The effect of complex posttraumatic stress and poverty on quality of life among adult survivors of childhood institutional maltreatment: evidence from survivors of the ‘Hyeongje Welfare Institution’ in South Korea

ABSTRACT

Background: Childhood institutional maltreatment (IM) is associated with both complex posttraumatic stress disorder (CTPSD) and poverty in adulthood life, each of which may impact an individual’s quality of life (QoL). To find implications for clinical practice and policy making for adult survivors with childhood IM experiences, it is necessary to conduct research examining their current QoL and identifying related factors.

Objective: By applying the model of the conservation of resources theory, we focused on how adulthood QoL can be indicated by childhood IM as well as the life outcomes of IM such as additional lifetime trauma, CPTSD, and poverty.

Methods: In a cross-sectional study, self-report data were collected from 127 adults who were survivors of the ‘Hyeongje’ childhood IM in South Korea. We conducted regression analyses of childhood IM experiences, trauma experiences after escape from the institution, current CPTSD symptoms, and current poverty experiences on current QoL.

Results: The duration of placement at the ‘Hyeongje’ (β = .24, p = .009) was associated with trauma experiences after escape from the institution. Trauma experiences after escape from the institution (β = .25, p = .007) were associated with CPTSD symptoms. CPTSD symptoms (β = .26, p = .005) were associated with poverty, and both CPTSD symptoms (β = −.52, p < .001) and poverty (β = −.26, p = .003) were negatively associated with current QoL.

Conclusions: Prolonged childhood IM brings about loss spirals by increasing an individual’s exposure to experiences of further cumulative trauma, CPTSD, and poverty. There is a need for due diligence-based policy making and public support from the government to help create upward spirals for QoL. This may include the imminent detection and rescue of children as well as providing a safe environment, offering multidisciplinary interventions including evidence-based treatment for CPTSD, and considering economic support including collective reparations.

HIGHLIGHTS

• Duration of placement at the ‘Hyeongje’ institution was associated with additional trauma experience after escaping the institution.

• Cumulative trauma after escape was associated with CPTSD symptoms; CPTSD symptoms were associated with poverty; and both CPTSD symptoms and poverty were associated with poor current QoL.

• Multidisciplinary interventions including evidence-based treatment for CPTSD and considering economic resources for childhood IM survivors would be crucial in increasing QoL.

Antecedentes: El maltrato institucional (MI) infantil se asocia tanto con el trastorno de estrés postraumático complejo (TEPTC) como con la pobreza en la vida adulta, de los cuales, ambos pueden afectar la calidad de vida (CdV) de un individuo. Para encontrar implicancias en la práctica clínica y la formulación de políticas para los adultos sobrevivientes con experiencias de MI en la infancia, es necesario realizar investigaciones que examinen su calidad de vida actual e identifiquen factores relacionados.

Objetivo: Al aplicar el modelo de la ‘teoría de la conservación de recursos,’ nos centramos en cómo la CdV adulta puede estar marcada por el antecedente de MI infantil, así como también los resultados de vida del MI, tales como el trauma adicional durante la vida, el TEPTC y la pobreza.

Métodos: En un estudio transversal, se recopilaron datos autoinformados de 127 adultos sobrevivientes de MI infantil en ‘Hyeongje’ de Corea del Sur. Realizamos análisis de regresión de las experiencias de MI en la infancia, experiencias traumáticas después de escapar de la institución, síntomas actuales de TEPTC y experiencias actuales de pobreza sobre la calidad de vida actual.

Resultados: La duración de la permanencia en ‘Hyeongje’ (β = .24, p = .009) se asoció con experiencias traumáticas posteriores a escapar de la institución. Las experiencias traumáticas después de escapar de la institución (β = .25, p = .007) se asociaron con síntomas de TEPTC. Los síntomas de TEPTC (β = .26, p = .005) se asociaron con la pobreza, y tanto los síntomas de TEPTC (β = −.52, p < .001) como la pobreza (β = −.26, p = .003) se asociaron negativamente con la CdV actual.

Conclusiones: El MI prolongado en la infancia provoca ‘espirales de pérdidas’ al aumentar la exposición del individuo a experiencias de trauma acumulativo adicionales, TEPTC y pobreza. Es necesario formular políticas basadas en una debida diligencia y contar con apoyo público del gobierno para ayudar a crear ‘espirales ascendentes’ para la CdV. Esto puede incluir la inminente detección y rescate de los niños, así como proporcionar un entorno seguro, ofrecer intervenciones multidisciplinarias que incluyan un tratamiento basado en evidencia para el TEPTC y considerar apoyo económico que incluya reparaciones colectivas.

KEYWORDS:

• Institutional maltreatment
• childhood abuse
• complex posttraumatic stress disorder
• poverty
• quality of life

PALABRAS CLAVE:

• Maltrato institucional
• abuso infantil
• trastorno de estrés postraumático complejo
• pobreza
• calidad de vida

1. Introduction

Institutional maltreatment (IM) may occur in childcare welfare institutions where prolonged abuse and neglect take place under a system of power. Childhood IM harms both the physical and mental health of a child and hinders positive development, eventually having a pervasive negative influence on adulthood life and the self (Knefel & Lueger-Schuster, 2013; Lueger-Schuster et al., 2018; Weindl & Lueger-Schuster, 2018). While previous studies have reported on limitations of and challenges in institutionalized care (Sherr et al., 2017), childhood IM is underrecognized worldwide (Carr et al., 2019).

The most known incident of IM in South Korean history is the pattern of human rights violations that occurred at ‘Hyeongje Social Welfare Institution’ (the ‘Hyeongje’); ‘Hyeongje’ means ‘brother’ in Korean. The ‘Hyeongje’ was operated in collusion with the government under the Ministry of Interior directive no. 410 from 1975 till 1987, during which more than 3800 adults and children were abducted from the streets and held captive in the institution; the victims were forced to labour in devastating environments and held captive while being subjected to numerous types of systematic physical, emotional, and sexual violence, including witnessing unknown numbers of deaths (I. Kim, 2019; J. Kim, 2015). Internationally, Wright (2017, 2023) documented 83 inquiries of historical abuse of children in institutions across 20 countries, reporting that past policies and practices of child welfare have failed to protect children and that a wide range of abuses have occurred. In South Korea, the ‘Hyeongje’ IM was first revealed in 1987, however, the inquiry was prevented at the time, likely due to the context of the non-democratic regime; in 2012, a survivor of the ‘Hyeongje’ made a public testimony, which was followed by numerous years of survivor advocacy that ultimately culminated in the South Korean government establishing an inquiry in 2020. The inquiry from the Truth and Reconciliation Commission of Republic of Korea (TRC Korea) found 191 people in August 2022 and an additional 146 people in June 2023 who were victims of human rights violation of the ‘Hyeongje’ IM (TRC Korea, 2022, 2023). In this study, we focused on their current adulthood quality of life (QoL).

The World Health Organization defines QoL as ‘an individual’s perception of their position in life in the context of the culture and value systems in which they live and in relation to their goals, expectations, standards and concerns.’ Previous research has shown that children in institutions had poor QoL (Gander et al., 2019; Van Damme-Ostapowicz et al., 2007). It should be noted that institutional upbringing itself is not a universal cause of detrimental effects on QoL; for instance, no difference in QoL was found among people from institutional care and age-matched peers in a pilot study in Russia (Chumakova et al., 2020). It is widely known that childhood adversities impact QoL; adolescents living in institutions have shown lower QoL than adolescents from the general population, which was attributed to their higher experiences of adversity (Greger et al., 2016). Childhood maltreatment and abuse have also been associated with lower QoL among both children and adult survivors (Affifi et al., 2007; Hoefnagels et al. 2020; Weber et al., 2016). Empirical studies have shown that childhood IM predicts later adulthood physical, mental, and psychosocial problems (Carr et al., 2019; Carr et al., 2020). However, it is important to note that a deficiency framework focusing solely on adversities may not fully explain QoL in later life.

The factors leading to poor QoL in the later life course of the survivors who experienced childhood IM can be well understood by applying the model of the conservation of resources theory (Hobfoll, 1988; Hobfoll & Schumm, 2002). According to Hobfoll and colleagues, ‘resources’ is defined as the interrelations of material resources and/or status, socially valued conditions, and characteristics that aid stress resistance and/or the acquisition of valued resources. It is such resources that determine individuals’ appraisals and coping capabilities in a situation. This model explains how limitations in life chances, inequality, and loss of personal and societal resources lead to cumulative ‘loss spirals’ that affect an individual's coping and life outcomes. The term ‘loss spiral’ emphasizes the force of resource loss, as an initial lack of resources increases one’s vulnerability to ongoing challenges, such as lacking opportunities for future resource gain, and diminishes one’s coping capabilities, thus contributing to subsequent losses in life (Hobfoll & Schumm, 2002).

In line with this theory, children who have been in institutions may develop resiliency when appropriate resource investments are made; a randomized controlled trial showed that early placement in foster family care improved the cognitive development of young children in institutions (Nelson et al., 2007). According to their follow-up data till adolescence, foster care had brought growth in self-regulation (Mukerji et al., 2021) and reduced risk for transdiagnostic psychopathology symptoms (Wade et al., 2018). Based on these discussions, the present work focused on how QoL can be indicated by IM and further resource losses in the life outcomes associated with IM, such as additional lifetime trauma after escaping the ‘Hyeongje,’ trauma-related symptoms, and poverty, with the aim of guiding resource investment for promoting recovery.

Previous studies have reported that childhood IM is a risk factor for trauma in later life (Lueger-Schuster et al., 2018; Thoma et al., 2021). The cumulative impact of IM and subsequent additional traumatic experiences may worsen mental health problems such as complex posttraumatic stress disorder (CPTSD), and eventually result in low QoL. In the present study, we focused on CPTSD, which is a diagnosis associated with prolonged childhood abuse, consisting of posttraumatic stress disorder (PTSD) and disturbances in self-organization (DSO) symptoms. PTSD include sub-symptoms of reexperiencing, avoidance, and heightened sense of current threat; and DSO include sub-symptoms of affect dysregulation, negative self-concept, and disturbed relationships (WHO, 2018). Cumulative maltreatment in children and adolescents living in institutions has been shown to predict a higher severity of CPTSD (Sölva et al., 2020), and adult survivors of IM showed high rates of CPTSD (Knefel & Lueger-Schuster, 2013). CPTSD is also known to be an important factor leading to poor QoL; Bondjers et al. (2019) reported that there is a high association between CPTSD and QoL, while a North Korean refugee study showed that a CPTSD class (high occurrence of both PTSD and DSO) had lower QoL than a PTSD only class (Baek et al., 2022). Previous studies on PTSD also have found this to be associated with low QoL (Araujo et al., 2014; Dhungana et al., 2021). Longitudinal studies have demonstrated that attenuation in PTSD symptoms is related to improved QoL (Giacco et al., 2013; Schnurr et al., 2006) and that PTSD predicted future QoL (Johansen et al., 2007).

Poverty was an important life consequence of IM (Thoma et al., 2021), which is another critical factor in predicting low QoL (Camfield & Skevington, 2008; Chan et al., 2021). Individuals with childhood maltreatment histories have been reported to have a high possibility of poor economic life outcomes in adulthood (Bunting et al., 2018; Mersky & Topitzes, 2010). A longitudinal study found that individuals who experienced childhood abuse showed lower employment, income, and assets compared to those with no such experience (Currie & Spatz Widom, 2010).

The association between poverty and mental health has been empirically supported in prior studies. Income has shown a stronger correlation with mental health problems than physical health problems (Mangalore et al., 2007). General mental health problems have been shown to increase 1.5–3 times in people with the lowest income level compared to the corresponding rates among people with high income (Lund et al., 2010). Poverty and mental health problems have also been shown to be closely related to each other in a 10-year longitudinal study in Korea (Hong et al., 2011). According to Conger and Donnellan (2007), the relationship between poverty and mental health can be explained by different perspectives, such as the social causation, social selection, and interaction model, and the interaction model has been supported by recent research. A study examining the temporal relationship between poverty and PTSD found associations in both directions (Holmes et al., 2022). Longitudinal studies support that children in poverty are at greater risk to be exposed to trauma and that exposure to trauma results in lower income, job loss, and decreased occupational skills (Bunting et al., 2018).

With the childhood IM survivors of the ‘Hyeongje’ as participants, we searched for the variables that are associated with poor QoL in current adulthood. We hypothesized that childhood IM experiences, including age at entrance, duration of placement, and number of multiple experiences of maltreatment, would be positively associated with the number of additional lifetime trauma experiences after escaping from the ‘Hyeongje’ and that these cumulative traumata would be associated with current CPTSD symptoms. We also hypothesized that cumulative trauma and CPTSD would be associated with current poverty. Lastly, we hypothesized that current CPTSD and poverty would be associated with current QoL. Figure 1 (supplementary material) depicts an exploratory model of our proposed hypothesis.

2. Method

2.1. Participants and procedure

This study was part of an extensive multidisciplinary research project on the Hyeongje Social Welfare Institution conducted by the Busan local government, as Busan is where the institution was located. This project was approved by the Dong-A University Institutional Review Board. Data was collected between December 2019 and February 2020. The survivors’ organization advertised the research, and 203 people initially volunteered. Among them, 149 agreed to participate, provided an informed consent and completed participation, and they received monetary rewards for participation. Inclusion for this study comprised having been captive when under the age of 18, and 127 people were eligible in total. An a priori power analysis by G*Power 3.1.9.6. assuming .95 revealed that an approximate sample size would be 129. The other 22 people were excluded from the current study because they were in the institution as adults, which is an exclusion criterion in this study, and data of all 149 people had been analysed in another unpublished report of the Busan City (Busan City, 2020).

2.2. Measures

2.2.1. The World Health Organization Quality of Life–Brief Version (WHOQOL-BREF)

To measure the current quality of life, the Korean version of the WHOQOL-BREF (Min et al., 2002) was used. In this tool, 26 items measure QoL along 4 domains, namely, physical health, psychological health, social relationships, and environmental domains. The Cronbach’s alpha values for each domain were .85, .89, .67, and .86, respectively, with an overall score of .95. For further analyses, we converted raw domain scores to a 4–20 scale (Whoqol Group, 1998).

2.2.2. Multiple experiences of childhood institutional maltreatment at ‘Hyeongje’

Twenty-four items were used to measure childhood IM at the institution. The research group that had been providing psychosocial support for the ‘Hyeongje’ IM survivors drew the items from the testimonies of the survivors who agreed to describe their experiences. Items were constructed based on four types of maltreatment: physical abuse (e.g. being beaten, forced body posture, forced labour), emotional abuse (e.g. death threats, mocking, disconnection from relationships), sexual abuse (e.g. verbal sexual insult, harassment, rape), and deprivation of basic rights (e.g. sleep deprivation, not receiving proper medical treatment, not having proper food). The total score was calculated by summing the number of items that the participants responded to as having directly experienced.

2.2.3. Life Events Checklist for Diagnostic and Statistical Manual of Mental Disorders–5 (LEC-5) Korean

LEC-5 Korean (Bae et al., 2008) is a self-report questionnaire that is used to assess traumatic experiences based on the Diagnostic and Statistical Manual of Mental Disorders (DSM-5). We used this to evaluate direct trauma experiences that occurred after the individuals left the institution. The total number of experiences was used for the analysis.

2.2.4. International Trauma Questionnaire–Korean (ITQ-K)

The International Trauma Questionnaire-Korean version (ITQ-K, H. Choi et al., 2021) was used to measure CPTSD symptoms. It consists of six items measuring PTSD (two items for each of three sub-symptoms: reexperiencing, avoidance, and heightened sense of current threat) with an additional three items to measure related functioning levels, as well as six items to measure DSO (two items for each of three sub-symptoms: affect dysregulation, negative self-concept, and disturbed relationships) along with an additional three items to measure related functioning levels. It is based on the CPTSD diagnosis of the International Classification of Diseases 11th version (ICD-11) (Cloitre et al., 2018). All items are responded to along a five-point Likert scale. The sum of 12 items measuring PTSD and DSO symptoms, which was intended to capture CPTSD symptoms, was used for analysis. Cronbach’s alpha in the current study was α = .84 for the PTSD items, α = .89 for the DSO items, and α = .90 for the total CPTSD items.

2.2.5. Poverty

To measure current poverty among survivors, nine items asked yes-or-no questions about the respondent’s past one-year experience. Items included delaying rent payment for housing, experiencing a forced eviction due to rent delay, not being able to pay utility bills, not being able to pay public utility fees, not being able to pay children’s education expenses, not being able to afford heating, having to give up medical treatment, having family members with credit delinquency, and having to suspend insurance benefits due to unpaid health insurance. The total score was used in the analysis. The Cronbach’s α for this study was .73.

2.3. Data analysis

Data was analysed using SPSS 28.0. First, we conducted a descriptive analysis of maltreatment experiences of the ‘Hyeongje’ and trauma experiences after escape. We then calculated the prevalence of a probable diagnosis of PTSD and CPTSD. According to the diagnostic algorithm designed by Cloitre et al. (2018), if any of the two items from each of the three sub-symptoms of PTSD in ITQ are responded to as ≥2, a probable diagnosis of PTSD can be made. If any of the two items from each of the three sub-symptoms of DSO in ITQ are responded to as ≥2 in addition to a probable PTSD, the probable diagnosis of CPTSD can be made. If an individual is diagnosed with CPTSD, a PTSD diagnosis cannot be made. After the descriptive analysis of the variables, a correlation analysis was conducted to examine the relationships between the variables followed by regression analyses using Bonferroni adjustment to examine the effects among the variables. The variables included age at entrance, duration of placement, total number of multiple IM experienced, total number of trauma experiences after escape, CPTSD symptoms, poverty, and QoL. While it may be necessary to examine the effect of each subtype of abuse (e.g. Hinchey et al., 2023), we decided to use the total number of multiple IM experienced rather than each subtype, because the majority of the participants (n = 112, 88.19%) reported having experienced all four types of maltreatment. For the proposed exploratory model in Figure 1 (supplementary material), we additionally analysed the indirect effects of each pathway using the process macro. The limited sample size indicates low power, so the results of the indirect effects analyses are tentative and therefore presented in the supplementary material.

There was a total of 74 (0.6%) missing data. Twenty-one (2.4%) computed variables were missing, mainly from the WHO-BREF score (1.2%). We used listwise deletion in each regression analysis (Graham, 2009), and the post hoc power analysis of the final regression analysis (n = 110) by G*Power 3.1.9.6. showed results of 0.86.

3. Results

3.1. Descriptive statistics

The majority were men (n = 121, 95.3%), a few were women (n = 6, 4.7%), and no participants reported their gender as transgender or gender diverse. The mean age of participants was 54.54 years old (SD = 4.75, ranging from 45 to 65 years old). In terms current of education levels, 69 participants had an elementary school education (54.3%), 21 graduated middle school (16.5%), 24 graduated high school (18.9%), and 12 (9.4%) had a college-level education or above. Fifty-two participants (41.0%) reported presently being in a marital relationship and 72 (56.7%) did not.

Regarding the social circumstances of the participants before their captivity in the ‘Hyeongje,’ which is presented in Table 1 (supplementary material), the majority were living with their parents (n = 70, 55.1%) while some were living on the street (n = 17, 13.4%), living with relatives (n = 16, 12.6%), or living in other institutions (n = 13, 10.2%). The mean age of respondents upon entering the institution was 11.46 years (SD = 3.17, ranging from 2 to 18 years). The mean duration of placement was 1371.78 days (SD = 1080.58, ranging from 20 to 5294 days). Of the participants, 34 (26.8%) reported receiving education while placed in the ‘Hyeongje’ whereas the rest had no educational experience at all.

Regarding post-social circumstances after escape from the institution, 49 participants (38.6%) were initially reunited with their family; however, due to various relational and financial issues, only 18 participants (14.2%) were able to stay with their family. Forty-six participants (36.2%) were placed in another institution. The majority of the participants did not have access to continuing education (n = 96, 75.6%). Nearly half of the participants experienced detention at least once in their lifetimes after escape (n = 64, 50.4%).

Table 2 (supplementary material) presents information on individual experiences of childhood IM including physical abuse, emotional abuse, deprivation of basic rights, and sexual abuse. Table 3 (supplementary material) shows information on lifetime trauma experiences after escape from the institution. In total, 98 individuals had traumatic experiences after IM escape (77.2%). There were 75 individuals (59.1%) who experienced at least one incident of poverty. The number and percentage of individuals who responded to each poverty item are presented in Table 4 (supplementary material).

Regarding current prevalence rates, 14 individuals (11.0%) had a probable PTSD diagnosis and 37 individuals (29.1%) had a probable CPTSD diagnosis. The mean total score for current QoL was 40.33 (SD = 12.92, ranging from 16.00 to 76.86).

Table 1 presents the correlations among variables. Younger age at entrance was correlated with the number of multiple childhood IM, CPTSD symptoms, and poverty. Duration of placement was correlated with the number of multiple childhood IM, trauma experiences after escape, and CPTSD symptoms. The number of multiple childhood IM was correlated with trauma experiences after escape, CPTSD, and poverty. Age at entrance, duration of placement, or multiple childhood IM did not show significant correlations with the QoL score. Meanwhile, the number of trauma experiences after escape from the institution, CPTSD symptoms, and poverty had negative correlations with the QoL score.

Table 1. Mean, standard deviation, and correlation coefficients with 95% confidence intervals between variables.

	1	2	3	4	5	6	7
1	–
2	−.324*** (−.472, −.159)	–
3	−.367*** (−.509, −.206)	.312*** (.146, .461)	–
4	−.175 (−.342, .002)	.308*** (.139, .460)	.262** (.089, .420)	–
5	−.240** (−.398, −.068)	.247** (.076, .405)	.223* (.050, .383)	.328*** (.160, .477)	–
6	−.258** (−.417, −.083)	.134 (−.045, .305)	.234* (.058, .396)	.260** (.083, .421)	.378*** (.214, .521)	–
7	.125 (−.058, .300)	−.064 (−.243, .119)	−.047 (−.226, .136)	−.184* (−.355, −.001)	−.583*** (−.691, −.449)	−.435*** (−.575, −.272)	–
Mean	11.46	1371.78	14.06	3.50	21.94	2.26	40.33
Standard deviation	3.17	1080.58	5.21	3.39	12.99	2.63	12.92
Skewness	−.297	1.090	−.539	1.383	.170	1.004	.125
Kurtosis	.294	1.299	−.349	2.031	−.980	−.279	−.502

Note: 1. Age at entrance; 2. Duration of placement (days); 3. Multiple childhood institutional maltreatment; 4. Trauma experience after escape from the institution; 5. Complex posttraumatic stress disorder symptoms; 6. Poverty; 7. Quality of life. 95% Confidence Intervals presented in parentheses.

*p < .05, **p < .01, ***p < .001.

3.2. Regression analyses of the variables on QoL

Table 2 presents the regression analyses. Using Bonferroni adjustment, the duration of placement (β = .24, p = .009) was associated with trauma experiences after escape from the institution. Trauma experiences after escape from the institution (β = .25, p = .007) were associated with CPTSD symptoms. CPTSD symptoms (β = .26, p = .005) were associated with poverty, and both CPTSD symptoms (β = −.52, p < .001) and poverty (β = −.26, p = .003) were negatively associated with current QoL.

Table 2. Regression analyses.

Outcome variables	Explanatory variables	R	Adj R^2	F	df	B	95% Confidence interval	β	t	p
Trauma experiences after escape from institution (n = 123)		.357	.106	5.799***	3, 119
	Age at entrance					−.040	(−.244, .164)	−.037	−.390	.697
	Duration of placement (days)					.001	(.000, .001)	.244	2.650	.009
	Multiple childhood institutional maltreatment					.114	(−.007, .236)	.175	1.859	.066
Complex posttraumatic stress disorder symptoms (n = 123)		.393	.126	5.378***	4, 118
	Age at entrance					−.410	(−1.178, .359)	−.098	−1.056	.293
	Duration of placement (days)					.001	(−.001, .004)	.112	1.196	.234
	Multiple childhood institutional maltreatment					.249	(−.216, .713)	.100	1.059	.292
	Trauma experiences after escape from institution					.949	(.266, 1.632)	.250	2.753	.007
Poverty (n = 118)		.427	.146	4.994***	5, 112
	Age at entrance					−.112	(−.270, .047)	−.132	−1.394	.166
	Duration of placement (days)					.000	(−.001, .000)	−.050	−.521	.603
	Multiple childhood institutional maltreatment					.059	(−.039, .158)	.117	1.197	.234
	Trauma experiences after escape from institution					.103	(−.037, .242)	.137	1.455	.148
	CPTSD symptoms					.055	(.016, .093)	.264	2.834	.005
Quality of life (n = 110)		.629	.360	11.209***	6, 103
	Age at entrance					.012	(−.722, .747)	.003	.033	.974
	Duration of placement (days)					.001	(−.001, .003)	.046	.541	.590
	Multiple childhood institutional maltreatment					.266	(−.193, .725)	.102	1.150	.253
	Trauma experiences after escape from institution					.077	(−.556, .710)	.021	.241	.810
	CPTSD symptoms					−.547	(−.731, −.362)	−.524	−5.884	<.001
	Poverty					−1.323	(−2.180, −.465)	−.260	−3.060	.003

Note: ***p < .001.

Table 5 and Figure 1 (supplementary material) present the results of the indirect effects of the pathways of the proposed exploratory model. The total standardized indirect effect was −.20 (95% CI [−.332, −.079]). Specifically, four indirect effects showed statistically significant results: The relationships between the total number of multiple childhood IM experienced and QoL were found to be significant indirectly through CPTSD symptoms (Ind 2; estimated effect = −.11), through trauma experience after escape from the institution and CPTSD symptoms (Ind 4; estimated effect = −.04), through CPTSD symptoms and poverty (Ind 6; estimated effect = −.02), and through trauma experience after escape from the institution, CPTSD symptoms, and poverty (Ind 7; estimated effect = −.01).

4. Discussion

Quality of life among the survivors of the ‘Hyeongje’ was indicated by CPTSD symptoms and poverty, which were associated with prolonged exposure to multiple childhood IM and cumulative trauma aftermath. These findings confirm previous research indicating that experiences of IM have serious effects on the well-being and health of individuals throughout adulthood (Carr et al., 2019; Carr et al., 2020). Consistent with the theoretical framework underlying this study (Hobfoll, 1988; Hobfoll & Schumm, 2002), while people maintain and acquire resources they can utilize and experience an upward spiral on physical and psychological well-being, prolonged exposure to childhood IM precipitates a continuous downward spiral of resource loss leading to additional trauma experiences, CPTSD, and poverty, ultimately resulting in low QoL. It is important to note that the negative effects of such experiences are not only limited to the adversities one underwent when institutionalized. The lack of resources after escape may further hinder one’s opportunities to promote resiliency and ultimately result in prolonged suffering later in life.

We supported that childhood IM increased the risk of re-victimization in later life, thus exacerbating its impact, as has been indicated by other findings (Knefel et al., 2019; Lueger-Schuster et al., 2018). It was also interesting in our study that it was the duration of the exposure, rather than the entrance age or the number of multiple types of IM, that was associated with additional trauma exposure after escaping the institution. This additional trauma exposure after escape was also associated with current CPTSD symptoms, which is consistent with a recent report indicating that childhood abuse histories are related to further revictimization and that PTSD symptoms of revictimization are associated with childhood abuse (Gama et al., 2021). This is also supported by Vang et al. (2019), who reported that being poly-victimized with both childhood and adulthood interpersonal trauma increases the risk for CPTSD.

This also aligns with the perspective of the resource conservation theory related to trauma. This indicates that early detection of childhood IM, imminent rescue, and providing subsequent protective responses are critical steps needed to cease further victimization and prevent related CPTSD. Our study shows that adequate resource investment was not in place for these participants after escape; only 14.2% of participants were able to remain reunited with their family, 36.2% were repeatedly replaced in other institutions, and 76% did not have access to education. Providing healthy environments (e.g. Mukerji et al., 2021; Nelson et al., 2007; Wade et al., 2018) following IM may increase the resiliency of such children and prevent further loss. We also suggest that psychosocial support for childhood IM survivors should target various lifetime trauma; therefore, it is necessary to increase our understanding that is difficult to distinguish a single index trauma for a CPTSD diagnosis, and that CPTSD may be a result of cumulative trauma.

Moreover, CPTSD was indicated to be critical to low QoL, as previous studies have shown that severe mental health issues are related to decreased QoL (e.g. Baek et al., 2022; Damnjanovic et al., 2011; Gander et al., 2019). This highlights the importance of emphasizing CPTSD treatment among interventions (e.g. Ford & Kidd, 1998; Haselgruber et al., 2020; Knefel et al., 2019) that are essential in breaking the cycle of loss spirals.

As Thoma et al. (2021) discussed the relationship between IM and poverty, more than half of our participants had at least one experience of monetary limitation in the previous year of our assessment, and these poverty experiences were associated with current QoL. Poverty was correlated with younger age at institution entrance, number of childhood IM experiences, trauma experiences after escape, and CPTSD; however, the regression analysis indicated that only CPTSD was indicative of poverty. Since this study cannot establish causality, then considering the perspective of Conger and Donnellan (2007), interventions targeting CPTSD might be needed to alleviate poverty-related problems; conversely, reducing poverty might influence the capability to regulate CPTSD symptoms, in addition to buffering loss spirals to low QoL. Addressing poverty has been pointed to as a crucial task for improving QoL (Murali & Oyebode, 2004). A one-year follow-up randomized controlled trial (Haushofer et al., 2020) reported that, compared to low-intensity psychological interventions, cash transfers helped mitigate violence and also had a spill-over effect of improving economic QoL. When low-intensity psychological interventions and cash transfers were both provided, life satisfaction was reported to be the highest. However, the effect of CPTSD clearly shows that monetary support would not be fully beneficial by itself. According to Holmes et al. (2022) discussing that poverty and PTSD affect each other, we may anticipate that chronic CPTSD would hinder the effectiveness of monetary interventions and that living conditions that perpetuate poverty would worsen CPTSD. The results of a previous meta-analysis have shown that the efficacy for ameliorating symptoms related to complex trauma is stronger when receiving intensive empirically supported treatments rather than other psychosocial interventions, even for people with low resources, thus highlighting the importance of bringing efforts to provide effective CPTSD treatments to people with low access to such interventions (Choi et al., 2020). This is also expected to be true for childhood IM survivors with less resources, such as those experiencing poverty.

Thus, intensive psychological treatment targeting CPTSD in addition to economic support, occupational rehabilitation, and increasing social awareness that supports the survivors’ social functioning would be best suited as a comprehensive approach. For instance, multidisciplinary treatment including medical service, psychotherapy, social treatment, and legal support has been shown to be effective for mental health, including PTSD and QoL, in a refugee population (Stammel et al., 2017). Meanwhile, an effectiveness study of a multidisciplinary intervention implemented in northern Kosovo which included cognitive behavioural therapy targeting PTSD symptoms was shown to bring about economic improvement in both employment and wage (Wang et al., 2016), thus indicating that such interventions may create positive upward spirals. While psychological interventions for PTSD are expected to be efficacious for CPTSD (Choi et al., 2020; Karatzias et al., 2019), these multidisciplinary approaches should be further examined to assess whether they are also effective for CPTSD.

4.1. Limitations

As a cross-sectional study, this work could not establish causal relationships or examine temporal pathways between variables. Moreover, as the measurements were retrospective self-reports, the accuracy of recalling past experiences or estimating CPTSD diagnoses could not be guaranteed. It is necessary to increase the construct validity in measuring childhood IM; for instance, while we measured neglect in terms of deprivation of basic needs, a more systematic measure of childhood neglect in addition to abuse should be considered. This is also true for poverty; more valid tools may be needed to assess it accurately. There are limitations in discussing how developmental loss spirals beginning from birth may affect current QoL. However, it should be mentioned that more than half of the children in our sample were living with their parents before placement in the ‘Hyeongje.’ Further, exploring the social circumstances of the participants after escape, the majority were unable to initially reunite or remain reunited with their family, and the majority also ceased education. This provides a clue that the ‘Hyeongje’ unlawfully captivated children in the institution against their will and implies that resource loss may have been precipitated at the ‘Hyeongje.’ Also, given the high prevalence of CPTSD and low education level among the participants, it is possible that IM hindered healthy developmental experiences and educational opportunities, which may also affect poverty and contribute to a low QoL. Future quantitative and qualitative research may explore the process of how IM leads to a loss spiral among the many different aspects of a person’s life. The exploratory model proposed herein should be reexamined with a larger sample in a future study.

According to TRC Korea, 191 people were found and confirmed to be victims during the period of our data collection, so the sample size of 127 may be adequate in some aspects. However, it falls short of being representative, as it is possible that TRC Korea could not contact those who are presumed to still be in institutions, not to mention those in detention centres. Moreover, the experiences of the ‘Hyeongje’ survivors cannot represent all people who were institutionalized, and the result of our research has limitations in terms of generalizability, which should be understood in the historical context of the ‘Hyeongje.’

Additionally, the participants of this study were predominantly men, which raises the possibility that social prejudice or sexism could have limited the disclosure and participation of women survivors in this research. It is assumed that approximately 10% of the victims of the ‘Hyeongje’ were women (TRC Korea, 2022), thus indicating the need for future research to consider gender disparities in research participation as well as methods to increase women participation. Future studies could employ diverse data collection methods and conduct experimental and longitudinal studies that can investigate whether interventions for CPTSD or economic support ultimately improve QoL, thus contributing to real-life changes in the lives of IM survivors.

4.2. Implications

This study identified key variables that may contribute to current QoL among adult childhood IM survivors. IM brings about loss spirals by increasing experiences of further cumulative trauma, CPTSD, and poverty in life. To understand the trajectory of life among childhood IM survivors, an extensive multidisciplinary perspective is needed to design a systematic intervention based on the due diligence approach of the government. Increasing QoL would be the aim of this intervention, where public support and policy decision-making would be critical to provide evidence-based treatments for CPTSD and relevant economic support. The Korean government should take accountability by not only continuing to investigate the ‘Hyeongje’ but also expanding the inquiries to other numerous IM. While the government is accountable for collective reparation, such as in the National Redress Scheme of Australia (Wright, 2023), the survivors in South Korea are seeking reparations individually. The enactment of a law for collective reparation would be crucial for establishing justice and providing resource gain in the current lives of the survivors. To provide consistent multidisciplinary care, the state should be responsible for increasing the budget for public support centres, which are currently limited in their ability to deliver the expected service. Moreover, seeking justice for survivors of the past is the foundation for preventing future IM. This means it is crucial to increase public awareness and acknowledgment of IM, investigate the failure to detect abuse in systems, and reform policy to proactively detect and protect people from IM.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The participants of this study did not give written consent for their data to be shared publicly, so due to the sensitive nature of the research supporting data is not available.

Additional information

Funding

This study was supported by the Busan City of South Korea.