Child Abuse & Neglect 76 (2018) 488–501

Contents lists available at ScienceDirect

Child Abuse & Neglect

journal homepage: www.elsevier.com/locate/chiabuneg

Child abuse and neglect in institutional settings, cumulative

T
lifetime traumatization, and psychopathological long-term
correlates in adult survivors: The Vienna Institutional Abuse Study
⁎
Brigitte Lueger-Schuster , Matthias Knefel, Tobias M. Glück, Reinhold Jagsch,
Viktoria Kantor, Dina Weindl
Faculty of Psychology, University of Vienna, Liebiggasse 5, 1010, Vienna, Austria

AR TI CLE I NF O AB S T R A CT

Keywords: Child maltreatment (CM) in foster care settings (i.e., institutional abuse, IA) is known to have
Institutional abuse negative effects on adult survivor’s mental health. This study examines and compares the extent
Childhood trauma of CM (physical, emotional, and sexual abuse; physical and emotional neglect) and lifetime
Foster care traumatization with regard to current adult mental health in a group of survivors of IA and a
Adult survivors
comparison group from the community. Participants in the foster care group (n = 220) were
Trauma sequelae
adult survivors of IA in Viennese foster care institutions, the comparison group (n = 234) con-
Mental health
sisted of persons from the Viennese population. The comparison group included persons who
were exposed to CM within their families. Participants completed the Childhood Trauma
Questionnaire, the Life Events Checklist for DSM-5, the PTSD Checklist for DSM-5, the
International Trauma Questionnaire for ICD-11, and the Brief Symptom Inventory-18 and com-
pleted a structured clinical interview. Participants in the foster care group showed higher scores
in all types of CM than the comparison group and 57.7% reported exposure to all types of CM.
The foster care group had significantly higher prevalence rates in almost all mental disorders
including personality disorders and suffered from higher symptom distress in all dimensional
measures of psychopathology including depression, anxiety, somatization, dissociation, and the
symptom dimensions of PTSD. In both groups, adult life events and some but not all forms of CM
predicted PTSD and adult life events partly mediated the association of PTSD and CM.
Explanations for the severe consequences of CM and IA are discussed.

1. Introduction

Child maltreatment (CM) is a worldwide phenomenon with severe consequences for survivors and for society, creating great
challenges for health care systems (Gilbert et al., 2012; Shaw & Jong, 2012). CM includes various types of violence such as sexual,
emotional and physical abuse and/or emotional and physical neglect (Vachon, Krueger, Rogosch, & Cicchetti, 2015). It is estimated
that 35% of the US population is exposed to some form of emotional abuse, followed by almost 16% of physical and 11% of sexual
abuse (Centers for Disease Control and Prevention (CDC), 2015). There is consensus that CM poses a great risk for the later adult’s
mental health including various problems such as posttraumatic stress disorder (PTSD), depression, anxiety, substance abuse, and
personality disorders to name only the most common negative outcomes (Kessler et al., 2010). The extent of these abusive acts may

⁎
Corresponding author.
E-mail addresses: brigitte.lueger-schuster@univie.ac.at (B. Lueger-Schuster), matthias.knefel@univie.ac.at (M. Knefel), tobias.glueck@univie.ac.at (T.M. Glück),
reinhold.jagsch@univie.ac.at (R. Jagsch), viktoria.kantor@univie.ac.at (V. Kantor), dina.weindl@univie.ac.at (D. Weindl).

https://doi.org/10.1016/j.chiabu.2017.12.009
Received 25 April 2017; Received in revised form 5 December 2017; Accepted 8 December 2017
Available online 22 December 2017
0145-2134/ © 2017 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/BY-NC-ND/4.0/).
B. Lueger-Schuster et al. Child Abuse & Neglect 76 (2018) 488–501

vary, both in quantity and in quality (Finkelhor, Ormrod, & Turner, 2009). Being a survivor of CM also comes with a higher risk of
revictimization in later life (Widom, Czaja, & Dutton, 2008). This seems also to be the case for children who were raised in in-
stitutions such as foster care settings (termed here Institutional Abuse - IA). For example in Austrian Catholic institutions, between
the late 1940s until the 1980s there are many reports that children were exposed to complex interpersonal childhood trauma in-
cluding various forms of abuse and neglect (Lueger-Schuster et al., 2013a, 2013b). In the present study, we aimed to assess the
trauma history and current mental health in a group of adult survivors of IA and to compare the associations of trauma and mental
health to a comparison group.
Ideally, institutions such as foster care homes should take care of children, who were confronted with childhood adversities in
their families of origin such as abuse, neglect, parental problems, or the loss of parents. They should offer safety, social support, and
healthy living conditions to aid the child’s recovery from a traumatic past (Brännström, Forsman, Vinnerljung, & Almquist, 2016).
However, institutions were and sometimes still are also characterized by psychosocial deficiencies, such as the frequent change of
caregivers and little emotional and social support, putting the child at risk of psychosocial deprivation (Merz & McCall, 2010). The
likelihood of CM in institutions may be increased by the separation from other social environments, a strong hierarchal form of the
organization with the child placed at the lowest level, the dependency on caregivers, who sometimes are not adequately trained to
cope with challenging behaviors, little access for the children to social support from outside the system, and a lack of functioning
control mechanisms to protect children’s rights (MacLean, 2003; Reilly, 2003). The experience of such institutional conditions may
create feelings of betrayal, powerlessness, and stigmatization with the potential to harm the child’s healthy development (Wolfe,
Jaffe, Jette, & Poisson, 2003). Betrayal trauma theory (Freyd, 1994) can not only be applied to individual perpetrators but also to
institutions (Smith & Freyd, 2014). Institutional betrayal is characterized by an environment in which CM is more likely, inadequate
responding to disclosure of CM, and the inability to escape from the abusive environment. Wright et al. (2016) found that institu-
tional betrayal independently predicted posttraumatic stress when controlling for lifetime trauma history. Findings supporting this
mechanism were also reported in cross-sectional national studies of IA that reported severe long-term effects on mental health of
survivors not only including various mental disorders, but also a broad spectrum of psychosocial impairments (Fitzpatrick et al.,
2010; Lueger-Schuster et al., 2013a, 2013b).
Despite these institution-specific adverse effects, the larger amount and extent of the abuse and neglect experienced in such
institutions may also account for its detrimental consequences. Survivors of IA were not only exposed to traumatic events in
childhood before foster care, but then often reported a cumulative and prolonged exposure to various forms of CM taking place at the
same time in the institution (Lueger-Schuster et al., 2013a, 2013b). Non-institutional CM seems more restricted in frequency and
variety of types of CM a single individual is exposed to compared to IA (cf. Centers for Disease Control and Prevention (CDC), 2010).
So far, there is some evidence for a dose-response relation for non-institutional CM between cumulative CM and mental health
problems such as PTSD, depression, but also physical pain, relational problems, and lower levels of occupational functioning (Steine
et al., 2017). Furthermore, a review found differential associations between types of early life stressors and adult psychopathology
reporting that physical, sexual, and emotional abuse as well as unspecific neglect showed clearer associations with later adult mental
disorders than for example physical neglect (Carr, Martins, Stingel, Lemgruber, & Juruena, 2013).
The path from of CM to adult psychopathology, however, is complex and children and adolescents who experienced CM are also at
risk to experience more life events during their adulthood (LaNoue, Graeber, Hernandez, de Warner, & Helitzer, 2012; Widom et al.,
2008). While there exists a large body of research on revictimizing life events, such as sexual revictimization in adult survivors of
sexual abuse in childhood (Classen, Palesh, & Aggarwal, 2005), the power of CM in predicting non-victimizing adult life events is less
well documented. However, there is some evidence suggesting that CM is also associated with a higher risk of exposure to non-
victimizing events, such as serious accidents (Widom et al., 2008). LaNoue et al. (2012) found that adult traumatization, including
victimizing and non-victimizing events, partially mediated the power of CM predicting depressive symptoms. Even though these
traumatizing events will likely contribute to adult psychopathology, it is not yet clear how adult life events mediate the association of
IA with mental health problems in adult survivors (Carr et al., 2010; Lueger-Schuster et al., 2013a, 2013b).
To our knowledge, no study has been conducted that directly compared childhood trauma and adult life events with regard to
current adult mental health in a sample of survivors of IA and a community sample including persons who were exposed to CA within
their families. Thus, we first aimed to compare the foster care group with the comparison group regarding trauma exposure and
expected to find higher rates within the group of adult survivors of childhood IA compared to the comparison group. We secondly
aimed to investigate the current mental health status of adult survivors of IA and to compare these rates to the comparison group.
Moreover, we expected to find a subgroup within the comparison group that reported high rates of CM. We then compared this
severe-maltreatment comparison subgroup to the foster care group in order to find specific associations of IA and mental health as
compared to CM and mental health. Finally, we aimed to find the predictive power of the five types of child maltreatment, as
identified by the CTQ (i.e., physical, emotional, and sexual abuse; physical and emotional neglect), for current psychopathological
distress, taking into account the effects of adult life events. We hypothesized that both, child maltreatment as well as adult life events
would predict adult psychopathology. We aimed to identify the extent to which adult life events mediated the prediction of psy-
chopathology by child maltreatment, i.e. in how far the effects of re-traumatization account for the correlation between child
maltreatment and adult psychopathology.

489
B. Lueger-Schuster et al. Child Abuse & Neglect 76 (2018) 488–501

2. Methods

2.1. Context

The Vienna Institutional Abuse Study (VIA-S) started in June 2014 with the aim to investigate long-term correlates of childhood
abuse (emotional, physical, and sexual) and neglect (emotional and physical) in institutions of the child welfare system of the City of
Vienna in a sample of adult survivors. The impact of the abuse and neglect on survivors’ later life was compared to a group of persons
from the Viennese population without a history of institutional abuse.
The participating survivors were mainly or partly raised in institutional foster care settings managed or supervised by the City of
Vienna over a period of more than 40 years, from the late 1940s to the late 1980s. The public attention to these crimes was attracted
when several newspaper articles were published starting in 2011 where adult survivors disclosed their experiences to journalists (e.g.
Hönigsberger & Schrenk, 2011). Although some of the survivors had tried for years to get support, attention or justice for the crimes
they endured during the childhood, the media reports finally created the momentum for a public awareness that caused todays
authorities to act. The City of Vienna mandated a panel of historians to study the history of CM in one specific Viennese foster care
institution coming to the conclusion that, apart from cases of severe maltreatment, corporal punishment happened on a daily base
and was administered as a routine correctional measures against the children (Schloss Wilhelminenberg; Kommission
Wilheminenberg, 2013). It also became apparent that this did not only happen in this one single institution, but also took place in all
foster care settings under the jurisdiction of the City of Vienna, in some cases even until the late 1980s. Consequently the authorities
established an independent victims’ protection commission, and the victim-protection organization “Weisser Ring (White Ring)” was
authorized to assess claims for compensation payments and/or therapy. Crimes reported by the survivors took place in these foster
care settings and all participating survivors had approached the independent victims’ protection organization. About 2700 in-
dividuals approached the White Ring between 2010 and March 2016 to undergo the clearing process, based on which the claims were
assessed. The clearing process comprised the initial contact to the White Ring, the retrieving and delivery of child welfare files from
the archives of the magistrate for youth and children of the City of Vienna (MA11), and up to ten psychological/psychiatric ex-
amination sessions to assess the extent of the experienced childhood trauma and long-term consequences with specially appointed
clinical psychologists, psychotherapists, and psychiatrists. A claims commission within the White Ring then assessed the report and
recommendation from the examination and the information from files stored in the archives to decide on the amount of compensation
and hours of therapy paid by the City of Vienna (in average 17,000 € per case).

2.2. Participants and procedure

Initially, 295 persons consented to take part in the VIA-S, of whom 220 successfully completed the interviews. Thirty-four persons
could not be contacted any more after giving their initial consent, and 25 persons withdrew their consent during the research process.
Sixteen participants were not able to finish their interview due to various reasons such being intoxicated (see Fig. 1 for a flow chart).
The majority of interviews was conducted in rooms of the Faculty of Psychology at the University of Vienna. Only a minor part took
place in prisons (n = 7, 3.2%) or in the homes of survivors if they were physically unable to visit the research facility (n = 24,
11.0%). Four trauma-specialized and experienced clinical psychologists, who were all part of the research team, conducted the
structured interviews with the survivors. Interviews comprised sociodemographic and biographical information, questions regarding
abuse and neglect in the family of origin (if applicable), abuse and neglect in institutional settings, various questionnaires regarding
psychopathology and various psychosocial aspects, and a shortened version of a structured clinical diagnostic interview (see Mea-
sures section for details). Interviews lasted between two and three hours including a break depending on the extent of traumatic
experiences and current psychopathology reported.
Only survivors who had finished the clearing process and who had received compensation payments (n = 1984) until February
2016, were invited to participate in the VIA-S. This ensured that only persons who had disclosed their traumatic childhood ex-
periences and were familiar with a psychological interview setting participated in the study. Consequently, (1) survivors knew
whether they could handle the possible distress caused by a trauma-related interview and (2) the independence of the study was
secured and survivors did not expect that the participation in the study influenced their financial claims, which could have created a
feeling of obligation. The information letter included information about the background of the study and its procedure, informed
consent forms, a prestamped return envelope, and the contact details of the research team at the University of Vienna. There were two
ways to participate: (1) active participation (interview with researchers and permission to analyze the files in the archives of the
MA11) and (2) passive participation (only the permission to analyze the files in the archives of the MA11 without being personally
contacted by the research team). The White Ring sent out the letters to the survivors, because of legal data protection reasons the
research team did not have any direct access to the addresses. Only if survivors sent back a signed consent form for active partici-
pation with their contact details to the research team they were contacted and invited to participate in a face-to-face interview. The
research team also set up an information website with information regarding the study and the procedure of participation (http://
heimstudie.univie.ac.athttp://heimstudie.univie.ac.at). The study was also announced in self-help groups and Internet-platforms
used by survivors. Although survivors could contact the research team without yet having received the information letter, they could
only participate if they had also finished the clearing process with the White Ring.
The comparison group was recruited by announcements in city district newspapers, notifications in supermarkets and large
building complexes, and by word of mouth. These announcements stated that the purpose of the assessment was to compare a group
of the Viennese population to a group of adult survivors of IA. We aimed to include only persons from the same socioeconomic

490
B. Lueger-Schuster et al. Child Abuse & Neglect 76 (2018) 488–501

Fig. 1. Participant flow chart.

background as survivors with similar age, who never lived in a foster care institution, to ensure comparability. As there was no
financial compensation for participation and because of the distance of potential participants to research and psychological topics,
recruitment of the comparison group was far more difficult than expected. Because of ethical reasons, participants in the comparison
group could not be compensated as participating survivors were not compensated either. Due to these recruitment problems, we also
administered online questionnaires and distributed them utilizing a Viennese newsletter platform, excluding the structured clinical
diagnostic. Online-participants could then volunteer to be contacted to also participate in the structured clinical interview. Clinical
psychology master students who were trained to conduct a structured clinical diagnostic interview carried out face-to-face interviews
in the comparison group (see Fig. 1 for details). Master students were instructed and supervised by a senior researcher and clinical
psychologist. To rule out a possible bias within the comparison group, we statistically compared the characteristics of both subgroups,
the face-to-face and the online comparison group. All participants gave written informed consent previous to participation in the
study. The study was approved by the Ethics Committee of the University of Vienna (No. 00071).

2.3. Participants

A total of 454 persons actively participated in the current study, 220 were adult survivors of child maltreatment in foster care

491
B. Lueger-Schuster et al. Child Abuse & Neglect 76 (2018) 488–501

Table 1
Sociodemographic characteristics.

a
Group Comparison

Foster Care Comparison face-to-face Comparison online Comparison total 1 2

Sample Size 220 143 91 234

Age M = 57.9 M = 59.8 M = 55.6 M = 58.2 0.003 0.789
SD = 9.5 SD = 10.5 SD = 9.6 SD = 10.3
Gender (female) 88 (40.0%) 94 (65.7%) 57 (62.6%) 151 (64.5%) 0.629 < 0.001
Relationship status
Divorced/single 110 (50.0%) 58 (40.6%) 48 (52.7%) 116 (45.9%) 0.138 0.869
Married/cohabiting 97 (44.1%) 73 (51.0%) 39 (42.9%) 112 (48.5%)
Widowed 13 (5.9%) 10 (7.0%) 3 (3.3%) 13 (5.6%)
Academic attainment
Compulsory school or less 65 (29.5%) 2 (1.4%) 2 (2.2%) 4 (1.7%) 0.104 < 0.001
Apprenticeship 108 (49.1%) 27 (18.9%) 18 (19.8%) 45 (19.2%)
Vocational school without A-level 31 (14.1%) 31 (21.7%) 9 (9.9%) 40 (17.1%)
A-level or higher 16 (7.3%) 83 (58.0%) 62 (68.1%) 145 (62.0%)
Employment status
Retired 59 (26.8%) 59 (41.3%) 22 (24.2%) 81 (34.8%) 0.065 < 0.001
Inability to work/early pension 59 (26.8%) 7 (4.9%) 3 (3.3%) 10 (4.3%)
Unemployed/social assistance 35 (16.0%) 9 (6.3%) 9 (9.9%) 18 (7.8%)
Long term sick leave 14 (6.4%) 2 (1.4%) 1 (1.1%) 3 (1.3%)
Imprisoned 7 (3.2%) – – –
Employed 46 (20.9%) 66 (46.2%) 55 (60.4%) 121 (51.1%)
Income per month M = 1221.1 M = 1651.7 M = 1898.9 M = 1768.0 0.118 < 0.001
SD = 736.3 SD = 864.8 SD = 697.4 SD = 796.3

Note. Number represent means and standard deviations for metric variables and counts and percentages for nominal variables; numbers may not add up to total group-
N in the comparison groups due to missings; ap-value of χ2-test/Fisher’s exact test/t-test; group comparison 1: face-to-face comparison group vs. online comparison
group; group comparison 2: foster care group vs. total comparison group.

institutions, 234 participated in the comparison group. The comparison group differed in several demographic variables from the
foster care group. See Table 1 for information on the demographic characteristics of the groups and group comparisons.

2.3.1. Survivors
All survivors lived in institutional foster care settings during their childhood and experienced maltreatment during this time. The
age of survivors ranged between 29 and 87 years (M = 57.9; SD = 9.5). The overall level of education of the research sample was
lower than the average Austrian population (Statistik Austria, 2016): the majority of the sample (63.2%) completed an apprenticeship
or a vocational school without an A-level degree (10–12 years of education without general qualification for university entrance;
Austrian population: 49.8%), 19.5% attended compulsory school (9 years of education; Austrian population: 19.1%), 4.1% received
an A-level education (12 years of education with general qualification for university entrance; Austrian population: 14.8%), 3.2% had
a university degree (> 15 years of education; Austrian population: 16.2%). 10.0% did not complete compulsory school (< 9 years of
education; Austrian population: < 1%).The average age of the first foster care placement was at 5.6 years (SD = 4.4) with a range of
0–16 years of age. All participants experienced various forms of abuse (emotional, physical, and sexual) and neglect (emotional and
physical) in the institutional settings. No maltreatment within the family of origin was reported only by n = 54 survivors (24.5%).
Survivors experienced between 0 and 16 traumatic events in their adulthood.

2.3.2. Comparison group

Of those in the comparison group, n = 143 persons attended a face-to-face interview (age range 40–86, M = 59.8; SD = 10.5) and
n = 91 persons participated in an online survey (age range 41–78, M = 55.6; SD = 9.6). The two subgroups in the comparison group
(online vs. face-to-face) differed significantly regarding their age (Table 1).

2.4. Measures

2.4.1. Brief symptom inventory 18 (BSI-18) plus subscale paranoid ideation

The BSI-18 is the shortest version of the SCL-90 R (Derogatis, 1994) and was also adapted for a German speaking population and
showed good psychometric properties (Franke et al., 2011). It measures the distress of the three psychological symptoms: anxiety,
depression, and somatization on a 5-point scale (from 0-not at all through 4-very much). It also derives a global measure of symptom
distress, GSI. Additionally we added the subscale “Paranoid Ideation” from the Brief Symptom Inventory (BSI, Franke & Derogatis,
2000) as this is an important factor in survivors of child maltreatment which is also related to anger and aggression. Cronbach’s α for
the BSI-18 was α = 0.92 (excluding paranoid ideation), Cronbach’s α for the subscales were: α = 0.82 for anxiety, α = 0.86 for
depression, α = 0.80 for somatization, and α = 0.74 for paranoid ideation.s

492
B. Lueger-Schuster et al. Child Abuse & Neglect 76 (2018) 488–501

2.4.2. Childhood trauma questionnaire (CTQ)

The CTQ (Bernstein & Fink, 1998) is a 28-item self-report questionnaire assessing on a 5-point scale (from 1-never through 5-very
often) the experienced frequency of different types of child maltreatment including sexual, physical, and emotional abuse and
emotional and physical neglect. We used the German version of the CTQ that has shown good psychometric properties (Wingenfeld
et al., 2010). In the foster care group, we assessed the frequency for both, events that took place during foster care and events in the
families, hence each item was rated twice. We therefore asked the participants to first rate each item in its original form (starting with
the phrase “When I was growing up…”), and then the same item in an adapted form (starting with the phrase “During foster care…”).
Items that particularly referred to family members were adapted (e.g. “People in my family said hurtful or insulting things to me” was
adapted to “People in my foster care home said hurtful or insulting things to me”). We then computed a mean value for the two
versions of each item and a cumulative child maltreatment index for all traumatic childhood events (i.e., the mean score over all
items). We decided not to calculate a scale wise sum score of the items but rather to calculate a mean score, because in this way the
results can be interpreted using the scale labels of the questionnaire (never to very often). Cronbach’s α for the cumulative child
maltreatment index was α = 0.90, for the intra-familial events α = 0.91, and for the events during foster care α = 0.85. In order to
provide a rating of the severity of the child maltreatment, we classified the average values of each type of child maltreatment using
norms of the German population (Häuser, Schmutzer, Brahler, & Glaesmer, 2011). We used norms from the German population
because Austrian norms do not exist and both countries share the same language and a very similar cultural and economic back-
ground.

2.4.3. Life events checklist for DSM-5 (LEC-5)

The LEC-5 (Weathers et al., 2013a) is a self-report measure assessing possible traumatic events that occurred throughout a
participant’s lifetime. It consists of 17 items of which 16 items refer to events potentially causing distress and PTSD and one addi-
tional item indicating any other stressful event. In the presented study, we used the German version of the LEC-5 (Ehring,
Knaevelsrud, Krüger, & Schäfer, 2014a, 2014b) and a dichotomous response pattern was given (yes = “experienced my own“, and/or
“witness happening to someone else’’; no = it doesn’t apply) for events happening in participant’s adulthood, after foster care. In
order to create a total score of traumatic life events in adulthood the items were summed up, resulting in a total score possibly
ranging from 0 to 17. Cronbach’s α for the LEC was α = 0.70.

2.4.4. International trauma questionnaire (ITQ)

The ITQ (Cloitre, Roberts, Bisson, & Brewin, 2013) is a 23-item self-report measure that was designed to capture the proposed
ICD-11 PTSD and Complex PTSD symptoms on a 5-point scale (from 0-not at all through 4-extremely). For the present study, we
translated the English version of the questionnaire into German (Knefel, Lueger-Schuster, & Maercker, 2013). The German version
was back-translated into English by a professional translator and then the German version was slightly changed following con-
sultation with the authors of the English version. The first seven items assess the ICD-11 PTSD symptoms on three dimensions: re-
experiencing (three items), avoidance (two items), and sense of threat (two items). The other 16 items assess the three disturbances in
self-organization (DSO) dimensions, which need to be present additionally to the PTSD symptoms for a Complex PTSD diagnosis.
Symptoms of affect dysregulation are assessed with nine items, negative self-concept symptoms with four items, and disturbances in
relationships with three items. The affect dysregulation dimension is again divided into deactivation symptoms and hyperactivation
symptoms. Dissociation in terms of depersonalization and derealization is addressed in the deactivation dimension (Lanius et al.,
2010). The English version of the ITQ showed good psychometric properties in initial evaluations (Karatzias et al., 2016, 2017).
Cronbach’s α in the present study was α = 0.84 for the PTSD items and α = 0.90 for the DSO items.

2.4.5. PTSD checklist for DSM-5 (PCL-5)

The PCL-5 (Weathers et al., 2013b) is a self-report questionnaire that includes all 20 PTSD symptoms as listed in the DSM-5. We
used the German version of the PCL-5 in the present study (Ehring et al., 2014a, 2014b). Items are queried on a 5-point scale of past-
month severity (from 1-not at all to 5-extremely). The German version of the PCL-5 showed good psychometric properties (Krüger-
Gottschalk et al. (2017)): high internal consistency (α = 0.95), high re-test reliability (r = 0.91), and a high correlation with the total
severity score of the Clinician-Administered PTSD Scale for DSM-5 (r = 0.77). Cronbach’s α in this study were α = 0.79–0.91 for the
four PCL-5 subscales and α = 0.94 for the total PCL-5 score.

2.4.6. Structured clinical interview for DSM-IV (SCID I + II)

In this study a shortened version of the German version of the SCID (Wittchen, Wunderlich, Gruschwitz, & Zaudig, 1997) was used
to determine the presence or absence of psychiatric disorders according to DSM-IV. Most common axis 1 and 2 disorders were
included, respectively disorders that had been reported in previous studies with survivors of institutional child abuse (Carr et al.,
2010; Lueger-Schuster et al., 2013a, 2013b). Axis 1 disorders included in the interview were major depression, dysthymia, alcohol
and substance abuse and dependency, anxiety disorders (panic disorder with and without agoraphobia, generalized anxiety disorder,
specific phobias, obsessive compulsive disorder, PTSD). Axis 2 disorders included the following personality disorders: avoidant,
compulsive, paranoid, schizotypal, borderline, and antisocial.

2.5. Statistical analysis

For descriptive analyses we reported means and standard deviations for continuous variables, and frequencies with percentages

493
B. Lueger-Schuster et al. Child Abuse & Neglect 76 (2018) 488–501

Fig. 2. Mediation model predicting PTSD symptom distress (left: foster care group, right: comparison group).

for categorical variables. We compared several groups during the analyses: To compare two groups, we used t-tests for continuous
data and χ2-tests for categorical data and to compare three groups (foster care group with intra-familial abuse experiences vs. face-to-
face comparison group vs. online comparison group; foster care group vs. total comparison group vs. severe maltreatment comparison
group) we used ANOVAs (F-tests) and planned contrasts for pair-wise comparisons, using SPSS 22 (IBM Corp, 2013). We used path
analysis to assess possible direct and indirect effects from CM on current posttraumatic psychopathology (PTSD symptoms and BSI
global severity index). Adult life events were defined as mediator (see Fig. 2). The path model was specified using robust maximum
likelihood estimation using the R-package lavaan (Rosseel, 2012). Missing values were treated with the full information maximum
likelihood method that is implemented in lavaan. We defined a just identified model which naturally leads to perfect model fit.
Therefore we did not report any fit indices. We primarily aimed to investigate the direct and indirect associations of CM with
symptom distress, mediated by adult life events. Confidence intervals of the mediation effects were calculated by bootstrapping
technique (number of bootstraps = 5000). The bivariate correlations of the included variables are reported in Table 6.

3. Results

3.1. Child maltreatment and adult life events

The severity of child maltreatment and the exposure rates of adult life events in both groups is reported in Table 2. Individuals in
the foster care group had significantly higher values in all types of child maltreatment than the comparison group (Table 3). Adult
survivors reported higher exposure rates with up to 13 out of 16 adult life events, although not all differences were significant. They
reported higher levels in both, revictimizing (e.g. physical assault) and non-revicitmizing (e.g. exposure to toxic substance) events. In
order to compare the situation before the foster care placement happened, we compared the CTQ reports from the foster care group
regarding their family of origin to the comparison group and again found that the foster care group experienced significantly more
child maltreatment within their family than the comparison group. The foster care group also experienced more adult life events than
the comparison group. The online comparison group reported higher values in the scales emotional abuse, emotional neglect, and
physical neglect than the face-to-face comparison group, as well as more adult life events (Table 3).

3.2. Prevalence of mental disorders and symptom distress

The prevalence of mental disorders are reported in Table 4. The most prevalent diagnoses were lifetime PTSD (56.4%) and
lifetime depression (51.6%) for the foster care group, and lifetime depression (27.9%) for the comparison group. The most prevalent
personality disorder was paranoid personality disorder (24.7%) in the foster care group and compulsive personality disorder in the
comparison group (5.2%). Almost all disorders were more prevalent in the foster care group than in the comparison group.
The results of the dimensional reports of psychopathological distress are reported in Table 5. Very similar to the diagnoses, the
foster care group reported more distress in all scales. The subsample of the comparison group, who reported the most severe child
abuse, also suffered from less severe psychopathological distress in many, but not in all scales than the foster care group (Table 5). It
is important to note that this comparison group subsample with the most severe child maltreatment still had significantly lower
values in the CTQ scales physical abuse, sexual abuse, emotional neglect, and physical neglect; however, they did not differ sig-
nificantly in the emotional abuse scale.

494
B. Lueger-Schuster et al. Child Abuse & Neglect 76 (2018) 488–501

Table 2
Characteristics of child maltreatment and adult life events.

Foster care group (%) Comparison group (%)

1 2 3 4 1 2 3 4

a
Child maltreatment
Emotional abuse 40.4/4.6 9.6/7.3 10.8/16.4 39.2/71.7 54.1 19.3 13.3 13.3
Physical abuse 42.2/5.9 7.2/6.4 7.2/11.4 43.4/76.4 74.7 8.6 8.2 8.6
Sexual abuse 73.3/36.5 4.8/6.4 6.7/10.5 15.2/46.6 72.1 12.0 9.0 6.9
Emotional neglect 18.1/1.8 15.1/3.7 7.8/7.3 59.0/87.2 46.8 30.5 10.3 12.4
Physical neglect 26.7/6.4 9.1/12.3 13.9/23.2 50.3/58.2 66.5 16.7 12.0 4.7

% yes a % yes
No maltreatment 7.7/0.0 27.5
One type of maltreatment 12.0/1.8 19.7
Two types of maltreatment 13.9/0.0 18.9
Three types of maltreatment 12.7/8.6 14.2
Four types of maltreatment 34.9/31.8 12.9
Five types of maltreatment 16.3/57.7 6.9

Adult life events

Natural disaster 28.6 29.7
Fire or explosion 24.5 19.0
Transportation accident 60.7 55.8
Serious accident 34.5 30.0
Exposure to toxic substance 15.0* 8.2
Physical assault 58.6*** 30.9
Assault with a weapon 42.3*** 12.9
Sexual assault 27.7** 16.7
Other unwanted or uncomfortable sexual experience 30.9 31.8
Combat or exposure to a war-zone 9.5 10.7
Captivity 9.6*** 1.3
Life-threatening illness or injury 45.7 41.2
Severe human suffering 57.7*** 39.5
Sudden violent death 35.5*** 20.2
Sudden accidental death 28.4** 16.3
Serious injury, harm, or death caused to someone else 19.1*** 3.9
Other 38.6 37.8

Note. Numbers represent the percentage of the sample that was exposed to a certain type and severity of child maltreatment and adult life events. Statistical
significance of differences is only reported for adult life events. For statistical significance of differences in CM, refer to Table 3; 1 = none to minimal, 2 = slight to
moderate, 3 = moderate to severe, 4 = severe to extreme (Häuser et al., 2011).
a
Numbers on the left represent exposure within a family context (before foster care placement, N = 166), numbers on the right represent exposure during foster
care; *p < .05; ** p < .01; *** p < .001.

Table 3
Child maltreatment, adult life events, and group comparisons.

Group Group comparison

b c
Foster Care Comparison t-test Contrasts

Intra-familiala Foster care Total Face-to-face Online Total 1 2

M (SD) M (SD) M (SD) M (SD) M (SD) M (SD) p p p

Emotional abuse 2.67 (1.44) 3.68 (0.98) 3.33 (1.04) 1.82 (0.91) 2.11 (1.03) 1.93 (0.96) < 0.001 < 0.001 0.034
Physical abuse 2.43 (1.43) 3.45 (1.14) 3.11 (1.11) 1.38 (0.71) 1.52 (0.88) 1.43 (0.78) < 0.001 < 0.001 0.183
Sexual abuse 1.53 (1.09) 2.47 (1.47) 2.14 (1.18) 1.31 (0.70) 1.27 (0.69) 1.30 (0.69) < 0.001 < 0.001 0.653
Emotional neglect 3.52 (1.35) 4.30 (0.73) 4.00 (0.83) 1.99 (0.95) 2.51 (1.05) 2.19 (1.02) < 0.001 < 0.001 < 0.001
Physical neglect 2.60 (1.21) 2.69 (0.77) 2.68 (0.78) 1.33 (0.48) 1.55 (0.60) 1.41 (0.54) < 0.001 < 0.001 0.004
CTQ total 2.55 (1.04) 3.32 (0.73) 3.05 (0.76) 1.57 (0.59) 1.79 (0.65) 1.65 (0.63) < 0.001 < 0.001 0.008
LEC sum – – 5.65 (3.10) 4.27 (2.90) 3.71 (2.93) 4.06 (2.92) < 0.001 – 0.166

Note. Data represent average item-level responses per dimension of the CTQ (ranging from one to five) and the number of experienced events according to the LEC
(possibly ranging from zero to 17); aN = 166; bt-tests comparing total foster care group values with total comparison group values: all t(451) > 5.65; cANOVAs with
equal variances for LEC data and unequal variances for CTQ data comparing intra-familial foster care group vs. face-to-face comparison group vs. online comparison
group: all F(2, 396) > 3.43, all p < .05; planned contrasts for CTQ: 1 = Foster care group intra-familial vs. total comparison group, 2 = face-to-face comparison group
vs. online comparison group; planned contrasts for LEC: 1 = Foster care group total sum of life events vs. total comparison group total sum of life events, 2 = face-to-
face comparison group vs. online comparison group.

495
B. Lueger-Schuster et al. Child Abuse & Neglect 76 (2018) 488–501

Table 4
SCID I and SCID II diagnoses.

c
Group Group comparison

a b
Diagnosis Foster care Comparison total

n (%) n (%) p

Mood disorders
Major depressive episode, current 31 (14.2) 0 (0.0) < 0.001
Major depressive episode, lifetime 113 (51.6) 43 (27.9) < 0.001
Dysthymia, current 19 (8.7) 1 (0.6) < 0.001
Alcohol or substance use disorders
Alcohol dependency, current 11 (5.0) 2 (1.3) 0.082
Alcohol dependency, lifetime 58 (26.5) 10 (6.5) < 0.001
Substance dependency, current 22 (10.0) 0 (0.0) < 0.001
Substance dependency, lifetime 33 (15.1) 0 (0.0) < 0.001
Anxiety disorders
Panic disorder, current and lifetime 58 (26.5) 9 (5.8) < 0.001
Social phobia, current and lifetime 37 (16.9) 3 (1.9) < 0.001
Specific phobia, current and lifetime 54 (24.7) 10 (6.5) < 0.001
Obsessive compulsive disorder, current and lifetime 23 (10.5) 2 (1.3) < 0.001
PTSD, current 78 (35.6) 1 (0.6) < 0.001
PTSD, lifetime 123 (56.4) 1 (0.6) < 0.001
Generalized anxiety disorder, current and lifetime 22 (10.0) 2 (1.3) < 0.001
Personality disorders
Avoidant personality disorder 38 (17.4) 3 (1.9) < 0.001
Compulsive personality disorder 34 (15.5) 8 (5.2) 0.001
Paranoid personality disorder 54 (24.7) 2 (1.3) < 0.001
Schizotypical personality disorder 6 (2.7) 1 (0.6) .247 d
Borderline personality disorder 50 (22.8) 0 (0.0) < 0.001
Antisocial personality disorder 18 (8.2) 0 (0.0) < 0.001

Note. a n = 219: sample size differs from total sample size due to incomplete SCID interviews; b n = 154: sample consists of n = 143 persons from the face-to-face
comparison group and n = 11 persons from the online comparison group who additionally participated in a face-to-face SCID interview; c Pearson’s χ2-test; d Fisher’s
exact test.

3.3. Predicting current psychopathological distress and adult life events

We conducted mediation models predicting current posttraumatic symptom distress (Fig. 2). In the foster care group, the model
explained 33.2% of the variability of PTSD symptom distress (R2 = 0.332). The types of CM covaried considerably, with correlation
coefficients ranging from r = 0.29 to r = 0.64. Only sexual abuse significantly predicted adult life events. PTSD was directly pre-
dicted by adult life events, emotional abuse, and sexual abuse. The effects of sexual abuse and physical neglect on PTSD were
mediated by adult life events (sexual abuse: estimator = 0.04, 95% CI = [.006,.071]; physical neglect: estimator = 0.04, 95%
CI = [.002,.107]) and the total effects of emotional abuse and sexual abuse, including the direct effect of each type of CM on PTSD
and the indirect effect via adult life events, were significant (emotional abuse: estimator = 0.33, 95% CI = [0.146,0.423]; sexual
abuse: estimator = 0.22, 95% CI = [0.060,0.265]). This means that physical neglect had no direct but an indirect effect on PTSD
symptom distress. No other mediation or total effect was significant.
In the comparison group, 24.5% of the variability of PTSD was explained by the model (R2 = 0.245). The CM types also covaried
considerably (r = 0.24–r = 0.76). In this group, emotional abuse significantly predicted adult life events, which in turn predicted
PTSD. Emotional neglect was the only type of CM that predicted PTSD (Fig. 2). The effect of childhood emotional abuse on PTSD was
mediated by adult life events (estimator = 0.05, 95% CI = [.004,.084]) and the total effect of emotional neglect was significant
(estimator = 0.28, 95% CI = [0.025,0.281]. In this group, emotional abuse had no direct, but an indirect effect on PTSD symptom
distress. No other mediation or total effect was significant.
We then estimated similar mediation models for both groups, with general psychopathological symptom distress as outcome
variable (BSI GSI index). The model explained 26.6% respectively 18.6% of the variability of general symptom distress in the foster
care group and in the comparison group (R2 = 0.266; R2 = 0.186, resp.). The pattern of mediation and total effects was the very same
as in the model predicting PTSD in both groups.

4. Discussion

The aim of this study was to provide an overview of the main findings of the Vienna Institutional Abuse Study (VIA-S) and to test
several hypothesis with regard to the severity of consequences on mental health of survivors of childhood maltreatment in institutional
foster care settings. We examined the extent of childhood trauma and adult life events with regard to current adult mental health in a
sample of survivors of IA and persons of a comparison group who were never in foster care but also included persons who experienced CM
within their families. In the following sections, we will discuss our findings according to the order of the presentation of our results.

496
B. Lueger-Schuster et al. Child Abuse & Neglect 76 (2018) 488–501

Table 5
Dimensional self-reported psychopathological distress and group comparisons.

Group

a b
Foster care Comparison Comparison - severe CM Group comparison

PCL-5 M (SD) M (SD) M (SD) 1 2

Intrusion 7.92 (6.35) 2.56 (3.26) 4.03 (4.05) < 0.001 < 0.001
Avoidance 3.19 (2.72) 1.27 (1.90) 1.95 (2.24) < 0.001 0.001
NACM 8.26 (6.80) 3.53 (4.50) 6.47 (5.83) < 0.001 0.039
Arousal 8.91 (5.43) 4.07 (4.07) 6.03 (4.88) < 0.001 0.001
PCL-5 sum score 28.28 (18.19) 11.43 (11.52) 18.48 (14.37) < 0.001 < 0.001

BSI
Depression 6.10 (6.13) 2.83 (4.44) 4.83 (5.64) < 0.001 0.118
Anxiety 6.78 (5.82) 2.94 (3.56) 4.32 (4.18) < 0.001 < 0.001
Somatization 6.31 (5.47) 2.85 (3.85) 3.83 (4.51) < 0.001 0.001
Paranoid Ideation 7.16 (4.80) 2.98 (3.44) 4.45 (3.63) < 0.001 < 0.001
BSI-18 GSI 19.20 (15.20) 8.62 (10.25) 12.98 (12.06) < 0.001 0.001

ITQ
Re-experiencing 5.31 (3.93) 2.29 (2.67) 3.40 (3.09) < 0.001 < 0.001
Avoidance 3.50 (2.77) 1.63 (2.25) 2.42 (2.42) < 0.001 0.003
Sense of threat 4.21 (2.68) 1.84 (2.09) 2.61 (2.36) < 0.001 < 0.001
Affect dysregulation hyperactivation 9.89 (4.67) 6.20 (3.82) 7.62 (3.96) < 0.001 < 0.001
Affect dysregulation deactivation 6.37 (4.81) 2.54 (3.32) 4.20 (3.87) < 0.001 0.001
Dissociation 3.47 (2.92) 1.26 (1.96) 2.03 (2.32) < 0.001 < 0.001
Negative self-concept 4.53 (4.57) 2.60 (3.28) 4.41 (4.03) 0.004 0.840
Disturbed relationships 4.57 (3.89) 2.17 (2.72) 3.80 (3.29) < 0.001 0.113
ITQ sum score 38.38 (20.59) 19.18 (14.66) 28.37 (16.61) < 0.001 < 0.001

Note. a n = 66: persons in comparison sample who reported moderate to severe or severe to extreme child maltreatment in at least two types of maltreatment; b
ANOVAs comparing foster care group vs. total comparison group vs. severe maltreatment comparison group: all F(2, 432) > 28.84, all p < .001; planned contrasts for
CTQ: 1 = Foster care group vs. lower maltreatment comparison group + severe maltreatment comparison group (=total comparison group). 2 = Foster care group vs.
severe maltreatment comparison group; Note that the Comparison-column in this table reports the results of the total comparison group, the results of the lower
maltreatment comparison group are not reported here

4.1. Child maltreatment and adult life events

Clearly, individuals in the foster care group reported significantly more CM exposure and severity in all types of CM than the
comparison group. However, these differences even persisted when comparing the foster care group to those participants of the
comparison group, who reported the most severe CM experiences within their families. The foster care group also experienced more
severe CM in their families of origin before being taken to foster care compared to the comparison group. This is also not surprising,
as children from problematic backgrounds and with traumatic experiences were those who were more likely to be taken into foster
care (cf. Greeson et al., 2011).
Multiple forms of CM, however, are reported throughout the general population without leading to be taken into foster care, thus
we expected to also find CM reports in the comparison group (Centers for Disease Control and Prevention (CDC), 2010; Finkelhor
et al., 2009). Our comparison group also reported various experiences of CM, however to a lesser extent than the foster care group.

Table 6
Bivariate correlations of variables used in mediation analysis for foster care group (upper triangle) and comparison group (lower triangle).

1 2 3 4 5 6 7 8 9 10 11

1 Emotional abuse .64** .41** .50** .54** .83** .15* .48** .39** −0.23** .28**
2 Physical abuse .63** .40** .52** .50** .82** .20** .37** .30** −0.10 < −0.01
3 Sexual abuse .34** .27** .25** .34** .64** .23** .34** .29** −0.26** 0.05
4 Emotional neglect .70** .57** .24** .57** .71** 0.10 .26** .27** −0.08 0.11
5 Physical neglect .46** .41** .19** .64** .75** .20** .36** .24** −0.18** 0.05
6 CTQ total .86** .70** .43** .91** .71** .21** .49** .41** −0.23** 0.13
7 LEC .31** .31** .25** .19** .16* .31** .31** .30** −0.08 −0.16*
8 PCL-5 .38** .37** .16* .39** .28** .42** .28** .82** −0.36** 0.08
9 BSI GSI .42** .31** .16* .42** .31** .44** .38** .66** −0.23** 0.02
10 Age −0.07 0.04 0.11 0.02 .14* 0.04 0.08 0.12 −0.01 −0.15*
11 Gender .13* −0.03 .22** 0.01 −0.02 0.06 −0.01 0.04 0.04 0.02

Note. Data are Pearson correlations coefficients (point-biserial correlation coefficient for gender); Abuse and neglect variables represent CTQ data; CTQ total -
Childhood trauma questionnaire total score; LEC - number of life events reported in the Life events checklist; PCL-5–total score of PTSD checklist for DSM-5; BSI GSI -
Global severity index of the Brief symptom inventory 18.
* p < .05; ** p < .01.

497
B. Lueger-Schuster et al. Child Abuse & Neglect 76 (2018) 488–501

Within the comparison group, the online comparison group disclosed higher rates in emotional abuse, emotional neglect, and
physical neglect compared with the face-to-face comparison group. Previous research also found increased disclosure online, which
may lead to a more accurate estimation of the true prevalence of CM within our comparison group (Buchanan, 2003).
The foster care group showed higher rates of adult life events. Previous research suggests several pathways from CM to adult re-
victimization such as risk-taking behavior, problematic alcohol use, or emotion dysregulation (Pratchett & Yehuda, 2011). However,
we also found higher rates in some non-victimizing events. Even though existing models on repeated traumatization focus on re-
victimizing events, it also appears likely that risk-taking behavior leads to greater risk of witnessing sudden violent or accidental
death. Furthermore, the lower socio-economic background of the IA-survivors as compared to the general population, increases the
likelihood of having to live in surroundings that bare higher chances of experiencing non-revictimizing events (Brady & Mathews,
2002). This lower socio-economic background may be seen to certain extent as being strongly influenced by the survivors’ traumatic
childhood (Brattström, Eriksson, Larsson, & Oldner, 2015).

4.2. Prevalence of mental disorders and symptom distress

Adult psychopathology in the foster care group included lifetime mood disorders, alcohol and substance dependency, anxiety
disorders, PTSD, and personality disorders. This is in accordance with previous research on CM (Kessler et al., 2010) and IA (Carr
et al., 2010), that also investigated these psychopathologies. However, when comparing our results in the foster care group with
similar previous studies, there was a differential picture regarding prevalence rates across disorders (Carr et al., 2010). Compared to
Carr et al. (2010), who studied survivors of IA in Ireland, we found higher prevalence rates of current and lifetime PTSD, borderline
personality disorder, and lifetime major depression. We found similar rates of current major depression, current and lifetime alcohol
dependency, avoidant and antisocial personality disorder. The rate of generalized anxiety disorder was lower in our sample compared
to Carr et al. (2010). Apart from random effects and sampling biases, a possible explanation for the higher prevalence rates may be
rooted in differences of institutional settings. While in the Irish sample all institutions where Catholic institutions, in our sample
survivors were in institutions that were under governmental control. However, it is unclear why participants from our study tend to
suffer from higher rates of mental disorders, and we suppose that factors following foster care, such as the possibility to engage in
treatment, may account for these differences. Studies comparing the consequences of IA across countries are still lacking.
Individuals in the comparison group showed significantly lower rates in almost all disorders and psychopathological symptoms.
Self-reported symptoms of PTSD, complex PTSD, and global distress yielded significant differences between the foster care group and
the subsample of the comparison group with severe CM in almost all aspects. Only negative alterations in cognitions and mood
(NACM), negative self-concept, and depression did not show a significant difference between these two groups. The variety of
disorders and symptoms has also been previously found for IA (Carr et al., 2010; Lueger-Schuster et al., 2013a, 2013b), however until
now no direct comparison has been made.

4.3. Predicting current psychopathological distress and adult life events

In line with our hypothesis, we found a significant dose-response relation between the extent of traumatization and the extent and
severity of psychopathology for both groups. Steine et al. (2017) showed that there was a significant dose-response relation of
cumulative CM with various psychopathological symptoms, but also social and occupational functioning. This is also reflected in our
results, showing more severe problems for the foster care group, underlining the special need of the survivors for support and mental
health care.
Our results also corroborate previous findings that stress the relevance of CM in private contexts for later adult psychopathology
(Fitzhenry et al., 2015). They further add to the existing research regarding the psychopathological burden of IA on the adult
survivors’ mental health (Carr et al., 2010; Lueger-Schuster et al., 2013a, 2013b), showing a broad spectrum of adverse psycho-
pathological sequelae in the foster care group. In our sample, sexual abuse and neglect predicted adult life events, PTSD, and distress
symptoms differentially in both groups. EA showed the strongest effect on adult psychopathology in the foster care group when
including all types of CM and adult life events in the model; the detrimental effect of EA has also been reported in other research (cf.
Norman et al., 2012). As expected, adult life events and SA also directly predicted symptom distress. Interestingly, emotional neglect
and adult life events were the only significant direct predictors in the comparison group, when all variables were included in the
model. While there was a mediation effect in the foster care group between SA and PN and adult life events, there was a mediation
effect between EA and adult life events in the comparison group. PN in the foster care group and EA in the comparison group are of
particular interest, because they both did not show a significant direct association with symptoms distress, but only indirectly via
adult life events. In a systematic review, Carr et al. (2013) found that results regarding PN as predictor of adult psychiatric disorders
are inconsistent. These inconsistencies might be clarified when adult re-victimization is taken into account. On the other side, the
impact of EA on symptom distress was mediated by adult life events in the comparison group and there was no direct effect. Also, in
the comparison group, other types of CM such as SA and PA that typically are associated with symptom distress, did not predict PTSD
in our study. We suppose that limitations in sample size and sampling procedure aspects can account for these results. Larger sample
sizes might detect smaller effects and the comparison sample in this study was a voluntary group that might be particularly resilient.
Despite the predictive power of adult life events, single types of CM were significant predictors in either one of the groups, but not
type of CM was a predictor in both groups. This supports the hypothesis that CM experienced in foster care settings differs in its
impact on adult mental health, not only in a dose-response fashion, but also in a qualitatively different pattern. A reason for these
differences can be the additional adversities that are inherent in abusive foster care settings, as stated by the institutional betrayal

498
B. Lueger-Schuster et al. Child Abuse & Neglect 76 (2018) 488–501

theory (Smith & Freyd, 2014). This theory argues that violence exposure within an institutional setting is particularly detrimental
because of the specific environment surrounding these events. Smith and Freyd (2013) found that college women who were exposed
to sexual violence suffered from more severe symptoms if they experienced institutional betrayal. In our study, sexual abuse was no
predictor of adult psychopathology in the comparison sample when all other traumatic events were included in the model. In
contrast, sexual abuse in institutions predicted adult psychopathology exceeding the predictive power of other traumatic events and
thus seems to have a unique impact that may be explained by institutional betrayal. This supports the hypothesis that the long term
consequences of sexual abuse in institutional settings go beyond the consequences of severe CM.
In the foster care group, only SA predicted adult life events, and in the comparison group, only EA predicted adult life events.
Previous research, however, found a relationship for all forms of CM (physical, sexual, and emotional) and neglect with adult life
events (Briere & Elliott, 2003; Widom et al., 2008), notably none of these studies was conducted for CM in institutions; in our study,
the bivariate analysis showed meaningful associations of several types of CM with adult life events in both groups. This finding
underlines the importance to consider different types of abuse and neglect simultaneously in the same model that might otherwise
cover differential associations based on high co-occurrence of various forms of CM (Vachon et al., 2015). Being re-victimized in
adulthood is not only associated with the experience of CM per se, but also with psychopathological symptom distress (Auslander,
Tlapek, Threlfall, Edmond, & Dunn, 2015). It is possible that there is a reciprocal relationship between the experience of traumatic
events and symptoms distress. However, as the present study did not include longitudinal data, we were unable to assess such a
relationship.
Taking all our results together, we could replicate the often reported associations of CM, adult life events, and adult mental health
problems. Above that, we found that CM in institutional settings is especially associated with adult psychopathology, which calls for
specific treatment strategies. Flanagan et al. (2009) report that adult survivors of CM in institutions who use non-avoidant coping
strategies were more resilient than those with avoidant coping strategies. Furthermore, resilient survivors were characterized by a
secure attachment style. This style is more unlikely to develop in the contexts IA-survivors had to face in their childhood (Muller,
Sicolo, & Lemieux, 2000). We conclude from our findings that adult survivors of IA suffer from a wide range of mental disorders and
thus need evidence-based psychological treatments (Schnyder, 2015). However, these treatments should be accompanied by tools to
reduce patterns enhancing the risk of repeated traumatization, enhance skills to cope with aversive social situations in a non-avoidant
way, and strengthen a secure attachment style (Cloitre & Schmidt, 2015; Fitzpatrick et al., 2010).

4.4. Limitations

Our study is limited by several factors. First of all we addressed a specific sample of survivors of IA, who were willing to
participate. Thus we are not able to generalize our results to the much larger group of survivors who did not follow the invitation to
participate. The severity and complexity of trauma reported by IA survivors is rather unique and may not be applied to the general
population. However, our results are comparable to other studies with survivors of IA. We must also consider a tendency of un-
derreporting due to the retrospective and self-reported nature of our data which has also been reported and discussed elsewhere
(Edwards, Holden, Felitti, & Anda, 2003). Second, the comparison group includes two formats of assessment, showing a tendency of
the online group to have a higher exposure in CA, which may be due to the more anonymous format of an online questionnaire (cf.
Buchanan, 2003). Additionally, the groups differed in some parameters: in the foster care group were more males and this group had
lower levels of education and average income. Third, we announced the comparison group as part of a larger study addressing the
long-term consequences of IA and it is therefore possible that this information has led to a priming effect within the comparison
group. Fourth, we did not assess dissociation with a diagnostic interview, even though it is a common response to traumatic ex-
periences. Finally, causal inference should be done with care, because a cross sectional study does not include information on
temporal sequencing.

4.5. Conclusion and outlook

So far, studies that compare IA and CM are rare. Future research in this field should integrate biological measures to test existing
considerations for the toxic impact of IA (Anda et al., 2006; O’Donovan et al., 2012). Furthermore, it needs to be elucidated in future
research in how far the specific dynamics within an institutional setting contribute to the noxious effect of IA, possibly beyond the
long lasting consequences of CM in non-institutional settings (Smith & Freyd, 2013). Longitudinal studies on the development of
psychopathology after the experience of CM and IA would aid a better understanding of the dynamics and processes. There is a need
for treatment studies with severely traumatized individuals such as survivors of IA to better understand the mitigation of psycho-
pathological symptoms that are rooted in the abusive childhood experiences within institutions. Finally, strategies of primary pre-
vention of CM in family contexts as well as in foster care contexts need to be systematically applied and evaluated (Mikton &
Butchart, 2009).

Funding

This work was funded by the Austrian Science Fund[FWF; grant number P 26584].

499
B. Lueger-Schuster et al. Child Abuse & Neglect 76 (2018) 488–501

References

Anda, R. F., Felitti, V. J., Bremner, J. D., Walker, J. D., Whitfield, C., Perry, B. D., & Giles, W. H. (2006). The enduring effects of abuse and related adverse experiences
in childhood. European Archives of Psychiatry and Clinical Neuroscience, 256, 174–186.
Auslander, W., Tlapek, S. M., Threlfall, J., Edmond, T., & Dunn, J. (2015). Mental health pathways linking childhood maltreatment to interpersonal revictimization
during adolescence for girls in the Child Welfare System. Journal of Interpersonal Violence, 1–23. http://dx.doi.org/10.1177/0886260515614561.
Bernstein, D. P., & Fink, L. (1998). Childhood trauma questionnaire: a retrospective self-reportPsychological Corporation [Manual].
Brännström, L., Forsman, H., Vinnerljung, B., & Almquist, Y. B. (2016). The truly disadvantaged? Midlife outcome dynamics of individuals with experiences of out-of-
home care. Child Abuse & Neglect, (67), 408–418. http://dx.doi.org/10.1016/j.chiabu.2016.11.009.
Brady, S. S., & Mathews, K. A. (2002). The influence of socioeconomic status and ethnicity on adolescents’ exposure to stressful life events. Journal of Pediatric
Psychology, 27, 575–583.
Brattström, O., Eriksson, M., Larsson, E., & Oldner, A. (2015). Socio-economic status and co-morbidity as risk factors for trauma. European Journal of Epidemiology, 30,
151–157.
Briere, J., & Elliott, D. M. (2003). Prevalence and psychological sequelae of self-reported childhood physical and sexual abuse in a general population sample of men
and women. Child Abuse & Neglect, 27, 1205–1222. http://dx.doi.org/10.1016/j.chiabu.2003.09.008.
Buchanan, T. (2003). Internet-based questionnaire assessment: appropriate use in clinical contexts. Cognitive Behaviour Therapy, 32, 100–109. http://dx.doi.org/10.
1080/16506070310000957.
Carr, A., Dooley, B., Fitzpatrick, M., Flanagan, E., Flanagan-Howard, R., Tierney, K., & Egan, J. (2010). Adult adjustment of survivors of institutional child abuse in
Ireland. Child Abuse & Neglect, 34, 477–489. http://dx.doi.org/10.1016/j.chiabu.2009.11.003.
Carr, C. P., Martins, C. M. S., Stingel, A. M., Lemgruber, V. B., & Juruena, M. F. (2013). The role of early life stress in adult psychiatric disorders: a Systematic Review
According to Childhood Trauma Subtypes. The Journal of Nervous and Mental Disease, 201, 1007–1020.
Centers for Disease Control, & Prevention (CDC) (2010). Adverse childhood experiences reported by adults -five states, 2009. Morbidity and mortality weekly report, 59,
MMWR1609–1613.
Centers for Disease Control and Prevention (CDC) (2015). Behavioral risk factor surveillance system survey ACE module data,2010 . atlanta, Georgia: U.S. department of
health and human services. Centers for Disease Control and Prevention. [Retrieved from:] https://www.cdc.gov/violenceprevention/acestudy.
Classen, C. C., Palesh, O. G., & Aggarwal, R. (2005). Sexual revictimization: A review of the empirical literature. Trauma, Violence & Abuse, 6, 103–129. http://dx.doi.
org/10.1177/1524838005275087.
Cloitre, M., & Schmidt, J. A. (2015). STAIR narrative therapy. In U. Schnyder (Ed.). Evidence based treatments for trauma-related psychological disorders. A practical guide
for clinicians (pp. 277–297). Cham: Springer International Publishing.
Cloitre, M., Roberts, N., Bisson, J., & Brewin, C. R. (2013). International trauma questionnaire for ICD-11 PTSD and CPTSD. [Research version].
Derogatis, L. R. (1994). SCL -90-R Symptom Checklist -90-R administration, scoring and procedures manual. Minneapolis: MN: National Computer Systems.
Edwards, V. J., Holden, G. W., Felitti, V. J., & Anda, R. F. (2003). Relationship between multiple forms of childhood maltreatment and adult mental health in
community respondents: results from the adverse childhood experiences study. American Journal of Psychiatry, 160, 1453–1460.
Ehring, T., Knaevelsrud, C., Krüger, A., & Schäfer, I. (2014a). Life events checklist für DSM-5 (LEC-5). [Deutsche Version, Retrieved from:] http://zep-hh.de/service/
diagnostik/.
Ehring, T., Knaevelsrud, C., Krüger, A., & Schäfer, I. (2014b). http://zep-hh.de/service/diagnostik/. PTSD checkliste für DSM-5 (PCL-5) [Deutsche Version, Retrieved
from:].
Finkelhor, D., Ormrod, R. K., & Turner, H. A. (2009). Lifetime assessment of poly-victimization in a national sample of children and youth. Child Abuse & Neglect, 33,
403–411.
Fitzhenry, M., Harte, E., Carr, A., Keenleyside, M., O’Hanrahan, K., White, M. D., & Browne, S. (2015). Child maltreatment and adult psychopathology in an Irish
context. Child Abuse & Neglect, 45, 101–107. http://dx.doi.org/10.1016/j.chiabu.2015.04.021.
Fitzpatrick, M., Carr, A., Dooley, B., Flanagan-Howard, R., Flanagan, E., Tierney, K., & Egan, J. (2010). Profiles of adult survivors of severe sexual, physical and
emotional institutional abuse in Ireland. Child Abuse Review, 19, 387–404. http://dx.doi.org/10.1002/car.1083.
Flanagan, E., Carr, A., Dooley, B., Fitzpatrick, M., Flanagan-Howard, R., Shevlin, M., & Egan, J. (2009). Profiles of resilient survivors of institutional abuse in Ireland.
International Journal of Child & Family Welfare, 12, 56–73.
Franke, G. H., & Derogatis, L. R. (2000). BSI: the brief sympton inventory von L.R. derogatis (Kurzform der SCL 90 R). [Deutsche Version. Göttingen : Beltz Test].
Franke, G. H., Ankerhold, A., Haase, M., Jäger, S., Tögel, C., Ulrich, C., & Frommer, J. (2011). Der Einsatz des Brief Symptom Inventory 18 (BSI-18) bei
Psychotherapiepatienten. [The usefulness of the Brief Symptom Inventory 18 (BSI-18) in psychotherapeutic patients]. Psychotherapie· Psychosomatik· Medizinische
Psychologie, 61, 82–86.
Freyd, J. J. (1994). Betrayal trauma: traumatic amnesia as an adaptive response to childhood abuse. Ethics & Behavior, 4, 307–329. http://dx.doi.org/10.1207/
s15327019eb0404_1.
Gilbert, R., Fluke, J., O’Donnell, M., Gonzalez-Izquierdo, A., Brownell, M., Gulliver, P., & Sidebotham, P. (2012). Child maltreatment: variation in trends and policies in
six developed countries. The Lancet, 379, 758–772. http://dx.doi.org/10.1016/S0140-6736(11)61087-8.
Greeson, J. K. P., Briggs, E. C., Kisiel, C. L., Layne, C. M., Ake, G. S., III, Ko, S. J., & Fairbank, J. A. (2011). Complex trauma and mental health in children and
adolescents placed in foster care: findings from the National Child Traumatic Stress Network. Child Welfare, 90, 91–108.
Häuser, W., Schmutzer, G., Brahler, E., & Glaesmer, H. (2011). Maltreatment in childhood and adolescence: results from a survey of a representative sample of the
German population. Deutsches Arzteblatt International, 108, 287–294. http://dx.doi.org/10.3238/arztebl.2011.0287.
Hönigsberger, G., & Schrenk, J. (2011). Kinderheim des Grauens: wir wurden alle vergewaltigt und verkauft.. Kurier, 15. [Retrieved from:] https://kurier.at/chronik/wien/
kinderheim-des-grauens-wir-wurden-alle-vergewaltigt-und-verkauft/733.814 October 15 .
Corp, I. B. M. (2013). IBM SPSS statistics for windows, version 22.0. Armonk, NY: IBM Corp.
Karatzias, T., Shevlin, M., Fyvie, C., Hyland, P., Efthymiadou, E., Wilson, D., & Cloitre, M. (2016). An initial psychometric assessment of an ICD-11 based measure of
PTSD and complex PTSD (ICD-TQ): Evidence of construct validity. Journal of Anxiety Disorders, 44, 73–79. http://dx.doi.org/10.1016/j.janxdis.2016.10.009.
Karatzias, T., Shevlin, M., Fyvie, C., Hyland, P., Efthymiadou, E., Wilson, D., & Cloitre, M. (2017). Evidence of distinct profiles of posttraumatic stress disorder (PTSD)
and complex posttraumatic stress disorder (CPTSD) based on the new ICD-11 trauma questionnaire (ICD-TQ). Journal of Affective Disorders, 207, 181–187. http://
dx.doi.org/10.1016/j.jad.2016.09.032.
Kessler, R. C., McLaughlin, K. A., Green, J. G., Gruber, M. J., Sampson, N. A., Zaslavsky, A. M., & Williams, D. R. (2010). Childhood adversities and adult psycho-
pathology in the WHO world mental health surveys. The British Journal of Psychiatry, 197, 378–385.
Knefel, M., Lueger-Schuster, B., & Maercker, A. (2013). Internationaler traumafragebogen, forschungsversion. [Unveröffentlichte deutsche Version].
Kommission Wilheminenberg (2013). Endbericht der kommission wilhelminenberg. wien. Retrieved from: http://www.kommission-wilhelminenberg.at/presse/jun2013/
Bericht-Wilhelminenberg-web_code.pdf.
Krüger-Gottschalk, A., Knaevelsrud, C., Rau, H., Dyer, A., Schäfer, I., Schellong, J., & Ehring, T. (2017). The German version of the Posttraumatic Stress Disorder
Checklist for DSM-5 (PCL-5): psychometric properties and diagnostic utility. BMC psychiatry, 17(1), 379.
LaNoue, M., Graeber, D., Hernandez, B. U., de Warner, T. D., & Helitzer, D. L. (2012). Direct and indirect effects of childhood adversity on adult depression. Community
Mental Health Journal, 48, 187–192. http://dx.doi.org/10.1007/s10597-010-9369-2.
Lanius, R. A., Vermetten, E., Loewenstein, R. J., Brand, B., Schmahl, C., Bremner, J. D., & Spiegel, D. (2010). Emotion modulation in PTSD: Clinical and neurobiological
evidence for a dissociative subtype. The American Journal of Psychiatry, 167, 640–647.
Lueger-Schuster, B., Kantor, V., Weindl, D., Knefel, M., Moy, Y., Butollo, A., & Glück, T. (2013a). Institutional abuse of children in the Austrian Catholic Church: types
of abuse and impact on adult survivors… current mental health. Child Abuse & Neglect, 38, 52–64.
Lueger-Schuster, B., Weindl, D., Kantor, V., Knefel, M., Jagsch, R., & Butollo, A. (2013b). Psychotraumatologische fragestellungen zu sexuellem missbrauch und gewalt in

500
B. Lueger-Schuster et al. Child Abuse & Neglect 76 (2018) 488–501

einrichtungen des landes niederösterreich. wien. [Retrieved from:] http://ppcms.univie.ac.at/index.php?id=2858.

MacLean, K. (2003). The impact of institutionalization on child development. Development and Psychopathology, 15, 853–884. http://dx.doi.org/10.1017/
S0954579403000415.
Merz, E. C., & McCall, R. B. (2010). Behavior problems in children adopted from psychosocially depriving institutions. Journal of Abnormal Child Psychology, 38,
459–470.
Mikton, C., & Butchart, A. (2009). Child maltreatment prevention: a systematic review of reviews. Bulletin of the World Health Organization, 87, 353–361.
Muller, R., Sicolo, L. A., & Lemieux, K. E. (2000). Relationship between attachment style and posttraumatic stress symptomatology among adults who report the
experience of childhood abuse. Journal of Traumatic Stress, 13, 321–332.
Norman, R. E., Byambaa, M., De, R., Butchart, A., Scott, J., & Vos, T. (2012). The long-term health consequences of child physical abuse, emotional abuse, and neglect:
a systematic review and meta-analysis. PLoS Medicine, 9, e1001349. http://dx.doi.org/10.1371/journal.pmed.1001349.
O’Donovan, A., Tomiyama, A. J., Lin, J., Puterman, E., Adler, N. E., Kemeny, M., & Epel, E. S. (2012). Stress appraisals and cellular aging: A key role for anticipatory
threat in the relationship between psychological stress and telomere length. Brain, Behavior, and Immunity, 26, 573–579. http://dx.doi.org/10.1016/j.bbi.2012.01.
007.
Pratchett, L. C., & Yehuda, R. (2011). Foundations of posttraumatic stress disorder: Does early life trauma lead to adult posttraumatic stress disorder? Development and
Psychopathology, 23, 477–491. http://dx.doi.org/10.1017/S0954579411000186.
Reilly, T. (2003). Transition from care: status and outcomes of youth who age out of foster care. Child Welfare, 82, 727–746.
Rosseel, Y. (2012). lavaan: an R package for structural equation modeling. Journal of Statistical Software, 48, 1–36.
Evidence based treatments for trauma-related psychological disorders: a practical guide for clinicians. In U. Schnyder (Ed.). Cham: Springer International Publishing.
Shaw, M., & de Jong, M. (2012). Child abuse and neglect: A major public health issue and the role of child and adolescent mental health services. The Psychiatrist, 36,
321–325.
Smith, C. P., & Freyd, J. J. (2013). Dangerous safe havens: institutional betrayal exacerbates sexual trauma. Journal of Traumatic Stress, 26, 119–124. http://dx.doi.org/
10.1002/jts.21778.
Smith, C. P., & Freyd, J. J. (2014). Institutional betrayal. The American Psychologist, 69, 575–587. http://dx.doi.org/10.1037/a0037564.
Statistik Austria (2016). Bildung in zahlen 2014/2015: schlüsselindikatoren und analysen. [Retrieved from:] http://www.statistik.at/web_de/statistiken/menschen_und_
gesellschaft/bildung_und_kultur/bildungsstand_der_bevoelkerung/index.html.
Steine, I. M., Winje, D., Krystal, J. H., Bjorvatn, B., Milde, A. M., Gronli, J., & Pallesen, S. (2017). Cumulative childhood maltreatment and its dose-response relation
with adult symptomatology: findings in a sample of adult survivors of sexual abuse. Child Abuse & Neglect, 65, 99–111. http://dx.doi.org/10.1016/j.chiabu.2017.
01.008.
Vachon, D. D., Krueger, R. F., Rogosch, F. A., & Cicchetti, D. (2015). Assessment of the harmful psychiatric and behavioral effects of different forms of child
maltreatment. JAMA Psychiatry, 72, 1135–1142. http://dx.doi.org/10.1001/jamapsychiatry.2015.1792.
Weathers, F. W., Litz, B. T., Keane, T. M., Palmieri, P. A., Marx, B. P., & Schnurr, P. P. (2013a). The life events checklist for DSM-5 (LEC-5). Scale available from the
National Center for PTSD at www.ptsd.va.gov.
Weathers, F. W., Litz, B. T., Keane, T. M., Palmieri, P. A., Marx, B. P., & Schnurr, P. P. (2013b). The PTSD checklist for DSM-5 (PCL-5). Scale available from the National
Center for PTSD at www.ptsd.va.gov.
Widom, C. S., Czaja, S. J., & Dutton, M. A. (2008). Childhood victimization and lifetime revictimization. Child Abuse & Neglect, 32, 785–796. http://dx.doi.org/10.
1016/j.chiabu.2007.12.006.
Wingenfeld, K., Spitzer, C., Mensebach, C., Grabe, H. J., Hill, A., Gast, U., & Driessen, M. (2010). The german version of the childhood trauma questionnaire (CTQ):
preliminary psychometric properties [Die deutsche version des childhood trauma questionnaire (CTQ), 60, 442–450. http://dx.doi.org/10.1055/s-0030-1247564 Erste
Befunde zu den psychometrischen Kennwerten]. Psychotherapie, Psychosomatik, medizinische Psychologie.
Wittchen, H.-U., Wunderlich, U., Gruschwitz, S., & Zaudig, M. (1997). Strukturiertes klinisches interview für DSM-IV: achse I: psychische Störungen/Achse II: persönlich-
keitsstörungen. Göttingen: Hogrefe.
Wolfe, D. A., Jaffe, P. G., Jette, J. L., & Poisson, S. E. (2003). The impact of child abuse in community institutions and organizations: Advancing professional and
scientific understanding. Clinical Psychology: Science and Practice, 10, 179–191.
Wright, N. M., Smith, C. P., & Freyd, J. J. (2016). Experience of a lifetime: study abroad, trauma, and institutional betrayal. Journal of Aggression. Maltreatment &
Trauma, 26, 50–68.

501