Chemlal and Russo BMC Women's Health (2019) 19:55

https://doi.org/10.1186/s12905-019-0751-0

RESEARCH ARTICLE Open Access

Why do they take the risk? A systematic

review of the qualitative literature on
informal sector abortions in settings where
abortion is legal
Sonia Chemlal1 and Giuliano Russo2*

Abstract
Background: Restrictive abortion laws are the single most important determinant of unsafe abortion, a major, yet
preventable, global health issue. While reviews have been conducted on the extent of the phenomenon, no study has so
far analysed the evidence of why women turn to informal sector providers when legal alternatives are available. This work
provides a systematic review of the qualitative literature on informal sector abortion in setting where abortion is legal.
Methods: We used the PRISMA guidelines to search Pubmed, Web of Science, Sciencedirect and Google Scholar
databases between January and February 2018. 2794 documents in English and French were screened for eligibility
against pre-determined inclusion and exclusion criteria. Articles investigating women’s reasons for aborting in the
informal sector in settings where abortion is legal were included. In total, sixteen articles were identified as eligible for this
review. Findings were reported following the PRISMA guidelines.
Results: The review highlights the diverse reasons women turn to the informal sector, as abortions outside of legal health
facilities were reported to be a widespread and normalised practice in countries where legal abortion is provided.
Women cited a range of reasons for aborting in the informal sector; these included fear of mistreatment by staff, long
waiting lists, high costs, inability to fulfil regulations, privacy concerns and lack of awareness about the legality of abortion
or where to procure a safe and legal abortion. Not only was unsafe abortion spoken of in terms of medical and physical
safety, but also in terms of social and economic security.
Conclusion: The use of informal sector abortions (ISAs) is a widespread and normalised practice in many countries
despite the liberalisation of abortion laws. Although ISAs are not inherently unsafe, the practice in a setting where it is
illegal will increase the likelihood that women will not be given the necessary information, or that they will be punished.
This study brings to the fore the diverse reasons why women opt to abort outside formal healthcare settings and their
issues with provision of abortion services in legal contexts, providing an evidence base for future research and policies.
Keywords: Unsafe abortion, Legal abortion, Informal sector abortion, women’s rights, Systematic reviews, Qualitative
research

* Correspondence: [email protected]
2
Centre for Primary Care and Public Health, Queen Mary University of
London, Yvonne Carter Building, 58 Turner street, London E1 2AB, UK
Full list of author information is available at the end of the article

© The Author(s). 2019 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0
International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and
reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to
the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver
(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
Chemlal and Russo BMC Women's Health (2019) 19:55 Page 2 of 11

Background sector abortions are not inherently unsafe, self-induced

Unsafe abortion is defined by the World Health Organ- abortion and over the counter abortion drugs can be
isation as any termination of pregnancy that is done ei- problematic if not enough or incorrect information is
ther by a person “lacking the necessary skills or in an provided to women. In Ethiopia, despite the liberalisa-
environment that does not conform to minimum med- tion of the abortion law in 2005, many terminations still
ical standards, or both” [1]. The phenomenon is a sig- take place outside health facilities [13]. In the run up to
nificant public health issue around the world, and a Irish abortion referendum the topic of informal abortion
major contributor to mortality in women of reproductive practices has made the headlines once again, as it
age. Around 8% of all maternal deaths have been attrib- emerged that informal practices and visits to abortion
uted to unsafe abortions [2], and the figure rises to 20% clinics in neighbouring countries have been common for
in countries where abortion is prohibited [3] . Post abor- those women denied safe and legal abortion services in
tion complications of unsafe abortion can be lengthy their own country [14].
and can lead to delays in patient’s recovery, which in- Although the epidemiology of the phenomenon has
creases risks of inappropriate treatment if a lay or formal attracted the interest of academic scholars [15, 16] in the
provider needs to complete that abortion [4]. Evidence past, little is known about the reasons behind women’s
also exists that taking ineffective medication or even effect- decisions to seek informal abortion services in circum-
ive medication that is in the wrong dosage, may not cause stances where such services are legally available through
morbidity but can be ineffective and lead to further delays the formal healthcare sector. This paper sets out to sys-
and expense [5]. When administered in accordance to the tematically review the qualitative evidence on the use of
WHO guidelines, pregnancy termination is one of the saf- informal services in countries where abortion is legal,
est existing medical procedures [6], as complication rates with the aim of understanding women’s motivations be-
have been estimated to occur only in around 1 per 1000 hind such choice, and provide an evidence base for pol-
abortions [7]. Yet, when performed unsafely, the procedure icies to regulate this potentially unsafe practice.
carries a high risk of complications such as haemorrhage,
renal failure, infertility and trauma to the bowel [7]. Methods
Significant differences exist in the way abortion is reg- To distinguish between countries where abortion is legal
ulated across the world [8]. In most cases, unsafe abor- and those where it is not, we adopted the definition by
tions are more common in areas with more restrictive The Center for Reproductive Rights [17] which catego-
abortion legislation, such as Sub Saharan Africa and rises countries and states into four groups reflecting to
Latin America [8]. In Africa, for example, the estimated their abortion law status: (a) Abortion is permitted either
mortality due to unsafe abortion in 2008 was highest in to save a woman’s life or outlawed altogether. Chile and
East and Western Africa at 13,000 and 9700 respectively, El Salvador are the only two countries where abortion is
where abortion laws are most restrictive [1]. Southern completely prohibited; (b) Abortion is permitted when it
Africa, home to some of the countries with the most lib- is necessitated to preserve the health of the pregnant
eral abortion laws such as South Africa, Mozambique woman. Some countries in this category also permit
and Zambia, had the lowest rates at 500 [1]. Abortions abortion on grounds of preserving the mothers’ mental
performed outside of facilities can also have legal conse- health; (c) Abortion is permitted on socioeconomic
quences for women and providers in most countries [9]. grounds, and; (d) Abortion is available on the request of
Although increased access to new drugs and technol- the mother without restriction as to reason. In this dis-
ogy such as misoprostol and mifepristone has made tinction as the categories move up, the grounds on
abortion less risky in the informal sector [10], what is which abortion is permitted become less restrictive.
often overlooked in public discourse is that access to in- Women wishing to terminate and an unwanted preg-
formation is often an issue in low-income settings, and, nancy may wish to do so in a legal, formal health facility
unsafe abortions, in particular those carried out in the or in the informal sector either through self-inducing or
informal sector, continue to burden women, families and visiting one of many informal sector providers, such as
societies in many countries where abortion is not legally traditional birth attendants. The literature shows that
restricted. Despite having one of the most liberal abor- women’s paths to seeking an informal sector abortion is
tion laws in Africa, women in Zambia still take substan- complex: multiple factors, motivations and concerns
tial risks to terminate unwanted pregnancies [11]. In come to play in the decision-making process [18–22].
Kenya and Tanzania policies inspired by past colonial The present review understands that women seeking
abortion laws, still hamper access to services in the pub- informal sector abortions may do so via one of two tra-
lic health care service [12]. In India abortion has been jectories; those who undertake the first trajectory turn to
legal for decades; yet the vast majority of abortions are the informal sector as their first point of call for a num-
conducted in the informal sector [3]; although informal ber of reasons, such as privacy concerns, high costs and
Chemlal and Russo BMC Women's Health (2019) 19:55 Page 3 of 11

lack of awareness of abortion laws (Fig. 1). Women who Abortion (ISA) through generating rich data on the mean-
undergo the second trajectory first attempt to seek a ings that women attach to their abortion seeking experi-
legal abortion, through a formal provider but face ence [24], and qualitative research is increasingly being
barriers and are forced to turn to the informal sector. recognised as having an important contribution to make
Both safe and unsafe outcomes are possible; if the infor- to evidence-based healthcare and in addressing policy re-
mal sector abortion provider carries out the abortion in lated questions. The Cochrane database of systematic re-
accordance to the WHO guidelines [1] then despite the views [25] and the Campbell library of systematic reviews
illegality of the procedure the outcome will be a safe [26] were thoroughly searched prior to the start of this
abortion with minimal medical risk. current review, to ensure that there were no past or
The central question addressed in this review asks: What on-going reviews on the chosen topic of study.
are the reasons women who live in setting where abortion
is legal choose to have an informal sector abortion? The The search strategy
process of this review was guided by the following research The PRISMA guidelines [27] were used to design the
objectives [1]: To identify all the primary, qualitative litera- search strategy extract relevant information from the
ture on the reasons why women who live in settings with papers. The databases Scopus, Google Scholar, Pubmed,
liberal abortion laws opt to have an informal sector abor- Web of Science and Sciencedirect were searched between
tion [2]; To extract, analyse and synthesise the relevant data the 30th of January 2018 and the 1st of February of the
on why women living in countries with liberal abortions same year. These databases were specifically selected for
laws end up having unsafe abortions in the informal sector their multidisciplinary nature. For example, Scopus focus
[3]; To provide an up to date, global review on the reasons covers a range of disciplines such as the social sciences,
women in countries with liberal abortion laws opt to have medicine, public health, humanities and women’s studies,
unsafe abortions in the informal sector, and [4]; To increase all of which concern the subject of unsafe abortion. Mul-
understanding on the barriers to accessing safe, legal abor- tiple databases were included to ensure that all the rele-
tion in countries where abortion is legal. vant articles are captured.
This review focuses on qualitative research as quantita- Liberati et al. [28] recommends reporting the full elec-
tive reviews have already explored [16, 23] the impact of tronic search for at least one database. A table has been
unsafe abortion, but in comparison, no review of the included under Additional file 1 showing the exact
qualitative literature on this issue has been carried out. search terms used for each database and results gener-
Qualitative research can be a useful tool for gaining a deep ated. The following search queries were inputted into
understanding of the phenomena of Informal Sector the majority of databases:

Macro-level: Safe: In accordance

Illegal but does not
The woman herself
with WHO guidelines
result in complications
Legal environment Pharmacists
Street vendors
Role and nature of religion Herbalists Unsafe: Not in Illegal and results in
Traditional birth accordance with WHO
Gender inequality
complications
attendants guidelines
Norms and acceptability of abortion

Access to information
Unsuccessful
Unsuccessful at attempt at inducing
Competency of providers
procuring a abortion
legal abortion

Individual-level:
Safe: In accordance Legal and does not
with WHO guidelines result in
Access to economic resources
Doctor, nurse or complications
midwife trained in
Individual preferences Unsafe: Not in
abortion provision
accordance with WHO
Previous experiences of abortion Legal but results in
guidelines
complications
Social network

Fig. 1 Informal sector abortion: A conceptual framework

Chemlal and Russo BMC Women's Health (2019) 19:55 Page 4 of 11

(“Informal sector abortion” OR “illegal abortion” OR (SC), allowing for the rapid elimination of ineligible arti-
“clandestine abortion” OR “unsafe abortion”) AND cles. Articles that did not give away enough information
(“legal abortion” or “abortion is allowed”) AND through their titles on the relevance of their content
(“factors” OR “reasons” OR “motivations” OR were reserved for step two, where abstracts were
“determinants” OR “motives”) assessed. In the final stage, the full texts of the
remaining records obtained and assessed for eligibility.
Synonyms such as ‘illegal’ and ‘clandestine’ and ‘motiva- Articles whose content did not meet the criteria were
tions’ and ‘reason’ were used to increase search results eliminated and the reasons for their exclusion were doc-
generated. Many terms, such as the phrase ‘Termination umented (Additional file 1). Finally, relevant information
of pregnancy’, were discarded after preliminary searches was extracted and analysed by the two authors.
showed they did not generate any additional useful results.
In some databases where very large numbers of search re- Results
sults were generated during the preliminary searches, add- The initial search yielded a total of 3179 records were
itional limits were placed, such as restricting the search yielded, 921 of which were generated through French
terms to the Titles, Abstracts and Keywords. term searches and 2258 from English search terms. A
405 duplicates were identified and removed through further 20 records were identified by snowballing, that
hand searching. Articles were ordered according to al- is, tracking and chasing down references in footnotes
phabetical order of study titles and then articles whose and bibliographies of the original articles and other re-
titles and dates published matched were excluded. In search documents. 2764 articles were excluded after
order to account for any article titles that may have been screening their titles and abstracts, leaving 30 full text
misspelled, this step was repeated but with the articles articles to be assessed for their eligibility using the pre-
being ordered according to their year of publication. defined criteria. 19 of these articles were excluded and
A pre-defined search criterion is one of the defining the details on which aspect of the criteria they did not
features of a systematic review [29] as it minimises bias fulfil are listed under Additional file 2. A further five
by including articles on the basis of the criteria rather studies were identified through searching the reference
than the authors’ preferences or search results [29]. The lists of relevant articles, resulting in a total of 16 studies
inclusion and exclusion criteria against which the studies being included in the final synthesis (Fig. 2).
generated by the search results were assessed covered
their topic, participants, settings, type of study, language Study characteristics
and publication date (Table 1). No limit was placed on A total of 16 studies, spanning twelve countries, were in-
time of publication as this varied by study depending on cluded in the qualitative synthesis. The full data extraction
when abortion has been legalised. table, complete with key study characteristics such as study
The records generated were examined for relevance in type, methods, purpose and main findings, is listed in
three stages. In the first, titles were assessed against the Additional file. The majority of studies were based in
inclusion and exclusion criteria by one of the authors Sub-Saharan Africa with the exception of Northern Ireland,

Table 1 Eligibility criteria

Criteria Included Excluded
Topic ● Unsafe, informal sector abortion ● Safe abortion
● Unsafe abortions that took place in a
formal setting
Type of participant ● Women or friends or family of women who have undergone an unsafe, ● Women who have not had an unsafe
informal sector abortion informal sector abortion
● Informal sector abortion providers ● Formal setting providers
Settings ● Country where abortion is legal either without restriction as to reason or ● Countries where abortion is only allowed
permitted in order to preserve health or on socioeconomic grounds- according to save a woman’s life or prohibited
to the Centre for Reproductive Rights’ world abortion laws. altogether
Type of publication ● Qualitative or mixed method ● Quantitative only
Type of study ● Original, primary research data ● Secondary data, reviews, opinion pieces
● Published literature and reports.
● Unpublished, grey literature
Language of ● English ● Other than English or French
publication ● French
Publication date ● Any date after legalisation/liberalisation of laws- varies by paper ● Papers published before abortion laws
were liberalised
Chemlal and Russo BMC Women's Health (2019) 19:55 Page 5 of 11

Fig. 2 PRISMA Flow diagram of the study search

Great Britain, Hong Kong and the United States. The specific exception for allowing abortion in order to pre-
publication dates of the included studies ranged from 1998 vent permanent damage to a mothers physical and mental
to 2018, although the majority were published after 2010. health [30], Northern Ireland was considered eligible for
Thirteen of the sixteen studies included recruited this review. The United States is also a unique case
participants who had specified either having undergone or whereby abortion has been legal since 1973 but is regu-
attempted an informal sector abortion either by recruiting lated at state level [31]. Therefore, the extent to which
women who presented at hospitals with post abortion abortion is restricted will vary from state to state [31].
complications or recruiting women via snowball sampling,
surveys and informal sector abortion organisations such as Women’s reasons for choosing to have an informal sector
Women on Web. Three studies focussed on members abortion
of the community who had either were ISA providers The studies surveyed widely confirmed the quantitative
or had friends or family who had experienced an ISA evidence that the practice of ISA is a widespread
(Additional file 2). phenomenon: The majority of the respondents in the
The sixteen studies covered a wide spectrum of abortion studies included reported being aware of women in their
laws, from the most liberal such as South Africa and communities who had undergone the termination proced-
Cambodia where abortion is available on request, to coun- ure clandestinely. Self-induction was found to be the pre-
tries where conditions for a legal abortion are more re- ferred method of terminating a pregnancy among
stricted such as Ethiopia and Kenya. Northern Ireland and respondents as it was perceived to be more natural, less in-
the United States were outliers in this study. Northern vasive and less medicalised [30, 32, 33], and comparable to
Ireland is notorious for having some of the most restrict- taking contraceptive pills or painkillers [34, 35] Among par-
ive abortion laws worldwide where abortion is rarely of- ticipants, there was widespread knowledge of the medical
fered on legal grounds [30]. However, as its laws make a risks associated with ISA. However, any fears over medical
Chemlal and Russo BMC Women's Health (2019) 19:55 Page 6 of 11

safety were outweighed by the reservations that women undergoing abortion are severe. Women in Great Brit-
had about abortions in formal health facilities [20]. ain, Kenya and Zambia reported fear of violence from
Ten key reasons for women opting to have an informal their families, being ostracised from the community and
sector abortion, emerged in the form of themes from the losing their livelihoods if they were dependent on those
literature surveyed, ranging from privacy, attitude of they wished to conceal their abortion from, such as their
healthcare staff, to costs involved and timeliness of the parents or their partner [11, 33, 37].
intervention (Table 2). We review the themes below Costs associated with legal abortion services was an
from the most to the least frequently mentioned. issue that came up in ten of the studies included [11, 19–
Concerns over privacy in legal health facilities were 22, 32, 33, 37, 39, 40] This applied to both countries where
listed as a reason for women turning to informal sector a cost is required for formal sector abortions and in con-
providers in thirteen of the studies [11, 20, 22, 30, 32– texts where abortion is provided free of charge or covered
38]. Women who chose to pursue informal sector abor- by insurance. In countries where a fee is required, legal
tion because of issues of privacy felt that the need to abortion was out of reach for women and girls from
conceal their abortion and protect their social security low-income backgrounds. In many cases women were
outweighed their physical safety needs. Formal health fa- dependent on the income of those who they wanted to
cilities were deemed to be unsafe if they failed to protect conceal their abortion from, such as their parents or part-
a woman’s social reputation. Formal sector abortions ner [11]. This also applied for when abortion is covered by
carried the risks of being seen and kept details of the insurance, as is the case in some US states, where women
women who sought abortion meaning that the women and girls were unable to use their insurance to acquire
could be easily identified as having had an abortion, abortion for fear that those with whom they shared their
whereas self-induction was preferred as it could be car- insurance would find out about their abortion [34]. Incor-
ried out in the privacy of one’s home [21]. The potential rect knowledge regarding the costs of legal termination of
consequences of confidentiality breaches for women abortion was also found to influence women’s abortion

Table 2 Reasons given to seek informal sector abortion in the literature reviewed
Source / Country Privacy Cost Knowledge Social Regulation Fear of Unwilling staff Self- Timeless- Distance
network mistreatment management ness
Koster-Oyekani, 1998 (Zambia) X X X X
Jewkes et al., 2005 X X X X X X X
(South Africa)
Hill et al., 2009 (Ghana) X X X X
Grossman et al., 2010 X X X X X X X
(United States)
Hung, 2010 (Hong Kong) X X
Rominski, Lori and X X X X
Morhe, 2017 (Ghana)
Hegde et al., 2012 X X
(Cambodia)
Marlow et al., 2014 (Kenya) X X X X
Izugbar, Egesa and X X X
Okelo, 2015 (Kenya)
Osur et al., 2015 (Kenya) X
Coast and Murray, X X X X X X
2016 (Zambia)
Gerdts et al., 2017 X X X X X X
(South Africa)
Kebede et al., 2017 X
(Ethiopia)
Aiken et al., 2018 X X X X
(Northern Ireland)
Aiken et al., 2018 X X X X X X
(Great Britain)
Aiken et al., 2018 X X X X X
(United States)
Chemlal and Russo BMC Women's Health (2019) 19:55 Page 7 of 11

seeking behaviour. Even when abortion was freely avail- states many abortion clinics were forced to shut down,
able, such as in the case of South Africa, women held the pushing women to travel longer distances for a legal abor-
perception that informal sector abortions would be tion or turn to the informal sector [32].
cheaper. The perception itself of unaffordability was a bar- Pregnant women’s’ fears of mistreatment by staff - that
rier to accessing safe and legal abortion. they will be judged, criticised, shamed and even possibly
The studies reviewed also showed that a lack of know- exposed - can be a deterrent for seeking a safe and legal
ledge of abortion laws, and a widespread perception that abortion. Fear of mistreatment by staff as a reason for
abortion is not legal, even though all of the studies in- women choosing to avoid having legal abortions in for-
cluded were based in contexts where abortion is permit- mal health facilities was mentioned in seven studies [22,
ted. Eleven of the studies included reported that one of 33, 35–39] In many cases this was due to women’s past
the reasons women opted for an informal sector abor- experiences or past experiences of their friends and fam-
tion was because they were unaware of the legality of ily having been mistreated by hospital staff. Mistreat-
abortion [11, 20, 22, 30, 32, 34–36, 38, 39, 41] In some ment included staff gossiping about their patients, being
cases, it was reported that this lack of knowledge was openly hostile, shaming the women and even accusing
exploited by health care workers who intentionally gave them of murder [35].
the perception that abortion was not legal [39]. Some Staff unwillingness to provide abortion or make a refer-
studies also pointed out that inability or confusion about ral. In many contexts where abortion is legal, providers
how to navigate the health system compounded the gen- may be unwilling to provide abortion for personal, religious
eral lack of knowledge of local abortion laws [42]. and cultural reasons. This was a finding in five of the stud-
A review of the included studies highlighted the import- ies included where staff unwillingness to provide an abor-
ant role that women’s social network played in shaping her tion and failure to make a referral was cited as a reason for
journey to procuring an informal sector abortion. Friends turning to the informal sector [11, 34, 35, 39, 41].
were found to be an important source of information on A preference for self-managed abortions at home was
abortion methods and in some cases even became in- cited as a reason in four studies [30, 32–34]. Women who
volved in the abortion attempt itself [20]. The role of so- had sought to self-induce their abortion described a pref-
cial networks in informing women about, and leading erence for their more private and familiar surroundings of
them to informal sector abortion was documented in their home. The home setting was considered to be less
seven of the included studies, indicating that this was a medicalised, more natural [44] and giving women a
major factor [11, 20, 22, 34, 35, 38, 43]. Knowing a public greater feeling of independence and control over their
sector health worker was seen as a factor in accessing bodies. At home oral abortifacients included misoprostol
abortion services in Kenya [39]. In addition to friends, and mifepristone, the standard for a medical abortion and
family, neighbours, teachers and even strangers whom the less effective and more dangerous methods such as over-
women had just met were also an important source of in- dosing on vitamin c and taking herbal concoctions. The
formation on ISA. In Siaya, Kenya, the effect of social net- use of self-induction was also used. Where women pre-
works on women’s decision to have an informal sector ferred not to seek abortion from legal facilities, for fear of
abortion was found to be greater in younger women [43]. stigma, privacy concerns or fear of mistreatment, the use
Regulation as a barrier to accessing safe, legal abortion of self-induction was used to avoid the social risks and
emerged as a theme in in seven studies [11, 19, 32, 34–37] harms associated with unsafe abortion [30, 32–34].
Women and girls reported the requirement of parental Timeliness of services was mentioned as an additional
consent as a barrier to access, as often it was their parents reason in three of the studies [22, 33, 35] as long waiting
that they were trying to hide the details of their abortion lists for regular abortion services were identified as a key
from. In Zambia the requirement of three doctors to ap- deterrent. Two of these studies were based in South Af-
prove the abortion in non-emergency cases posed issues, rica but published twelve years apart. This suggests that
particularly in rural areas and regions that did not have long waiting lists have been an issue affecting South Af-
enough doctors to meet this criterion. This was the find- rican women’s chances of procuring a safe and legal
ing of two studies; published sixteen years apart, suggest- abortion for many years. In Great Britain, long waiting
ing that this has been an on-going problem for many lists often meant that women were no longer eligible for
years [11, 36]. In Hong Kong, laws that punished men a medical abortion and would be obliged to undergo a
who had sex with minors deterred young girls below the surgical termination [33]. This reason alone was enough
ages of sixteen to seek legal abortion for fear that their to put women to the informal sector.
partners would be prosecuted [19]. In the United States Distance combined with a lack of transport was cited in
where abortion regulation is decided at state level, state three of the reviewed studies [30, 32, 33]. This was also
regulation played a major role in limiting women’s access closely linked to costs as the greater the distance they
to legal abortion [32]. For instance, in more conservative would have to travel for an abortion the greater the costs
Chemlal and Russo BMC Women's Health (2019) 19:55 Page 8 of 11

associated with transport. This was compounded by the 34, 35, 37, 40]. In comparison, more invasive methods
fact that most abortion clinics require two or more ap- such as the insertion of foreign objects [20] through the
pointments to administer the pills and follow up for com- vagina or physical methods such as intensively massaging
plications. For women in Northern Ireland, travelling to the abdomen [37] were less frequently cited.
nearby England for an abortion could pose issues for their
privacy. Some women cited finding it difficult to keep pri- Discussion
vate the true reason for their travels [30]. To the best of our knowledge, this is the first systematic
review to be conducted on the qualitative evidence around
Attributes of informal sector abortions the reasons why women who live in settings where abor-
A key finding of this study is the existence of two path- tion is legal end up having informal sector abortions, using
ways to seeking an informal sector abortion. In the first predominantly unsafe and ineffective methods. ISAs were
category, women first attempted to seek a legal abortion reported in low-income as well as high-income countries;
at a formal health facility and only turned to the infor- long waiting lists, costs, and lack of awareness about the
mal sector when their attempts failed [11, 19, 22, 35]. legality of abortion appeared to be the dominant concern
Women in this trajectory were either not referred by the in the former countries, while privacy concerns and lack
doctor at the clinic, not made aware of the legality of of insurance coverage were quoted in some high-income
abortion or were put off by the price once they hand studies. Among the study participant’s, unsafe abortion
reached the clinic [19, 35]. was not spoken of solely in terms of medical and physical
The second trajectory involves women who directly safety, but also in terms of social and economic security.
sought an informal sector abortion without first trying Abortion facilities that did not protect women’s anonymity
to seek a legal service [20, 34, 38] For women in this were deemed to be unsafe. This was mainly due to the fact
pathway, ISA was regarded as the normal trajectory for that societal attitudes to abortion were largely negative
seeking abortion. Formal sector abortion was the last re- and that severe repercussions, such as loss of livelihood
sort after multiple ineffective clandestine abortion at- and being shunned by their community, would result if it
tempts or for post abortion complication treatments. were discovered that they had procured an abortion [37].
Women were aware of the medical dangers of ISA in all Whilst ISA’s are not unsafe by default, in most of the stud-
of the studies except one [41]. Although all of the women ies included in this review women reported using unsafe
had sought out an informal sector abortion they were well methods such as taking herbal concoctions, using contra-
aware of the medical risks that this form of abortion car- ceptives and painkillers or inserting foreign objects into
ried. All of the women interviewed by Kebede et al. [40] the uterus. Although some studies did report the use of a
said they were aware of the potential complications of in- combination of mifepristone and misoprostol, which when
formal sector abortions. However, fears of physical dan- taken appropriately can be used to safely induce an early
gers were outweighed by fears over their social safety and term abortion, the methods employed to induce an abor-
desperation to terminate their pregnancy [20]. tion in the informal sector were reported to be often un-
One of the broad themes emerging review in this study safe, suggesting that without the provision of the right
is the great stigma that is attached to abortion, in particu- information, ISAs carry an increase risk of harm to health.
lar if the pregnancy is the result of premarital sex. The A few limitations should be considered when inter-
practice of abortion could result in strong repercussions preting the findings from this review. Two studies meet-
from the woman’s community and friends and family [37]. ing our criteria could not be located, despite our best
The repercussions of stigma were especially important for attempts to contact the authors [45, 46]. Language bar-
women who did not have an independent income as the riers were another restriction of our search, as only Eng-
potential loss of support of their family or partner could lish and French articles were included, potentially
lose them their means of livelihood [40]. Many respon- introducing a language bias. A third limitation was the
dents demonstrated feelings of internalised stigma exclusion of grey literature, such as reports and confer-
through the feelings of guilt and shame that they attached ence abstracts, which may have introduced an element
to their abortion experience [37, 40] Respondents felt that of publication bias. Furthermore, we did not search for
they had took part in a deviant or atypical practice [37]. terms such as ‘self-abortion’ and ‘self-managed abortion’,
The widespread use of oral abortifacients was another although we believe these would have been picked up by
key theme that emerged in this review. The use of oral our search for ‘self-induction’ and ‘self-use’. Despite
methods of inducing abortion, such as new drugs like mi- these shortcomings, we believe this review represents a
soprostol. Herbal mixtures, painkillers, hormonal prepara- meaningful contribution to the existing knowledge of
tions and household cleaning products, were the most why women keep taking the risk of undertaking informal
commonly reported method of abortion in the informal abortion practices even when legal options are available.
sector in the majority of the included studies [11, 20, 22, A final but important limitation is the wide
Chemlal and Russo BMC Women's Health (2019) 19:55 Page 9 of 11

heterogeneity in the abortion laws of the countries in- aimed at communities where abortion is legal. A signifi-
cluded in this review, as different countries have differ- cant proportion of unsafe abortions in the informal sector
ent laws and regulations and stipulations regulating the could have been averted if women’s’ social network had
practice, which makes it extremely difficult to categorise been excluded from the decision making process. Educa-
countries on the basis of their abortion laws. Judging tion campaigns could be targeted at education communi-
from its official regulation of the practice, Northern ties on the potential medical dangers of informal sector
Ireland for example, may be considered a country with abortions and legal consequences in many settings where
relatively little abortion restrictions, particularly consid- ISA’s are criminalised and could carry a prison sentence.
ering the exceptions to protect woman’s mental health; The use of medical abortion pills for self-induction, as a
however, women’s access to abortion services is in prac- harm reduction programme is another area of interest
tice severely limited in the country. highlighted in this study. When women are provided with
Health workers’ competence [47] was an important information on recommended dosages and administration
factor pushing women towards the informal sector. Pre- of the drug, the evidence shows that self-induction using
vious experiences of mistreatment, such as shaming and misoprostol can be safe [44]. Our findings regarding
general hostility by health workers or hearing of other women’s preference for self-managed abortions for early
women’s experiences of mistreatment, was a strong de- stage pregnancies are consistent with recent studies on
terrent to seeking a legal abortion. Some health workers women’s views on the acceptability of home managed
were also unwilling to provide abortions or make a refer- abortions [5, 49]. Governments should recognise this is
ral, thereby forcing women to turn to the informal sec- rapidly becoming an option for women seeking abortion.
tor. Studies that have investigated health workers’ However, the issue of cultural acceptability should be a
willingness to provide abortion in a number of countries focus for policy makers when designing healthcare ser-
where abortion is legal show that overwhelmingly health vices; for instance, one study carried out in the United
workers oppose abortion and are unwilling to provide Kingdom in 2010 found that Asian women are more likely
the service. As in the case of a study from South Africa to find self-managed abortions more acceptable than hos-
reporting staff refusing to care for abortion clients, often pital based ones [49].
on religious and moral grounds [48]. Future research into potential interventions should
Women’s knowledge of abortion laws and country spe- also place an emphasis on addressing long waiting lists
cific regulations was a key factor in the selection of ISA. and high costs of services which often force women to
Despite the legality of abortion and the broad grounds choose between having a late term abortion or turning
on which it is permitted, regulations such as the require- to the informal sector. Regulations should be
ment for more than one doctor’s approval in areas where re-evaluated to ensure that they are reasonable in their
there is a shortage of doctors, may restrict access to legal requirements and that they do not jeopardize a woman’s
abortion. This was evident in two studies from Zambia right to anonymity.
published sixteen years apart [11, 36]. Long waiting lists
were cited as a factor by three studies, two of which are Conclusion
from South Africa, suggesting that is a persistent issue Unsafe abortions induced in the informal sector remain
in South Africa. Women’s’ lack of knowledge on the le- a major public health challenge in countries where abor-
gality of abortion, cited in eleven out of sixteen studies, tion is legal. Following the PRISMA guidelines, we con-
appeared to be an important determinant of their ducted a systematic review of the qualitative studies
abortion-seeking decisions. Costs were often mentioned published in English and French investigating the fea-
as a barrier to access in our studies; as recent work tures of the practice and the reasons behind women’s
seems to show that informal abortions can end up being risky choice when safer alternatives are available. We
as expensive as – or even more costly – than similar ser- screened over 2700 publications to identify a total of six-
vices in the public sector [22], it should be noted that teen relevant studies exploring the practice in seven
‘perceived costs’ could often being confused with ‘real low-and middle-income countries. These studies re-
costs’ in information-poor settings. vealed that women’s’ reasons for seeking informal sector
Whilst it is beyond the scope of this review to provide a abortions were diverse in high- and low-income settings,
comprehensive list of recommendations for addressing in- and their abortion seeking trajectories complex. Some
formal sector abortions, it has highlighted areas for poten- themes such as the issue of long waiting lists and regula-
tial intervention. For example, the significance of tions were found to be largely country specific.
women’s’ social networks in influencing their abortion tra- A number of gaps were identified in the literature,
jectories was an important point raised in this review. This which may represent areas for future research on the sub-
suggests that there is potential to address the issue of ISA ject. Firstly, given their importance, further primary quali-
through education and awareness raising campaigns tative studies appear to be needed on the phenomenon of
Chemlal and Russo BMC Women's Health (2019) 19:55 Page 10 of 11

informal sector abortion in more regions of the world Received: 20 July 2018 Accepted: 28 March 2019
where it is recognised that ISA is an issue, such as South
East Asia, South America, North Africa and the Middle
East. Research would also be needed on how to address References
the issues with formal sector abortion services identified 1. WHO. Unsafe abortion. Global and regional estimates of the incidence of
in this study, such as privacy, affordability and timeliness. unsae abortion and associated mortality in 2008. Sixth ed; 2011. p. 1–56.
2. Sedgh G, Bearak J, Singh S, Bankole A, Popinchalk A, Ganatra B, et al.
Future research on ISA should aim to understand the Abortion incidence between 1990 and 2014: global, regional, and
characteristics of women who prefer self-managed abor- subregional levels and trends. Lancet. 2016 Jul;388(10041):258–67.
tions over hospital-based ones. 3. Warriner I, Shah IH. Preventing unsafe abortion and its consequences:
priorities for research and action. New York: Guttmacher Institute; 2006. p.
Despite its limitations, this review shows the import- 1–254.
ance of collating and analysing systematically the infor- 4. Melese T, Habte D, Tsima BM, Mogobe KD, Chabaesele K, Rankgoane G, et
mation available on the drivers of unsafe abortions, and al. High levels of post-Abortion complication in a setting where Abortion
service is not legalized. PLoS One. 2017 Jan;12(1):e0166287.
suggest future research into the safety of ISA’. 5. Kapp N, Blanchard K, Coast E, Ganatra B, Harries J, Footman K, et al.
Developing a forward-looking agenda and methodologies for research of
self-use of medical abortion. Contraception. 2018 Feb;97(2):184–8.
Additional file 6. Mkulichi A. Unsafe Abortion, Its Determinants and Associated Factors: The
Case of Malawi. A Systematic Literature Review. Thesis, Georgia State
Additional file 1: Search strategy for the systematic literature review. University. 2017; 1–37.
Description of the methodology used to search the literature, keywords, 7. Lim LM, Singh K. Termination of pregnancy and unsafe abortion. Best Pract
and steps. (DOCX 23 kb) Res Clin Obstet Gynaecol. Elsevier Ltd. 2014;28(6):859–69.
8. Myers JE, Seif MW. Global perspective of legal abortion - trends analysis and
Additional file 2: Data extraction form. Raw tabulated data and
accessibility. Best Pract Res Clin Obstet Gynaecol. Elsevier Ltd. 2010;24(4):457–66.
information extracted from each of the references retrieved. (DOCX 38 kb)
9. Berer M. Abortion Law and Policy around the world. Health Hum Rights.
2017 Jun;19(1):13–27.
10. Ngo TD, Park MH, Shakur H, Free C. Comparative effectiveness, safety and
Abbreviations
acceptability of medical abortion at home and in a clinic: a systematic
ISA: Informal Sector Abortion; LMIC: Low- and Middle-Income Countries;
review. Bull World Health Organ. 2011 May;89(5):360–70.
PRISMA: Preferred Reporting Items for Systematic Reviews and Meta-
11. Coast E, Murray SF. “These things are dangerous”: understanding induced
Analyses; WHO: World Health Organization
abortion trajectories in urban Zambia. Social Science and Medicine, (153),
2016;(February).
Acknowledgements 12. Brookman-Amissah E, Moyo JB. Abortion Law reform in sub-Saharan Africa:
N/A no turning Back. Reprod Health Matters. 2004;12(sup24):227–34.
13. Gebrehiwot Y, Fetters T, Gebreselassie H, Moore A, Hailemariam M, Dibaba
Funding Y, et al. Changes in morbidity and Abortion Care in Ethiopia after Legal
No funding was received for this research. Reform: National Results from 2008 and 2014. Int Perspect Sex Reprod
Health. 2016;42(3):121–30.
14. Bardon S. Fact check: have more than 170,000 Irish women travelled abroad
Availability of data and materials for an abortion? [internet]. 2018. [cited 2018 Nov 16]. Available from: https://
Data sharing is not applicable to this article as no datasets were generated www.irishtimes.com/news/politics/fact-check-have-more-than-170-000-irish-
or analysed during the current study. women-travelled-abroad-for-an-abortion-1.3481581
15. Aghaei F, Shaghaghi A, Sarbakhsh P. A systematic review of the research
Authors’ contributions evidence on cross-country features of illegal abortions. Heal Promot
SC and GR elaborated the original idea and designed the study. SC designed Perspect. 2017;7(3):117–23.
the methodology for the review. SC drafted the manuscript. GR revised the 16. Gerdts C, Vohra D, Ahern J. Measuring Unsafe Abortion-related mortality: a
manuscript. Both authors revised, read and approved the final manuscript. systematic review of the existing methods. PLoS One. 2013;8(1):1–14.
17. Center For Reproductive Rights. The World’s Abortion Laws 2018 [internet].
2018 [cited 2018 Nov 11]. Available from: http://worldabortionlaws.com/map/
Ethics approval and consent to participate 18. Jewkes RK, Gumede T, Westaway MS, Dickson K, Brown H, Rees H. Why are
N/A women still aborting outside designated facilities in metropolitan South
Africa? BJOG An Int J Obstet Gynaecol. 2005;112(9):1236–42.
Consent for publication 19. Hung S. Access to safe and legal abortion for teenage women from deprived
N/A backgrounds in Hong Kong. Reprod Health Matters. 2010;18(36):102–10.
20. Hill ZE, Tawiah-Agyemang C, Kirkwood B. The context of informal abortions
in rural Ghana. J Women's Health (Larchmt). 2009;18(12):2017–22.
Competing interests 21. Grossman D, Holt K, Pena M, Lara D, Veatch M, Cordova D, et al. Self-
The authors declare that they have no competing interests. induction of abortion among women in the United States. Reprod Health.
2010;16(31):99–107.
22. Gerdts C, Raifman S, Daskilewicz K, Momberg M, Roberts S, Harries J. Women ’
Publisher’s Note s experiences seeking informal sector abortion services in Cape Town, South
Springer Nature remains neutral with regard to jurisdictional claims in Africa: a descriptive study. BMC Womens Health. 2017;17(1):1–10.
published maps and institutional affiliations. 23. Dastgiri S, Yoosefian M, Garjani M, Kalankesh LR. Induced Abortion: a
systematic review and induced Abortion: a systematic review and meta-
Author details analysis. Mater Sociomed. 2017;29(1):58–67.
1
Faculty of Epidemiology and Population Health, London School of Hygiene 24. Pope C, Mays N. Qualitative methods in health research. In: Pope C, Mays N,
and Tropical Medicine, London, UK. 2Centre for Primary Care and Public editors. Qualitative research in health care. Third ed. Oxford: Blackwell
Health, Queen Mary University of London, Yvonne Carter Building, 58 Turner Publishing Ltd; 2007. p. 1–11.
street, London E1 2AB, UK.
Chemlal and Russo BMC Women's Health (2019) 19:55 Page 11 of 11

25. Cochrane Library. Cochrane database of systematic reviews [internet]. [cited 49. Lohr PA, Wade J, Riley L, Fitzgibbon A, Furedi A. Women’s opinions on the
2018 Apr 25]. Available from: https://www.cochranelibrary.com/cdsr/about- home management of early medical abortion in the UK. J Fam Plan Reprod
cdsr. Heal Care. 2010;36(1):21–5.
26. Campbell Collaboration. The Campbell Collaboration Online Library
[Internet]. 2018 [cited 2018 Apr 25]. Available from: https://www.
campbellcollaboration.org/library.html
27. Moher D, Liberati A, Tetzlaff J, Altman DG, Altman D, Antes G, et al.
Preferred reporting items for systematic reviews and meta-analyses: the
PRISMA statement. PLoS Med. 2009;6(7):1–6.
28. Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gøtzsche P, et al. The PRISMA
statement for reporting systematic reviews and meta-analyses of studies
that evaluate health care interventions: Explanation and elaboration. PLoS
Med. 2009;6:e1000100.
29. McCrae N, Blackstock M, Purssell E. Eligibility criteria in systematic reviews: a
methodological review. Int J Nurs Stud. 2015:1269–76.
30. Aiken ARA, Padron E, Broussard K, Johnson D. The impact of Northern
Ireland’s abortion laws on women’s abortion decision-making and
experiences. BMJ Sex Reprod Heal. 2018;0(1):1–7.
31. Guttmacher Institute. An overview of Abortion Laws [internet]. 2018 [cited
2018 Nov 16]. Available from: https://www.guttmacher.org/state-policy/
explore/overview-abortion-laws
32. Aiken ARA, Broussard K, Johnson DM, Padron E. Motivations and
experiences of people seeking medication Abortion online in the United
States. Perspect Sex Reprod Health. 2018;50(4):1–7.
33. Aiken ARA, Guthrie KA, Schellekens M, Trussell J, Gomperts R. Barriers to accessing
abortion services and perspectives on using mifepristone and misoprostol at
home in Great Britain. Contraception. Elsevier Inc. 2018;97(2):177–83.
34. Grossman D, Holt K, Peña M, Lara D, Veatch M, Córdova D, et al. Self-
induction of abortion among women in the United States. Reprod Health
Matters. 2010;18(36):136–46.
35. Jewkes RK, Gumede T, Westaway MS, Dickson K, Brown H. Why are women
still aborting outside designated facilities in metropolitan South Africa?
BJOG. 2005;112(September):1236–42.
36. Koster-oyekan W. Why Resort To Illegal Abortion in Zambia? Findings of a
Community-Based Study in Western Province. Soc Sci Med. 1998;46(10):1303–12.
37. Izugbara CO, Egesa C, Okelo R. “High profile health facilities can add to your
trouble”: women, stigma and un/safe abortion in Kenya. Soc Sci Med
Elsevier Ltd. 2015;141:9–18.
38. Rominski SD, Lori JR, Morhe ESK. “My friend who bought it for me, she has
had an abortion before.” the influence of Ghanaian women’s social
networks in determining the pathway to induced abortion. J Fam Plan
Reprod Heal Care. 2017;43(3):216–21.
39. Marlow HM, Wamugi S, Yegon E, Fetters T, Wanaswa L, Msipa-Ndebele S.
Women’s perceptions about abortion in their communities: perspectives
from western Kenya. Reprod Health Matters. Reproductive Health Matters.
2014;22(43):149–58.
40. Kebede MT, Middelthon AL, Hilden PK. Negotiating the social and medical
dangers of abortion in Addis Ababa: an exploration of young, unmarried
women’s abortion-seeking journeys. Health Care Women Int Taylor &
Francis. 2017;39(2):1–22.
41. Hegde S, Hoban E, Nevill A. Unsafe abortion as a birth control method:
maternal mortality risks among unmarried Cambodian migrant women on
the Thai-Cambodia border. Asia-Pacific J Public Heal. 2012;24(6):989–1001.
42. Assifi AR, Berger B, Tunçalp Ö, Khosla R, Ganatra B. Women’s awareness and
knowledge of Abortion Laws: a systematic review. PLoS One. 2016;11(3):1–14.
43. Osur J, Orago A, Mwanzo I, Bukusi E. Social networks and decision making for
clandestine Unsafe abortions: evidence from Kenya. Afr J Reprod Health. 2015;
19(1):34–43.
44. Hyman A, Blanchard K, Coeytaux F, Grossman D, Teixeira A. Misoprostol in
women’s hands: a harm reduction strategy for unsafe abortion.
Contraception. Elsevier Inc. 2013;87(2):128–30.
45. Likwa RN, Whittaker M. The characteristics of women presenting for abortion and
complications of illegal abortions at the university teaching hospital, Lusaka,
Zambia: an explorative study. Afr J Fertil Sexual Reprod Heal. 1996;1(1):42–9.
46. Kebede S, Jira C, Mariam D. A survey of illegal abortion in Jimma hospital,
south western Ethiopia. Ethiop Med J. 2000;38(1):35–42.
47. Kak N, Burkhalter B, Cooper M. Measuring the competences of healthcare
providers. Qual Assur Proj. 2001;1(1):1–28.
48. Harries J, Stinson K, Orner P. Health care providers’ attitudes towards
termination of pregnancy: a qualitative study in South Africa. BMC Public
Health. 2009;9:1–11.