Health consequences of unsafe abortion in Colombia, 1989–2008

Abstract

The number of Colombian women hospitalized for the treatment of induced abortion complications increased from 57 679 in 1989 to 93 336 in 2008; the hospitalization rate also rose: from 7.2 to 9.1 cases per 1000 women aged 15-44 years. Factors that likely underlie the increase include improved access to postabortion care (although 1 in 5 women still do not obtain the care they need) and the growing role of misoprostol, often used incorrectly and to some extent replacing the use of surgical abortion by doctors. Efforts are evidently needed to improve access to safe abortion and effective contraception.

Full text

International Journal of Gynecology and Obstetrics 118, Supplement 2 (2012) S92–S98
CONSEQUENCES OF UNSAFE ABORTION ON MORBIDITY AND MORTALITY
Health consequences of unsafe abortion in Colombia, 1989–2008
Elena Prada a,*, Susheela Singhb, Cristina Villarrealc
aConsultant to Guttmacher Institute, Bogotá, Colombia
bGuttmacher Institute, New York, NY, USA
cFundación Oriéntame, Bogotá, Colombia
ARTICLE INFO ABSTR ACT
Keywords:
Abortion complications
Abortion morbidity
Access to care
Misoprostol
Postabortion care
The number of Colombian women hospitalized for the treatment of induced abortion complications increased
from 57 679 in 1989 to 93 336 in 2008; the hospitalization rate also rose: from 7.2 to 9.1 cases per 1000
women aged 15–44 years. Factors that likely underlie the increase include improved access to postabortion
care (although 1 in 5 women still do not obtain the care they need) and the growing role of misoprostol, often
used incorrectly and to some extent replacing the use of surgical abortion by doctors. Efforts are evidently
needed to improve access to safe abortion and effective contraception.
© 2012 Published by Elsevier Ireland Ltd. on behalf of International Federation of Gynecology and Obstetrics.
1. Introduction
Unsafe pregnancy terminations are a harsh reality in countries
where induced abortion is highly restricted. Unsafe abortion is a
major cause of death and disability among women worldwide, and
the treatment of abortion complications represents an important
burden to health systems and to the economic well-being of
households [1]. Data on maternal mortality in Colombia are scarce
and outdated, and information on live births (basic information
for the denominator) is of poor quality. Recent estimates of the
country’s maternal mortality ratio in 2008 vary from 75 maternal
deaths per 100 000 live births according to the Ministry of Social
Protection [2] to 85 per 100000 estimated by the World Health
Organization (WHO) [3]. In the mid-1990s, unsafe abortion was
the second leading cause of maternal mortality, accounting for
16% of maternal deaths in the country [4]; by 2007, abortion
was responsible for 9% of all maternal deaths [5], indicating that
its contribution to maternal mortality fell by about 50% between
1994 and 2007. Nonetheless, WHO’s most recent estimate of 780
maternal deaths in 2008 [3] translates to roughly 70 Colombian
women dying each year from the easily preventable cause of unsafe
abortion.
Given some striking changes in the abortion methods that
women in many parts of the world are using – particularly their
growing use of misoprostol – continued research into the health
consequences of unsafe abortion remains necessary. Increased
use of misoprostol could theoretically bring about less severe
complications than those associated with the more invasive, unsafe
methods used in the past. On the other hand, women may be
substituting this method for the safer abortion procedures provided
* Corresponding author: Elena Prada, Carrera 9, #97-06, Bogotá, Colombia. Tel:
+57 1 691 7993.
E-mail address: eprada@guttmacher.org (E. Prada).
0020-7292/$ – see front matter © 2012 Published by Elsevier Ireland Ltd. on behalf of International Federation of Gynecology and Obstetrics.
by doctors and qualified midlevel providers. The present study
assessed changes in the health consequences of unsafe abortion in
Colombia over the past 2 decades. The only previously available
national statistics date back to the early 1990s, when a study on the
incidence of abortion and abortion morbidity was carried out for
the first time in the country [6]. The current study thus fills a gap
in our understanding of abortion morbidity. It draws on national
government hospitalization statistics for the year 1989, surveys
of health professionals knowledgeable about abortion practices in
Colombia, and a survey of healthcare facilities to obtain a picture of
the current state of induced abortion in the country.
2. Background
In the past 2 decades, national surveys have documented a
rise in contraceptive use and a decline in average family size
across the country, although these trends are strongest in urban
regions such as Bogotá and among more wealthy women [7].
Yet, unplanned childbearing continues to be prevalent, and the
unintended pregnancy rate was higher in 2008 than it was in
1989 [8]. These trends argue that some proportion of women with
unintended pregnancies will continue to seek abortion, even if it is
from an untrained provider or under unsafe conditions because of
their motivation to avoid a mistimed or unwanted birth.
Some important changes in the organization of the country’s
health sector may affect the situation regarding access to postabor-
tion care and the consequences of unsafe abortion. From 1986 to
1990, the health system in Colombia was decentralized, giving more
responsibility to local levels for the provision of health services and
empowering communities to demand services that they had not
had before. Then in 1993, Law100 had the effect of integrating
the social security and public health sectors and creating universal
access to health care. These 2 major changes contributed to large
increases in the population covered by the national health insur-

E. Prada et al. /Internation al Journal of Gynecology and Ob stetrics 118, Su pplement 2 (2012) S92– S98 S93
ance plan. According to the Ministry of Social Protection, healthcare
coverage rose from 23% in the mid-1990s to 88% by 2008 [2]. More
recent studies, however, show that even with this large expansion
in coverage, inequities still exist, and rural people and poor ur-
ban people still have much lower coverage than their respective
counterparts [9].
Another major change occurred on May 10, 2006, when the
Constitutional Court of Colombia liberalized abortion, allowing
voluntary terminations in cases of rape and incest, and fetal
malformation, or when the life or health of the woman is in
danger. This change has the potential to affect the safety of
abortion in Colombia. However, in 2009, 3 years after Colombia’s
high court liberalized abortion, official data show that very few
women had had a legal pregnancy termination [10]. Women are
facing obstacles at both the individual level (many doctors, on
the grounds of conscientious objection, are reluctant to perform
abortions) and the institutional level (women are often required to
provide certifications and documents not mandated under the law).
As a result, the recent abortion ruling has as yet not greatly affected
the health consequences of unsafe abortion, and will do so only if
barriers to the provision of legal abortion services are overcome
[11].
3. Data sources and methodology
Data for the year 1989 come from a national study of abor-
tion morbidity, which used official government statistics of the
number of women hospitalized for the treatment of postabortion
complications. However, the completeness of government statistics
deteriorated substantially after the reform of the healthcare system
in 1993, as the collection of healthcare data was decentralized [12].
For this reason, the current study could not use existing official
statistics. Instead, a Health Facilities Survey (HFS) of hospitals likely
to provide postabortion care was conducted to obtain representa-
tive national data on the number of women treated for postabortion
complications in 2008.
The 2009 data come from 2 surveys: the HFS just described
(although the reference year for this study is 2008) and a Health
Professionals Survey (HPS) of key informants knowledgeable about
the conditions of abortion provision in the country. The HFS,
which is based on a nationally representative sample of 300 public
and private health facilities, provides information on the annual
number of women treated in facilities for complications of abortion,
both induced and spontaneous. Applying the Abortion Incidence
Complications Method (AICM) to these hospitalization numbers, an
indirect methodology that has been used in many countries over
the past 2 decades [6], the total number of women treated for
postabortion complications in Colombia in 2008 was estimated. The
HPS was carried out in 1992 and 2009. Underreporting of abortions
is very high in all countries, particularly those with restrictive
laws. Therefore, the study relied on health professionals’ opinions
and perceptions rather than on information obtained from women
themselves. Although face-to-face interviews may be more direct,
the high level of underreporting with that format cancels out their
possible accuracy.
Both surveys are based on purposive samples (30 professionals
in 1992, and 102 in 2009). A large proportion of respondents
in both years were medical doctors (including both public- and
private-sector physicians), and the remainder (33% in 1992 and 54%
in 2008) were professionals from other fields, including researchers,
health administrators, policy analysts, and advocates. Respondents
were selected because of their knowledge about the conditions of
abortion provision in Colombia.
These 2 surveys capture comparable estimates of the conditions
under which abortion is provided and the extent to which unsafe
abortion occurs in Colombia. They also document professionals’
perceptions of the probability of women experiencing abortion
complications and the likelihood that women with complications
will obtain care at a medical facility. Because a woman’s place of
residence and economic status may influence her access to and
choice of an abortion provider, questions were asked separately
about each of 4 different subgroups of women: urban poor, urban
nonpoor, rural poor, and rural nonpoor. Given the lack of reliable
information on income in Colombia, women’s education was used
as a proxy variable for poverty: poor women were defined as
those who had attained 7 or fewer years of education, whereas
nonpoor were defined as those who had attained 8 or more
years. Two new topics were added to the 2009 HPS: the use of
misoprostol and respondents’ attitudes about the 2006 court ruling
that partially decriminalized abortion. In the 2009 HPS, 2 separate
questions were asked about provision of induced abortion. The
first question sought estimates of the proportion of all women
obtaining induced abortions who used misoprostol, regardless of
the source from which the medication was obtained. The second
question asked respondents to estimate the proportion of women
having non-misoprostol abortions who went to each of 5 types of
providers (medical doctors, certified midwives/nurses, pharmacists,
traditional midwives, and the woman herself). This approach was
used because of the difficulty of asking respondents to estimate
for 10 categories that would be expected to sum up to 100%
(misoprostol and non-misoprostol) for each of 5 provider types.
Separate questions were also asked about the proportion of
women who might develop a medical complication requiring care
in a medical facility if they used misoprostol (regardless of the
source from which this method was obtained) and then separate
questions for each of the 5 types of abortion providers, including
self-induced. These questions were also asked about each of the 4
subgroups of women mentioned before: urban poor, urban nonpoor,
rural poor, and rural nonpoor.
For comparison with the earlier HPS survey, information ob-
tained on sources of abortion provision in the 2009 survey was
grouped into 4 categories: (1) medical doctors and gynecologists;
(2) midlevel health professionals: nurses, pharmacists, and certified
midwifes; (3) traditional providers: traditional midwives, and the
woman herself; and (4) misoprostol (obtained from a range of
sources), a category that exists only in 2009.
4. Results
4.1. Trends in the safety of induced abortion, 1992–2009
Over the past 2 decades, there has been a significant shift
in how Colombian women obtain abortions (Fig. 1). In 1992,
health professionals estimated that medical doctors were the main
Fig. 1. Percentage distribution of women by source of abortion, 1992–2009. Source:
Health Professionals Survey, 1992 and 2009.

S94 E. Prada et al. /Internation al Journal of Gynecology and Ob stetrics 118, Su pplement 2 (2012) S92– S98
providers (37%), while 31% of women obtained abortions from
midlevel health professionals, and 32% from traditional providers.
By 2009, experts estimated that half (50%) of women obtaining
abortions used misoprostol. The remaining half is almost evenly
distributed among 3 types of providers: medical doctors (18%),
midlevel health professionals (16%), and traditional providers (16%).
These estimates suggest that there has been a large shift in the
proportion of women obtaining abortions from the 3 main groups
of providers identified in the early 1990s, to using misoprostol;
the provision of abortions by physicians is also estimated to
have declined. This shift has been more pronounced among 3 of
the 4 socioeconomic subgroups: in 2009, misoprostol (obtained
from a variety of sources) accounted for 40%–59% of abortions
in 3 subgroups (rural nonpoor, urban poor, and urban nonpoor)
(Table 1). In the case of both urban and rural nonpoor women,
all of the movement was from physician-provided non-misoprostol
abortions to abortions obtained through the use of misoprostol
(which can be obtained from a variety of sources; information is
not available on the extent to which this is self-administration as
opposed to use under the supervision of providers).
Poor urban women probably experienced a net improvement
in safety, given their shift out of the 2 least safe categories
of providers into the category using misoprostol (the proportion
of their abortions provided by midlevel health professionals or
traditional providers dropped from 78% to 46%), as misoprostol
became the method used by 47% of this group by 2009 (Table 1).
There was some shifting away from physician-provided abortions as
well, with the proportion in this group dropping from 22% to about
8%. Poor rural women saw a similar trend of a likely improvement
in safety, since proportions in the 2 least safe provider categories
(traditional providers and self-induced) declined, as the proportion
using misoprostol, a method that is likely to have less serious
complications, increased to about 25%. The lowest proportion of
women relying on misoprostol in 2009 was among poor rural
women, who are less likely to know about this drug and where
to obtain it, as well as being less able to afford the cost of the
medication.
4.2. Trends and differentials in estimated probability of complications
Despite the important shift estimated by HPS respondents
toward safer abortion sources/methods, particularly the increased
use of misoprostol and decreased use of least safe providers,
respondents perceived that there was a substantial increase in
women’s experience of abortion complications.
The overall proportion of women thought to experience abortion
complications went from 29% in 1992 to 33% in 2009 (Fig. 2). The
large increase in the use of misoprostol, accompanied by a decline
in the use of physicians, for the reasons explained before, are
important factors contributing to the observed increase in abortion
complications between 1992 and 2009.
Over the period in question, the estimated probability of compli-
cations increased for each of the 3 categories of providers for which
there are comparable data for these 2 years (Fig. 2). It increased
from 55% to 58% among traditional providers and from 31% to 35%
among midlevel health professionals. These increases are not very
large, and given the fact that the source of the data is respondents’
perceptions, some variability would be expected between groups
of respondents and over time. The estimated probability of com-
plications also increased for physicians, from 5% to 11%. The fact
that a much larger proportion of respondents were medical doctors
in 1989 (67%) than in 2009 (46%) may have influenced this result,
as medical doctors’ perceptions of their own likelihood of causing
abortion complications is likely to be underestimated.
Half of abortion complications in 2009 were the result of
misoprostol use (16% of the 33%). Although misoprostol has a high
rate of effectiveness if administered correctly (85%–90%) [13], it is
likely that the same situation as found in Mexico [14] – where many
women do not know how to use the medication and even providers
such as pharmacists do not know the exact doses required – may
also be occurring in Colombia.
The systematic relationship between socioeconomic status and
the probability of complications among women using misoprostol
(lowest among urban nonpoor women – 28% – and highest among
poor rural women – 47%; Table 2) supports the expectation that
more prosperous and better educated women will have more
accurate information about use of the method and will be more
successful in using it.
In both 1992 and 2009, within each source of abortion category,
socioeconomic status was directly related to the estimated prob-
ability of an abortion complication occurring (Table 2). For all 4
socioeconomic subgroups, the lowest probabilities of complications
were estimated for women who obtained their abortions from
physicians and the highest for women who went to traditional
practitioners. For misoprostol users in 2009, the probability of a
complication was estimated as being close to that of women in all
4 population groups who went to midlevel health professionals.
4.3. Trends in proportions of women treated for abortion complications
Because of reforms in the country’s healthcare system, access to
postabortion care in Colombia improved significantly over the past
2 decades, enabling more women to obtain postabortion care [2,15].
However, some women still do not obtain the treatment they need.
By 2009, according to the HPS, an estimated 21% of all women with
abortion-related complications in Colombia failed to obtain medical
treatment. Nevertheless, this proportion is much lower than in
1992, when it was perceived to be 46% (Fig. 3). In contrast, for
poor rural women the situation has not changed at all, and the gap
between the lack of access to care of this group (49% in 2009) and
that of the other socioeconomic groups (11%–31%) has widened in
the past decade. In 1992, 48% of poor rural women did not receive
the postabortion medical care they needed compared with 40%–
51% of the other 3 subgroups of women. It should be emphasized
that these estimates indicate that access to postabortion care has
been improving and that differentials have narrowed among most
socioeconomic groups. However, it is discouraging that inequities
among those most in need are now more pronounced than 2
decades ago.
4.4. National trends in abortion complications
When abortions are performed by skilled professionals in en-
vironments that conform to medical standards, abortion complica-
tions are minimal [16]. However, the number of Colombian women
treated in public and private health facilities for complications
of any type of abortion (spontaneous or induced) increased from
76 905 in 1989 to 115 325 in 2008 (Table 3). Using comparable
indirect estimation techniques for both years, it was calculated
that the number of women experiencing abortion-related morbidity
due o nly to uns afe abo rtion rose from 57 679 in 1989 to 93 336
in 2008. The overall annual rate of treatment of induced abortion
complications increased by 26%, from 7.2 hospitalization cases per
1000 women aged 15–44 years in 1989 to 9.1 in 2008. This rate
increase is net of population growth. The most likely reasons are
improvements in access to postabortion care, the increased use of
misoprostol over the past 2 decades and the increase (of 8%) in
the country’s abortion rate between 36 and 39 per 1000 women
aged 15–44 years. The first 2 of these trends are documented by
the HPS surveys. Table 4 presents the number of women treated for
induced abortion complications and the rate per 1000 women aged
15–44 years by region. Pacífica had the highest rate of facility-based

E. Prada et al. /Internation al Journal of Gynecology and Ob stetrics 118, Su pplement 2 (2012) S92– S98 S95
Table 1
Respondents’ estimates of the percent distribution of women having induced abortions according to source of abortion by women’s socioeconomic status and place of
residence, Colombia, 1992 and 2009
Source 1992 2009
Urban non poor Rural non poor Urban poor Rural poor Total Urban non poor Rural non poor Urban poor Rural poor Total
Total 100 100 100 100 100 100 100 100 100 100
Misoprostol U U U U U 58.6 39.7 46.6 25.2 48.8
MD/Gynecologist 79.6 49.6 21.6 8.2 37.2 25.1 16.5 7.5 4.4 16.6
Midlevel health professionals a15.0 32.6 39.4 37.6 30.9 11.8 25.8 20.7 20.4 16.7
Traditional providers b5.3 17.7 39.0 54.3 31.9 4.5 18.0 25.2 50.0 17.9
Abbreviations: MD, medical doctor; U, unavailable at that time.
aIncludes nurse, pharmacist, and certified midwife.
bIncludes traditional midwife and woman herself.
Source: Health Professionals Surveys, 1992 and 2009.
Fig. 2. Proportion of all women who experience abortion complications per source of abortion, 1992–2999. Source: Health Professionals Survey, 1992 and 2009.
Table 2
Respondents’ estimates of the proportion of women likely to experience complications among those women obtaining an abortion from each source, according to women’s
place of residence and socioeconomic status, Colombia, 1992 and 2009
Source 1992 2009
Urban non poor Rural non poor Urban poor Rural poor Urban non poor Rural non poor Urban poor Rural poor
National 8.9 19.2 36.4 44.2 24.4 43.7 33.3 53.2
Misoprostol U U U U 27.6 33.8 40.9 47.2
MD/Gynecologist 4.2 6.0 6.9 10.4 9.9 12.1 13.6 16.9
Midlevel health professionals a24.2 26.1 31.9 32.5 29.0 34.0 41.6 43.5
Traditional provider b44.6 50.6 52.9 58.3 47.8 51.2 60.0 63.4
Abbreviations: MD, medical doctor; U, unavailable at that time.
aIncludes nurse, pharmacist, and certified midwife.
bIncludes traditional midwife and woman herself.
Source: Health Professionals Surveys, 1992 and 2009.
Fig. 3. Percentage of women who have abortion complications and who do not receive medical care, according to socioeconomic group, 1992–2009. Source: Health
Professionals Survey, 1992 and 2009.
postabortion treatment, followed closed by Bogotá (16 and 13 per
1000 women, respectively), whereas Oriental had the lowest rate
(4 cases per 1000 women). The high rate in Pacífica region is likely
due to more dangerous conditions of abortion provision than to
an increase in access to healthcare services. This region contains
3 of the 4 poorest Departments (there are 32 Departments, or
geographic subdivisions, in Colombia) in the country [17] and has
fewer health facilities than does Bogotá.
4.5. Status of postabortion services
Based on information from the Ministry of Social Protection,
in October 2008, there were about 1100 facilities likely to pro-
vide postabortion care. Of this number, 6 in 10 did not provide
postabortion care in 2008; of the remaining facilities, 23% offered
postabortion care to women both as inpatients and outpatients, 14%
as outpatients only, and 2% as inpatients only (Table 5). The region

S96 E. Prada et al. /Internation al Journal of Gynecology and Ob stetrics 118, Su pplement 2 (2012) S92– S98
Table 3
National trends in abortion complications, Colombia, 1989–2008
National 1989 2008 % change
No. of women hospitalized for abortion
complication (any type) 76905 115 325 50
No. of women hospitalized for induced
abortion 57679 93 336 62
Induced abortion hospitalization rate
(per 1000 women aged 15–44 years) 7.2 9.1 26
No. of women aged 15–44 years 7 968 123 10 222 960 28
Source: For 1989, Ministry of Health official statistics; for 2009, Health Facilities
Survey.
Table 4
Number of women treated for induced abortion complications and rate per 1000
women aged 15–44 years by region, Colombia, 2008
Region No. of women treated for Rate of facility-based treatment
complications of induced per 1000 women aged 15–44 years
abortions
Total 93336 9.1
Region
Bogotá 23928 13.1
Pacífica 28129 15.6
Atlántica 20838 9.7
Central 13533 4.9
Oriental 6 908 3.7
Source: Health Facilities Survey, Colombia, 2009.
Table 5
Percentage of facilities that offer inpatient and outpatient postabortion care, and
average annual number of postabortion patients treated among those facilities
offering any care by region, Colombia, 2008
Region % that offer abortion care
Inpatient only Outpatient only Both Neither
Total 1.6 13.5 23.1 61.9
Region
Atlantic 2.2 27.3 35.2 35.2
Bogotá 3.6 15.9 24.6 55.8
Central 0.4 0.4 13.0 86.3
Eastern 0.0 1.5 26.0 72.5
Pacific 2.5 24.7 20.1 52.7
Source: Health Facilities Survey, 2009.
with the lowest proportion of health facilities likely to provide both
types of postabortion care was the Central region (13%), while the
region with the highest was the Atlántica region (35%). Among
facilities that provide postabortion services (38% of the total), the
average annual caseload of patients treated for induced abortion
in 2008 was 221, with no major difference between public- and
private-sector facilities (Table 6). As expected, tertiary health facil-
ities and larger facilities (those with more than 100 beds) treated
a higher-than-average number of cases during 2008: 325 and 520
patients, respectively. This is not surprising given that the most
common procedure used to treat abortion complications is dila-
tion and curettage (D&C), a procedure likely to require anesthesia
and the presence of an anesthesiologist, services that small and
remote health facilities often do not have. The common practice in
small health facilities is to stabilize women and then refer them
to the nearest health facility with an adequate infrastructure and
the appropriate human resources. The HFS shows that almost all
abortion complications in Colombia are treated by D&C. Of all facil-
ities that reported providing postabortion care, 93% said they most
commonly used D&C, and just 7% used manual vacuum aspiration
(MVA – data not shown). It is important to note that although the
Ministry of Social Protection guidelines [18] for both postabortion
care and legal abortions, which are based on WHO guidelines [19],
recommend that MVA be used to treat incomplete abortions at
Table 6
Average annual case load of patients treated for induced abortion according to
facility size, ownership, and number of beds, Colombia, 2008
Characteristics Average No. of patients % distribution
caseload treated
% of facilities that offer PAC 38 – –
Annual average caseload 221 93 336 100.0
Ownership
Public 224 45 921 49.2
Private 220 46 854 50.2
Other a10 19 0.02
No information on ownership b254 542 0.58
No. of beds
0–10 42 4200 4.5
11–30 128 19 413 20.8
31–100 286 29 214 31.3
101–732 520 40508 43.4
Facility size
Tertiary level 325 23 054 24.7
Secondary level 395 56 935 61.0
Primary level 69 9240 9.9
Others 56 4107 4.4
Abbreviation: PAC, postabortion care.
aNGO, only 2 cases.
bOnly 2 cases.
Source: Health Facilities Survey, Colombia, 2009.
15 weeks of gestation or less; in practice, the recommendation is
not observed by health professionals in most health facilities in
Colombia. A baseline study on barriers to access to legal abortion
services found that in late 2007, just 11% of providers had MVA
equipment [20]. Inadequate availability of MVA equipment and lack
of training in its use are possible reasons why medical doctors rely
on and prefer D&C over all other types of procedures. The technique
of D&C was the most common in both public and private facilities,
hospitals and clinics, and most regions, with the exception of Pací-
fica, where 23% of facilities reported that they most commonly used
the recommended method of MVA.
5. Discussion
5.1. Key findings
Between 1992 and 2008, there were large increases in Colom-
bian women’s use of misoprostol to end unintended pregnancies
and concomitant reductions in their use of relatively less safe
abortion providers (traditional midwives) as well as declining use
of safe providers (medical doctors). Over the same period, abortion
complications increased by 14% and the rate of hospitalization for
the treatment of abortion-related complications rose by 26%.
A number of factors can help explain the increase in abortion
complications and in the need for postabortion care: a likelihood
that the dramatic rise in reliance on misoprostol (the method now
used by half of all women) has been accompanied by some degree
of ineffective or incorrect use, as more and more women obtain
the medication from informal sources; incorrect instructions to
women to go to a health facility to have the abortion completed
once bleeding begins, even though this may not have been needed;
a substantial movement away from the use of physicians, whose
safety record is usually high; improved access to treatment among
large numbers of women who now have healthcare coverage as
a result of reforms in the country’s healthcare system; and an
associated increase in morbidity resulting from the 8% increase in
Colombia’s overall abortion rate between 1989 and 2008 [8].
It is still not known where most Colombian women obtain miso-
prostol. The only known published study, carried out in 13 hospitals
between 1999 and 2002, found that among postabortion patients
who reported having induced their abortion with misoprostol, 70%

E. Prada et al. /Internation al Journal of Gynecology and Ob stetrics 118, Su pplement 2 (2012) S92– S98 S97
reported that they obtained the medication from pharmacies [21].
As the pills have become less expensive and easier to find on
the black market, the drug now appears to be widely available
and easy to purchase through informal outlets. The HPS results
and some press reports [22–24] seem to confirm this impression.
In fact, given a large decline in sales through formal distribution
channels (by 87% between 2002 and 2007 [25]), it seems likely that
misoprostol is now mostly obtained through informal sources – a
trend that could contribute to poor knowledge about how to use it
correctly.
Despite impressive improvements in coverage, following exten-
sive health system reforms, 1 in 5 Colombian women in need of
postabortion care (and half of poor rural women) still go untreated
[26]. Many factors may be playing a role in preventing women
from obtaining postabortion care. These include long distance from
a source of health care, lack of economic resources to cover trans-
portation and treatment costs, lack of empowerment to seek care,
and some women’s preferences for seeking care from a pharmacy
or a traditional source, or to treat themselves. In some other cases,
women may be afraid of mistreatment at the health facility, or
the facility that they can get to lacks the resources or personnel
necessary to provide postabortion treatment. Therefore, additional
efforts should be made to improve access to treatment for poor
rural women and to learn why this group’s healthcare coverage has
not improved.
It is also sobering to discover that the most widely used
mode of treatment for incomplete abortions in Colombia is still
D&C when other less invasive and more cost-effective techniques
recommended by WHO, such as MVA and misoprostol, are available
and approved in the country. The findings also point to an
urgent need to train healthcare providers, pharmacists, and women
themselves in the correct use of misoprostol.
Overall, however, the results of this study could be construed
as more positive than negative. Now more Colombian women with
abortion-related complications than in the past get the care they
need, and misoprostol, which women can use in complete privacy,
is largely replacing more expensive and often less safe methods.
And with time, as women learn to use misoprostol correctly, the
worst consequences of unsafe abortions in Colombia may well start
to disappear.
5.2. Limitations
There are methodologic limitations to both the data and ana-
lytic approach given the challenges of undertaking research on a
stigmatized and clandestine behavior. This study does not provide
information on the type of abortion complications and their treat-
ment. Instead, it gives only the total count of women admitted to
health facilities. Analysis of trends in 1 of the outcomes assessed
in this study, the rate of treatment for postabortion complications,
may be affected by the different sources of data for 1989 and
2008: data for 1989 came from official hospital discharge statistics,
whereas the 2008 data came from a nationally representative sur-
vey of health facilities. A further limitation is that information on
proportions of women with complications and proportions getting
care relies on professionals’ perceptions and opinions, because it
is not possible to obtain such data through community-based sur-
veys of women, given the high level of underreporting of abortion
experience by women in face-to-face interviews.
5.3. Conclusion
The large increases in Colombian women’s use of misoprostol
combined with increases in the postabortion treatment rate are
most likely the result of providers’ inadequate knowledge of
protocols for use of this medication and women’s incorrect use
of the method. These findings strongly suggest that there is great
need for providing more accurate information to both women and
providers on how the method works, how it should be used, and
when to seek medical care to decrease the unnecessarily high
complication rates.
Acknowledgments
This research was supported by a grant from the Dutch Ministry
of Foreign Affairs.
Conflict of interest
The authors declare that they have no conflicts of interest.
References
[1] Singh S, Wulf D, Hussain R, Bankole A, Sedgh G. Abortion Worldwide: A
Decade of Uneven Progress. New York, NY, USA: Guttmacher Institute; 2009.
[2] Ministerio de la Protección Social, Organización Mundial de la Salud. Situación
de Salud en Colombia: Indicadores Básicos, 2008. Bogotá, Colombia: INS; 2008.
[3] World Health Organization, UNICEF, United Nations Population Fund (UNFPA),
World Bank. Trends in Maternal Mortality 1990 to 2008: Estimates Developed
By WHO, UNICEF, UNFPA and the World Bank. Geneva, Switzerland: WHO;
2010.
[4] Ministerio de Salud. Mortalidad Materna en Colombia. Bogotá, Colombia:
Imprenta Nacional; 1996;18.
[5] Departamento Administrativo Nacional de Estadística (DANE). 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). Available at: http://www.dane.gov.co/index.php?option=com_
content&view=article&id=205&Itemid=119. Accessed May 7, 2012.
[6] Singh S, Wulf D. Estimated levels of induced abortion in six Latin American
countries. Int Fam Plann Perspect 1994;20(1):4–13.
[7] Profamilia. Encuesta Nacional de Demografía y Salud (ENDS), Colombia, 2010.
Bogotá, Colombia: Profamilia; 2011.
[8] Prada E, Biddlecom AE, Singh S. Induced abortion in Colombia: new esti-
mates and change between 1989 and 2008. Int Perspect Sex Reprod Health
2011;37(3):114–24
[9] Flórez CE, Tono TM. La Equidad en el Sector Salud: Una Mirada de Diez Años.
Bogotá, Colombia: Fundación Corona and Fundación Ford; 2002. No. 6.
[10] Ministerio de la Protección Social. Situación de los Derechos Sexuales y
Reproductivos de las Mujeres en Colombia. Bogotá, Colombia: Ministerio de la
Protección Social; 2010.
[11] Moloney A. Unsafe abortions common in Colombia despite law change. Lancet
2009;373(9663):534.
[12] Uribe Mallarino C. Borrón y cuenta nueva: las estadísticas en Colombia se
reinventan a sí mismas. Univ Humaníst 2007;63:91–108.
[13] Faúndes A, Fiala C, Tang OS, Velasco A. Misoprostol for the termination of
pregnancy up to 12 completed weeks of pregnancy. Int J Gynecol Obstet
2007;99(Suppl 2):S172–7.
[14] Lara D, Abuabara K, Grossman D, Diaz-Olavarrieta C. Pharmacy provision of
medical abortifacients in a Latin American city. Contraception 2006;74(5):394–
9.
[15] Yepes FJ, Ramírez M, Cano MG, Bustamante R. Aiming for equity in Colombia’s
health system reform: achievements and continuing challenges. In: Haddad
S, Baris E, Narayana D, eds. Safeguarding the Health Sector in Times of
Macroeconomic Instability: Policy Lessons for Low- and Middle-Income Coun-
tries, Trenton, NJ, USA: Africa World Press; and Ottawa, Canada: International
Development Research Centre; 2008.
[16] Henshaw SK. Unintended pregnancy and abortion: a public health perspective.
In: Paul M, Lichtenberg E, Borgatta E, Grimes D, Stubblefield P, eds. A
Clinician’s Guide to Medical and Surgical Abortion. New York, NY, USA:
Churchill Livingstone; 1999:11–22.
[17] Garcia Cano P. Indicadores Sociales Departamentales. Bogotá, Colombia: De-
partamento Nacional de Planeación; 2008. No. 37.
[18] Ministerio de la Protección Social. Norma Técnica para la Atención de la
Interrupción Voluntaria del Embarazo. Bogotá, Colombia: Ministerio de la
Protección Social; 2006.
[19] World Health Organization. Safe Abortion: Technical and Policy Guidance for
Health Systems. Geneva, Switzerland: WHO; 2003.
[20] Guerrero Moreno H, 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: Ministerio de la Protección
Social, Universidad Nacional de Colombia; 2007.
[21] Gómez-Sánchez PI, Escandón I, Gaitán-Duarte H. Evaluating post-abortion care
in 13 Colombian hospitals [in Spanish]. Rev Salud Pública 2007;9(2):241–52.

S98 E. Prada et al. /Internation al Journal of Gynecology and Ob stetrics 118, Su pplement 2 (2012) S92– S98
[22] El Tiempo. De los cerca de 350.000 abortos que se calculan al año, 12%
corresponderían a los tres casos especiales. El Tiempo. May 14, 2006. p. 1–7.
[23] Colmenares E. Método abortivo más extendido entre las jóvenes bogotanas es
una droga para la úlcera. El Tiempo. March 19, 2007.
[24] Colmenares E. The abortion pill. La pastilla del aborto bogotano. El Tiempo.
March 20, 2007.
[25] Fernandez MM, Coeytaux F, Ponce de León RG, Harrison DL. Assessing the
global availability of misoprostol. Int J Gynecol Obstet 2009;105(2):180–6.
[26] Prada E, Singh S, Remez L, Villarreal C. Unintended Pregnancy and Unsafe
Abortion: Causes and Consequences. New York, NY, USA: Guttmacher Institute;
2011.