Psychological Trauma: Theory, Research, Practice, and Policy 2016, Vol. 8, No. 2, 180 –187
© 2015 American Psychological Association 1942-9681/16/$12.00 http://dx.doi.org/10.1037/tra0000037
Adverse Childhood Experiences, Mental Health, and Quality of Life of Chilean Girls Placed in Foster Care: An Exploratory Study Annina Seiler
Stefanie Kohler and Martina Ruf-Leuschner
University Hospital Zurich, Zurich, Switzerland
University of Konstanz
Markus A. Landolt This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
University Children’s Hospital Zurich, Zurich, Switzerland In Latin America, little research has been conducted regarding exposure to adverse childhood experiences (ACEs), mental health, and health-related quality of life (HRQoL) among foster children. This study examined the association between ACEs and mental health, posttraumatic stress disorder (PTSD), and HRQoL in Chilean foster girls relative to age-matched Chilean family girls. Data were obtained from 27 Chilean foster girls and 27 Chilean girls ages 6 to 17 years living in family homes. Standardized self- and proxy-report measures were used. Foster girls reported more ACEs than controls in terms of familial and nonfamilial sexual abuse and both emotional and physical neglect. Girls living in foster care had a significantly higher rate of PTSD, displayed greater behavioral and emotional problems, and reported a lower HRQoL. Analysis confirmed the well-known cumulative risk hypothesis by demonstrating a significant positive association between the number of ACEs and PTSD symptom severity and a significant negative association with HRQoL. Chilean foster girls endured more ACEs that impair mental health and HRQoL than age-matched peers living with their families. These findings have implications for out-of-home care services in Latin America, highlighting the need to implement not only appropriate trauma-focused treatments but also appropriate prevention strategies. Keywords: adverse childhood experiences, foster children, mental health, posttraumatic stress disorder, health-related quality of life
al., 2009). Additional family-related risk factors, including domestic violence, family disruption, parental mental illness, poverty, and general life stress commonly co-occur with such familyrelated problems and often make out-of-home placement necessary for the child’s safety (Edwards, Holden, Felitti, & Anda, 2003). Today, associations between exposure to ACEs and physical and mental health problems in foster children and adolescents are well documented (Greeson et al., 2011; Oswald, Heil, & Goldbeck, 2010). In particular, research has consistently demonstrated that girls in foster care who have endured persistent maltreatment and neglect are especially vulnerable to developing behavioral problems and psychological disorders (McMillen et al., 2005). Symptoms of posttraumatic stress disorder (PTSD) and the occurrence of internalizing and externalizing behavioral problems have commonly been reported in maltreated children living in foster care in the United States and Europe (Greeson et al., 2011; McMillen et al., 2005; Oswald et al., 2010). Moreover, cumulative exposure to ACEs and traumatic events, also known as the building block effect, is assumed to be a robust predictor of adverse developmental outcomes in maltreated youths. Prior research reviews have consistently shown that the greater the number of ACEs, the greater the child’s risk of having behavioral problems and mental disorders (Appleyard, Egeland, van Dulmen, & Sroufe, 2005; Edwards et al., 2003; Raviv, Taussig, Culhane, & Garrido, 2010). Notably, children exposed to ACEs develop not only mental health problems but also an impaired health-related quality of life (HRQoL) (Jud, Landolt, Tatalias, Lach, & Lips, 2013; Villalonga-
In Chile in 2011, an estimated 10.4% of children and adolescents lived outside their family home, of whom more than half (54%) were girls (Servicio Nacional de Menores, Chile [SENAME], 2012). Before entering foster care, many of these children endured complex and multiple adverse childhood experiences (ACEs) such as emotional and physical neglect and/or physical and sexual abuse that occurred within their family or caregiving system (Hopenhayn et
This article was published Online First April 27, 2015. Annina Seiler, Department of Psychiatry and Psychotherapy, University Hospital Zurich, Zurich, Switzerland; Stefanie Kohler and Martina RufLeuschner, Department of Psychology, Clinical Psychology and Behavioral Neuroscience Unit, University of Konstanz; and Markus A. Landolt, Department of Psychosomatics and Psychiatry, University Children’s Hospital Zurich, Zurich, Switzerland, and Department of Child and Adolescent Health Psychology, Institute of Psychology, University of Zurich. We thank the girls who participated in this study and the directorate of the children’s home in Chile for allowing us to conduct the study. We would like to express our gratitude to Dr. Markus Zwicky and Gabriela Grimm, who initiated the study, established contact with the Chilean foster home, and generously contributed to the funding of this project. We also thank Dr. Maria Teresa Diez and Jeanine Schaelin for their help conducting the field study in Chile and Dorothea Isele for her methodological support. Correspondence concerning this article should be addressed to Markus A. Landolt, Department of Psychosomatics and Psychiatry, University Children’s Hospital Zurich, Steinwiesstrasse 75, CH-8032 Zurich, Switzerland. E-mail: [email protected] 180
This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
ADVERSE CHILDHOOD EXPERIENCES AND MENTAL HEALTH IN FOSTER GIRLS
Olives et al., 2010). HRQoL is a multidimensional construct composed of the individual’s perception of his or her functioning in a variety of aspects of life, including physical, psychological, and social domains (Mazur & Mierzejewska, 2003). Only two previous studies, one in Australia (Sawyer, Carbone, Searle, & Robinson, 2007) and one in Serbia (Damnjanovic, Lakic, Stevanovic, & Jovanovic, 2011), have examined the HRQoL of children and adolescents living in foster care. Both studies identified substantially poorer HRQoL in these youths compared to the general population. Moreover, symptoms of anxiety and depression accounted for significant variations in many domains of HRQoL in foster children. There is little research on the mental health of girls living in foster care in Latin America. However, cross-cultural research is important because the risk factors associated with child and adolescent mental health problems vary across different continents, especially between developed and developing countries (Fatori, Bordin, Curto, & de Paula, 2013). Evidence-based research on mental health and HRQoL in children living in foster care in Latin America is important to improve the quality of mental health services in foster care and to help professionals implement appropriate trauma-focused treatment in this region. To bridge these existing research gaps, the aims of the present exploratory study were threefold. First, we sought to examine whether Chilean girls living in foster care report more ACEs than girls living with their families. Second, we aimed to compare rates of PTSD and behavioral problems and HRQoL in these two groups. Third, we wanted to examine associations between ACEs and both behavioral problems and HRQoL among girls living in foster care. Based upon previous findings, we expected higher rates of ACEs in girls living in foster care versus family homes. Moreover, we assumed that foster home girls would have more mental health problems and lower HRQoL. We also hypothesized that the number of different types of ACEs would be positively associated with PTSD symptom severity and the number of behavioral problems and negatively associated with HRQoL.
Method Participants In February 2012, 27 girls living in a Chilean children’s home for sexually abused or emotionally and physically neglected girls were assessed regarding ACEs, mental health, and HRQoL. The children’s home was located in a rural region in the southern part of Chile and was under the direction of SENAME, a Chilean Youth Welfare Agency that aids child victims and children in conflict with the law (SENAME, 2012). The institution is largely funded by private donations, collections, fundraising events, and the agency’s own funds. Eight caregivers were responsible for taking care of a maximum of 30 girls between the ages of 3 and 18 years. Most of the girls had had to be separated from their biological parents for a variety of reasons that included violence, physical and sexual abuse, and emotional and/or physical neglect. The children’s home sought to reunify each girl with her family as soon as the family situation was considered stable again. Twenty-seven age-matched (⫾6 months) Chilean girls living in family homes served as a control group. They were recruited from a holiday swimming class in a small city close to the foster home.
181
This daily swimming class is held each year during the summer holidays. The course is aimed at children of low-income families who cannot afford to go on holidays or children whose parents have to work. To be eligible as controls, these girls had to live with their biological mother. For the sake of convenience, the girls living in foster homes shall be hereafter called “foster home girls”; those in the comparison group will be called “controls.”
Procedure Permission to conduct the study in the orphanage was granted by the directorate, who signed informed consents for all the girls in foster care. For girls living with their families, written informed consent was obtained from the mothers. Girls 14 years of age or older in both study groups signed the informed consent form independently. Furthermore, all girls who decided to participate in our study were informed verbally about the contents and goals of the research project and about their right to withdraw from the assessment at any time. All girls were informed that, according to Chilean laws, in case of reported ongoing maltreatment, the directorate would be notified in order to decide whether to inform local authorities. In the case of severe mental health problems, a referral to a local health professional was initiated upon consultation with the directorate of the orphanage. All girls gave their consent to participate in the study. All girls between the ages of 6 and 17 years were interviewed. In addition, proxy reports were obtained to assess emotional– behavioral difficulties, either from the girl’s mother (controls) or by any one of the orphanage caregivers who knew the girl well enough. Household income was assessed from the parents of control girls or was retrieved from the records for foster home girls. Records were also assessed for any psychotropic treatment. Standardized face-to-face interviews were conducted in a quiet, private room by two trained graduate students of psychology, one attending the University of Zurich, Switzerland, and the other attending the University of Konstanz, Germany. Both students were fluent in Spanish and required no interpreter during the interview. To ensure that all the girls could express their own views openly and confidentially, they were interviewed without their caregivers or parents present. As remuneration, each girl received a bracelet.
Measures Maltreatment and Abuse Chronology of Exposure. ACEs were recorded using the Maltreatment and Abuse Chronology of Exposure (MACE) scale (Teicher & Parigger, 2011). The MACE scale documents different types of abuse and neglect over the first 18 years of life. It consists of 75 yes–no questions that are divided into 10 subcategories. For analysis, two scores can be computed. The MACE summation score follows dimensional logic and is calculated by adding the summation scores for each of the 10 MACE subscales. For each subscale, items are standardized so that summation values range between 0 and 10. Therefore, the MACE summation score ranges between 0 and 100, indicating the overall severity of maltreatment experienced. In contrast, the MACE multi score follows categorical logic and indicates the presence of different types of ACEs. The MACE multi score is the number of MACE subscales for which a predetermined subscale threshold is
This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
182
SEILER, KOHLER, RUF-LEUSCHNER, AND LANDOLT
reached, so that values range between 0 and 10. The following subscales and specific cutoff values were used: (a) familial and nonfamilial sexual abuse (eight items; cutoff: 2); (b) parental verbal abuse (four items; cutoff: 3); (c) parental nonverbal emotional abuse (five items; cutoff: 3); (d) parental physical abuse (six items; cutoff: 4); (e) emotional neglect (10 items; cutoff: 5); (f) physical neglect (five items; cutoff: 3); (g) witnessing interparental physical abuse (four items; cutoff: 2); (h) witnessing physical abuse of a sibling (four items; cutoff: 2); (i) peer-initiated emotional abuse (five items; cutoff: 4); and (j) peer-initiated physical abuse (five items; cutoff: 2). At the time of our data collection, the child version of the instrument was not yet available. For this reason, we used the official English adult version. In doing so, the adult version was translated into Spanish by the authors of this study, back-translated from Spanish to English by a Chilean native speaker, and revised by a Chilean psychiatrist and psychologist to optimize understandability. By administering the MACE scale in face-to-face interviews, children could ask questions if they had difficulty understanding certain questions. Also, items could be explained to the child in an age-appropriate manner. In the current study, the translated MACE scale displayed acceptable-to-good internal consistency, with Cronbach’s alpha values ranging from .73 to .87 for the different subscales and a Cronbach’s alpha of .88 for the MACE summation score. Post-traumatic stress symptoms. The presence of posttraumatic stress symptoms was assessed by means of the UCLA Post-Traumatic Stress Disorder Index (UCLA PTSD Index) for DSM–IV, a widely used self-report tool to assess PTSD in children and adolescents (Steinberg, Brymer, Decker, & Pynoos, 2004). The instrument is organized into three sections in order to (a) screen for 13 traumatic experiences (TEs); (b) evaluate DSM–IV criteria for trauma exposure (Criteria A1 and A2); and (c) assess the frequency at which DSM–IV PTSD symptoms occur (Criteria B, C, and D), as well as a time Criterion E and the presence of functional impairment (Criterion F). If a child does not mention exposure to any traumatic event in the first section of the UCLA PTSD Index, further questions are skipped. For symptom severity, the frequency at which each symptom occurs is rated on a 5-point Likert scale that ranges from 0 (none of the time) to 4 (most of the time). A total PTSD severity index is computed, indicating symptom severity. Furthermore, the UCLA PTSD Index yields a dichotomous PTSD diagnosis in accordance with DSM–IV criteria. Psychometric properties of the UCLA PTSD Index have been shown to be good, with excellent internal consistency and good convergent validity. For the current study, a Spanish version for children (Rodriguez, Steinberg, & Pynoos, 1999) was used, which was further adapted to Chilean Spanish and checked for comprehensibility by a Chilean child psychiatrist. In the current study, the UCLA PTSD Severity Index demonstrated good internal consistency (Cronbach’s alpha ⫽ .84), similar to the figure reported in the original manual. Health-related quality of life. HRQoL was assessed by means of the Spanish version of the KIDSCREEN-27 (RavensSieberer, 2004). This is a well-validated, standardized measure of HRQoL in children and adolescents 8 to 18 years of age that has a version for both the child and the caregiver. In this study, we merely used the self-report version. Four foster girls who completed the questionnaire were younger than 8 years old. The KIDSCREEN-27 is composed of 27 items assessing the following
five HRQoL dimensions recalled over 1 week: (a) physical wellbeing, (b) psychological well-being, (c) parental relations and autonomy, (d) social support, and (e) school. As the foster home girls had limited contact with their parents, their assessment of “parent relationships and autonomy” referred to their relationship with the caregivers at the foster home. Either the frequency or intensity of each attitude is assessed on a 5-point scale for the last week. Higher scores indicate better HRQoL. The global KIDSCREEN-10 index that was developed from the KIDSCREEN-27 was computed as an indicator of general HRQoL (Ravens-Sieberer, 2004). In the current study, T-values were computed for the interpretation of data from both study groups and compared against international norm data. In our sample, internal consistencies of the KIDSCREEN-27 were acceptable to good for the KIDSCREEN-10 index (␣ ⫽ .80) and for the subscales for physical (␣ ⫽ .71) and psychological well-being (␣ ⫽ .82). However, internal consistencies for the remaining subscales were only moderate (␣ ⫽ .62– .67). Mental health. The Spanish version (Abad & Forns, 1991) of the Child Behavior Checklist (CBCL) was used for proxy reports by parents and caregivers on the girls’ behavioral problems over the preceding 6 months (Achenbach & Rescorla, 2000). The CBCL consists of 120 items, each scored on a 3-point Likert scale (0 ⫽ absent, 1 ⫽ occurs sometimes, 2 ⫽ occurs often). Computations were conducted on the basis of two broadband scales designated as internalizing and externalizing problem scales, with a total problems score derived from all questions. Higher scores indicate greater psychological maladjustment. In the present study, internal consistencies for the internalizing (␣ ⫽ .90), externalizing (␣ ⫽ .96), and total behavioral problems scales (␣ ⫽ .96) were excellent and similar to the original Spanish version (Abad & Forns, 1991). Analyses were conducted by means of raw scores because no Chilean norms were available for computing T-scores.
Data Analysis Data were analyzed using the statistical package SPSS version 21 for Macintosh (SPSS Inc., Chicago, IL). Nonparametric tests were applied due to ordinal data and violation of the normality assumption using the Mann–Whitney U test for continuous variables and Pearson’s chi-squared analysis for dichotomous variables. Spearman’s rank correlation coefficients were computed to determine the strength of relationships between the variables. Correlations between .40 and .69 were assumed to be moderate, and those between .70 and .89 were considered strong. Moreover, to generate some appreciation regarding the clinical significance of our results, effect sizes were computed according to Cohen (1988) and categorized as small (.2–.5), intermediate (.5–.8), or large (⬎.8).
Results Characteristics of the Sample Demographic characteristics for the two study groups are given in Table 1. The average age of both study groups was 10.2 years. Foster home girls came significantly more often from families with a household income below the Chilean minimum wage of 180,000 Chilean Pesos compared to controls. Moreover, foster home girls
ADVERSE CHILDHOOD EXPERIENCES AND MENTAL HEALTH IN FOSTER GIRLS
183
This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
Table 1 Demographic Characteristics of the Two Study Groups Foster home girls (n ⫽ 27)
Controls (n ⫽ 27)
10.06 2.98 100%
10.30 3.12 100%
Age Mean SD School attendance School grade ⱕFourth grade ⱖFifth grade Siblings Half brothers and sisters On psychotropic treatment Family income less than minimum wage Note.
59.2% 40.7% 55.6% 63.0% 26% 17 (62.9%)
37.1% 62.9% 85.2% 7.4% 0 0
Test statistics
p
U ⫽ 341.0
.68
U ⫽ 305.0
.30
2 ⫽ 5.68 2 ⫽ 18.27 2 ⫽ 54.00 2 ⫽ 4.00
© 2015 American Psychological Association 1942-9681/16/$12.00 http://dx.doi.org/10.1037/tra0000037
Adverse Childhood Experiences, Mental Health, and Quality of Life of Chilean Girls Placed in Foster Care: An Exploratory Study Annina Seiler
Stefanie Kohler and Martina Ruf-Leuschner
University Hospital Zurich, Zurich, Switzerland
University of Konstanz
Markus A. Landolt This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
University Children’s Hospital Zurich, Zurich, Switzerland In Latin America, little research has been conducted regarding exposure to adverse childhood experiences (ACEs), mental health, and health-related quality of life (HRQoL) among foster children. This study examined the association between ACEs and mental health, posttraumatic stress disorder (PTSD), and HRQoL in Chilean foster girls relative to age-matched Chilean family girls. Data were obtained from 27 Chilean foster girls and 27 Chilean girls ages 6 to 17 years living in family homes. Standardized self- and proxy-report measures were used. Foster girls reported more ACEs than controls in terms of familial and nonfamilial sexual abuse and both emotional and physical neglect. Girls living in foster care had a significantly higher rate of PTSD, displayed greater behavioral and emotional problems, and reported a lower HRQoL. Analysis confirmed the well-known cumulative risk hypothesis by demonstrating a significant positive association between the number of ACEs and PTSD symptom severity and a significant negative association with HRQoL. Chilean foster girls endured more ACEs that impair mental health and HRQoL than age-matched peers living with their families. These findings have implications for out-of-home care services in Latin America, highlighting the need to implement not only appropriate trauma-focused treatments but also appropriate prevention strategies. Keywords: adverse childhood experiences, foster children, mental health, posttraumatic stress disorder, health-related quality of life
al., 2009). Additional family-related risk factors, including domestic violence, family disruption, parental mental illness, poverty, and general life stress commonly co-occur with such familyrelated problems and often make out-of-home placement necessary for the child’s safety (Edwards, Holden, Felitti, & Anda, 2003). Today, associations between exposure to ACEs and physical and mental health problems in foster children and adolescents are well documented (Greeson et al., 2011; Oswald, Heil, & Goldbeck, 2010). In particular, research has consistently demonstrated that girls in foster care who have endured persistent maltreatment and neglect are especially vulnerable to developing behavioral problems and psychological disorders (McMillen et al., 2005). Symptoms of posttraumatic stress disorder (PTSD) and the occurrence of internalizing and externalizing behavioral problems have commonly been reported in maltreated children living in foster care in the United States and Europe (Greeson et al., 2011; McMillen et al., 2005; Oswald et al., 2010). Moreover, cumulative exposure to ACEs and traumatic events, also known as the building block effect, is assumed to be a robust predictor of adverse developmental outcomes in maltreated youths. Prior research reviews have consistently shown that the greater the number of ACEs, the greater the child’s risk of having behavioral problems and mental disorders (Appleyard, Egeland, van Dulmen, & Sroufe, 2005; Edwards et al., 2003; Raviv, Taussig, Culhane, & Garrido, 2010). Notably, children exposed to ACEs develop not only mental health problems but also an impaired health-related quality of life (HRQoL) (Jud, Landolt, Tatalias, Lach, & Lips, 2013; Villalonga-
In Chile in 2011, an estimated 10.4% of children and adolescents lived outside their family home, of whom more than half (54%) were girls (Servicio Nacional de Menores, Chile [SENAME], 2012). Before entering foster care, many of these children endured complex and multiple adverse childhood experiences (ACEs) such as emotional and physical neglect and/or physical and sexual abuse that occurred within their family or caregiving system (Hopenhayn et
This article was published Online First April 27, 2015. Annina Seiler, Department of Psychiatry and Psychotherapy, University Hospital Zurich, Zurich, Switzerland; Stefanie Kohler and Martina RufLeuschner, Department of Psychology, Clinical Psychology and Behavioral Neuroscience Unit, University of Konstanz; and Markus A. Landolt, Department of Psychosomatics and Psychiatry, University Children’s Hospital Zurich, Zurich, Switzerland, and Department of Child and Adolescent Health Psychology, Institute of Psychology, University of Zurich. We thank the girls who participated in this study and the directorate of the children’s home in Chile for allowing us to conduct the study. We would like to express our gratitude to Dr. Markus Zwicky and Gabriela Grimm, who initiated the study, established contact with the Chilean foster home, and generously contributed to the funding of this project. We also thank Dr. Maria Teresa Diez and Jeanine Schaelin for their help conducting the field study in Chile and Dorothea Isele for her methodological support. Correspondence concerning this article should be addressed to Markus A. Landolt, Department of Psychosomatics and Psychiatry, University Children’s Hospital Zurich, Steinwiesstrasse 75, CH-8032 Zurich, Switzerland. E-mail: [email protected] 180
This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
ADVERSE CHILDHOOD EXPERIENCES AND MENTAL HEALTH IN FOSTER GIRLS
Olives et al., 2010). HRQoL is a multidimensional construct composed of the individual’s perception of his or her functioning in a variety of aspects of life, including physical, psychological, and social domains (Mazur & Mierzejewska, 2003). Only two previous studies, one in Australia (Sawyer, Carbone, Searle, & Robinson, 2007) and one in Serbia (Damnjanovic, Lakic, Stevanovic, & Jovanovic, 2011), have examined the HRQoL of children and adolescents living in foster care. Both studies identified substantially poorer HRQoL in these youths compared to the general population. Moreover, symptoms of anxiety and depression accounted for significant variations in many domains of HRQoL in foster children. There is little research on the mental health of girls living in foster care in Latin America. However, cross-cultural research is important because the risk factors associated with child and adolescent mental health problems vary across different continents, especially between developed and developing countries (Fatori, Bordin, Curto, & de Paula, 2013). Evidence-based research on mental health and HRQoL in children living in foster care in Latin America is important to improve the quality of mental health services in foster care and to help professionals implement appropriate trauma-focused treatment in this region. To bridge these existing research gaps, the aims of the present exploratory study were threefold. First, we sought to examine whether Chilean girls living in foster care report more ACEs than girls living with their families. Second, we aimed to compare rates of PTSD and behavioral problems and HRQoL in these two groups. Third, we wanted to examine associations between ACEs and both behavioral problems and HRQoL among girls living in foster care. Based upon previous findings, we expected higher rates of ACEs in girls living in foster care versus family homes. Moreover, we assumed that foster home girls would have more mental health problems and lower HRQoL. We also hypothesized that the number of different types of ACEs would be positively associated with PTSD symptom severity and the number of behavioral problems and negatively associated with HRQoL.
Method Participants In February 2012, 27 girls living in a Chilean children’s home for sexually abused or emotionally and physically neglected girls were assessed regarding ACEs, mental health, and HRQoL. The children’s home was located in a rural region in the southern part of Chile and was under the direction of SENAME, a Chilean Youth Welfare Agency that aids child victims and children in conflict with the law (SENAME, 2012). The institution is largely funded by private donations, collections, fundraising events, and the agency’s own funds. Eight caregivers were responsible for taking care of a maximum of 30 girls between the ages of 3 and 18 years. Most of the girls had had to be separated from their biological parents for a variety of reasons that included violence, physical and sexual abuse, and emotional and/or physical neglect. The children’s home sought to reunify each girl with her family as soon as the family situation was considered stable again. Twenty-seven age-matched (⫾6 months) Chilean girls living in family homes served as a control group. They were recruited from a holiday swimming class in a small city close to the foster home.
181
This daily swimming class is held each year during the summer holidays. The course is aimed at children of low-income families who cannot afford to go on holidays or children whose parents have to work. To be eligible as controls, these girls had to live with their biological mother. For the sake of convenience, the girls living in foster homes shall be hereafter called “foster home girls”; those in the comparison group will be called “controls.”
Procedure Permission to conduct the study in the orphanage was granted by the directorate, who signed informed consents for all the girls in foster care. For girls living with their families, written informed consent was obtained from the mothers. Girls 14 years of age or older in both study groups signed the informed consent form independently. Furthermore, all girls who decided to participate in our study were informed verbally about the contents and goals of the research project and about their right to withdraw from the assessment at any time. All girls were informed that, according to Chilean laws, in case of reported ongoing maltreatment, the directorate would be notified in order to decide whether to inform local authorities. In the case of severe mental health problems, a referral to a local health professional was initiated upon consultation with the directorate of the orphanage. All girls gave their consent to participate in the study. All girls between the ages of 6 and 17 years were interviewed. In addition, proxy reports were obtained to assess emotional– behavioral difficulties, either from the girl’s mother (controls) or by any one of the orphanage caregivers who knew the girl well enough. Household income was assessed from the parents of control girls or was retrieved from the records for foster home girls. Records were also assessed for any psychotropic treatment. Standardized face-to-face interviews were conducted in a quiet, private room by two trained graduate students of psychology, one attending the University of Zurich, Switzerland, and the other attending the University of Konstanz, Germany. Both students were fluent in Spanish and required no interpreter during the interview. To ensure that all the girls could express their own views openly and confidentially, they were interviewed without their caregivers or parents present. As remuneration, each girl received a bracelet.
Measures Maltreatment and Abuse Chronology of Exposure. ACEs were recorded using the Maltreatment and Abuse Chronology of Exposure (MACE) scale (Teicher & Parigger, 2011). The MACE scale documents different types of abuse and neglect over the first 18 years of life. It consists of 75 yes–no questions that are divided into 10 subcategories. For analysis, two scores can be computed. The MACE summation score follows dimensional logic and is calculated by adding the summation scores for each of the 10 MACE subscales. For each subscale, items are standardized so that summation values range between 0 and 10. Therefore, the MACE summation score ranges between 0 and 100, indicating the overall severity of maltreatment experienced. In contrast, the MACE multi score follows categorical logic and indicates the presence of different types of ACEs. The MACE multi score is the number of MACE subscales for which a predetermined subscale threshold is
This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
182
SEILER, KOHLER, RUF-LEUSCHNER, AND LANDOLT
reached, so that values range between 0 and 10. The following subscales and specific cutoff values were used: (a) familial and nonfamilial sexual abuse (eight items; cutoff: 2); (b) parental verbal abuse (four items; cutoff: 3); (c) parental nonverbal emotional abuse (five items; cutoff: 3); (d) parental physical abuse (six items; cutoff: 4); (e) emotional neglect (10 items; cutoff: 5); (f) physical neglect (five items; cutoff: 3); (g) witnessing interparental physical abuse (four items; cutoff: 2); (h) witnessing physical abuse of a sibling (four items; cutoff: 2); (i) peer-initiated emotional abuse (five items; cutoff: 4); and (j) peer-initiated physical abuse (five items; cutoff: 2). At the time of our data collection, the child version of the instrument was not yet available. For this reason, we used the official English adult version. In doing so, the adult version was translated into Spanish by the authors of this study, back-translated from Spanish to English by a Chilean native speaker, and revised by a Chilean psychiatrist and psychologist to optimize understandability. By administering the MACE scale in face-to-face interviews, children could ask questions if they had difficulty understanding certain questions. Also, items could be explained to the child in an age-appropriate manner. In the current study, the translated MACE scale displayed acceptable-to-good internal consistency, with Cronbach’s alpha values ranging from .73 to .87 for the different subscales and a Cronbach’s alpha of .88 for the MACE summation score. Post-traumatic stress symptoms. The presence of posttraumatic stress symptoms was assessed by means of the UCLA Post-Traumatic Stress Disorder Index (UCLA PTSD Index) for DSM–IV, a widely used self-report tool to assess PTSD in children and adolescents (Steinberg, Brymer, Decker, & Pynoos, 2004). The instrument is organized into three sections in order to (a) screen for 13 traumatic experiences (TEs); (b) evaluate DSM–IV criteria for trauma exposure (Criteria A1 and A2); and (c) assess the frequency at which DSM–IV PTSD symptoms occur (Criteria B, C, and D), as well as a time Criterion E and the presence of functional impairment (Criterion F). If a child does not mention exposure to any traumatic event in the first section of the UCLA PTSD Index, further questions are skipped. For symptom severity, the frequency at which each symptom occurs is rated on a 5-point Likert scale that ranges from 0 (none of the time) to 4 (most of the time). A total PTSD severity index is computed, indicating symptom severity. Furthermore, the UCLA PTSD Index yields a dichotomous PTSD diagnosis in accordance with DSM–IV criteria. Psychometric properties of the UCLA PTSD Index have been shown to be good, with excellent internal consistency and good convergent validity. For the current study, a Spanish version for children (Rodriguez, Steinberg, & Pynoos, 1999) was used, which was further adapted to Chilean Spanish and checked for comprehensibility by a Chilean child psychiatrist. In the current study, the UCLA PTSD Severity Index demonstrated good internal consistency (Cronbach’s alpha ⫽ .84), similar to the figure reported in the original manual. Health-related quality of life. HRQoL was assessed by means of the Spanish version of the KIDSCREEN-27 (RavensSieberer, 2004). This is a well-validated, standardized measure of HRQoL in children and adolescents 8 to 18 years of age that has a version for both the child and the caregiver. In this study, we merely used the self-report version. Four foster girls who completed the questionnaire were younger than 8 years old. The KIDSCREEN-27 is composed of 27 items assessing the following
five HRQoL dimensions recalled over 1 week: (a) physical wellbeing, (b) psychological well-being, (c) parental relations and autonomy, (d) social support, and (e) school. As the foster home girls had limited contact with their parents, their assessment of “parent relationships and autonomy” referred to their relationship with the caregivers at the foster home. Either the frequency or intensity of each attitude is assessed on a 5-point scale for the last week. Higher scores indicate better HRQoL. The global KIDSCREEN-10 index that was developed from the KIDSCREEN-27 was computed as an indicator of general HRQoL (Ravens-Sieberer, 2004). In the current study, T-values were computed for the interpretation of data from both study groups and compared against international norm data. In our sample, internal consistencies of the KIDSCREEN-27 were acceptable to good for the KIDSCREEN-10 index (␣ ⫽ .80) and for the subscales for physical (␣ ⫽ .71) and psychological well-being (␣ ⫽ .82). However, internal consistencies for the remaining subscales were only moderate (␣ ⫽ .62– .67). Mental health. The Spanish version (Abad & Forns, 1991) of the Child Behavior Checklist (CBCL) was used for proxy reports by parents and caregivers on the girls’ behavioral problems over the preceding 6 months (Achenbach & Rescorla, 2000). The CBCL consists of 120 items, each scored on a 3-point Likert scale (0 ⫽ absent, 1 ⫽ occurs sometimes, 2 ⫽ occurs often). Computations were conducted on the basis of two broadband scales designated as internalizing and externalizing problem scales, with a total problems score derived from all questions. Higher scores indicate greater psychological maladjustment. In the present study, internal consistencies for the internalizing (␣ ⫽ .90), externalizing (␣ ⫽ .96), and total behavioral problems scales (␣ ⫽ .96) were excellent and similar to the original Spanish version (Abad & Forns, 1991). Analyses were conducted by means of raw scores because no Chilean norms were available for computing T-scores.
Data Analysis Data were analyzed using the statistical package SPSS version 21 for Macintosh (SPSS Inc., Chicago, IL). Nonparametric tests were applied due to ordinal data and violation of the normality assumption using the Mann–Whitney U test for continuous variables and Pearson’s chi-squared analysis for dichotomous variables. Spearman’s rank correlation coefficients were computed to determine the strength of relationships between the variables. Correlations between .40 and .69 were assumed to be moderate, and those between .70 and .89 were considered strong. Moreover, to generate some appreciation regarding the clinical significance of our results, effect sizes were computed according to Cohen (1988) and categorized as small (.2–.5), intermediate (.5–.8), or large (⬎.8).
Results Characteristics of the Sample Demographic characteristics for the two study groups are given in Table 1. The average age of both study groups was 10.2 years. Foster home girls came significantly more often from families with a household income below the Chilean minimum wage of 180,000 Chilean Pesos compared to controls. Moreover, foster home girls
ADVERSE CHILDHOOD EXPERIENCES AND MENTAL HEALTH IN FOSTER GIRLS
183
This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
Table 1 Demographic Characteristics of the Two Study Groups Foster home girls (n ⫽ 27)
Controls (n ⫽ 27)
10.06 2.98 100%
10.30 3.12 100%
Age Mean SD School attendance School grade ⱕFourth grade ⱖFifth grade Siblings Half brothers and sisters On psychotropic treatment Family income less than minimum wage Note.
59.2% 40.7% 55.6% 63.0% 26% 17 (62.9%)
37.1% 62.9% 85.2% 7.4% 0 0
Test statistics
p
U ⫽ 341.0
.68
U ⫽ 305.0
.30
2 ⫽ 5.68 2 ⫽ 18.27 2 ⫽ 54.00 2 ⫽ 4.00
