Reasons why women have induced abortions: a synthesis of findings from 14 countries.

OBJECTIVE: Many reasons inform women's reproductive decision-making. This paper aims to present the reasons women give for obtaining induced abortions in 14 countries. STUDY
DESIGN: We examined nationally representative data from 14 countries collected in official statistics, population-based surveys, and facility-based surveys of abortion patients. In each country, we calculated the percentage distribution of women who have abortions by main reason given for the abortion. We examined these reasons across countries and within countries by women's sociodemographic characteristics (age, marital status, educational attainment, and residence). Where data are available, we also studied the multiple reasons women give for having an abortion.
RESULTS: In most countries, the most frequently cited reasons for having an abortion were socioeconomic concerns or limiting childbearing. With some exceptions, little variation existed in the reasons given by women's sociodemographic characteristics. Data from three countries where multiple reasons could be reported in the survey showed that women often have more than one reason for having an abortion.
CONCLUSION: This study shows that women have abortions for a variety of reasons, and provides a broad picture of the circumstances that inform women's decisions to have abortions. IMPLICATIONS: Future research should examine in greater depth the personal, social, economic, and health factors that inform a woman's decision to have an abortion as these reasons may shed light on the potential consequences that unintended births can have on women's lives.

1. Introduction

A growing body of research has examined the reasons women seek an abortion. Many of these studies are based on convenience samples of women from specific subgroups (i.e. ever-married or students) or women seeking abortions or postabortion care at certain health facilities [1-8]. Thus, findings may not represent all women seeking abortions and may instead reflect women who have access to or require facility-based care. Other studies have examined why women obtain abortions during a particular trimester of pregnancy [9,10]. Women who have abortions after the first trimester, however, may be motivated by different reasons from those seeking first-trimester abortions.

The most recent review of women’s reasons for obtaining abortions focused on developed countries [8]. Although a prior review included developed and developing countries, that study was conducted almost 20 years ago, and the reasons why women have abortions may have changed [7,8]. This paper attempts to fill this gap in the literature by using more recent data to categorize women’s reasons for having abortions. Knowledge of these reasons will provide a broad picture of the circumstances surrounding women’s abortion decision-making and demonstrate some of the potential consequences unintended births can have on women’s lives.

2. Material and methods

2.1. Data sources

We identified data from 14 countries across three types of data sources (Table A.1): official statistics,[1] population-based surveys of reproductive-age women (15–49 years) and facility-based data collected from abortion patients. These countries represent diverse contexts that could inform reasons reported for having an abortion. For each country, we present information on geographic region, income classification, abortion law restrictiveness, total fertility rate, and modern contraceptive prevalence (Table 1).

Table A.1

Description of data sources used to examine reasons why women have abortions by country

Country	Coverage	Type of data	Year of data	Data source
Armenia	Nationally representative	Population-based	2010	Demographic and Health Survey
Azerbaijan	Nationally representative	Population-based	2006	Demographic and Health Survey
Belgium	Nationally representative	Official statistics	2011	Commission Nationale d’Evaluation Interruption de Grossesse
Congo Republic	Nationally representative	Population-based	2011–12	Demographic and Health Survey
Gabon	Nationally representative	Population-based	2012	Demographic and Health Survey
Georgia	Nationally representative	Population-based	2010	Reproductive Health Survey
Ghana	Nationally representative	Population-based	2007	Maternal Health Survey
Jamaica	Nationally representative	Population-based	2002	Reproductive Health Survey
Kyrgyz Republic	Nationally representative	Population-based	2012	Demographic and Health Survey
Nepal	Nationally representative	Population-based	2011	Demographic and Health Survey
Russia	Nationally representative	Population-based	2011	Reproductive Health Survey
Turkey	Nationally representative (ever-married women)	Population-based	2003	Demographic and Health Survey
Sweden	13 urban and rural abortion clinics in middle-third of Sweden	Facility-based	2009	Makenzius et al., Repeat induced abortion - a matter of individual behavior or societal factors? A cross-sectional study among Swedish Women, The European Journal of Contraception and Reproductive Health Care, 2011, 16:369–377
United States	11 abortion providers	Facility-based	2004	National Survey of Women Biggs MA, Gould H, Foster DG. Understanding why women seek abortions in the US. BMC Women’s
United States	30 abortion providers in 21 states	Facility-based	2008–2010	Health. 2013; 13:29.

Table 1

Select country-level geographic, economic, political, and fertility characteristics by country and year of survey

Country and year of survey	Geographic region	Country income classificationa	Circumstances under which abortion is legalb	Total fertility rate (TFR)c	Modern contraceptive prevalence (%)d
Armenia, 2010	Central Asia	Lower-middle income	Without restriction as to reason	1.6	27.2
Azerbaijan, 2006	Central Asia	Lower-middle income	Without restriction as to reason	2.0	14.3
Belgium, 2011	Europe	High income	Without restriction as to reason	1.8	69.1
Congo Republic, 2011–2012	Sub-Saharan Africa	Lower-middle income	To save the woman’s life or prohibited altogether	5.0	20.0
Gabon, 2012	Sub-Saharan Africa	Upper-middle income	To save the woman’s life or prohibited altogether	4.0	19.4
Georgia, 2010	Central Asia	Lower-middle income	Without restriction as to reason	1.8	34.7
Ghana, 2007	Sub-Saharan Africa	Low income	To preserve health	4.3	15.7
Jamaica, 2002	Latin America & Caribbean	Lower-middle income	To preserve health	2.5	66.2
Kyrgyz Republic, 2012	Central Asia	Lower-middle income	Without restriction as to reason	3.2	33.7
Nepal, 2011	South Asia	Low income	Without restriction as to reason	2.5	43.2
Russia, 2011	Central Asia	Upper-middle income	Without restriction as to reason	1.6	55.0
Sweden, 2009	Europe	High income	Without restriction as to reason	1.9	NA
Turkey, 2003a	Central Asia	Lower-middle income	Without restriction as to reason	2.3	42.5
United States, 2004	North America	High income	Without restriction as to reason	2.1	68.1

NA = Not available.

Categories based on the World Bank’s classification of income groups (https://datahelpdesk.worldbank.org/knowledgebase/articles/906519-world-bank-country-and-lending-groups): low-income countries have a gross national income (GNI) per capita less than $1025; lower-middle income countries have GNI per capita between $1026 and $4035; upper-middle income countries have GNI per capita between $4036 and $12,475; and high-income countries are those with a GNI per capita of $12,476 or more. Classifications based on GNI per capita from the year of data collection.

Classifications reflect country abortion laws at the time of data collection. These categories are, from most to least restrictive: to save the woman’s life or prohibited altogether; to preserve health; socioeconomic grounds; and without restriction as to reason. Data are from the Center for Reproductive Rights, “The World’s Abortion Laws”: http://worldabortionlaws.com/map/.

TFR at the time of data collection. Estimates are from the World Bank: http://data.worldbank.org/indicator/SP.DYN.TFRT.IN.

Prevalence at the time of data collection except in the following countries: Belgium (2010), Ghana (2006), Jamaica (2003), and United States (2002). Data compiled from the World Bank: http://databank.worldbank.org/data/reports.aspx?source=2&series=SP.DYN.CONU.ZS&country=.

For 11 countries, we had population-based data collected by the Demographic and Health Surveys (DHS) and Reproductive Health Surveys (RHS), standardized surveys designed to collect comparable data across countries. Both are interviewer-administered and conducted in-person. Women who reported an abortion in the 5-year period before the survey were included in analyses; if more than one abortion was reported, only information about the most recent abortion was included. The primary reason for having an abortion was captured using the following close-ended question: “What was the main reason you decided to have this abortion?”[2] In Turkey, open-ended responses were collected.

One source of US data was from women obtaining abortions at 11 large providers in 2004 using a self-administered survey [12]. The first question, used to determine the main reason for having an abortion, stated: “Please describe briefly why you are choosing to have an abortion now. If you have more than one reason, please list them all, starting with the most important one first.” Subsequent close-ended questions, related to specific reasons, were used to code women’s multiple reasons for having an abortion.[3]

We extracted data from two peer-reviewed articles that included data about reasons from self-administered surveys conducted in abortion facilities in Sweden and the US [13,14]. The Swedish data were collected from abortion-seeking patients in 13 urban and rural clinics across the middle one-third of the country. Multiple responses were allowed, but the main reason for having an abortion was not collected. The US data were collected between 2008 and 2010 from abortion patients at 30 facilities in 21 states across all regions. These data are included alongside the 2004 US data to provide a more recent snapshot of women’s multiple reasons for seeking an abortion. The US and Swedish data sources are not nationally representative.

2.2. Categories of reasons

Each data source categorized reasons for having an abortion. Although the same categories did not exist across all countries, reasons were similarly enough worded to construct overarching categories. In some cases, we collapsed two or more similar categories into a single category. For analyses of the main reason given, we created seven categories: wants to postpone/space childbearing[4], wants no (more) children, socioeconomic concerns, partner-related, too young or parents/others object, risk to maternal health and risk to fetal health. All other reasons were included in the ‘other’ category. The specific reasons comprising each category are listed by country in Table A.2.

Table A.2

Survey-specific reasons for having an abortion that comprise each reported Main Reason category by country and year of survey

Country and year of survey	Main reason category
	Wants to postpone/space childbearing	Wants no (more) children	Socioeconomic concerns	Partner-related reasons	Too young; parent(s) or other(s) object to pregnancy	Risk to maternal health	Risk to fetal health	Other
Armenia, 2010	Spacing next pregnancy	Did not want (any more) children	Socioeconomic reasons	Partner did not want the child	–	Health of mother	Risk of birth defects	Sex selection (wanted a boy), sex selection (wanted a girl), other
Azerbaijan, 2006	Spacing next pregnancy	Did not want (any more) children	Socioeconomic reasons	Partner did not want the child	–	Health of mother	Risk of birth defects	Child’s sex selection, other
Belgium, 2011	No desire for a child at the moment	Woman is too old, family complete	Student, financial problems, professional situation, housing problems	Recently broke up, partner did not want the child, casual relationship, new relationship, not married, couple problems, family/friends have problems with the relationship, divorce	Woman is too young	Health problems of mother, mental health problems	Health problems of unborn child	Rape, incest, political refugee, other
Congo Republic, 2012	Too short birth interval	Too old to have a child, has many children	Lack of money, to keep with schooling, to keep working	Husband/partner does not like to have any more children	Too young to have a child, fear of parents	Health problems	–	Other
Gabon, 2012	Too short birth interval	Too old to have a child, has many children	Lack of money, to keep with schooling, to keep working	Husband/partner does not want to have any more children	Too young to have a child, fear of parents	Health problems	–	Other
Georgia, 2010	Want to postpone childbearing	Want no (more) children	Socioeconomic reasons	Partner objected to pregnancy	–	Pregnancy was life or health threatening	Risk of birth defects	Other
Ghana, 2007	Wanted to delay childbearing, wanted to space childbearing	–	No money to take care of baby, wanted to continue schooling, wanted to continue working	Did not love the father, did not want to stay w/the father, partner did not want child/denied paternity, father of child died	Too young to have child, not ready to be a mother, to avoid shame, afraid of parents, parents insisted	Health of mother	Risk of birth defect	Because of rape, no one to help me look after the child, other
Jamaica, 2002	–	Respondent did not want (any) children	Could not afford to have another child	Partner did not want (any) children, did not have a partner	–	Pregnancy was life threatening	Risk of birth defects	Don’t know, other
Kyrgyz Republic, 2012	Spacing next pregnancy	Respondent did not want (any more) children	Socioeconomic reasons	Partner did not want child, not married	–	Health of mother	Risk of birth defect	Sex selection/wanted boy, sex selection/wanted girl, other
Nepal, 2011	Wanted to delay childbearing, wanted to space child	Did not want (any more) children	No money to take care of baby, wanted to continue schooling, wanted to continue working	Did not love the father, partner did not want child, father of child died	Too young to have a child, not ready to be a mother	Health of mother	Risk of birth defect	Child’s sex, b/c of rape, to avoid shame, no one to look after child, other
Russia, 2011	Birth spacing	She did not want another child	Could not afford another child	Partner opposed to another child	–	Health reasons	–	Other
Turkey, 2003	Just delivered/had little child	Like to limit, has enough children, woman too old	Economic problems (lack of money, unemployment), economic and other problems, she is working/nobody to look after the child	Husband was in the army/was abroad, husband does not want it, problems within the family/planning to divorce, husband too old/disabled, not married	–	Health problem - woman, health problem - woman and fetus	Health problem - fetus, usage of medicine during pregnancy	Does not want the child/too early for a child, problems w/pregnancy (probability of spontaneous abortion), unwanted pregnancy, unintended pregnancy, IUD failure, not able to take care of children (unspecified), other
United States, 2004	Not ready for a(nother) child, timing is wrong	Have completed my childbearing, have other people/children depending on me, children are grown	Can’t afford a baby now; unemployed; can’t afford basic needs of life; can’t leave job to take care of baby; would have to find a new place to live; not enough financial support from partner; partner unemployed; on welfare; would interfere w/education or career plans; I’m a student or planning to study	Don’t want to be single mother, am having relationship problems, husband or partner wants me to have an abortion, partner abusive	Don’t feel mature enough to raise a(nother) child, feel too young, parents want me to have an abortion	Physical problem w/my health	Possible problems affecting the health of the fetus	Was a victim of rape, don’t want people to know I had sex or got pregnant, other

We were unable to create overarching categories of reasons from the peer-reviewed articles. Combining categories would overestimate the percentage of women reporting a singular reason because some women reported more than one reason.

2.3. Measures

For 13 countries, we calculated the percentage distribution of main reasons given for obtaining an abortion.[5] For countries with subgroups of women that had data about reasons for at least 100 women, we presented these data according to women’s age at abortion,[6] marital status, educational attainment, and residence. Lastly, we extracted or calculated the percentage of women citing each reason in the US and Sweden.

2.4. Analyses

We conducted original data analyses using Stata version 14 and applied survey-specific sample weights where appropriate. For countries without publically available data, we extracted the relevant information from published articles and reports [13-17].

3. Results

3.1. Main reasons for abortion

In six of the 13 countries for which we had data on the main reason, the most commonly reported reason for having an abortion was socioeconomic concerns, cited by a plurality of women (ranging from 27% to 40%) (Table 3). In five countries, limiting childbearing was the most frequently reported reason, ranging from 20% in Nepal to 64% in Azerbaijan. In Belgium, the most frequently cited reasons were partner-related (23%) and socioeconomic concerns (23%), and in Kyrgyz Republic, risk to maternal health was commonly reported (44%).

Table 3

Percentage of women citing each reason for seeking an abortion, various countries and years

	Sweden, 2009	United States, 2004	United States, 2008–2010
Wants to postpone childbearing	–	27	–
Wants no (more) children	21	47	–
Not financially prepared	32	56	40
Interferes with future opportunities (education, work)	–	54	20
Partner-related	32	55	31
Too young; parent(s) or other(s) object to pregnancy	1	25	–
Risk to maternal health	8	11	–
Risk to fetal health	1	11	–
Risk to maternal/fetal health	–	–	12
Not the right time for a baby	60	–	36
Need to focus on other children	–	–	29
Not emotionally or mentally prepared	–	–	19
Want a better life for the baby than she could provide	–	–	12
Not independent or mature enough for a baby	–	–	7
Influences from family or friends	–	–	5
Don’t want a baby or place baby for adoption	–	–	4
Other	1	25	1
N	773	1160	954

Note: Women may report more than one reason.

– = Not applicable. Response option not given.

3.2. Main reasons for abortion by women’s sociodemographic characteristics

We examined the three most frequently cited main reasons cited for having an abortion according to women’s sociodemographic characteristics (Figs. 1–4). In all countries, except Nepal, we observed minimal variation by age at abortion (Fig. 1). While similar proportions of Nepalese women cited socioeconomic concerns, younger women frequently reported wanting to postpone/space children and partner-related concerns, and older women commonly reported limiting childbearing and other reasons.

Fig. 1

Most frequently cited main reasons given for having an abortion by age at abortion.

Fig. 4

Most frequently cited main reasons given for having an abortion by residence.

In three of the four countries with data on reasons for abortion by marital status, some variation exists in the most frequently cited main reasons (Fig. 2). Across all three African countries, married women were more likely to cite postponing/spacing children while unmarried women were more likely to report being too young or family objections (Fig. 2).

Fig. 2

Most frequently cited main reasons given for having an abortion by marital status.

The distribution of the most frequently cited main reasons by educational attainment is similar within countries, except in Nepal and Ghana (Fig. 3). In Nepal, less educated women cited partner-related concerns and more educated women reported postponing/spacing childbearing. In Ghana, less educated women cited partner concerns as a key reason whereas more educated women reported being too young or family objections.

Fig. 3

Most frequently cited main reasons given for having an abortion by educational attainment.

Overall, the most frequently cited main reasons for having an abortion did not vary by residence except in Armenia, Nepal and Kyrgyz Republic (Fig. 4). In Armenia, risk to maternal health and postponing/spacing childbearing were commonly reported in urban and rural areas, respectively, and in Nepal, socioeconomic concerns and other reasons were more frequently cited in urban and rural areas, respectively. Urban women in Kyrgyz Republic more commonly reported partner-related concerns while rural women reported limiting childbearing.

3.3. Multiple reasons for abortion

We examined multiple reasons reported by women seeking abortions in Sweden and the US (Table 3). In the Swedish study, women reported seeking abortions because of a desire to postpone childbearing (60%), partner-related concerns (32%), and not being financially prepared (32%).

In the 2004 US data, 89% of women reported more than one reason for having an abortion; slightly more than half reported at least four reasons. The most frequently cited reasons were lack of financial preparedness (56%), partner-related (55%), and interference with future opportunities (54%). These results were comparable with the findings on women’s main reason for having an abortion using data in the same survey (Table 2), in that lack of financial preparedness was also frequently cited as a main reason. While partner-related concerns and interference with future opportunities were frequently reported when multiple reasons were allowed, they were less likely to be cited as the primary reason. The recent US data from 2008–2010, presented in Table 3, show that the top three motivations for seeking an abortion were not being financially prepared (40%), not the right time for a baby (36%), and partner-related concerns (31%).

Table 2

Percentage distribution of women who had an abortion by main reason given for seeking abortion, various countries and years

Country and year of survey	Wants to postpone/space childbearing	Wants no (more) children	Socioeconomic concerns	Partner-related	Too young; parents/others object to pregnancy	Risk to maternal health	Risk to fetal health	Other	Total	N
Armenia, 2010	9.7	47.8	17.9	3.8	–	9.8	2.8	8.3	100	456
Azerbaijan, 2006	3.6	64.1	10.1	9.6	–	7.0	1.1	4.5	100	1368
Belgium, 2011	18.4	13.3	22.9	23.0	11.1	2.9	1.1	7.3	100	29,431
Congo Republic, 2011–2012	22.6	5.2	30.4	15.2	12.3	8.1	–	6.3	100	1231
Gabon, 2012	12.6	6.3	39.5	15.4	9.8	9.8	–	6.6	100	847
Georgia, 2010	18.1	51.1	20.2	1.5	–	7.8	–	1.4	100	2054
Ghana, 2007	21.8	–	39.8	15.2	12.8	4.4	1.1	5.0	100	558
Jamaica, 2002	–	11.2	30.0	6.5	–	23.4	2.5	26.4	100	102
Kyrgyz Republic, 2012	14.6	15.9	10.7	9.7	–	44.4	1.8	2.9	100	423
Nepal, 2011	19.0	19.7	13.6	13.4	2.2	9.6	2.9	19.6	100	506
Russia, 2011	10.0	24.0	33.0	17.0	–	11.0	–	5.0	100	a
Turkey, 2003b	17.3	30.1	24.3	4.5	–	7.4	7.2	9.1	100	1689
United States, 2004	24.7	19.1	27.2	8.2	7.6	3.8	2.7	6.8	100	957

– = Not applicable. Response option not given.

Number of women is not available in data source.

Based on sample of ever-married women.

4. Discussion

This study highlights the many personal, social, economic and health factors that inform a woman’s decision to have an abortion. Socioeconomic concerns or limiting childbearing were the most frequently cited reasons in most of the countries in our study. Few generalizations can be made about these reasons by women’s sociodemographic characteristics, partly because only a handful of countries had sufficient data to examine patterns. Our study echoes prior findings that information on only the main reason may not fully demonstrate the circumstances that surround women’s abortion decision-making [12-14].

In many of the countries where socioeconomic concerns or limiting childbearing were most frequently cited, a substantial proportion of abortions occurred among young, unmarried women [18,19]. These women may obtain abortions because they lack the financial means to raise a child or feel that having a child would interfere with future opportunities. In the remaining countries, socioeconomic concerns was still ranked highly as a reason for abortion, emphasizing the widespread influence of financial circumstances in women’s reproductive decision-making. In the five countries where limiting childbearing was most frequently cited, prior findings suggest that women begin childbearing early and reach their desired family size at relatively young ages [17,20,21]. As a result, women may spend more years trying to avoid pregnancy while still fecund, and be at higher risk for unintended pregnancies.

Only three of the included data sources collected information on multiple reasons. Findings from the US, where data existed on both primary and multiple reasons, suggest that many factors influence a woman’s decision to have an abortion and collecting data on only one reason may overlook the web of circumstances that contribute to the decision-making process. For example, when asked for their main reason, women may choose to focus on either the distal (financial concerns) or proximal (delay childbearing) drivers for obtaining their abortion. Focusing solely on women’s primary reason could result in misleading or simplistic conclusions about their abortion decision-making. Future surveys should ask women to report all or a certain number of reasons (in rank order) or incorporate qualitative approaches to help illuminate women’s underlying reasons for having an abortion.

This study contains several limitations. In 11 of the 14 countries, the timeframe for abortions was 5 years prior to the survey, while marital status, educational attainment, and residence were measured at the time of the survey. Thus, women’s characteristics at the time of their abortion may differ from those at the time of the survey; this likely affects younger women disproportionately. Moreover, recall bias may affect women’s responses, especially for those who obtained an abortion further in the past. Future research should consider using panel data to overcome this limitation.

Differential underreporting of abortions in population- based surveys, especially where abortion is highly restricted [22], may bias findings of women’s reasons for obtaining abortions. Women who report their abortion might have different reasons for seeking one, compared to those who do not report them. The fact that very few unmarried women reported abortions in Asia suggests that differential underreporting was present in this region.

Our study analyzes data from an array of survey instruments with varying data collection procedures and interviewers, which may have affected the quality of responses. Furthermore, population-level data were collected using face-to-face interviews, which could have prompted socially desirable responses to abortion-related questions. Circumstances surrounding the interview, such as how the questions were asked, respondent’s comfort level with the interviewer, and location of the interview, as well as the woman’s stage of life (e.g., marital status) could affect responses to these questions. Additionally, the social acceptability of particular reasons may vary across countries, influencing which reasons women are more and less likely to report.

Lastly, our study is a descriptive comparison of the reasons why women have abortions. The categories of reasons are necessarily simplified at two levels — country questionnaire and analysis. Rather than fully reflect the reasons why women have abortions, our findings may reflect the structured response categories listed in the questionnaires and our classification of these categories.

Despite these limitations, the evidence suggests that women have abortions for a variety of reasons, and these appear to vary across countries and, sometimes, within countries by women’s sociodemographic characteristics. Cross-country differences may be indicative of varied cultural contexts; access to quality and affordable health care; desired number, timing and spacing of births and levels of social stigma and support for childbearing. These findings provide a broad picture of the circumstances that may influence women’s decisions to have an abortion and the potential consequences of unwanted childbearing.