Introduction
Background
Reproductive justice, a framework introduced by Black feminists in the 1990s, emphasizes the rights to have children, to not have children, and to parent in safe, supportive environments [
1,
2]. Women have different reasons for choosing abortion, and their social and economic backgrounds may influence the risk of physical and psychological sequelae after the procedure. The legal and social contexts surrounding induced abortion differ across countries. As of April 2023, 88 countries broadly allow abortion for indications other than saving a woman’s life or health. Abortion rates are not significantly different between countries where abortion is highly restricted and those where it is broadly legal [
3]. The World Health Organization guidelines for abortion care recommend removing regulatory, policy, and programmatic barriers that hinder access to, and the timely provision of, safe abortion care [
4].
In settings where socially indicated abortions are illegal, those seeking abortion services face higher risks of using unsafe methods, lacking standard practice guidelines, and receiving care from unskilled providers. Although generally considered safe, therapeutic abortion and spontaneous miscarriage can lead to various complications [
5]; most of these are minor, such as pain, bleeding, infection, and uterine perforation [
6–
8].
Despite persistent beliefs about negative mental health effects of induced abortion, prior research has shown that severe adverse reactions are rare when the procedure is performed in early pregnancy without legal restrictions [
9]. Serious mental health problems after abortion are thought to be triggered or aggravated by the circumstances that led the woman to seek an abortion rather than by the procedure itself [
9]. Limited evidence is available regarding the prevalence and associated factors of severe physical and psychological sequelae after abortion in settings with legal restrictions on abortions without medical indications. In 2012, the Constitutional Court of the Republic of Korea ruled that the existing anti-abortion laws were constitutional [
10]. Numerous civil requests were subsequently filed to change the laws, and the 2012 decision was reversed in 2019 to decriminalize abortion [
11]. However, to date, there has been no institutionalization of safe and respectful abortion services in the Republic of Korea [
12]. Women have therefore continued to lack legislative protection, facing ongoing legal uncertainty since 2019 [
13]. Such restrictions on access to abortion care may contribute to mental health inequities among women, particularly affecting those of lower socioeconomic status [
14].
Objectives
This study assessed trends and risk factors for severe physical and psychological sequelae after abortion in contemporary women in the Republic of Korea.
Materials and Methods
Study Design and Setting
We conducted a cross-sectional online study of women aged 13 to 64 years in the Republic of Korea, termed the 2022 Sex and Reproductive Health Survey. Our target sample included 1,000 adolescents (13–18 years) and 3,500 adults (19–64 years). Quotas were set by geographic region (17 metropolitan cities and provinces), age group (1-year intervals for adolescents and 10-year intervals for adults), and education level (high school or below vs. university or above), based on resident registration data from May 2022. The sampling frame was a female survey panel of 231,246 individuals provided by a professional survey company. From this panel, 37,896 potential participants were randomly selected via stratified sampling on these characteristics and invited to complete an online questionnaire. In total, 4,651 responded; after excluding unreliable responses (e.g., identical repetitive answers or inconsistencies in age and menstrual status), 4,552 were retained for analysis [
15].
Participants
Among the 4,552 participants (1,019 adolescents and 3,533 adults), we excluded those who did not respond affirmatively to the question about abortion history. This yielded 914 women who reported a history of induced abortion and were eligible to complete a questionnaire about their abortion experience.
Variables and Measurement
The survey employed structured questionnaires to capture the total number of abortions, age at the most recent abortion, primary indication, hardships prior to receiving the procedure, subsequent sequelae, and the year of the most recent abortion. Given Korean Constitutional Court decisions regarding the criminalization (2012) and decriminalization (2019) of induced abortion, we grouped respondents by the year of their most recent abortion into 4 periods: 1980–1992, 1993–2002, 2003–2012, and 2013–2022. Because the legal and social circumstances have not changed substantially since 2019, we combined the 2020–2022 and 2013–2019 cohorts [
13]. Regarding the indication for the respondent’s most recent abortion, the question was, “What was the primary indication of your latest abortion?” with response options: (1) unwanted pregnancy, (2) maternal indication (medically contraindicated pregnancy), (3) fetal indication (fetal malformation), (4) non-supportive environment, (5) forced by partner or parents, (6) pregnancy by rape, and (7) others. For analysis, we grouped indications into 3 categories: “social” (unwanted pregnancy or non-supportive environment), “medical” (maternal or fetal indications), and “forced” (forced by partner/parents or pregnancy by rape). The primary challenge at the time of abortion was assessed with the question, “What was the biggest challenge when you had your latest abortion?” Response options were: (1) limited knowledge of the procedure, (2) limited information about facilities providing the service, (3) fear of breaking the law, (4) expenses associated with the procedure, (5) concerns regarding sequelae, (6) social stigma, (7) rejection by the facility, and (8) others. We coded hardships as “medical” (concerns regarding sequelae) or “non-medical” (all other hardships).
The sequelae of abortion were assessed using the question, “Did you experience any sequelae after the abortion?” Women could choose from the following options: (1) severe physical sequelae, including uterine perforation and pelvic inflammatory disease; (2) mild physical sequelae, including irregular vaginal bleeding, dysmenorrhea, and vaginitis; (3) severe psychological sequelae, including depression and suicidal ideation; (4) mild psychological sequelae, including anxiety, guilt, and depressed mood; and (5) no sequelae. The survey allowed multiple responses, enabling women to report both severe physical and severe psychological sequelae. Demographic and clinical information, including age at the time of abortion, education level, household income, urbanization level of the residential address, and presence of gynecological conditions at the time of the survey, was collected via self-administered questionnaires. Monthly household income was categorized into 10 levels in increments of Korean won (KRW) 1,000,000 (approximately United States dollar [USD] 700), ranging from <KRW 1,000,000 to ≥KRW 9,000,000.
Statistical Methods
We described the characteristics of the women, stratified by the year of their most recent abortion, using summary statistics. We examined temporal trends, the composition of non-medical hardship, and the prevalence of severe physical and psychological sequelae after the most recent abortion. The abortion-related factors of interest included in the models were age at the time of abortion, primary challenge at the time of abortion, indication for abortion, and total number of induced abortions. We fitted logistic regression models including all covariates to assess associations between abortion-related factors and severe sequelae. Covariates included age at the survey, year of abortion, urbanization level of the residential address, education level, average monthly household income (<1,626, 1,626–3,252, 3,253–4,878, or >4,878 USD converted from KRW; the response options were <2, 2–4, 4–6, or ≥6 million KRW), total number of abortions, presence of gynecological disease, primary indication for the most recent abortion, and access-related barriers. For the adjusted odds ratios (aORs) of physical and psychological sequelae, we restricted the analyses to those whose most recent abortion occurred in 2003 or later, given the potential for recall bias. Statistical analyses were performed using SAS ver. 9.4 (SAS Institute Inc.).
Ethics Statement
This study was approved by the Institutional Review Board (IRB) of the Korea Institute for Health and Social Affairs (IRB No: 2022-052) and the IRB of the Korea Disease Control and Prevention Agency (IRB No: 2021-04-02-1C-A).
Results
Participants
The lifetime prevalence of induced abortion was 20.1% (914/4,552). Among these women, the most recent abortion occurred in 1980–1992 for 22%, 1993–2002 for 38%, 2003–2012 for 25%, and 2013–2022 for 14% (
Table 1). About 70% of women whose most recent induced abortion occurred between 1980 and 1992 were aged 60 years or older. Those whose latest abortion took place between 1993 and 2002 were mostly in their 50s, whereas most women whose latest abortion occurred between 2003 and 2012 were in their 40s.
Descriptive Data
The average age at the time of abortion increased over time. Most women lived in urban areas, and more than half did not have a college or university education. The proportion of university graduates was higher in more recent cohorts, reflecting demographic shifts in women’s educational attainment and age at childbirth. The median number of abortions was 1 across all 4 time periods. The most frequent indication for abortion was social, and across all periods, approximately half of the women or more reported a primary challenge in the non-medical category at the time of their most recent abortion.
Outcome Data and Main Results
Over the study period, the overall prevalence was 23.7 per 1,000 women for severe physical sequelae and 73.6 per 1,000 women for severe psychological sequelae. Among women whose most recent abortion occurred in 1980–1992, 1% and 3% reported severe physical and severe psychological sequelae, respectively (
Figure 1). Among those whose most recent abortion occurred in 2013–2022, severe physical sequelae were reported in 5% and severe psychological sequelae in 15%.
After adjustment for individual risk factors, the risk of severe psychological sequelae was higher in 2013–2022, after the Constitutional Court decided that the existing criminal codes were constitutional, than in the prior period (aOR, 2.31; 95% confidence interval [CI], 1.09–4.91) (
Table 2). An age younger than 30 years was associated with a higher risk of psychological sequelae (aOR, 6.75; 95% CI, 1.73–26.4), though estimates for severe physical sequelae were imprecise due to low prevalence (aOR, 4.20; 95% CI, 0.64–27.37). Non-medical hardship barriers at the time of abortion were associated with a lower risk of severe physical (aOR, 0.21; 95% CI, 0.06–0.72) and severe psychological (aOR, 0.45; 95% CI, 0.21–0.95) sequelae. Compared with abortions performed for medical indications, abortions for social indications were associated with a lower risk of severe physical (aOR, 0.36; 95% CI, 0.06–2.11) and severe psychological (aOR, 0.19; 95% CI, 0.07–0.48) sequelae. A higher total number of abortions was associated with an increased risk of severe physical sequelae (aOR, 1.66; 95% CI, 1.13–2.44), whereas the association between total number of abortions and severe psychological sequelae was close to null.
Discussion
Key Results
The lifetime prevalence of induced abortion in this study is comparable to figures reported in the United States (24.7%), Lesotho (35.25%), South Africa (32.13%), Namibia (26.95%), and Zimbabwe (23.06%) [
16,
17]. An increased risk of severe physical and psychological sequelae was observed in a sample of contemporary women in the Republic of Korea. Young age at the time of abortion was a risk factor for severe psychological sequelae, whereas social indications for abortion were associated with lower risk of both severe physical and psychological sequelae. The total number of abortions was a risk factor for severe physical sequelae but not for severe psychological sequelae. Our findings imply that eliminating social hardships and protecting against the negative social implications of abortion are important for preventing severe sequelae following induced abortion.
The frequency and severity of sequelae after induced abortion differ according to the operational definition used for measurement and gestational age [
18–
20]. The estimated complication rates across all health care sources were approximately 2% for medication abortions, 1.3% for first-trimester aspiration abortions, and 1.5% for second-trimester or later abortions. After surgical abortion, the reported prevalence of physical complications ranges from 0.4% to 19.7% [
21–
23]. The prevalence of depression was 22.5% in a sample of Chinese women seeking an abortion in the first trimester [
24]. A longitudinal study in New Zealand showed that 1.5% to 5.5% of overall mental disorders could be attributed to abortions [
25]. A recent review observed that women who were denied abortion access demonstrated greater baseline anxiety, and perceived abortion-related stigma was associated with increased psychological distress [
26]. Although our measurement of severe sequelae was based on self-reported data, the prevalence of physical and psychological sequelae in our study population largely aligns with previous studies. The prevalence of severe sequelae was generally lower than that reported in prior research [
27]. Unlike the Republic of Korea, other countries have legal systems that broadly allow induced abortion on social and economic grounds; thus, the prevalence of severe sequelae reported before 2021 may have been underestimated in our study. Because of the national family planning campaign in the Republic of Korea from the 1960s to the 1980s, abortion was widely accessible despite being generally illegal [
28–
30]. When the national policy shifted to childbirth promotion as fertility declined, the government of the Republic of Korea revived enforcement of the criminal code on abortion in 2012 [
28]. Given this strong national stance on banning abortion, many women seeking abortion may have been forced to undergo the procedure in less safe environments [
31], leading to a higher risk of severe physical and psychological sequelae during this period than in the past.
The inverse association between social indications for abortion and severe psychological sequelae underscores the importance of supporting women’s decisions to minimize the psychological impact of induced abortion. Evidence indicates that induced abortion itself does not cause deleterious psychological effects, particularly when the reason for termination is insufficient support for the pregnancy [
32]. In contrast, women who undergo the procedure against their will, for a medical indication, or under pressure from others are more likely to experience subsequent mental health problems [
33]. A woman whose pregnancy is terminated for social reasons may feel that she has fully exercised her reproductive rights even without legal protections for induced abortion [
34].
Limitations
Given the cross-sectional design, our findings should be interpreted with caution due to several limitations. First, differential recall bias is possible: women who underwent induced abortion several decades ago may not accurately recall their experiences of severe sequelae. However, our subgroup analysis restricted to more recent abortions showed similar results, suggesting that recall bias may have been minimal. Second, because some women may feel guilt or discomfort discussing their experiences of induced abortion, they may have responded inaccurately and therefore been excluded from the analysis. Third, the demographic variables used in our explanatory models may not reflect women’s circumstances at the time of abortion, introducing potential bias in the risk estimates. Prospective studies that collect contemporaneous demographic data, preexisting health conditions, and the quality of medical services provided are needed to confirm our findings. Fourth, outcomes were based on self-reported responses, which limits interpretability. Fifth, combining the years 2013–2022 into a single analytic category without distinguishing the impact of the 2019 Constitutional Court ruling might have underestimated the risk of sequelae. Because the number of events during 2013–2022 was relatively small, model fitting was challenging, especially with multiple covariates.
Interpretation and Generalizability
The findings of this study may have limited the generalizability due to the specific characteristics of our study population. Nevertheless, leveraging a large sample of women in the Republic of Korea, our study provides valuable insights that meaningfully contribute to a field where current knowledge remains limited.
