionals interviewed in the HPS. Experts estimated that, as of mid-2009, a dose of misoprostol sufficient to induce an abortion (four 200 mcg tablets) would cost as little as 36,000 Colombian pesos (US$17*D), if provided by a midwife, to as much as 86,000 pesos (US$39), if provided by a physician.27 Prices charged by pharmacists, nurses and sellers on the black market fell in the middle (57,000–63,000 pesos, or US$26–29). Black market sales of the drug have likely been increasing since 2000,11 with concomitant decreases in sales in the formal sector. (Formal retail sales plummeted between 2006 and 2010.32)

For first-trimester clandestine abortions using any method or procedure, physicians are estimated to charge an average of 393,000 pesos (US$179), most likely for surgical procedures performed in clinics or offices.27 The next most costly abortions are those provided by nurses (on average, D

*Calculated at a May 2009 exchange rate of 2,200 Colombian pesos to one U.S. dollar (source: Banco de la República, Tasa de Cambio Representativa del Mercado, Tabla de datos de la grafica: serie del año previo y del año actual, no date, <http://www.banrep.gov.co/series‐estadisticas/see_ts_cam.htm#trm>, accessed June 10, 2010).

23

164,000 pesos, or US$75), followed by abortions offered by pharmacists or midwives (roughly 90,000 pesos, or US$40 each). In Colombia, as is true the world over, women with economic means are always able to minimize their health risks by paying higher prices to obtain a safe clandestine abortion.

It is important to put the above prices into perspective. In rural areas, the cost of terminating an unwanted pregnancy is far from low, even when performed by an untrained provider. For example, a poor rural woman would likely pay an average of 68,000 pesos (US$31) to a traditional midwife for an abortion (induced by any method or technique).27 Considering that the minimum legal monthly salary in 2009 was 496,900 pesos (about US$225),33 this financial burden—on top of possible stigmatization and health complications—is very heavy indeed.

What about abortions that fit the legal criteria? A crucial difference between the 1989 and 2008 studies was the latter’s inclusion of questions concerning legal abortion. The landmark May 2006 court ruling was accompanied by Decree 4444, issued by the Ministerio de la Protección Social (the former Ministry of Health), which established the regulatory apparatus for compliance.34 The ministry also published guidelines for high-quality legal abortion services (Resolution 4905 of 200635); these were adapted from Safe Abortion: Technical and Policy Guidance for Health Systems, published by WHO in 2003.36 As might be expected, the partial decriminalization of abortion met with vigorous opposition, and many in the medical, religious and political establishments mounted efforts to delay or derail its implementation.

One such effort was the injunction issued by the Consejo de Estado (one of the four organs of the judicial branch) in October of 2009 to temporarily suspend Decree 4444. The injunction questioned the legal authority of the national government, through a ministry, to regulate a Constitutional Court ruling.37 As of June of 2011, when this report was written, the decree remains suspended. However, the overall legal framework guaranteeing women’s constitutional right to terminate a pregnancy in the three specified circumstances remains in place. Consequently, so do the obligations and responsibilities of providers to offer legal services.

24

The HFS found that just 322 legal procedures were performed in Colombia in 2008.*E,38 Together, Bogotá and Pacífica accounted for more than four-fifths of them and very small proportions occurred in each of the remaining three regions. HFS respondents reported that twofifths of legal procedures were performed using the surgical method of dilation and curettage (D&C) on its own, even though legal abortion guidelines recommend that D&C be used only if manual vacuum aspiration (MVA) and misoprostol are unavailable.39 (Although the legal abortion guidelines adapted from the WHO specifically recommend the medication abortion protocol of mifepristone along with misoprostol, mifepristone has yet to be approved by INVIMA.) Just one-fifth of legal abortions reported in the HFS involved MVA, a less invasive, less complex surgical method that does not require general anesthesia. The remaining procedures used misoprostol alone or misoprostol with D&C.

In a country of more than 10 million women of reproductive age, the negligible number of legal procedures confirms that women meeting the legal criteria face serious obstacles to getting a legal abortion. Multiple reasons explain this situation. To begin with, many eligible women were unlikely to have been aware of the criteria in 2008, just two years after the ruling. Most importantly, only 11% of the nationally representative sample of health facilities with the capacity to provide legal abortion services offered them in 2008 (Figure 2.3).38 The proportion providing legal abortions was 2–3 times as high in Bogotá as in the other four regions (23% vs. 8–12%). It is important to note that there was little difference between public- and private-sector facilities in the proportions offering legal procedures (10% and 12%, respectively).

Why were so few eligible facilities not offering legal abortions two years after the court’s decision? Some of the most common reasons cited by HFS respondents were a lack of equipment and infrastructure (55%), lack of demand for legal abortion (29%) and lack of trained personnel E

*The numbers of legal abortions reported to the Ministerio de la Protección Social for 2008 and 2009 are 230 and 331 procedures, respectively. (Source: Information provided by Empresas Promotoras de Salud and Direcciones Territoriales de Salud to the Dirección General de Salud Pública, 2006 through 2010.) Of the 657 legal abortions with data by legal criteria that were officially reported for 2008–2010, 57% met the criterion of grave fetal abnormality, 27% were indicated because of rape or incest, and 16% were needed to preserve the life or health of the pregnant woman. In 2008, the year of our study, 28% of reported legal abortions with data by age were obtained by young women aged 19 and younger.

25

(13%; not shown).38 Furthermore, 14% of HFS respondents pointed to conscientious objection by all staff at the facility, although such institutional claims of conscientious objection are strictly forbidden by the language of the 2006 decision.

Onerous bureaucratic obstacles and inadmissible assertions of institution-wide conscientious objection have also contributed to legal procedures being delayed or denied. Obstructions to obtaining an abortion early in pregnancy can raise the medical risks involved and inevitably lead to even greater barriers, since very few doctors are trained in late abortion techniques. A review of reported cases revealed that some women had been denied a legal abortion by more than one facility. In one highly publicized case, a 13-year-old rape victim was denied an abortion in 2007 by at least six successive health facilities; she was eventually forced to carry the pregnancy to term.40 This type of outcome directly contradicts the court’s ruling and the requirement of the Ministerio de la Protección Social that doctors who conscientiously object to providing legal abortions must refer women to providers who have agreed to perform them.

The detailed guidelines on the provision of legal abortion mean that women who obtain one— though very few in number—likely undergo safe, uncomplicated procedures. That is not the case for the many women who are unable to obtain a safe legal abortion, either because they do not meet the criteria or because they encounter the barriers outlined above. As the following chapter shows, clandestine abortions performed by untrained providers or induced through incorrect use of misoprostol can lead to health consequences that harm women and their families, and place unnecessary strain on the health system.

26

Methodology for estimating abortion incidence This 2009 application of the Abortion Incidence Complications Method (AICM) yields data for 2008 that are directly comparable to the AICM results from Colombia for 1989, enabling analysis of trends over time. The methodology provides national estimates of the number of induced abortions occurring each year (in 2008 and 1989, respectively), the annual number of abortions per 1,000 women (abortion rate) and the number of abortions per 100 live births (abortion ratio). For 2008 only, these measures are available for five major regions of the country. The methodology also yields national-level estimates of the annual number of postabortion complications treated in health facilities per 1,000 women (treatment rate) for both years, 1989 and 2008.

for induced abortion complications over a one-year period, and the proportion of all women having an induced abortion who are treated for complications. The first measure was obtained from the Health Facilities Survey (HFS).

Respondents were asked about the characteristics of their facility and the services it provided, including postabortion care; the procedures used to treat abortion complications; and whether the facility kept statistics on such procedures according to version 10 of the International Classification of Diseases codes. We obtained these data for the previous three years whenever possible. To take into account the likelihood that postabortion caseloads would fluctuate throughout the year, we used the HFS to obtain information on two reference periods: the past month and a typical month. By averaging these data and multiplying them by 12 we arrived at an estimate of the total number of postabortion patients over a full year.

Using abortion, pregnancy and population data, we estimated the rates of unintended pregnancy and pregnancy outcomes (i.e., the proportion ending in planned births, unplanned births, abortions and miscarriages). In applying this methodology, two essential pieces of data were needed: the number of women treated in a health facility

Additionally, HFS respondents were asked whether the facility provided legal abortions. If yes, the key informant was asked the number of legal abortions performed over the past year and the type of procedure used.

27

Facilities that did not provide legal abortions were asked their reasons for not providing the service. Using these HFS data, we estimated the number of women treated for postabortion complications (either from a miscarriage or an induced abortion) in two steps: First, we used data from clinical studies to estimate the number of pregnant women who would have experienced a late miscarriage (at 13– 21 weeks’ gestation, because only women miscarrying late would likely need care; late miscarriages are equal to 3.41% of all reported live births1). Second, because not all women needing facility-based treatment for a late miscarriage succeed in obtaining it, we assumed that the proportion obtaining care for late miscarriages is the same as the proportion of women who give birth in a health facility. We then subtracted from the total of all postabortion cases those women treated for complications from late miscarriages, to arrive at the number receiving treatment for complications of induced abortion only. Of the total 115,000 women treated in 2008 for complications from pregnancy losses, an estimated 22,000 were treated for late miscarriages and 93,000 were treated for complications of an induced abortion.

The second measure—the proportion of women who have an induced abortion who would receive facility-based treatment for complications—comes from the Health Professionals Survey (HPS). We used this information to calculate a multiplier, or adjustment factor, to account for women having an abortion who do not receive facility-based treatment, either because they do not develop complications or because they do not obtain needed care. We then estimated the total number of induced abortions as the product of the number of women treated for induced abortion complications and the multiplier. Estimation of the multiplier builds in two important factors—whether a woman is poor or nonpoor* and whether she lives in an urban or rural area. In addition, for the 2008 estimates, multipliers were calculated for each region in the study, except Oriental. Using the responses from the HPS, we estimated that approximately 23% of all women having an abortion are likely to receive treatment in a health facility. The national-level multiplier is the inverse of this proportion, 100/23.33 = 4.29. This means that

approximately one out of every four women who have an induced abortion in Colombia

28

are treated for complications in health facilities. The HPS instrument used to collect the 2008 data closely parallels the instruments used in the first study in Colombia and in Guttmacher Institute studies carried out in other countries.2 The current application required two main modifications, however: First, the most recent Colombian questionnaire includes questions on misoprostol and its cost. Thus, among the data used to calculate the multiplier was the distribution of all women who had had an abortion, according to whether it was induced using misoprostol (obtained from any source), and whether it was induced using a different method by four provider types or by the woman herself. The multiplier also depends on the likelihood that each type of abortion will lead to complications and the likelihood that women will receive treatment for their complications. The second modification was to include questions on the 2006 court ruling and its implementation in the questionnaire.

abortion provision in the country, the proportions of women needing facility-based postabortion care and the probability that these women would obtain such care—are based on the professionals’ perceptions, not on empirical fact. Moreover, each of the approaches for collecting data on abortion complications—the one used in 1989 and in the current study—has problems. For 1989, the total count of induced abortion complications came from official hospital discharge statistics for all public-sector hospitals. We adjusted that total official count by 20% to account for the omission of private facilities. The 1989 official hospital statistics may also suffer from underreporting or miscoding.

Because official Colombian hospital discharge data became much less complete over time, we used HFS data instead of official statistics for 2008. Yet, as mentioned earlier, the HFS data have a margin of error because they rely on estimates made by key informants and because they are based on a sample survey. Further, the facilities’ official data may suffer from diagnosis

Limitations The analytical approach and data have some limitations. Key data—on the conditions of

misclassification and incompleteness.

29

In addition, the current HFS asked each facility to provide a copy of its available statistics for the past three years, or for the last year only, so we could assess data availability and perform a data-quality check between data collected in our survey and the facility’s official records. Only 46% of the HFS facilities were able to provide such data. Among those that did, the number of facilities’ officially reported 2008 postabortion cases is fairly close to the value obtained through the HFS for those facilities (20% lower). This could be because the oneyear periods covered by the two sources of morbidity data are somewhat different: The facilities’ official records are for the calendar year 2008. By contrast, the HFS interviews, which were conducted in March and April of 2009, asked for data on admissions in the past month and in a typical month, which were then averaged and multiplied by 12 to provide a best estimate for 2008.
References 1. Bongaarts J and Potter R, Fertility, Biology and Behavior, New York: Academic Press, 1983. 2. Singh S, Prada E and Juarez F, The abortion incidence complications method: a quantitative technique, in Singh S, Remez L and Tartaglione A, eds., Methodologies for Estimating Abortion Incidence and Abortion-Related Morbidity: A Review, New York: Guttmacher Institute; and Paris: International Union for the Scientific Study of Population, 2010.
*We used education as a proxy for poverty, since income data are unreliable and are also inadequate for measuring regional differences in poverty. Women with seven or fewer years of schooling were considered poor, and those with eight or more were considered nonpoor. Weights for the percentage of women in each region according to the four subgroups—poor rural, nonpoor rural, poor urban and nonpoor urban—were based on data from the 2010 ENDS.

30

Figure 2.1. Abortion rates are higher in Bogotá and Pacífica than in other regions of Colombia. Figure 2 1 Abortion rates are higher in Bogotá and Pacífica than in other regions of Colombia

Total

39

Bogotá g

66

Pacífica

54

Atlántica

42

Central

24

Oriental

18

0

20

40

60

80

100

No. of abortions per 1,000 women aged 15–44

Source Reference 21. Source Reference 21

31

Figure 2.2. Misoprostol abortions, from any source or provider, are the most common type of abortions among most subgroups of women.
100

25

% among women who have an abortion

40 75 50 47 59

50 16

50

18 25 26 21 25 17 8 5 12

25

16 21 18

0 Total

4

Poor Nonpoor Poor Nonpoor rural women rural women urban women urban women Misoprostol abortions All providers/sources Other types of abortions, by provider
Traditional midwife/woman herself Nurse/pharmacist Physician

Note: The following are examples of methods other than misoprostol mentioned by the interviewed professionals, by type of provider: physician—surgical methods (D&C, manual vacuum aspiration and electrical aspiration); nurse—oxytocin injections, insertion of catheters, saline washes; pharmacist—hormone injections, incorrect use of emergency contraception, high doses of oral contraceptives, oxytocin; traditional midwife—insertion of sharp objects, herbal infusions or pastes, vaginal insertion of herbs, massages; woman herself—striking the abdomen, vaginal insertion of sharp objects or herbal pastes, intentional falls, high doses of oral contraceptives.

Source Reference 27.

32

Figure 2.3. Legal abortion services are scarce throughout the country.
Total

11

89

Bogotá Pacífica Atlántica Central Oriental

23 9 8 9 12 91 92 91 88

77

Public Private

10 12 0 20
% offering legal procedures % not offering legal procedures

90 88 40 60 80 100

% of health facilities, 2008 (N=1,091)*

* Denotes facilities considered likely to provide postabortion care and legal abortion services. Note Number of cases is weighted. Source Reference 38.

33

Chapter 3. Abortion Complications and Postabortion Care When safe legal abortions are performed by trained professionals under hygienic conditions, fewer than 0.3% lead to complications that require facility-based care.41 The situation in Colombia, where very few legal procedures take place, is far different. Nonetheless, abortionrelated maternal deaths have declined dramatically. The proportion of maternal deaths caused by abortion fell by nearly half in recent years, from 16% in 199442 to 9% in 2007.43 Based on the WHO estimate that 780 maternal deaths occurred in Colombia in 2008,44 this means that about 70 women die each year from unsafe abortion, perhaps the most preventable cause of maternal mortality. Of course, these numbers are rough approximations, as maternal mortality is notoriously difficult to assess. As of 2008, Colombia’s maternal mortality ratio was estimated by the Ministerio de la Protección Social to be 75 maternal deaths per 100,000 live births45 and by WHO to be a similar 85 per 100,000.46

Unsafe abortion continues to threaten Colombian women’s health Although abortion-related maternal deaths have experienced an encouraging decline (likely due in part to increased use of misoprostol47), unsafe procedures still frequently result in debilitating complications, some of which may have lifelong consequences. Overall, an estimated total of 132,000 women suffer complications from clandestine abortions, which are performed in conditions where safety cannot be assured. That is, the HPS respondents familiar with the current context of abortion estimated that 33% of women obtaining an illegal abortion in Colombia experience complications that require treatment*F (Figure 3.1).27 This overall complication rate represents an increase from the 29% estimated for 1989, most likely because proportionally fewer physician-provided surgical abortions were performed in 2008.48 In other words, many of the women who had sought relatively safe abortions from physicians in the past have now shifted to using misoprostol (from a range of providers or sources). Unfortunately, misoprostol is F

*The questionnaire item asking about complications reads: “In this study, by “abortion complications” we mean those complications that would need medical attention in a health facility. Such complications include both severe cases (sepsis or uterine perforation) and “incomplete abortions,” which are characterized by heavy bleeding, and are likely to pose less of a risk to the woman’s health, but nonetheless could require medical attention in a health facility.”

34

commonly used incorrectly in the country, leading to an unexpectedly high complication rate of 32%. (We are unable to quantify the overall decline in all physician-provided abortions because we do not know how many misoprostol abortions were provided by physicians in 2008.)

Given misoprostol’s proven effectiveness and safety when used correctly and during the recommended weeks of gestation,9,10 this rate is higher than it should be for a number of reasons. These include ineffective doses and administration; use beyond the recommended maximum week of pregnancy (the ninth); and misinformation and inadequate knowledge among both patients and providers about when medical attention is needed. (Some providers reportedly give direct instructions to women to go to a health facility as soon as the drug’s normal mechanism of action—heavy bleeding—begins.49)

Among women whose abortions involve methods other than misoprostol, those who self-induce by inflicting injury on themselves or ingesting or inserting damaging substances are thought to have the highest rate of complications (65%).27 The complication rate is estimated to be lowest (11%) among those who obtain surgical abortions performed by physicians. Women whose nonmisoprostol abortions are performed by a traditional midwife are less fortunate, as an estimated 54% develop complications that require medical care.

The less expensive an abortion, the riskier it usually is. Thus, poor women, especially those living in rural areas, who cannot afford to pay for a safe, high-quality procedure, are thought to be more likely than nonpoor women to develop complications.15 Poor women also tend to delay their abortion for many reasons, and abortions can be riskier the longer they are delayed, particularly when they are performed by untrained or inexperienced providers. The proportion of women having a clandestine abortion who experience complications depends on where they live and their poverty status, and experts believe that it varies from 24% among nonpoor urban women to 53% among poor rural women (Figure 3.2).27 The proportion of induced abortions that lead to complications is highest in Pacifica (40%) and lowest in Bogotá (25%), with Central and Atlántica having rates that are very close to the national average (31–32%; not shown).27

35

Who is likely to get treatment? Not all women who experience postabortion complications obtain the medical care they need from a formal source; many simply prefer to try to treat themselves to keep their abortion secret. To name just a few other possible reasons from worldwide research, women may be more comfortable going to a traditional practitioner, live too far from formal health services, are prevented from seeking care by a partner, or fear being mistreated or reported to the authorities by health personnel.15,50,51 Nationally, an estimated 21% of women with abortion-related complications fail to obtain the treatment they need (Figure 3.3).27 Higher proportions of poor rural women than of all other women are thought to go without care when they experience complications (49% vs. 11–31%). This troubling pattern persists among women in all regions, highlighting the persistent inequity in rural areas throughout the country.

What about the women who do receive care? Roughly 93,000 women were treated for complications of induced abortion in 2008, which translates to a rate of nine women treated per every 1,000 women of reproductive age (Table 3.1).38 This represents about a 26% increase from the 1989 rate of seven per 1,000. Counterintuitively, this increase is likely driven by a combination of positive developments: Improvements over the past two decades to the national health care infrastructure and insurance system have likely increased access to medical care, enabling more women to seek postabortion care.45,52 In addition, the introduction and spread of misoprostol likely led to an increase in the number of women presenting for care in facilities— both because of incorrect use24,53 and because many women are unfamiliar with how the method works and seek care as soon as the drug’s normal mechanism of action (heavy bleeding) starts.

The 2008 rate of facility-based, postabortion treatment ranged from four cases per 1,000 women of reproductive age in Oriental to 16 per 1,000 in Pacífica.38 Differences between regions highlight the difficulties of interpreting variations in treatment rates in Colombia. For example, Pacífica contains three of the four poorest departments in the country,54 has far fewer health facilities than Bogotá and has a somewhat lower abortion rate than Bogotá (54 vs. 66).38 Nonetheless, Pacífica has a slightly higher complication treatment rate than Bogotá (15 cases

36

treated per 1,000 women vs. 13 per 1,000). Pacífica’s high rate likely reflects more dangerous abortion conditions, rather than greater health facility use. Indeed, health professionals surmised that abortions are far more likely to lead to complications in Pacífica (40%) than in Bogotá (25%).

Characteristics of postabortion care When complications occur, women need to seek treatment without delay. However, among the roughly 1,100 facilities in Colombia that could potentially provide postabortion care, six in 10 did not do so in 2008. The proportion of all potential providers offering any postabortion care was lowest in Central (14%) and highest in Atlántica (65%). Public and private facilities were equally likely to provide this essential service. Among facilities providing postabortion services, the average annual caseload in 2008 was 222 patients, with no difference between public- and private-sector facilities. Facilities in Bogotá and Central had much larger caseloads than the national average. As expected, larger tertiary-care facilities also had a higher-than-average caseload (405 patients), partially because they receive referrals from small and often distant facilities that lack necessary infrastructure.

Nationally, when HFS respondents were asked which of the two main methods they used to treat abortion complications, the vast majority (93%) reported they most commonly used the surgical method of D&C; the remaining 7% most commonly used MVA.38 The D&C procedure is more invasive, takes longer and consumes more resources than MVA. Furthermore, WHO—whose guidelines form the basis for the Ministerio de la Protección Social’s specifications for legal abortion39—recommends that when there are no serious symptoms such as sepsis or trauma, MVA should be used to treat incomplete abortions that take place relatively early in pregnancy (at 16 weeks’ gestation or earlier),36 which is when the majority of abortions in Colombia likely occur.55 Nonetheless, D&C was described as the most preferred technique by 73% of HFS respondents.38 Colombian physicians’ longstanding preference for D&C (which dates from the mid-1970s56), combined with a widespread lack of adequate MVA training and equipment, likely contribute to the high prevalence of D&C. In fact, just 11% of facilities surveyed in late 2007

37

had MVA equipment at that time.57 Overreliance on D&C is found in both public and private facilities, in hospitals and clinics, and across most regions. Only in Pacífica did a sizable minority of facilities (23%) report that they most commonly offered MVA for postabortion care.38

Complications are undoubtedly less severe today than they were in the late 1980s, when unsafe abortions more commonly involved invasive methods such as the insertion of sharp objects, which often led to sepsis and trauma of the reproductive organs. Despite the probable lessening in severity, unsafe abortion and its complications clearly persist. Women continue to suffer, and the health system continues to spend scarce resources treating conditions that are almost entirely preventable by providing broad access to quality abortion and postabortion services and, of course, by avoiding unintended pregnancy in the first place.

38

Figure 3.1. Women who obtain nonmisoprostol abortions from untrained providers are at higher risk for complications.
Total Misoprostol All providers/sources

33

32

Other abortions* Woman herself Traditional midwife Pharmacist Nurse Physician

65 54 41 28 11

0

10

20

30

40

50

60

70

80

90

100

Among women who have an abortion, % who develop complications
* Induced by a method other than misoprostol.

Source Reference 27.

39

Fig 3 2 High p p ti fp th p p i b ti pli ti Figure 3.2. Higher proportions of poor women than nonpoor women experience abortion complications.

100
Amon wom en who ha e a bo ion who experience compli ations Am ng w men w o h ave an ab rti n, % w o e pe en e c mp ca on

80

60

53 44

40

33

33 24

20

0 Total Poor Poor rural Poor Poor urban Nonpoor Nonpoor rural Nonpoor Nonpoor urban

Source Reference 27. f

40

Figure 3.3 . Especially high proportions of poor women who experience abortion complications go Figure 3 3 Especially high proportions of poor women who experience abortion complications go without needed care. without needed care
100

Am on wo men wi h c om lic tio s, % who do not get car Among w m n with comp cat on % who d t g t c re

80

60
49 40 31 21 20 11 22

0
Total Poor rural Poor rural Poor urban Poor urban Nonpoor rural Nonpoor urban Nonpoor rural Nonpoor urban

Source Reference 27. Source Reference 27

41

Table 3.1. Numbers and rates of women treated in health facilities for complications of induced abortion, by region, 2008 Region Total Bogotá Pacífica Atlántica Central Oriental No. of women treated 93,336 23,928 28,129 20,838 13,533 6,908 Postabortion treatment rate* 9 13 16 10 5 4

*No. of women treated for abortion complications per 1,000 women 15–44 Source Reference 38.

Chapter 4. The Root Cause of Induced Abortion: Unintended Pregnancy Except for the tragic exceptions of grave fetal abnormalities or serious maternal illness, the vast majority of induced abortions result from unintended pregnancy (i.e., a pregnancy that came too soon or was not wanted at all because the woman already had the number of children she desired).15 Pregnancies resulting from rape or incest are unintended by definition. The immediate cause of all other unintended pregnancies is nonuse, misuse or failure of a contraceptive method, yet that does not tell us why a woman wanted to avoid pregnancy to begin with. Women may want to postpone a pregnancy or stop childbearing for myriad reasons: They cannot afford to raise a child, have not yet finished school, are in an unstable relationship, have a partner who cannot or will not support a child, or have reached their desired family size—to name some major ones.58

Unintended pregnancy is common throughout Colombia An estimated 1,357,600 pregnancies occurred in Colombia in 2008 (Appendix Table 1).21 This total number includes 764,300 pregnancies ending in live births and 593,300 ending in miscarriages and induced abortions. According to these findings, an estimated two-thirds (67%) of all pregnancies are unintended. The proportion unintended is below the national average in Oriental (61%), but above it in Bogotá (74%) and Pacífica (73%).

Colombia’s number of unintended pregnancies in 2008, 911,900, translates to an annual rate of 89 unintended pregnancies per 1,000 women aged 15–44 (Figure 4.1).21 This rate is one-quarter higher than the average for all of Latin America and the Caribbean (72 unintended pregnancies per 1,000 women).15 Moreover, the 2008 unintended pregnancy rate represents a 6% increase from 1989, when it was 84 per 1,000.21 Regional results show that the highest unintended pregnancy rate is in Bogotá (113 per 1,000 women, compared with 67–104 per 1,000 women in the remaining four regions). Some 74% of all pregnancies in Bogotá are unintended, despite the high level of modern method*G use there (77%6), suggesting that motivation to successfully time births and have small families is especially strong in that region.

G

*Sterilization (male and female), the pill, the IUD, injectables, implants, the male condom and spermicides.

43

What do Colombian women do when faced with unintended pregnancy? More than two-fifths (44%) of unintended pregnancies each year end in induced abortions (Appendix Table 1). When we look instead at all pregnancies in 2008 (both intended and unintended), similar proportions end in unplanned births (29%), induced abortions (29%) and planned births (27% ). The remaining pregnancies are lost through miscarriage (15%; Figure 4.2). As countries pass through the stages of fertility transition, preferred family size tends to get smaller and the desire to have births precisely when they are wanted grows stronger. Therefore, the proportion of pregnancies that are unintended may increase even with higher levels of contraceptive use and the resulting decrease in pregnancies overall.

In Colombia, the proportion of pregnancies that were unintended increased between 1989 and 2008 from 52% to 67%, even as women had far fewer pregnancies (a decline from 163 to 133 pregnancies per 1,000 women).21 Colombian women’s wanted fertility*H has consistently declined (from 2.2 children in 19904 to 1.6 in 2010).5 This decline in wanted fertility has led to increases in the proportions of pregnancies ending in both unplanned births (from 23% in 1989 to 29% in 2008) and abortions (from 22% to 29%).*I These proportions may increase further— and abortion rates may rise—unless women and couples succeed in using modern methods more effectively and consistently.

What explains the high levels of unintended pregnancy and unplanned births? Strong preferences for smaller families. As noted above, when women’s desires to limit family size and space births increase faster than their effective and consistent use of modern contraception, the inevitable result is a rise in unintended pregnancy. The proportion of married women (in a legal or consensual union) wanting to delay or space births changed little in recent years (from 20% in 19907 to 17% in 20106). Meanwhile, the proportion of women in union wanting to stop having children altogether (which includes women protected by sterilization) increased from 64% in 1990 to 70% in 2010. These trends provide evidence of Colombian H

*Wanted fertility is the theoretical number of children a woman would have if she could avoid any unwanted births. A birth is defined as unwanted if the woman’s number of living children at the time of conception equaled or exceeded her ideal family size. I *A relatively stable 7–9% of pregnancies were unintended and ended as miscarriages in each year.

44

women’s growing desire for smaller families. Moreover, data from the most recent Encuesta Nacional de Demografía y Salud (ENDS) indicate that strong motivation to limit the number of children will continue into the future: As of 2010, the younger the interviewed woman, the smaller her ideal family size.5

Whereas preferences for smaller families are widespread, some women have a much harder time achieving their preferences than others. As mentioned earlier, if all unwanted births could be avoided, Colombian women would have 1.6 children instead of the current 2.1. As of 2010, women in the poorest quintile (who are also the least educated) had the largest gap between wanted and actual family size (Table 4.1).5 These women probably lack adequate access to modern contraceptives and the information and means to use them correctly and consistently. If the poorest women could avoid having unwanted births altogether, they would have just over two children instead of just over three, a difference of more than one child; by contrast, the richest women (and the most educated), who are best able to act on their preferences, would have 1.2 children, a number just below what they currently have.

Characteristics of current method use. Colombia has a very high level of overall contraceptive use, yet persistently high rates of unintended pregnancy indicate that not all women use effective methods or use their method consistently and correctly. Moreover, it is likely that traditional method use, user failure, nonuse and method discontinuation all contribute to unintended pregnancy in the country. In 2010, 73% of women in union aged 15–49 were using a modern method, and modern use varied little by region (except in Atlántica, where prevalence was much lower than average—65%; Appendix Table 2).6 Roughly 6% of women in union throughout the country currently rely on traditional methods,*J which have far higher failure rates than modern methods.59 Encouragingly, traditional method use has declined consistently over time.6,7 Atlántica stands out as the only region to experience an increase in traditional use (from 5% in 19907 to 7% in 20106). And although the exact role that emergency contraception may play in averting unintended pregnancy (and possibly) abortion J

*Periodic abstinence, withdrawal, lactational amenorrhea and folk methods.

45

cannot be quantified, the proportion of sexually active unmarried women ever having used the method increased from 4% in 200060 to 28% in 20106 (not shown). This is an important indication that women are becoming more aware of this indispensable method to prevent pregnancy after unprotected intercourse.

The specific mix of methods used can affect the rate of unintended pregnancy. For example, traditional methods such as periodic abstinence or withdrawal make up 8% of all contraceptive use, and these methods have very high typical-use failure rates (24–27% of their users become pregnant within one year in Colombia). K,61 Supply methods, such as the condom and the pill, account for an additional 18% of all use; these methods have failure rates of 8–10% in Colombia.

When sexually active, fecund women want to avoid pregnancy but fail to use a method, they are considered to have an unmet need for contraception. Unmet need heightens women’s vulnerability to unintended pregnancy and, by extension, to unsafe abortion. (Only in Bogotá is high unmet need not paired with high levels of unintended pregnancy; in that region, even though the level of unmet need is very low, it appears that strong motivation to time births and have small families have kept unintended pregnancy high.) Unmet need among Colombian women in union dropped sharply between 1990 and 2010 (from 11%7 to 7%6). However, it continues to be disproportionately high among the women who are least able to support large families or pay for a safe abortion: The poorest women are roughly twice as likely as the richest women to have an unmet need for contraception (11% vs. 5%).6

Another factor that puts women at risk of unintended pregnancy is stopping the method they are currently using. In 2005, 44% of women using contraceptive methods other than permanent sterilization discontinued during the first year of use, mostly because of their method’s inconvenience, failure or side effects.62 Excluding women who stopped use to become pregnant, were infecund, had sex infrequently or were separated from their husbands, the discontinuation

K

*These country‐specific failure rates provide a valuable measure of the relative differences between failure rates of different contraceptive methods. They are most likely underestimates because method and user failures that result in induced abortion are likely highly underreported.

46

rate for Colombia in 2005 was 37%, the second highest rate among the eight countries*L studied.62

Increased sexual activity and unmet need among young women. The increase in the proportion of young unmarried women who are sexually active has undoubtedly had an impact on current levels of unintended pregnancy. In 2010, 31% of unmarried women aged 15–24 were sexually active (i.e., had had sex in the past three months), up from 8% in 1990.6,7 And although fertility has declined substantially among women in all other age-groups since 1990, it only recently started to decline among 15–19-year-old women. For example, their number of births increased from 70 per 1,000 in 19904 to 85 per 1,000 in 200060 and to 90 per 1,000 in 2005,63 but then declined to 84 per 1,000 in 2010.5

Unfortunately, there are no data that quantify young women’s probability of resorting to an abortion when faced with an unintended pregnancy, so we cannot calculate overall unintended pregnancy rates for this age-group. However, given that adolescents and young women, especially those in urban areas, are increasingly expected to attend school and participate in the workforce, the proportion that resolve unintended pregnancy through abortion has likely increased in recent years. Although specific data are unavailable for Colombia, WHO estimates that as of 2000, 15–19-year-old women accounted for 14% of all unsafe abortions in Latin America and the Caribbean, and young women aged 20–24 accounted for an additional 29%.64

Unmet need for contraception among single, sexually active 15–24-year-old women, although still low in absolute terms, rose from 5% in 200065 to 8% in 20106 (no comparable data are available for 1990). This trend stands in stark contrast to the unchanged level of unmet need among 15–24-year-olds in union. Moreover, method discontinuation has been shown to be associated with age in Colombia: Young women aged 15–24 are significantly more likely than older women to discontinue a method during their first year of use while they still need one.62

L

*Armenia, Bangladesh, Colombia, Dominican Republic, Egypt, Indonesia, Kenya and Zimbabwe.

47

Use of a modern contraceptive method is currently—and has always been—lowest among the youngest women in union (15–19-year-olds), compared with all other women. Their level of unmet need is 20%, the highest of any age-group.5 Indeed, adolescents in a union stand out as the only group to not experience large declines in unmet need over the past two decades. At the same time, the proportion of recent births (and current pregnancies) among all 15–19-year-old women has risen from 27% in 19904 to 64% in 2010.5 Given that young women’s desire to control the timing of their births and have smaller families continues to outpace their effective use of contraceptives, their reliance on abortion may have risen and may continue to rise.

A growing displaced population. The decades-long armed insurgency, along with ongoing drugand gang-related violence, has forced the displacement of at least 3.3–4.9 million Colombians.17 Pacífica’s coastal area has had to contend with especially high numbers of fleeing citizens. These millions of Colombians have had to endure brutality, loss of their livelihoods and estrangement from their social support networks.18 Displaced women’s inadequate health insurance coverage (in 2005, 42% were not enrolled in any plan,66 compared with roughly 32% of all women67), high poverty level (78% of displaced households live in abject poverty), limited access to contraceptive services, and heightened exposure to rape and other sexual violence have all increased their vulnerability to unintended pregnancy.17

According to a 2005 survey of 1,097 women aged 13–49 who were displaced by armed conflict, 66% of those currently pregnant said that they had not planned to be.66 This proportion is notably higher than the 59% found among all women interviewed in the 2005 ENDS.67 Given displaced women’s precarious situation and their higher than average level of current unintended pregnancy, their risk of resorting to unsafe abortion is also likely to be higher than average.

48

Figure 4.1. The unintended pregnancy rate in the region of Bogotá is the highest in the country.

Total

89

44

133

Bogotá Pacífica Atlántica Central Oriental 67 67

113 104 102 39 44 106 111 38

40 142 57

153

159

No. of pregnancies per 1,000 women aged 15–44, 2008
Unintended pregnancy rate Intended pregnancy rate

Source Reference 21.

49

Figure 4.2. Over the past two decades, the proportion of pregnancies ending in Fig re 4 2 O er the past t o decades the proportion of pregnancies ending in unplanned births increased by more than one quarter. unplanned births increased by more than one‐quarter.

1989
Miscarriages Mi i 8% Intended Intended pregnancies p g d Unplanned pl births 23% Planned births Planned births 40% Unintended Unintended pregnancies i

Induced Induced abortions 22%

i i Miscarriages 7%

Miscarriages 6% I t d d Intended pregnancies

2008
U i d d Unintended pregnancies Unplanned p births 29%

Planned births 27%

Mi i Miscarriages 9%

Induced abortions 29%

Source Reference 21. Source Reference 21

50

Table 4.1. Wanted and actual fertility rates among Colombian women aged 15–49, by wealth quintile, 2010 Wealth quintile Wanted total fertility rate* 1.6 2.1 1.7 1.6 1.4 1.2 Actual total fertility rate Gap (births in excess of woman's ideal family size) 0.5 1.1 0.8 0.5 0.3 0.2

Total Lowest (poorest) Second Third Fourth Fifth (wealthiest)

2.1 3.2 2.5 2.1 1.7 1.4

*This indicator measures the potential impact of avoiding unwanted births, which are defined as those that are conceived after a woman's number of living children meets or exceeds her ideal number of children. Source Reference 5.

51

Chapter 5. Conclusions and Implications Until recently, all Colombian women faced with an unintended pregnancy were forced to either carry it to term or seek a clandestine, often unsafe abortion. The partial decriminalization of induced abortion in 2006 marked an important step toward improving the health and lives of Colombian women, setting in motion a promising series of developments that created a rightsbased framework to guarantee safe abortion under certain circumstances. The ruling not only lifted restrictions but put into place best practices and training requirements to assure the delivery of safe services by competent medical personnel. In May of 2009, the Constitutional Court’s Sentencia T-388/09 forbade judges to conscientiously object to hear cases involving providers’ refusals to offer abortion services.68 This ruling also ordered a national education campaign to inform the public, including students, about the partial decriminalization and the original 2006 decision’s basis in internationally recognized treaties guaranteeing women’s sexual and reproductive rights.

Despite these positive developments, women’s health in Colombia is still imperiled by limited access to safe abortion. About 400,400 induced abortions occur annually, and only 322 are legal procedures performed in health facilities. We are unable to determine the number of women who meet a criterion but still resort to clandestine abortion (because their request for a legal procedure is denied or authorization is delayed). However, we do know that the vast majority of abortions in Colombia continue to be performed under conditions where safety cannot be assured. These findings strengthen the robust evidence from around the world showing that legally restricting abortion does not eliminate it but pushes the practice underground, making it less safe.15,22

Although the rate of induced abortion did not change substantially over the past two decades, today’s unprecedentedly high number of women of reproductive age (who were born when population growth was much higher) means that the medical system now has to contend with the health consequences of roughly two-fifths more clandestine abortions now than it did a few decades ago. The consequences are evident in the higher numbers of women treated for complications in formal facilities—from an estimated 58,000 in 198955 to 93,000 in 2008.38 Treatment of complications uses up scarce medical resources that could be far better spent

52

elsewhere. The overwhelming reliance on the resource-intensive D&C method for postabortion care, instead of the less complex methods of MVA or misoprostol, adds to the unnecessary and avoidable health-system costs.

The prevalence of unintended pregnancy—the main factor driving abortion—has risen with women’s growing preferences for smaller families. Colombia’s trend toward higher rates of unintended pregnancy runs counter to the 20% decline in this measure in all developing countries between 1995 and 2008.15 The situation likely stems both from the desire for smaller families outpacing the adoption of effective contraception, and from the difficulties couples continue to have practicing contraception effectively and consistently. The combined strengthening motivation to avoid giving birth and steep declines in fertility have resulted in a current ratio of 52 abortions for every 100 live births, compared with 35 per 100 two decades ago.21 Women’s resolve to have fewer births and have them when they want is so strong in Bogotá that each year, abortions and live births occur in almost equal numbers in that region.

Roughly one-third of all women who obtain an induced abortion experience health complications that require treatment in a health facility. However, an estimated one-fifth of women needing care go without it. This gap is even larger among poor rural women, for whom half of postabortion complications go untreated. Almost two-thirds of health facilities that could provide postabortion care do not, suggesting that there is a need—and an opportunity—to expand the number of facilities offering such care, particularly in rural areas. The widespread reliance on D&C, which requires anesthesia, in and of itself limits the availability of postabortion care, since it reduces the pool of facilities capable of providing it.

When used correctly, misoprostol is a safe and effective method of inducing an abortion. But the estimated complication rate for misoprostol abortions in Colombia—32%—is well above what would be expected given the method’s clinical failure rate of 10–15%,9 which suggests that many women and providers are using it incorrectly. Research conducted in Mexico shows that pharmacy vendors who sell the drug have poor knowledge of how it works and rarely recommend an effective medication regimen (dose and timing) or specify the appropriate weeks of gestation indicated for use.69 Moreover, many vendors simply direct women to seek medical

53

attention as part of the process, even though such care is not really necessary in most cases. Similar problems are likely occurring in Colombia.

How can Colombia move forward? Implementation of the 2006 court decision has faced daunting political and administrative challenges. As a result, only an estimated 322 women obtained a legal abortion in 2008.38 Several factors likely contribute to this extremely low number: the relatively short lapse of time since the decision, women’s lack of knowledge about the changed legal situation and the availability of legal abortion, the narrow interpretation of the health exception by some providers and government officials, and providers’ unwillingness to perform legal abortions. As of June 2011, when this report was written, the decree that guides the implementation and enforcement of the ruling remains suspended. Although the 2006 Constitutional Court ruling is still valid, it lacks the essential regulatory apparatus to sanction noncompliance. However, unintended pregnancy—the main factor behind the 400,400 clandestine abortions that occur annually—is amenable to immediate intervention. Below, we outline some specific steps to reduce unintended pregnancy, improve access to legal abortion services, and increase the coverage and quality of postabortion care.

Strengthen contraceptive services Couples and providers need better information about correct and consistent use of contraception. Contraceptive use is currently widespread; what is needed is more consistent and more effective use. Providers themselves must be better informed if they are to help couples improve use their current modern method or switch from traditional to more effective ones. Women need better counseling on how to use their method continuously and correctly; the full support and cooperation of male partners must also be fostered. Furthermore, special attention needs to be directed to women who are not using a method despite not wanting a pregnancy (the 7% of women in union with an unmet need for contraception) and those who are currently using a traditional method (another 6%).5

54

Access to emergency contraception should be expanded. The use of emergency contraception, which has been available in the country since 2000,70 can go a long way toward reducing the numbers of unintended pregnancies and subsequent abortions. The current agreement (Acuerdo 008 of 2009) authorizing all medicines and procedures in the national health package (Plan Obligatorio de Salud, or POS) restricts the plan’s coverage of emergency contraception to victims of rape and to the “at-risk adolescent population.”71 All women who have had unprotected sex should be given the option of using this method.

Targeted interventions are needed to meet the needs of groups at high risk of unintended pregnancy. The proportion of single young women who are sexually active is increasing. To ensure that they have the means to avoid unplanned births, policymakers responsible for health and education at local and national levels should consider expanding current sex education campaigns, especially in rural areas. Health authorities should also work to identify and address barriers to providing confidential, youth-friendly sexual and reproductive health services. In addition, outreach programs are needed to target young people who are not attending school. Women who are displaced by the ongoing armed conflict and other types of violence also need special attention. Given their high level of poverty and acute difficulties in preventing unintended pregnancy during a time of upheaval, their need for improved access to reproductive health information and services is especially pressing.

Improve postabortion care services Coverage and quality of postabortion services should be improved. Roughly one-fifth of women experiencing abortion-related complications currently go without treatment. Guaranteeing that women receive confidential and respectful treatment is key to overcoming the reluctance and fear of mistreatment that deters some from seeking care. In addition, several steps would help ensure that treatment is accessible to women who need it. These include training and equipping personnel in health posts and centers to provide postabortion care to women living in small towns; training midlevel and paramedical staff in relevant skills and techniques (including how to recognize when to refer patients to other providers); and shifting away from the more invasive

55

and resource-consuming D&C toward MVA, to expand both the range of staff capable of providing care and the types of facilities able to offer it.

The quality of care can also be improved. To prevent repeat abortions, postabortion care providers must provide women with contraceptive counseling and a highly effective method of their choice. Furthermore, although Colombia has adopted clear guidelines for the provision of legal abortion39 and for treating miscarriages,72 it has no comparable guidelines for facility-based treatment of complications from unsafe abortions. A possible solution would be for the Ministerio de la Protección Social to consider preparing and adopting guidelines based on accepted standards of comprehensive postabortion care73 to improve the quality of treatment and minimize the harm caused by unsafe abortion.

Providers need more accurate information about caring for women who have used misoprostol. Misoprostol can be a safe, effective and low-cost way of inducing early abortion. According to the estimates of HPS respondents, the drug is used in roughly half of all clandestine abortions in Colombia. However, use of the drug seems to have spread faster than has accurate information about how it works and how it should be administered. Medical providers, staff in pharmacies and the general public need to be more fully informed about the ways in which misoprostol is commonly misused in the country.

Postabortion care providers should be trained to use MVA whenever feasible. Our study shows clear underuse of MVA, the recommended procedure for treating cases of incomplete abortion. In general, providers need to be more receptive to training in MVA, which does not require general anesthesia. The current reliance on D&C leads to an unnecessary waste of scarce resources. Now that the country has officially approved the use of misoprostol for postabortion care,31 this drug provides another low-cost alternative to D&C, but more training in how to best use it to treat complications is needed.

56

Improve implementation of the 2006 court decision and the provision of legal abortion Continue spreading awareness of this important ruling. Although the 2010 ENDS suggests that knowledge of the three criteria for legal abortion is becoming more widespread,5 many women, particularly those in rural areas, likely remain unaware of them and are thus unable to exercise their right to a legal abortion. The specifics of the ruling also need to be widely disseminated to the professionals who are responsible for implementing and enforcing it, such as medical and paramedical workers, members of the judiciary, and staff at agencies that assist victims of rape and domestic violence. Public education campaigns are essential to enabling women to exercise their legal right to safely terminate pregnancies that meet the specified criteria. Studies gauging the extent of public knowledge would help direct such public education campaigns to where they are needed most.

Barriers to legal abortion warrant special attention. Institutional and individual barriers to legal abortion are many, and they are especially common among poor women and rural women with limited access to health facilities. Reports of providers denying women legal abortions, and of judges refusing to take up such cases, suggest that many women who meet the legal criteria end up having a clandestine abortion. It is essential that medical and judicial professionals know the limits to conscientious objection established by the 2006 decision. Denying access to legal abortion services to women who qualify for them is a flagrant violation of women’s constitutional rights.

Mechanisms are needed to ensure that medical guidelines are followed. Members of the medical profession are obligated to follow the safe abortion guidelines issued by the Ministerio de la Protección Social. Nonetheless, the HFS results show that legal abortions are twice as likely to be performed by D&C than by MVA. The POS includes the use of these two procedures (but not of medications) to induce legal abortions, so physicians who provide covered services should be up to date in approved procedures. To that end, training in the correct use of procedures to safely induce legal abortions, and safely treat complications resulting from clandestine abortions, needs to be incorporated into medical curricula and professional development.

57

Collection of data on legal abortion should be improved. The health system is not adequately tracking the number of legal abortions performed in the country. Indeed, estimates offered by health entities differ widely. An improved system needs to be put into place to both accurately quantify the number of legal procedures and monitor the extent to which health facilities are complying with the law.

In-depth research is needed on the three legal criteria. The 2006 court ruling established that women were entitled to a legal abortion in three specific circumstances. It is important that providers and women understand how the ruling changed the legal situation, and that the current legal criteria be interpreted in their full meaning. We also need to know why so many Colombian women seek out clandestine abortions. Their reasons are likely similar to what has been found among women worldwide, in both legal and illegal settings, where social and economic reasons predominate.58,74 (Indeed, one early study conducted among Colombian women who had undergone a clandestine abortion found that the most common reason for doing so was because of economic problems.75) Research into women’s reasons is needed to inform discussions about whether current legal criteria are adequate to meet women’s needs.

Final reflections The current trend toward more unintended pregnancies and unplanned births calls for unified efforts to enable more effective contraceptive use, which will lower the need to resort to induced abortion. A wide array of stakeholders must act to improve the current situation on three main fronts—contraceptive services, postabortion care, and the provision of legal abortion services: • Providers—including physicians, nurses, pharmacists, and personnel at health institutions and associations—can increase the provision and improve the quality of essential contraceptive services and abortion care. • The Ministerio de la Protección Social, medical schools, and program planners in the public and private sectors and from Nongovernmental Organizations (NGOs), can train providers in recommended medical practices for performing legal procedures and treating complications from clandestine abortions.

58

•

The Ministerio de la Protección Social and program officials must assure the steady availability of medical supplies and equipment needed for contraceptive services, and for the provision of legal abortions and quality postabortion care (including MVA kits).

•

The Ministry of Education and officials responsible for private- and public-school systems should provide young people with the knowledge and skills they need to protect their reproductive health and receive the services they have a right to use.

•

Government agencies, NGOs and activists must monitor the provision of legal abortion services to ensure that women are informed of their rights and are able to act on them, and that members of the medical and judicial establishments adhere to the limits of conscientious objection established by the Constitutional Court.

•

NGOs, activists and women’s organizations must work to keep the issue of unsafe abortion in the public eye.

Only when all these forces come together can significant progress be made in reducing unintended pregnancy and improving the provision of both legal abortion and postabortion care. The tangible effects of these efforts will be fewer induced abortions (especially unsafe ones), a less burdened health system, and healthier women and families.

59

REFERENCES 1. Law 599, which issues the Penal Code with changes from Sentencia C-355/06, 2006; <http://www.alcaldiabogota.gov.co/sisjur/normas/Norma1.jsp?i=6388>, accessed May 30, 2010. 2. Mesa por la Vida y la Salud de las Mujeres, A 4 años de la despenalización parcial del aborto en Colombia, 2007, <http://www.despenalizaciondelaborto.org.co/A-4-ANOS-DE-LADESPENALIZACION>, accessed June 1, 2010. 3. Palacio D et al., Plan de Choque para la Reducción de la Mortalidad Materna, Bogotá, Colombia: Ministerio de la Protección Social (MPS), 2004. 4. Asociación Probienestar de la Familia Colombiana (Profamilia), Encuesta de Prevalencia, Demografía y Salud, Colombia, 1990, Bogotá, Colombia: Profamilia; and Columbia, MD, USA: Institute for Research Development/Macro International, 1991. 5. Profamilia, Encuesta Nacional de Demografía y Salud (ENDS), Colombia, 2010, Bogotá, Colombia: Profamilia, 2011. 6. Special tabulations of data from the Colombian Encuesta Nacional de Demografía y Salud, 2010. 7. Special tabulations of data from the Colombian Encuesta de Prevalencia, Demografía y Salud, 1990. 8. Consorcio Latinoamericano Contra el Aborto Inseguro (CLACAI) and Ipas, Misoprostol y Aborto con Medicamentos en Latinoamérica y el Caribe, Chapel Hill, NC, USA: Ipas, 2010. 9. Faúndes A et al., Misoprostol for the termination of pregnancy up to 12 completed weeks of pregnancy, International Journal of Gynecology and Obstetrics, 2007, (99):S172–S177. 10. Moreno-Ruiz N.L. et al., Alternatives to mifepristone for early medical abortion, International Journal of Gynecology and Obstetrics, 2007, 96(3):212–218.

60

11. Gynuity Health Projects, Opciones para la Introducción del Aborto con Medicamentos en Brasil, Colombia, México y Perú, New York: Gynuity Health Projects, 2007. 12. Colmenares E, Método abortivo más extendido entre las jóvenes bogotanas es una droga para la úlcera, El Tiempo, Mar. 19, 2007, <http://www.eltiempo.com/archivo/documento/CMS3484520>, accessed May 30, 2011. 13. Colmenares E, La pastilla del aborto bogotano, El Tiempo, Mar. 20, 2007, <http://www.eltiempo.com/archivo/documento/MAM-2421465>, accessed May 30, 2011. 14. Special tabulations of data from Departamento Administrativo Nacional de Estadística (DANE), Proyecciones Nacionales y Departamentales de Población, 2005–2020, Bogotá, Colombia: DANE, 2011. 15. Singh S et al., Abortion Worldwide: A Decade of Uneven Progress, New York: Guttmacher Institute, 2009. 16. Department of Economic and Social Affairs, United Nations Population Division, World Urbanization Prospects: The 2007 Revision Population Database, 2010, <http://esa.un.org/unup/p2k0data.asp>, accessed Mar. 23, 2010. 17. Norwegian Refugee Council, Internal Displacement: Global Overview of Trends and Developments in 2009, Geneva: Internal Displacement Monitoring Centre, 2010. 18. Consejo Nacional de Política Económica y Social (CONPES), Lineamientos de la política de generación de ingresos para la población en situación de pobreza extrema y/o desplazamiento, Documento CONPES, Bogotá, Colombia: CONPES, 2009, No. 3616. 19. De los cerca de 350.000 abortos que se calculan al año, 12% corresponderían a los tres casos especiales, El Tiempo, May 14, 2006, pp. 1-6–1-7. 20. Moloney A, Unsafe abortions common in Colombia despite law change, Lancet, 2009, 373(9663):534.

61

21. Prada E, Biddlecom A and Singh S, Induced abortion in Colombia: new estimates and change between 1989 and 2008, International Perspectives on Sexual and Reproductive Health, 2011, 37(3)114–124. 22. World Health Organization (WHO), Unsafe Abortion: Global and Regional Estimates of Incidence of Unsafe Abortion and Associated Mortality in 2003, fifth ed., Geneva: WHO, 2007. 23. Sedgh G et al., Induced abortion: estimated rates and trends worldwide, Lancet, 2007, 370(9595):1338–1345. 24. Juarez F et al., Estimates of induced abortion in Mexico: What’s changed between 1990 and 2006? International Perspectives on Sexual and Reproductive Health, 2008, 34(4):158–168. 25. Singh S, Prada E and Kestler E, Induced abortion and unintended pregnancy in Guatemala, International Perspectives on Sexual and Reproductive Health, 2006, 32(3):136–145. 26. Ferrando D, El Aborto Clandestino en el Perú: Hechos y Cifras, 2002, Lima, Peru: Centro para las Mujeres Flora Tristán and Pathfinder International/Peru, 2002. 27. Guttmacher Institute, unpublished data from the Health Professionals Survey, application of the Abortion Incidence Complications Method (AICM), Colombia, 2009. 28. Fernandez MM et al., Assessing the global availability of misoprostol, International Journal of Gynecology and Obstetrics, 2009, 105(2):180–186. 29. Gómez-Sánchez PI, Escandón I and Gaitán-Duarte H, Evaluación de la atención integral al pos-aborto en 13 hospitales de Colombia, Revista de Salud Pública, 2007, 9(2):241–252. 30. Instituto Nacional de Vigilancia de Medicamentos y Alimentos (INVIMA), MPS, Sala Especializada de Medicamentos y Productos Biológicos de la Comisión Revisora, Acta No. 20, Bogotá, 2007. 31. INVIMA, MPS, Sala Especializada de Medicamentos y Productos Biológicos de la Comisión Revisora, Acta No. 37, Bogotá, 2009.

62

32. Intercontinental Marketing Services (IMS), Targeting de farmacias para el mercado de misoprostol en Colombia, México y Perú, powerpoint presentation, 2010. 33. Banco de la República de Colombia, Series estadísticas producción, salarios y empleo: salario mínimo legal diario, 2010, <http://www.banrep.gov.co/seriesestadisticas/see_prod_salar.htm#3>, accessed July 7, 2010. 34. MPS, Decreto Número 4444 de 2006, por el cual se reglamenta la prestación de unos servicios de salud sexual y reproductiva, Bogotá, 13 de diciembre, 2006. 35. MPS, Resolución Número 4905 de 2006, por la cual se adopta la Norma Técnica para la atención de la Interrupción Voluntaria del Embarazo, Bogotá, 14 de diciembre, 2006. 36. WHO, Safe Abortion: Technical and Policy Guidance for Health Systems, Geneva: WHO, 2003. 37. Injunction issued by Consejo de Estado, Sala de lo Contencioso Administrativo, Sección Primera, Bogotá, session of October 15, 2009, <http://www.scribd.com/doc/21464787/AutoConsejo-de-Esado-Suspension-decreto-4444>, accessed May 30, 2011. 38. Guttmacher Institute, unpublished data from the Health Facilities Survey, application of the AICM, Colombia, 2009. 39. MPS, Norma Técnica para la Atención de la Interrupción Voluntaria del Embarazo (IVE), Adaptada de “Aborto sin Riesgo: Guiá Técnica y de Políticas para Sistemas de Salud,” Bogotá, Colombia: MPS, 2006. 40. Corte Constitucional, Sentencia T-209/08, Acción de tutela interpuesta Gladys Belén Arias Becerra contra la EPS Coomeva y el Hospital Universitario Erasmos Meoz de Cúcuta, Bogotá, 2008. 41. Henshaw SK, Unintended pregnancy and abortion: a public health perspective, in: Paul M et al., eds., A Clinician’s Guide to Medical and Surgical Abortion, Philadelphia, PA, USA: Churchill Livingstone, 1999, pp. 11–22.

63

42. Ministerio de Salud et al., Mortalidad Materna en Colombia, Bogotá, Colombia: Ministerio de Salud, 1996. 43. DANE, Estadísticas vitales, defunciones 2007: información estadística Cuadro 7, defunciones maternas por grupos de edad, departamento de residencia, y lista de causas agrupadas 6/67 CIE-10 (basada en la lista 6/66 de OPS), <http://www.dane.gov.co/index.php?option=com_content&task=category§ionid=16&id=53 5&Itemid=1035>, accessed May 30, 2011. 44. WHO et al., Trends in Maternal Mortality 1990 to 2008: Estimates Developed by WHO, UNICEF, UNFPA and the World Bank, Geneva: WHO, 2010. 45. MPS and WHO, Situación de Salud en Colombia: Indicadores Básicos, 2008, Bogotá, Colombia: Instituto Nacional de Salud, 2008. 46. WHO et al., Trends in Maternal Mortality, 1990 to 2008: Estimates Developed by WHO, UNICEF, UNFPA and the World Bank, Geneva: WHO, 2010. 47. Harper CC et al., Reducing maternal mortality due to elective abortion: potential impact of misoprostol in low-resource settings, International Journal of Gynecology and Obstetrics, 2007, 98(1):66–69. 48. Prada E, Singh S and Villarreal C, Health consequences of unsafe abortion in Colombia, 1989–2008, paper presented at the Seminar on the Health, Social and Economic Consequences of Unsafe Abortion at the annual meeting of the International Union for the Scientific Study of Population (IUSSP), San Juan del Río, Mexico, Nov. 10–12, 2010. 49. Lara D et al., Pharmacy provision of medical abortifacients in a Latin American city, Contraception, 2006, 74(5):394–399. 50. Jagwe-Wadda G, Moore M and Woog V, Abortion morbidity in Uganda: evidence from two communities, Occasional Report, New York: Guttmacher Institute, 2006, No. 26.

64

51. Saenz de Tejada S, Prada E and Ball G, Morbilidad por aborto en Guatemala: una visión de la comunidad, Informe Ocasional, New York: Guttmacher Institute, 2006, No. 27. 52. Yepes FJ et al., Aiming for equity in Colombia’s health system reform: achievements and continuing challenges, in: Haddad S, Baris E and Narayana D, eds., Safeguarding the Health Sector in Times of Macroeconomic Instability: Policy Lessons for Low- and Middle-Income Countries, Trenton, NJ, USA: Africa World Press; and Ottawa, Canada: International Development Research Centre, 2008. 53. Wilson K, Garcia S and Lara D, Misoprostol use and its impact on measuring abortion incidence and morbidity, in: Singh S, Remez L and Tartaglione A, eds., Methodologies for Estimating Abortion Incidence and Abortion-Related Morbidity: A Review, New York: Guttmacher Institute; and Paris: IUSSP, 2010. 54. Garcia Cano P, Indicadores Sociales Departamentales, Bogotá, Colombia: Departamento Nacional de Planeación, 2008. 55. Singh S and Wulf D, Estimated levels of induced abortion in six Latin American countries, International Family Planning Perspectives, 1994, 20(1):4–13. 56. Higuera-Cáceres G et al., Manejo hospitalario del aborto incompleto, estudio comparativo del curetaje uterino versus la aspiración por vacío, Monografías de la Corporación Regional de Población, 1989, Vol. 16(Capítulo III), p. 18. 57. Guerrero Moreno H and Bolívar Vargas MC, Línea de Base de las Barreras de Acceso a la Prestación de Servicios para la Interrupción Voluntaria del Embarazo: Informe Ejecutivo, Bogotá, Colombia: MPS and Universidad Nacional de Colombia, 2007. 58. Bankole A, Singh S and Haas T, Reasons why women have induced abortions: evidence from 27 countries, International Family Planning Perspectives, 1998, 24(3):117–127 & 152. 59. Hatcher RA et al., Contraceptive Technology, 19th ed., New York: Ardent Media, 2007.

65

60. Profamilia, Salud Sexual y Reproductiva: Resultados Encuesta Nacional de Demografía y Salud, 2000, Bogotá, Colombia: Profamilia, 2000.

61. Department of Economic and Social Affairs, Population Division, Levels and Trends of Contraceptive Use as Assessed in 2002, New York: United Nations, 2004. 62. Bradley S, Schwandt SM and Shane K, Levels, trends and reasons for contraceptive discontinuation, DHS Analytical Studies, Calverton, MD, USA: ICF Macro, 2009, No. 20. 63. Profamilia, Salud Sexual y Reproductiva en Colombia: Resultados Encuesta Nacional de Demografía y Salud, 2005, Bogotá, Colombia: Profamilia; and Calverton, MD, USA: Macro International, 2005. 64. Shah I and Ahman E, Age patterns of unsafe abortion in developing country regions, Reproductive Health Matters, 2004, 12(24 Suppl.):9–17. 65. Special tabulations of data from the Colombia Encuesta Nacional de Demografía y Salud, 2000. 66. Profamilia, Salud Sexual y Reproductiva en Zonas Marginadas: Situación de las Mujeres Desplazadas, Bogotá, Colombia: Profamilia, 2006. 67. Special tabulations of data from the Colombia Encuesta Nacional de Demografía y Salud, 2005. 68. Corte Constitucional, Sentencia T-388/09, Acción de tutela instaurada por BB actuando en representación de su compañera permanente AA contra SaludCoop E.P.S., Bogotá, 2009. 69. Billings DL et al., Pharmacy worker practices related to use of misoprostol for abortion in one Mexican state, Contraception, 2009, 79(6):445–451.

66

70. INVIMA, Resolución No. 266285, del 14 de septiembre de 2000, por medio del cual se concede a Profamilia el registro sanitario para importar y distribuir en forma exclusiva el producto POSTINOR 2, por el término de 10 años, Bogotá, 2000. 71. Mesa por la Vida y Salud de las Mujeres, Situación de las Prestaciones de Salud Sexual y Reproductiva ante los Cambios Recientes en los Planes de Beneficios, Bogotá, Colombia: Mesa por la Vida y Salud de las Mujeres, 2010. 72. Dirección General de Promoción y Prevención, Guía de Atención de las Complicaciones Hemorrágicas Asociadas al Embarazo, Bogotá, Colombia: Ministerio de Salud, no date. 73. Postabortion Care Consortium, PAC model, no date, <http://www.pacconsortium.org/site/PageServer?pagename=Themes_PAC_Model_2>, accessed Mar. 23, 2010. 74. Finer LB et al., Reasons U.S. women have abortions: quantitative and qualitative perspectives, Perspectives on Sexual and Reproductive Health, 2005, 37(3):110–118. 75. Villarreal J, Embarazo indeseado y aborto, Tribuna Médica, 1993, 88(3):149–153.

67

 APPENDIX TABLE 1. Measures of abortion and pregnancy among Colombian women, by region, 2008 Indicator ABORTION No. of women treated in health facilities for complications of miscarriages and induced abortions No. of women treated in facilities for complications of induced abortions Region Total 115,325 93,336 Bogotá 28,023 23,928 Pacífica 32,010 28,129 Atlántica 27,506 20,838 Central 19,223 13,533 Oriental 8,563 6,908

No. of induced abortions (medium range estimate) Abortion rate (no. of abortions per 1,000 women aged 15–44) Abortion ratio (no. of abortions per 100 live births)

400,412 39.2 52.4

117,422 65.6 97.3

96,515 53.5 76.7

92,284 42.1 45

62,360 23.8 35.9

31,831 17.5 22.9

Abortion complication treatment rate (no. of women treated per 1,000 women aged 15–44) PREGNANCY Total no. of pregnancies No. intended No. unintended Total pregnancy rate (no. of pregnancies per 1,000 women aged 15–44) Intended Unintended PERCENTAGE DISTRIBUTIONS All pregnancies by planning and outcome % unintended % ending in live births % ending in abortions % ending in miscarriages % intended % ending in live births % ending in miscarriages Unintended pregnancies only, by outcome % ending in abortions % ending in live births % ending in miscarriages Total
Note Numbers may not add up to totals because of rounding. Source Reference 21.

9.1

13.4

15.6

9.5

5.2

3.8

1,357,659 445,762 911,897 133 44 89

273,992 71,255 202,737 153 40 113

257,070 69,567 187,504 142 38 104

347,721 124,582 223,140 159 57 102

276,964 101,475 175,489 106 39 67

201,911 79,109 122,802 111 44 67

67.2 28.9 29.5 8.7 32.8 27.4 5.5

74.0 22.4 42.9 8.8 26.0 21.7 4.3

72.9 26.4 37.5 9.0 27.1 22.6 4.5

64.2 29.1 26.5 8.5 35.8 29.9 6.0

63.4 32.2 22.5 8.7 36.6 30.5 6.1

60.8 36.2 15.8 8.8 39.2 32.7 6.5

43.9 43.1 13.0 100.0

57.9 30.2 11.9 100.0

51.5 36.1 12.4 100.0

41.4 45.4 13.2 100.0

35.5 50.8 13.7 100.0

25.9 59.6 14.5 100.0

68

APPENDIX TABLE 2. Social, demographic and reproductive characteristics of women of reproductive age, by region and area of residence, Colombia, 1990, 2000, 2005 and 2010 Characteristic ALL WOMEN Social and demographic characteristics % living in urban areas (2005) % with more than completed primary (2010) % in two poorest wealth quintiles (2010) Fertility Total fertility rate (lifetime births per woman) 1990 2000 2005 2010 Wanted total fertility rate (lifetime births per woman)* 1990 2000 2005 2010 Total Region Bogotá Area of residence Oriental Urban Rural

Pacífica

Atlántica Central

74.4 76.1 35.8

99.8 88.1 3.6

67.4 73.5 41.7

71.9 75.9 53.6

72.4 73.4 35.6

67.2 70.9 41.9

na 82.8 19.0

na 51.4 98.3

2.9 2.6 2.4 2.1 2.2 1.8 1.7 1.6

2.4 2.4 2.2 1.9 2.0 1.9 1.6 1.5

2.9 2.6 2.3 2.0 2.4 1.6 1.6 1.5

3.6 2.7 2.7 2.6 2.8 2.0 1.9 2.0

2.6 2.5 2.3 1.9 1.9 1.7 1.5 1.3

2.8 2.8 2.6 2.2 2.1 1.8 1.7 1.6

2.5 2.3 2.1 2.0 2.1 1.7 1.5 1.5

3.8 3.8 3.4 2.8 2.7 2.3 2.1 1.9

Planning status of births in the five years before the study† % mistimed (wanted later) 1990 16.2 2000 28.2 26.3 2005 2010 27.9 % unwanted 1990 19.9 2000 23.2 2005 27.2 2010 23.5 % unplanned (mistimed plus unwanted) 1990 36.1 2000 51.4 2005 53.5 51.4 2010

18.6 31.0 30.8 30.0 14.8 16.7 27.6 20.8 33.4 47.7 58.4 50.8

20.6 25.4 26.7 28.1 15.7 29.0 26.9 25.8 36.3 54.4 53.6 53.9

15.3 27.5 27.7 28.7 19.1 20.7 24.3 20.7 34.4 48.2 52.0 49.4

13.3 26.3 22.1 27.4 23.1 26.6 32.4 23.9 36.4 52.9 54.5 51.3

15.0 31.8 25.5 25.7 25.1 21.3 24.3 26.9 40.1 53.1 49.8 52.6

18.5 29.9 27.0 28.5 16.7 21.4 26.1 21.8 35.2 51.3 53.1 50.3

11.6 24.4 24.8 26.4 26.1 27.0 29.5 28.0 37.7 51.4 54.3 54.4

Prenatal and delivery care among births occurring in the five years before the survey % of women receiving professional prenatal care‡ 78.3 82.0 94.4 1990 2000 90.8 95.5 88.5 2005 93.5 98.0 90.9 2010 97.0 98.2 96.5

71.9 90.2 89.9 95.5

82.9 88.3 93.3 97.6

87.8 93.0 96.5 97.8

87.6 93.7 95.8 98.0

71.0 83.3 87.5 94.0

69

% of deliveries occurring in a health facility 1990 2000 2005 2010

76.3 87.5 92.0 95.4

93.4 97.2 99.3 99.5

72.1 80.4 86.5 90.5

64.7 85.7 89.6 95.3

80.5 86.6 91.3 96.1

76.6 89.2 94.3 96.8

84.8 94.4 97.5 98.4

59.7 70.6 78.0 87.5

Contraceptive use, unmet need for contraception and reproductive preferences among women in union (formal and consensual) % using any contraceptive method 66.1 74.9 65.7 53.5 65.5 73.6 69.1 59.1 1990 76.9 80.6 76.2 70.8 76.7 82.5 77.6 75.1 2000 2005 78.2 82.5 78.2 70.4 79.8 81.4 78.8 76.7 2010 79.1 80.8 80.4 71.5 81.7 81.5 79.0 79.3 % using a modern method§ 1990 54.6 61.8 54.8 48.4 53.1 57.0 57.7 47.5 2000 63.3 68.7 64.2 57.3 62.8 66.4 65.8 56.9 2005 67.5 74.1 66.9 59.0 69.4 69.6 68.8 64.1 2010 72.8 75.7 73.5 65.0 75.3 75.2 73.0 72.2 % using a traditional method** 1990 11.5 13.0 10.9 5.0 12.4 16.7 11.4 11.6 13.6 11.9 12.1 13.5 13.9 16.1 11.9 18.2 2000 10.7 8.4 11.3 11.4 10.4 11.8 10.0 12.5 2005 2010 6.3 5.1 6.9 6.5 6.4 6.3 6.0 7.1 % with unmet need for contraception†† 1990 11.1 7.9 11.8 18.3 9.6 7.5 9.2 15.6 2000 6.2 4.5 7.6 9.0 5.5 3.7 5.4 8.2 2005 5.8 3.6 5.5 9.6 4.9 4.4 4.8 8.2 2010 6.9 5.4 7.7 10.2 5.4 6.0 6.6 8.0 % wanting a(nother) child later‡‡ 1990 19.7 22.1 18.3 19.2 18.9 20.7 20.7 17.4 2000 16.7 18.9 15.3 19.9 14.1 15.7 17.3 15.3 16.3 18.1 15.8 18.0 13.9 16.4 16.6 15.5 2005 2010 17.0 18.6 14.6 19.0 15.4 17.5 17.2 16.5 % wanting no more §§ 1990 64.0 62.0 67.9 63.6 64.4 62.1 62.1 68.4 2000 69.5 66.0 72.2 65.0 71.3 73.0 67.7 74.3 2005 71.2 71.0 71.7 67.7 73.9 71.4 70.0 74.3 2010 70.1 67.1 74.7 65.6 73.2 70.4 69.2 72.8 ADOLESCENTS AND YOUNG ADULT WOMEN Adolescent fertility rate (no. of births per 1,000 15–19‐year‐olds) 1990 70.0 2000 85.0 2005 90.0 2010 84.0

43.0 69.0 88.0 64.2

100.0 90.0 95.0 84.6

90.0 92.0 88.0 96.3

62.0 83.0 95.0 82.4

47.0 87.0 83.0 83.3

63.0 71.0 79.0 73.0

90.0 134.0 128.0 122.0

70

% of 15–19‐year‐olds who are already mothers or are currently pregnant 1990 12.8 11.5 2000 19.1 16.7 2005 20.5 22.6 2010 19.5 17.5 Median age*** At first sex 1990 20.2 20.8 2000 18.8 18.9 2005 17.9 18.1 2010 17.6 17.5 At first union 1990 21.5 22.1 2000 21.4 22.6 2005 21.8 23.0 2010 21.5 23.8 At first birth 1990 22.6 23.6 2000 21.8 22.5 2005 21.6 22.0 2010 17.6 22.8

14.7 23.0 20.3 20.5

17.5 18.9 19.0 20.2

11.6 19.4 21.5 19.2

8.9 17.2 18.8 19.1

11.8 16.9 18.5 17.3

16.2 26.2 26.9 26.7

18.9 18.3 17.6 17.2 20.7 21.3 22.4 21.2 22.0 21.2 21.1 21.3

19.0 19.0 18.5 18.2 19.7 20.3 21.2 20.6 21.2 21.5 21.8 21.7

21.4 18.7 17.8 17.4 22.9 21.9 22.2 21.2 23.8 22.1 21.4 20.9

20.7 18.8 17.8 17.6 21.7 21.6 21.1 20.7 22.4 21.6 21.1 21.0

20.7 18.9 18.1 17.7 22.1 22.0 22.5 22.2 23.4 22.4 22.1 22.0

19.3 18.1 17.2 17.2 20.4 19.7 19.9 19.3 21.5 20.0 19.8 19.8

Note na=not applicable. *Number of births a woman would have if she avoided all unwanted births, which are defined as those conceived after a woman had already achieved her reported ideal family size. †Refers to whether births were mistimed (i.e., wanted later) or unwanted (i.e., occurring when a woman did not want to have additional children). ‡Defined as care from doctors and nurses. The data refer to prenatal care for the most recent birth among women who had a birth in the five years before the survey. §Pill, IUD, injectable, female and male sterilization, implant, male condom and spermicides. **Periodic abstinence, withdrawal, lactational amenorrhea and folk methods. ††A woman has an unmet need for contraception if she is in a union, able to become pregnant (and is not currently pregnant or amenorrheic), does not want to have a child in the next two years or wants to stop childbearing, and is not using any method of contraception. ‡‡Includes women who reported wanting to have a child later or who were unsure of timing/undecided. §§Includes women who are sterilized and those whose partners are. ***Medians are calculated among women aged 25–29.

Sources % living in urban areas—Departamento Administrativo Nacional de Estadística (DANE), Censo General 2005, Nivel Nacional, Cuadro C1, Población Ajustada al 30 de Junio de 2005 por área según departamentos y municipios, Bogotá, 2008, pp. 471–496, <http://www.dane.gov.co/censo/files/libroCenso2005nacional.pdf>, accessed May 6, 2010. For all demographic and health survey data, by survey year—1990 (references 4 and 7); 2000 (references 60 and 65); 2005 (references 63 and 67); and 2010 (references 5 and 6).

71